Going to an OpenSource ECM World....

Maven + Alfresco : Delivery and Assembly

2010-04-28T02:49:00.000-07:00

Hello everyone!

It is time to talk about the delivery format of your project. You tell me, he use for many post AMP format and now he wants to give us something else? Yeap it's true...!

Depending on your projects, your distribution of roles, your teams and your business practices, the delivery format can be simple or very very complex!

Take the classic case, on the one side we have a development team and in the other side we have an operation (production) team. The first one is responsible for creating the product, the other is in charge of installing the product and verify that it works throughout its lifetime.

The question: What is the exchange format of product between these two teams?
Ps: we will not consider here all the documentation from the production team ...

Let's make a list of possible cases:

AMP(s): For the development team is the simple format. Indeed it is the output format of our project! So you have just to get AMP file and send it! Of course if our project consists of several AMPs, we will send all AMPs. In this case, the production team is responsible for generating the WAR application using the script provided by Alfresco.
A WAR: Here the development team is responsible for creating the war. They have to integrate different modules in the default Alfresco war. The production team have to simply install the application.
AMP + WAR: In this scenario, the production team want to have the application and AMPs.
Other: Here you can invent anything that goes through your head. If you work in big projects, or in a large size companies, I think you have very beautiful examples!

As you can see, examples are not lacking! It is rather the opposite. The question :

How can we create these new delivery formats in our AMP project ?

The idea as usual is to run a Maven command and wait for the generated file(s). Never forget that the developer is supposed to be lazy ^ ^.

In lines below, I will present two approaches to generate anything other than your AMP. These approaches are based on a new project called war-alf-osecm.

Note: This approach is not very consistent with the philosophy but Maven does the work desired.

Creating a new project AMP

Defining Structure
Create a directory with the name: alf-war-osecm and add a pom.xml with value below


 xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
 4.0.0
 fr.opensourceecm
 alf-war-osecm
 pom
 0.1
 Alfresco AMP Open Source ECM
 Open Source ECM - Extension

Integrating the tool into a repository
In this project, we will use AlfrescoModule Management Tool . This module is available in the bin directory of Alfresco standard installation . It is also available here: http://wiki.alfresco.com/wiki/Community_Edition_file_list_3.3.
So initially, we will install this file in our maven repository with the command:


call mvn install:install-file  -Dfile=alfresco-mmt.jar -DgroupId=org.alfresco.sdk -DartifactId=mmt  -Dversion=3.3 -Dpackaging=jar -DgeneratePom=true -DcreateChecksum=true

Note : We perform this command in the directory where Alfresco module management tool file is.
Due to this integreation, It will be possible to find it through the classic mechanism of dependency .

Create a new Alfresco WAR
Now, as the title suggests, I want to generate a WAR with all AMPs. It will integrate all AMPs in a WAR Alfresco. This process will be divided into 4 parts.

Setting Properties: In this part, we will initialize the properties that we used during other parts.
Import pre-requisite: We will install Alfresco war and tool into our working directory.
Import of AMP: In our working directory, we will create a subdirectory that will serve to contain all AMPs to integrate.
War Generation : Here, we will launch a command line to integrate AMPs
(Optional) Install war in maven repository: After generation, you can optionally add the generated war in our maven repository.

Setting Properties
We will define four properties

Work.dir: Working directory
Amps.dir: AMP directory
Mmt.jar: Module Management Tool filename
Alfresco.war: Alfresco war filename



    livraison
    amps
    mmt.jar
    alfresco.war

Import pre-requisites
Via the plugin maven-dependency-plugin we will use the goal copy during the prepare-package phase. To simplify, we want to copy a set of dependencies (alfresco.war and tool) in our working directory.



    org.apache.maven.plugins
    maven-dependency-plugin
    
        
            war-copy-requisites
            prepare-package
            
                copy
            
            
                
                    
                        org.alfresco.sdk
                        alfresco-community-war
                        3.3
                        war
                        ${project.build.directory}/${work.dir}
                        ${alfresco.war}
                    
                    
                        org.alfresco.sdk
                        mmt
                        3.3
                        jar
                        ${project.build.directory}/${work.dir}
                        ${mmt.jar}

Import AMP
In this stage, we will add a dependencies block in our pom.xml. In our case, dependencies contains 2 dependencies :

fr.opensourceecm:alf-osecm-am:amp:0.0.5
org.alfresco.sample:SDK-CustomWizard:amp:1.0

Our pom.xml looks like



    
        fr.opensourceecm
        alf-amp-osecm
        0.0.5
        amp
    
    
        org.alfresco.sample
        SDK-CustomWizard
        1.0
        amp

Then, we continue using maven-dependency-plugin. We will realize the goal copy-dependencies in the phase of prepare-package. More simply, we want to copy all MPAs (defined in length) in a subdirectory of work amps.dir ().



    org.apache.maven.plugins
    maven-dependency-plugin
    
        
            war-copy-amp
            prepare-package
            
                copy-dependencies
            
            
                amp
                ${project.build.directory}/${work.dir}/${amps.dir}
                false
                true
                true

War Generation
Via the plugin exec-maven-plugin, we will realize the goal exec during the package phase. In other words, we want to achieve the same behavior as apply-amps.bat file (or .sh) but via Maven.



    org.codehaus.mojo
    exec-maven-plugin
    1.1
    
        
            war-create
            package
            
                exec
            
        
    
    
        java
        
            -jar
            ${project.build.directory}/${work.dir}/${mmt.jar}
            install
            ${project.build.directory}/${work.dir}/${amps.dir}
            ${project.build.directory}/${work.dir}/${alfresco.war}
            -directory

Now it is possible to run the command


mvn clean package

You can now find in your work.dir a new alfresco.war containing all AMP dependencies defined in our pom.xml.

(Optional) Install the war locally
If you want to install this war in your local repository (and thus be able to create dependencies for testing or integration purposes), here is the configuration to add to your build phase:



    org.apache.maven.plugins
    maven-install-plugin
    2.3
    
        
            war-install
            install
            
                install-file
            
            
                ${project.build.directory}/${work.dir}/${alfresco.war}
                ${project.groupId}
                ${project.artifactId}
                ${project.version}
                war
                test
                true
                true

With the command


mvn clean install

you can install your new war. Of course, you can change various settings to suit your needs ...

Note: You can do the same deploy phase if you change some parameters... ^ ^

Create a custom delivery format
Now we'll look at creating a custom delivery format. In my case, I wish

Create a zip file containing
A root folder : groupId.artifactId
A sub-folder amps containing my amps
A sub-folder application containing alfresco.war (with the amp already include)

To do the job, I'll use maven-assembly-plugin. This plugin is used to create custom distribution formats. Configuration of this plugin is done in 2 steps:

Create a description file
Adding configuration in pom.xml

Create a description file
This description file defines the expected result ie what are the directories to create, file formats, rights, encoding ... We can actually define a lot of parameters!

In our case, we will create osecm-assembly.xml in alf-war-osecm\src\assembly directory. This file contains the following configuration:


  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
  xsi:schemaLocation="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0 http://maven.apache.org/xsd/assembly-1.1.0.xsd">
  production
  
    zip
  
  ${project.groupId}.${project.artifactId}.${project.version}
  
   
    target/${work.dir}
    unix
    /application
    
     ${alfresco.war}
    
   
  
  
    
      /${amps.dir}
      
        org.alfresco.sample:SDK-CustomWizard
        fr.opensourceecm:alf-amp-osecm

As you can see, it takes the configuration expected in my requirements.

Adding to the pom.xml
Now we must declare in the pom.xml this new file. So we add maven-assembly-plugin with the configuration below



    maven-assembly-plugin
    
        
            src/assembly/osecm-assembly.xml

Start assembly
Finally, I can run the command :


mvn clean assembly:assembly

This command assemble my custom distribution format. In the target directory, you can find your custom file called : alf-war-osecm-0.1-production.zip ready to be sent!

Note: This plugin may be greedy in memory. To resolve this problem increase values in MAVEN_OPTS environment variable.

Maven + Alfresco : Multi-AMP relationship

2010-04-28T01:02:00.000-07:00

Hello World!

This time I'll try to explain multi-AMP relationships and how they can be managed with Maven and Alfresco...

Introduction

The problem is as follows:

In a multi-AMP project, how we organize those relationships in our Maven project(s) ?

Suppose we want to create a project with 3 AMPS: A, B, C

Above all, what are the relationships between these three AMPS? Are they independent, interdependent ? What are relations induced?

Then, what is the delivery format of our application? Is it a WAR, a single AMP, a set of AMP, a particular format of delivery...

According to the answers of each question, we have to create different policy! and of course, we have to create different configuration of Maven projects!

In general, I identify several use cases.

Simple relationships: In other words it is the same dependency that we've seen since we began our projects. In our case, modules A, B and C are independent, but our production server needs these three AMP to match our user's needs. For example, we want to integrate Alfresco RM module, SharePoint module and our own module including our personalization.
Dependencies (weak and strong): In this case, one module needs another module to work. A simple example is the association that may exist between a model of data and actions associated with this data model. If the model doesn't exist and if is not installed then B can't work. If B is changed so it changes the behavior of A.

How to define (and test!) these relationships in our project?

Simple relationships

First, we will use our historical alf-amp-osecm module.

Currently, our module has only dependencies to libraries already present in the Alfresco war. There are no problem because when we start our jetty server with Alfresco war.

But now I want to test my module with the presence of another modules.

Today for example, I'll take the module SDK-TaggingSample (previously created - cf. Post here). The idea is to

add SDK-TaggingSample module into Alfresco.war,
package my alf-amp-osecm module
run jetty to test the intregation and interaction between these modules

To add SDK-TaggingSample module, we will modify the pom.xml of alf-amp-osecm and the declaration below between dependencies tags in the webapp Profile.



   org.alfresco.sample
   SDK-TaggingSample
   amp
   1.0

At this stage, we might say, it's good ! We can launch our application with the maven command:


mvn clean integration-test -P webapp,clean-jetty -DskipTests=true

and we will have an Alfresco started with our two modules. Well in fact no... That would be too easy!

The AMP format is a custom format which is not well recognized by Maven and related plugins. Maven recognized this format thanks to maven-amp-plugin. But here we add an AMP dependency! The question is : Does Maven know how to integrate this AMP into exploded war necessary to jetty server ? The answer is obviously NO!

The solution lies in the declaration of a new plugin: maven-dependency-plugin. This plugin as its name suggests can manipulate dependencies and artifacts and therefore AMP! We will configure this plugin to use our dependency and install AMP into our exploded war.

How ? We have to add the configuration below. It has to be incorporated in webapp profile (in my case, it is now in the project alf-sdk. re-factoring required! Nevertheless the declaration of dependency is in my pom.xml)



   org.apache.maven.plugins
   maven-dependency-plugin
   
       
           unpack-amps
           process-resources
           
               unpack-dependencies
           
           
               amp
               ${build.directory}/${webapp.name}
               META*
           
       
   
   
       
           org.alfresco.maven.plugin
           maven-amp-plugin
           3.0.2

Note that this plugin will take all the dependencies of AMP type and will execute the goal: unpack-dependencies in the directory $ (build.directory) / $ (webapp.name). In other words, it will install AMP as would (almost) Alfresco Module Management Tool.

Next, we must add a testRessource block in webapp Profile .



   true
   .
   
       module.properties
   
   alfresco/module/${project.groupId}.${project.artifactId}

This block is necessary when we want to integrate several AMP in a war (which is our case here).

Now we can run the command:


mvn clean integration-test -P webapp,clean-jetty -DskipTests=true

and check (after a few minutes) our working AMPS!

Technology is beautiful sometimes ;o)

Dependencies

Now we will focus on dependencies. In this case, I don't want my Alfresco app starts if my module doesn't have all necessary dependencies. If this case happens, we would have an unstable application that could damage data! (I obviously don't wish it...).

How to define these strong dependencies between my modules?

Due to significant changes in my code, I want to create a strong link between my alf-amp-osecm and SDK-TaggingSample. To do this, there is a mechanism defined in the AMP and available via module.properties (located at the root of our project).
At the end of this file, simply add the line below to create a dependency:


module.depends.org.alfresco.sample.SDK-TaggingSample=0.1-*

Note the simple formalism:

Module.depends: to define a dependency module
groupId.artifactId or module name : to identify the module
XX - YY: To determine version range of the module to be taken into account (between XX and YY). In our exemple, our module depends on SDK-TaggingSample from version 0.1. (Note the possibility to use wildcard character that means any)

For more information: wiki.alfresco.com

You can now start the application via this Maven command:


mvn clean integration-test -P webapp -DskipTests=true

As you can see, nothing change! It's good but what we would like here is to verify that our mechanism works well. For this you have two choices.

Change the version number in the file module.properties. For example: module.depends.org.alfresco.sample.SDK-TaggingSample = 99.99-*
Remove SDK-TaggingSample dependency in dependencies block in our pom.xml.

In both cases, restart the application and you will see in the console a nice stacktrace! It prohibits that you start the application if the dependency is not respected!

That's all for today!

Mavenize well!

PS :
alf-sdk sources are available here:
http://www.box.net/shared/exremqsr0m

alf-amp-osecm sources are available here:
http://www.box.net/shared/dzige0nchr

Maven + Alfresco : Alfresco SDK Migration

2010-04-27T04:35:00.000-07:00

Welcome Again (maybe...)!

To continue my series about Maven and Alfresco practice, I propose today a straightforward exercise. Your mission, if you accept it, is to change Alfresco SDK samples to our custom AMP format !

What for ? Clearly... It's only for practice. With this exercice, you can see the benefits and limitations of this approach. This approach allow you to understand how maven build, test and integrate Alfresco AMP.

For information about SDK, you can find this url very useful : http://wiki.alfresco.com/wiki/Alfresco_SDK_3.3

Currently I have successfully migrated following samples:

SDK Custom Dialog
JSP Custom SDK
Custom Log SDK
Custom SDK Wizard
Tagging SDK Sample

Each sample inherit the project alf-sdk. That's why you can execute commands below to:

Run unit tests: mvn clean test
Install the project in the local repository: mvn install
(Reset)Initiate the database /: mvn clean -P clean-jetty
Restart the application: mvn integration-test -P webapp

If you followed my previous posts, you should be able to do the same thing. Otherwise please see SOURCES of each project that I made.

See you soon for new adventures !

Maven + Alfresco : Jetty, Boostrap and Profil

2010-04-25T13:25:00.001-07:00

Hello everyone!

This time I'll show you how you can bootstrap (initialize) data during the installation of our AMP and how we can test it easily.

1. Bootstrap
First of all, there are several ways to create a bootstrap

by XML files,
programmatically (via a java class).

The advantage of the first approach is its simplicity! Indeed, a bootstrap XML is nothing more than the equivalent of the import / export available in Alfresco Explorer. You know the green arrows on a folder in Details screen ... Anyway... Nodes are exported as XML and then (re) integrate into a (new) Alfresco instance. A typical example is the transition between a test server to a production server. On the DEV server, the development team has created a specific project tree, which must be installed on the PROD server automatically . This must be done during the first installation of our AMP on the PROD server.

The limitation of this approach is that only object creation is possible. You can not delete or change nodes...

The advantage of the second approach is its richness! Indeed, we have access to all Alfresco services plus (possibly) services of our project. We can do whatever you want .. You can reorganize a tree, perform a mass change of nodes or transform some contents ...

The disadvantage : you need to know programming :o/ and know how Alfresco service works.

In following I am going to explore the first approach. I may focus on the second approach in a future post ..

1.1 Boostrap via XML
So first, I'll explain the approach by an XML file.
In our example of bootstrapping, I want to create four spaces: a root space : BASE DOCUMENTAIRE (don't forget I'm a little french) + 3 sub-spaces PROJET 1 PROJET 2 PROJET 3. These spaces have been previously created in my Alfresco TEST instance. Now I want to boostrap my project with this data. In this manner, I can share data with any instance of Alfresco.

1.2 Export
First I export spaces in my Alfresco TEST (via the export action), then I download locally the ACP file generated in order to extract the XML file. My XML file is called spaces.xml (orginal isn't it ...).

For more information on the IMPORT / EXPORT: http://wiki.alfresco.com/wiki/Export_and_Import

1.3 Integration in the project
I will add this file to the new directory /alf-amp-osecm/src/main/config/bootstrap. Currently, this file is orphaned and will not be taken into account. We must declare it in Alfresco (and in Spring of course...).
For this, we will create a new file called bootstrap-context.xml in the directory /alf-amp-osecm/src/main/config/context














 
  
   /${spaces.company_home.childname}
   alfresco/module/fr.opensourceecm.alf-amp-osecm/bootstrap/spaces.xml

To summarize how to understand this file?
Literally, that's what he means:
We create a Spring bean identified by osecm.bootstrapSpaces . It inherits module.baseComponent. This bean is used to import spaces. It applies to the module fr.opensourceecm.alf-amp-osecm (our AMP), since version 0.0.4. Treatment of bean (hence the import) is required only once (so only during the first start). It takes as a parameter "bootstrapViews". This will allow us to create our tree structure as a subspace of {spaces.company_home.childname}

Simple isn't it ? :o)

1.4 Creating the project
Now we have our bootstrap. We just have to test it ! We can make a package of our AMP via the maven command


mvn clean install

In target folder, you have alf-amp-osecm-0.0.4.amp. It's possible to integrate it into your alfresco.war and test it.

Copy/paste the file alf-amp-osecm-0.0.4.amp to the amps directory in your Alfresco installation. Then run the installation script : apply_amps.bat in your root Alfresco installation folder. Follow instructions, wait few seconds and after that you have a new alfresco.war in tomcat/webapps folder.

Now you can run Alfresco and verify the integration of your custom module.

For more information : http://wiki.alfresco.com/wiki/AMP_Files

2. Starting and Testing Alfresco via Maven

If you've followed all of preceding posts on integrating Alfresco + Maven, you realized that you spend a lot of time to incorporate an AMP in Alfresco... Copy / paste is well but is a bit repetitive... The idea to improve the process would execute a maven command to start Alfresco !

How?
Well through the use of new plugins: maven-jetty and war!

Jetty is a lightweight HTTP container (just the same style as Tomcat ...) which will enable us to launch Alfresco. Maven-war allow us to create the war of our application.

Simply, we want to create an exploded alfresco.war in a subdirectory of target folder (with maven-war plugin) and launch it as a web application (with the Jetty server).

Here we go!

2.1 Profile
In fact the majority of the configuration work is done at our project's pom.xml. We will create what is called a PROFILE. A profile is a set of maven action made optional or not by the user during the execution of a command. It can be seen as a conditional statement controlled by the user. In this case, we try to start an Alfresco to verify that bootstrap has worked. We also try to verify if our AMP works.

However, before setting up our profile, we need to create some properties files.

2.2 Jetty-env.xml
Above all, we create jetty-env.xml in /alf-amp-osecm/src/test/properties/local. This file configure connection to the database via JNDI.





jdbc/dataSource

 
  jdbc:mysql://localhost/${alfresco.db.name}
  ${alfresco.db.username}
  ${alfresco.db.password}

2.3 Application.properties
if you say new environment inevitably you say new repository configuration. We will create a new file for this called application.properties and we will add it to the directory: /alf-amp-osecm/src/test/properties/local
This file contains a set of key / value which can be used by the pom.xml and by dev.properties file.


webapp.name=alfresco-jetty
alfresco.data.jetty=alf_data_jetty
alfresco.data.location=./alf_data_jetty
alfresco.db.driver=org.gjt.mm.mysql.Driver
alfresco.hibernate.dialect=org.hibernate.dialect.MySQLInnoDBDialect
alfresco.db.url=localhost
alfresco.db.name=alfjetty
alfresco.db.username=alfresco33
alfresco.db.password=alfresco
webapp.log.level=error
webapp.log.dir=
module.log.level=debug

2.4 Dev.properties
Here we will create a file called dev.properties and we'll add it to the directory: /alf-amp-osecm/src/test/properties/local. This file create the link between variables defined in application.properties and dev-context.xml. When creating our war, this file will overwrite our dev.properties previously defined.


dir.root=${alfresco.data.location}
index.recovery.mode=AUTO
integrity.failOnError=true
db.test.name=${alfresco.db.name}
db.username=${alfresco.db.username}
db.password=${alfresco.db.password}
db.host=${alfresco.db.host}
db.port=${alfresco.db.port}
db.driver=${alfresco.db.driver}
db.url=jdbc:mysql://${db.test.host}:${db.test.port}/${db.test.name}
hibernate.dialect=${alfresco.hibernate.dialect}

2.5 properties
Now we can attack the configuration of our pom.xml. First we will declare two property : env and webapp.name. Those properties are defined after project declaration block and before the dependencies block .


 
local
alfresco-jetty

2.6 Create Profile
As the title indicates, we will now create a block profile called webapp. Its main default goal is jetty: run-exploded (ie launch exploded war as a webapp application into jetty). This block is to add at the end of our pom.xml but before the project closing tag.




 webapp
 
  
   src/test/properties/${env}/application.properties
  
  jetty:run-exploded
  
   
                 ... Following points...
    
  
             
                 
                     true
                     src/test/resources
                     
                         alfresco/desktop/*.*
                     
                 
                 
                     true
                     src/test/properties/${env}
                     alfresco/extension
                     
                         dev.properties
                     
                 
             
         
         
             
                 org.alfresco.sdk
                 alfresco-community-war
                 3.3
                 war

Note several points:
First we added FILTER via the file application.properties. It will allow us to use all the variables defined in this file in the rest of the pom.xml and more specifically in our profile.

Then we addedalfresco-community-war (ie the application Alfresco Explorer) as a dependency. Of course, we have previously added to our Maven repository local via a maven command :


call mvn install:install-file -Dfile=alfresco.war -DgroupId=org.alfresco.sdk -DartifactId=alfresco-community-war -Dversion=3.3 -Dpackaging=war

After that we added a testResource. It allow us

to ignore desktops actions into our AMP (As reminder desktop actions are simply necessary for unit testing in Eclipse)
to declare repository configuration defined by dev.properties.

2.6 Configure Maven-War plugin
This plugin will allow us to deploy the alfresco war and to add our configuration files.


     
         org.apache.maven.plugins
         maven-war-plugin
         2.0.2
         
             
                 it
                 package
                 
                     exploded
                 
                 
                     true
                 
             
         
         
             false
             ${project.build.directory}/${webapp.name}
             
                 false
             
             licenses/**
             
                 
                     ${project.build.testOutputDirectory}
                     WEB-INF/classes
                     true
                     
                         **
                     
                 
                 
                     src/test/properties/${env}
                     WEB-INF
                     true
                     
                         jetty*

We note in resources block that we

Add all configuration files in src/test/resource/alfresco/* (equivalent to ${project.build.testOutputDirectory}

Add jetty-env.xml for JNDI database setting

2.8 Configure Jetty plugin
This plugin will enable us to launch the alfresco application and to declare sources of our project.


     
         org.mortbay.jetty
         maven-jetty-plugin
         6.1.22
         
             
                 it
                 integration-test
                 
                     run-exploded
                 
                 
                     /${webapp.name}
                     ${project.build.directory}/${webapp.name}
                     10
                     
                         
                             8080
                             60000
                         
                     
                 
             
         
         
             
                 mysql
                 mysql-connector-java
                 5.0.3
                 provided

Don't forget the dependence with mysql!

2.9 Launch application
Now you are ready to launch your application via a simple command


 mvn clean integration-test -P webapp

For information, we use a profile by adding the parameter -P followed by the profile name. If you do not want to play the tests, add the parameter -DskipTests = true at the end of the command.

A few minutes later if all goes well, you can open a browser and type the address http://localhost:8080/alfresco-jetty/

You realize then the presence of a space called BASE DOCUMENTAIRE!

Well Done!

3. Extras

Generally, a bootstrap test does not work the first time! If you want to retry again, it would be useful to reset our application (and your repository).

Now you are familiar with profiles and configuration plugin, so I want to share this extra profile : clean-jetty. Its main goal is to erase your database, stores and lucenes indexes.
It's your turn to place it judiciously in the pom.xml and eventually to improve it!



 clean-jetty
 
     
         
             maven-clean-plugin
             2.4
             
                 true
                 
                     
                         target
                         
                             war/**
                         
                         
                             **
                         
                     
                     
                         alf_data_jetty
                         
                             oouser/**
                         
                     
                 
             
         
         
             org.codehaus.mojo
             sql-maven-plugin
             1.4

             
                 
                     mysql
                     mysql-connector-java
                     5.0.3
                 
             
             
                 org.gjt.mm.mysql.Driver
                 jdbc:mysql://localhost/
                 root
                 admin
                 true
             
             
                 
                     drop-db-jetty
                     pre-clean
                     
                         execute
                     
                     
                         true
                         drop database IF EXISTS alfjetty 
                     
                 
                 
                     create-db-jetty
                     clean
                     
                         execute
                     
                     
                         true
                         create database alfjetty default character set utf8 collate utf8_bin

Hope you enjoy this post. Rendez vous to the next chapter!

A ++

All sources of alf-amp-osecm can be downloaded at:
http://www.box.net/shared/bpcx8kl7dn

Maven + Alfresco : Unit Test

2010-04-21T06:13:00.000-07:00

Hello world!

In this post, I'll try to share my approach to unit tests with Maven during AMP creation.
Of course if I say TEST, I say CODE! So prepare your eclipse !

Objectives :
As a developer and integrator my goal is multiple :

First of all I want to run unit tests in Eclipse. I want to keep the debugger. It helps me a lot during coding!
Then I want to run all my unit tests via Maven. So when I want to create a version of my AMP, it has to pass all tests! I think it's a good practice (^^)

So how?
If you followed my 2 previous posts here and here, we have in our possession two AMP (well. .. 2 projects MAVEN to be exact ...).

First, alf-sdk. It is a kind of PARENT-AMP which contains all dependencies of Alfresco SDK and all plugins to create an AMP.
Second, alf-amp-osecm. This is our AMP. It inherit alf-sdk.

We will modify these 2 projects to implement unit tests.

1. Updating ALF-SDK

First we will put some news in our parent pom. The main changes occur at build tag.

Maven-compiler-plugin and maven-resources-plugin. They received parameters for encoding.
Maven-eclipse-plugin has been installed. We will see in the next paragraph why ...




4.0.0
fr.opensourceecm
alf-sdk
pom
3.3C
Maven - Alfresco SDK 3.3C
Super POM Maven Alfresco SDK 3.3C 



....





true
src/main/resources

**README-*



true
src/main/config
alfresco/module/${groupId}.${artifactId}

**README-*







org.apache.maven.plugins
maven-eclipse-plugin
2.8


org.springframework.ide.eclipse.core.springnature


org.springframework.ide.eclipse.core.springbuilder





org.apache.maven.plugins
maven-resources-plugin
2.2

UTF-8




org.apache.maven.plugins
maven-compiler-plugin

1.6
1.6
UTF-8
true




org.alfresco.maven.plugin
maven-amp-plugin
3.0.2
true


false

alfresco/extension/**,alfresco/module/**

Once modified, you can install and deploy the artifact locally with a simple command maven :


mvn clean install

In our repository, we now have access to the module fr.opensourceecm:alf-sdk:pom:3.3C
Projects inheriting from this artifact will automatically receive the updates.

2. Updating our AMP

Now we'll create an Alfresco action and a test. As reminder, an action is a unit of work on Alfresco Repository. It can be run via a rule or directly by the user via Alfresco Explorer.

In our case, the action simply performs the creation of a sub-space.
The test class will start a background Alfresco (creation of a store), create a space, launch the action on this space and check the creation of one new sub-space.

2.1 Adding an action Alfresco
In the directory /alf-amp-osecm/src/main/java/fr/opensourceecm/alf/amp/osecm/actions / I add the action: sampleActionExecuter.java


package fr.opensourceecm.alf.amp.osecm.actions;

import java.util.List;

import org.alfresco.model.ContentModel;
import org.alfresco.repo.action.executer.ActionExecuterAbstractBase;
import org.alfresco.service.cmr.action.Action;
import org.alfresco.service.cmr.action.ParameterDefinition;
import org.alfresco.service.cmr.model.FileFolderService;
import org.alfresco.service.cmr.model.FileInfo;
import org.alfresco.service.cmr.repository.NodeRef;

public class sampleActionExecuter extends ActionExecuterAbstractBase{

private FileFolderService fileFolderService;

private final String FOLDER_TEST_NAME = "FolderTest";

@Override
protected void executeImpl(Action arg0, NodeRef actionUponRef) {
List listNodeRef = fileFolderService.listFolders(actionUponRef);
fileFolderService.create(actionUponRef, FOLDER_TEST_NAME + "_" +  listNodeRef.size(), ContentModel.TYPE_FOLDER);
}

@Override
protected void addParameterDefinitions(List arg0) {
// TODO Auto-generated method stub

}

public void setFileFolderService(FileFolderService fileFolderService) {
this.fileFolderService = fileFolderService;
}
}

Note : To create our subspace, we use the FileFolderService and the name of our space will be the concatenation between FolderTest and number of folder present.

2.2 Adding a unit test
In the directory /alf-amp-osecm/src/test/java/fr/opensourceecm/alf/amp/osecm/actions, I add the test: sampleActionExecuterTest.java


package fr.opensourceecm.alf.amp.osecm.actions;

import junit.framework.Assert;

import org.alfresco.model.ContentModel;
import org.alfresco.service.cmr.action.Action;
import org.alfresco.service.cmr.repository.NodeRef;
import org.alfresco.util.BaseAlfrescoTestCase;
import org.junit.Test;

public class sampleActionExecuterTest extends BaseAlfrescoTestCase {

protected NodeRef companyHomeRef;
protected NodeRef rootFolderTestRef;
protected final String rootFolderTestName = "ROOTFOLDERTEST";

@Test
public void testIntegrerFluxCreationParAction() throws Throwable {

// Création d'un répertoire de test dans le store de test.
rootFolderTestRef = serviceRegistry.getNodeService().createNode(
rootNodeRef, ContentModel.ASSOC_CHILDREN,
ContentModel.TYPE_FOLDER, ContentModel.TYPE_FOLDER).getChildRef();

//Lancement de l'action de création d'un sous-répertoire.
Action action = serviceRegistry.getActionService().createAction("createFolderTest-action");
serviceRegistry.getActionService().executeAction(action,rootFolderTestRef);

//Vérification de la présence du sous-répertoire.
Assert.assertEquals("ERREUR LORS DE LA RECHERCHE", 1, serviceRegistry.getNodeService().getChildAssocs(rootFolderTestRef).size());
}
}

Here I draw your attention to the inheritance. sampleActionExecuterTest extends the class BaseAlfrescoTestCase. It's a util class which allow to start a spring/alfresco context. In my class, I just create a test to start my work and verify the result.

2.3 Configuring the pom.xml
Our pom.xml will of course changed after this integration of Java code. In fact, Maven must know dependencies needed to compile classes. In this sample, we need

Alfresco-repository-3.3 for Alfresco services
Org.springframework.beans-3.0.0 to create Spring context
Junit for unit testing.

Our pom.xml becomes:




fr.opensourceecm
alf-sdk
3.3C

4.0.0
alf-amp-osecm
jar
0.0.3
Alfresco AMP Open Source ECM
Open Source ECM - Extension



org.alfresco.sdk
alfresco-repository-3.3
3.3
community
provided


org.alfresco.sdk
org.springframework.beans-3.0.0
3.3
community
provided


junit
junit
4.7
test

IMPORTANT NOTE: The packaging is JAR! Don't put it into AMP (it's a little "trick" to configure the project in Eclipse ...)

2.4 Configuring the Eclipse workspace
Generally (and if you've followed instructions above) you have never opened Eclipse for the previous manipulations! I agree ... is difficult. But you have to know that a Maven project generated is not directly integratable in Eclipse! In fact it is not planned for! There is no .classpath, .project file or .settings for configuring your project in Eclipse.
But Maven has a plugin to automatically create all of these configuration files. This is the maven-eclipse-plugin defined previously in the parent pom.

To create this configuration, type a command in the root directory of the maven project :


mvn eclipse:eclipse

Automatically, maven-eclipse.xml, .project and .classpath files have been created as well as .settings directories and ExternalToolBuilders (and eventually target).

Note: For users of Spring IDE, the project will have the nature of project Spring by default.

Note: If you have already joined the project in eclipse, or if you want to delete the configuration eclipse, type the following command:


mvn eclipse:clean

2.5 Importing the project in eclipse
You can now launch an eclipse (3.5R2 Galilleo for me) and import the project into your workspace. The project now looks like the image below:

Note: If you do not have this configuration, make sure you launch the command eclipse while the packaging is JAR. Indeed, if you execute this command with amp packagin the plugin will build another tree.

2.6 (Re) Definition of spring beans
We will now modify the declaration of spring beans for each part of our AMP to improve readability. For this, we will specialize a directory : context .

So in the \alf-amp-osecm\src\main\config, create a directory context and create 3 files:

action-context.xml
model-context.xml
webclient-context.xml

action-context.xml

model-context.xml






alfresco/module/fr.opensourceecm.alf-amp-osecm/model/ecmModele.xml

This statement, as its name suggests will allow us to define the data model. Note that in my previous post, I replaced fr.opensourceecm.alf-amp-osecm by (groupId) and (artifactId). But here, I want Eclipse to be able to understand after a compilation. Indeed Maven is the only one capable of replacing these values! So we must declare this value ...

Note: In general, groupId and artifactId must be defined at the beginning of your project.
Note: I take advantage of refactoring to simplify the name of my data model.

Now we will create the file
webclient-context.xml






alfresco/module/fr.opensourceecm.alf-amp-osecm/ui/ecm-web-client-config.xml

This statement, as its name suggests will allow us to define the configuration of the web client.
Note: I take advantage of refactoring to simplify the name of my web client config.

In the \alf-amp-osecm\src\main\config, edit the file:
module-context.xml

This file will serve to reference all statements necessary for our Spring module. In our case, to summarize, our module have : a data model, an action and a web client configuration.

2.7 Definition of our test environment
Now we will define and configure our Alfresco repository. Ie we will define the configuration settings of the database and where we put our folder (content-store and lucene indexes).

For this purpose in the directory /alf-amp-osecm/src/test/resources/alfresco/extension/ (create the missing directories if necessary) create the file

dev.properties


dir.root=./alf_data
index.recovery.mode=AUTO
integrity.failOnError=true
db.name=alf33Test
db.username=alfresco33
db.password=alfresco
db.host=localhost
db.port=3306
db.driver=org.gjt.mm.mysql.Driver
db.url=jdbc:mysql://${db.host}:${db.port}/${db.name}
hibernate.dialect=org.hibernate.dialect.MySQLInnoDBDialect

This file defines all properties of our test environment. In my case, I already installed and configured a MySQL database.

For more information on how to configure an environment, please download the following pdf: Manual Configuration and Installation of Alfresco 3.3

Then in the same directory, you must create the file
dev-context.xml







  classpath:alfresco/repository.properties
  classpath:alfresco/domain/transaction.properties
  
  
  classpath*:alfresco/module/*/alfresco-global.properties
  
  classpath*:alfresco-global.properties
  classpath:alfresco/extension/dev.properties



SYSTEM_PROPERTIES_MODE_OVERRIDE




  hibernate.dialect
  hibernate.query.substitutions
  hibernate.jdbc.use_get_generated_keys
  hibernate.default_schema

This bean will identify the properties file previously created. This is used only during maven test phase

NOTES
In this step, you can add a directory /alf-amp-osecm/src/test/resources/alfresco/desktop containing the default desktop ations Alfresco (available in the sdk). Apparently it can't have them in my environment ...

In this step, you can also add a log4j.properties file in the directory /alf-amp-osecm/src/test/resources to determine levels of logs.

2.8 Changing the pom.xml
We now have an environment ready. You just have to change the packaging tag in pom.xml to AMP.

2.9 Starting the Packaging

Now your project should look like the image below:

The project is complete, you can start packaging via the command


mvn clean install

If all goes well, you see the famous and coveted BUILD SUCCESSFULL and if you pay attention to the console, it display the results of tests like

Tests run: 1, Failures: 0, Errors: 0, Skipped: 0

If so you have your first AMP tested by Maven.

CONGRATULATIONS!

In the target directory of the project, you have packaged your AMP: alf-amp_osecm.0.0.3.amp. You can install it on your local Alfresco instance.

2.10 Starting the test via Eclipse
As I said at the beginning, I also want to use Eclipse to debug my application if necessary. I execute my JUnit test and verify the result.

Goals achieved!

That's all for today!
If you have any questions, remarks or bug / improvement to report for example, feel free to post a comment.

For this part, I wish to thank G. Tournier for his help and advice.
Without him Alfresco nodes would probably still a big mystery to me.

Stay Tuned !

PS

All sources of alf-amp-osecm can be downloaded at:
http://www.box.net/shared/374lms3ogv

All sources of alf-sdk can be downloaded at:
http://www.box.net/shared/oevx14hy8n

Maven + Alfresco : SDK, Dependencies, Inheritance

2010-04-21T04:03:00.000-07:00

Hello,

This time, I want to discuss about the integration of Alfresco SDK and dependencies in creation . The goal of this post is to explain how we can manage AMP dependency with Alfresco SDK.

If you are not familiar with dependency and maven, have a look at this adress

Actually, there's two mode to manage dependency.

Eclipse project mode or Build Path Dependency
Maven Mode or Maven Dependency Mechanism

Eclipse mode
In Eclipse mode, dependency are managed by links between Eclipse Projects. It's your responsability to define your builpath and classpath via Eclipse settings. In our case, we have a link between our eclipse AMP Project and Alfresco SDK. In details, this is the procedure to add SDK dependency with our project :

First, you have to download Alfresco SDK and unzip it in a directory. (let's say ALF_SDK_HOME)
Then in Eclipse, you have to perform the operation "Import Existing Project", usually AlfrescoEmbedded SDK project, in the directory ALF_SDK_HOME\lib
Always in Eclipse, you have to configure AMP Project Build Path. Most of the time, you have to open "Configure Build Path" form and add AlfrescoEmbedded SDK in "Project" tab.
Once all these steps accomplished, you can use the SDK in Eclipse. All links and dependencies are now available for auto-completion and compilation in Eclipse.

Maven Mode
In this mode, you have all AlfrescoEmbedded SDK dependencies integrated in a Maven 2 repository. It could be your local, your project, your company or a third party maven 2 repository. It doesn't really matter. You have just to know where to find it!

For information, Alfresco has recently open its own repository at Alfresco Nexus / Maven. Nevertheless, all AlfrescoEmbedded SDK dependencies are not available. But day after day it's going to be better ! Thanks at @Gabriele Columbro to his work http://mindthegab.com/

Now the job is to search, define and identify in our project pom.xml all AMP dependencies with the n-tuple : groupId, artifactId, version, classifier, packaging. According to this parameters, we can retrieve all dependency you want! When the pom.xml is complete and comprehensive, it becomes possible to compile and test the application with Maven.

But that's not enough... I'm a developper and I WANT(!!) my project into Eclipse ! How can I do if all is manage with Maven ??? Simple answer : Maven can generate Eclipse configuration thanks to a maven-eclipse-plugin. With just on Maven command : mvn eclipse:eclipse, you can (re)generate Eclipse configuration. But it's note the right time...

Anyway, it is this second method I will try to present.

How to set up Alfresco SDK in a Maven 2 repository
and use it to build an AMP

Note and warning: This procedure is far (very far even ...) to be the cleanest or most aesthetic method. It does not necessarily meet the very basic principles of Maven.
That's why it should be used with caution!

1. Integrating Alfresco SDK locally

As you know, Alfresco SDK includes all dependencies needed to run an Alfresco and to use it.

Problem : How can we add this dependency into our Maven repository ?
Via a small project written in Java and compiled using Maven (if you want to make Maven as well do it anywhere ^^ ).

Principle?
List all dependencies present in AlfrescoEmbedded SDK and integrate them into our repository via a script (. Bat or. Sh)

Code?


package fr.opensourceecm;

import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.File;
import java.io.FileNotFoundException;
import java.io.FileWriter;
import java.io.IOException;
import java.io.InputStreamReader;
import java.util.ArrayList;

public class AlfrescoMavenSDKCreator {

   public static String isWindows = "O";

   public static String pathToSDK = "D:\\opt\\JM-Alfresco\\SDK\\3.3C\\lib\\server";
   public static String sdkGroupId = "org.alfresco.sdk";
   public static String sdkVersion = "3.3";
   public static String sdkClassifier = "community";
   public static String mavenLocation = "call mvn";

   private static final String ln = System.getProperty("line.separator");

   private static void execute() {

      String nomFichier = null;

      FileWriter fout1 = null;
      FileWriter fout2 = null;
      try {

         showVariables();

         System.out.println("Voulez-vous modifier ces paramètres (O/N) ? ");
         BufferedReader in = new BufferedReader(
               new InputStreamReader(System.in));
         String reponse = in.readLine();
         if (reponse.toUpperCase().equals("O")) {
            System.out.println("Windows (O/N) ? ");
            isWindows = in.readLine();
            System.out.println("PATH SDK ? ");
            pathToSDK = in.readLine();
            System.out.println("Version d'Alfresco ? ");
            sdkVersion = in.readLine();
            System.out.println("GroupID ? ");
            sdkGroupId = in.readLine();
            System.out.println("emplacement de Maven (binaire) ? ");
            mavenLocation = in.readLine();
            System.out.println("classifier (labs/enterprise) ? ");
            sdkClassifier = in.readLine();
         }

         if (isWindows.toUpperCase().equals("O")) {
            nomFichier = "maven-sdk-install-files.bat";
         } else {
            nomFichier = "maven-sdk-install-files.sh";
         }

         File file1 = new File("maven-dependencies.txt");
         fout1 = new FileWriter(file1, false);
         BufferedWriter out1 = new BufferedWriter(fout1);

         File file2 = new File(nomFichier);
         fout2 = new FileWriter(file2, false);
         BufferedWriter out2 = new BufferedWriter(fout2);

         if (isWindows.toUpperCase().equals("O")) {
            out2.write("cd " + pathToSDK + ln);
            out2.write(ln);
         } else {
            out2.write("#!/bin/bash" + ln);
            out2.write(". /etc/profile" + ln);
            out2.write(ln);
            out2.write("cd " + pathToSDK + ln);
            out2.write(ln);
         }

         ArrayList listingFiles = getFiles(pathToSDK);

         for (File file : listingFiles) {

            String librairieNameJar = file.getName();
            String librairieName = librairieNameJar.substring(0,
                  librairieNameJar.length() - 4); // moins ".jar"

            out1.write("" + ln);
            out1.write("  " + sdkGroupId + "" + ln);
            out1.write("  " + librairieName + "" + ln);
            out1.write("  " + sdkVersion + "" + ln);
            out1.write("  " + sdkClassifier + "" + ln);
            out1.write("  test" + ln);
            out1.write("" + ln);

            out2.write(mavenLocation + " install:install-file" + 
                  " -Dfile=" + file.getPath().substring(pathToSDK.length() + 1) + 
                  " -DgroupId=" + sdkGroupId + 
                  " -DartifactId=" + librairieName + 
                  " -Dversion=" + sdkVersion + 
                  " -Dpackaging=jar" +
                  " -Dclassifier=" + sdkClassifier + 
                  " -DgeneratePom=true" +
                  " -DcreateChecksum=true" + ln) ;

            out1.flush();
            out2.flush();
         }
         fout1.close();
         fout2.close();

         System.out.println("Les fichiers ont été générés : ");
         System.out.println(file1.getAbsolutePath());
         System.out.println(file2.getAbsolutePath());
      } catch (FileNotFoundException e) {
         e.printStackTrace();
      } catch (IOException e) {
         e.printStackTrace();
      }
   }

   private static void showVariables() {
      System.out.println("This utils is going to generate two files");
      System.out.println("");
      System.out.println("Le programme va s'ex�cuter avec les paramtres suivants :");
      System.out.println("- pathSDK : " + pathToSDK);
      System.out.println("- Windows : " + System.getProperty("os.name"));
      System.out.println("- GroupID Alfresco : " + sdkGroupId);
      System.out.println("- version d'Alfresco : " + sdkVersion);
      System.out.println("- classifier : " + sdkClassifier);
      System.out.println("- emplacement de Maven (binaire) : " + mavenLocation);
   }

   private static ArrayList getFiles(String root) {
      ArrayList listeFichiersJar = new ArrayList();
      getFilesRec(listeFichiersJar, root);
      return listeFichiersJar;
   }

   private static void getFilesRec(ArrayList listeFichiersJar, String root) {
      File f = new File(root);
      File[] listFiles = f.listFiles();
      for (int i = 0; i < listFiles.length; i++) {
         if (listFiles[i].isDirectory()) {
            getFilesRec(listeFichiersJar, listFiles[i].toString());
         } else if (listFiles[i].getName().endsWith(".jar")) {
            listeFichiersJar.add(listFiles[i]);
         }
      }
   }

   public static void main(String[] args) {
      (new AlfrescoMavenSDKCreator()).execute();
   }
}

Warning : My code formatter (in this blog) sometimes modify or delete some part of the code. If you want to be sure to have the correct one, download the source of this project at the end of this post.

Running this small program from the command line interface or via Eclipse create two files.

The first file (a. Bat) can install dependencies in the repository. This file launch a serie of call maven as :

call mvn install:install-file -Dfile=alfresco-core-3.3.jar -DgroupId=org.alfresco.sdk -DartifactId=alfresco-core-3.3 -Dversion=3.3 -Dclassifier=community -Dpackaging=jar

We can notice from this sample, there's a "problem" in version number! Indeed, my script will generate an artifact name like : alfresco-core-3.3-3.3-community.jar.
It's a little bit non conform with Maven principle... That's why I consider this method as not aesthetic... but clearly the fastest!
If you want to resolve this problem, you have multiple choice.

First, identify one by one each dependency (and repository). Next, create files... if you have time to spend...
Second, you have to wait that Alfresco will add all dependency in his official repository or create a pom.xml with them.
Third, it's a mix between all of preceding solutions.

The second file creates all dependencies declarations. It can be easy to integrate them into a pom.xml This file contains a list of element below :



org.alfresco.sdk
alfresco-core-3.3
3.3
community
test

All dependencies have TEST scope. Indeed Alfresco already has all of its libraries. It is therefore not necessary to add them at the creation of our AMP. By cons it is needed to test the module.

Now how to use dependencies?
A strong point of Maven is its inheritance ability between artifact (plugin). It's the same as Java Classes inheritance. On object can inherit from its parent object.

In this context, one idea is to consider Alfresco SDK as the father of our draft AMP. Thus All SDK dependencies can be available for the project son.

2. Create a Super Pom Alfresco SDK

1. Creating a Maven 2 project
Via this command line in a directory :


mvn archetype:create -DgroupId=fr.opensourceecm -DartifactId=alf-sdk -Dpackagename=fr.opensourceecm.alf.sdk

you can create the whole tree of a jar Maven 2 project type.

2. Cleaning
I delete subdirectories from src/main/java and src/test/java to remove default generated class .

3. Changing pom
Now I take the content of maven-file dependencies.txt and I add it between &alt;depedencies> tags in my pom.xml. Next, I modify the defintion of my artifact to create : fr.opensourceecm: alf-sdk: pom: 3.3C.
Indeed, I know this project will become a parent project, so it must be automatically package as pom. Finally I add the plugin configuration for the build phase.
This produces the file below :



4.0.0
fr.opensourceecm
alf-sdk
pom
3.3C
Maven - Alfresco SDK 3.3C
Super POM Maven Alfresco SDK 3.3C 




alfresco-public-snapshots
http://maven.alfresco.com/nexus/content/groups/public-snapshots

true
daily



alfresco-public
http://maven.alfresco.com/nexus/content/groups/public




alfresco-public
http://maven.alfresco.com/nexus/content/groups/public


alfresco-public-snapshots
http://repository.sourcesense.com/nexus/content/groups/public-snapshots

true
daily






org.alfresco.sdk
alfresco-core-3.3
3.3
community
test


org.alfresco.sdk
alfresco-deployment-3.3
3.3
community
test

...

org.alfresco.sdk
mysql-connector-java-5.1.7-bin
3.3
community
test


org.alfresco.sdk
dom4j-1.6.1
3.3
community
test


org.alfresco.sdk
ehcache-core-2.0.0
3.3
community
test

...

org.alfresco.sdk
org.springframework.beans-3.0.0
3.3
community
test


org.alfresco.sdk
org.springframework.context-3.0.0
3.3
community
test

...





true
src/main/resources


true
src/main/config
alfresco/module/${groupId}.${artifactId}






org.apache.maven.plugins
maven-compiler-plugin
2.2

1.6
1.6




org.alfresco.maven.plugin
maven-amp-plugin
3.0.1-SNAPSHOT
true


false

4. Installation and Deployment Project
A simple command maven is enough to create our artifact and deploy it in our local repository :


mvn clean install

In our repository, we now have access to the module fr.opensourceecm:alf-sdk:pom:3.3C

Important !
If you use version 3.3C (like me here), we don't have config.jar in Alfresco SDK. In SDK it's a folder and not a jar. For information, this folder contains all configuration files for Alfresco. So it's clearly important to have it! This means, you have to create a jar of this directory and put it as a dependency in our pom.xml.



org.alfresco.sdk
alfresco-config
3.3
community
test

3. Update alf-amp-osecm project

1. Changing the pom.xml
Now we can modify our AMP Project pom.xml to take into account the parent. Note that we added the parent block in first place to define inheritance.




fr.opensourceecm
alf-sdk
3.3C

4.0.0
alf-amp-osecm
amp
0.0.2
alf-amp-osecm
Open Source ECM - Extension

We realize now that the pom is shorter. The information about dependencies and plugins have been inherited from the project SDK. That's the magic of inheritance!

2. AMP Creation
Now we have everything to create our module. You have just to execute one command in your project folder :


mvn clean install

In target folder, you have alf-amp-osecm-0.0.2.amp. It's possible to integrate it into your alfresco.war and test it.

3. Integration
Copy/paste the file alf-amp-osecm-0.0.2.amp to the amps directory in your Alfresco installation. Then run the installation script : apply_amps.bat in your root Alfresco installation folder. Follow instructions, wait few seconds and after that you have a new alfresco.war in tomcat/webapps folder.

For more information : http://wiki.alfresco.com/wiki/AMP_Files

4. Starting Alfresco
Now you can run Alfresco and verify the integration of your custom module.

WELL DONE !

Here I think, how we can do more with Maven Alfresco.

In the next post will explore how to set up tests with Maven.

PS

All sources of alf-sdk can be downloaded at:
http://www.box.net/shared/2zpgm3fcag

All sources of alf-sdk-osecm (for creating dependency files) can be downloaded at:
http://www.box.net/shared/fe6ftxeb4y

All sources of alf-amp-osecm be downloaded at:http://www.box.net/shared/s966vj1xqz

Maven + Alfresco : Creating AMP

2010-04-21T01:53:00.000-07:00

Hello everyone,

Recently I just finished the excellent Maven book written by A.Heritier and N.De Loof.
This is the kind of book I love because it is not just a catalog of functions but a big use case with idea and thinkings! So if you have not yet, I suggest you buy it at this address or any other good bookstore that has Technology / Computers shelf. Sadly it's only in French... It's a good ocasion to start learning french :)

I digress ... what I would like to share today is the integration between Maven and Alfresco. I think it's time to have a little reflection on this topic during several posts.

If you use a little bit Maven and Alfresco, you know there's something on the subject.
In fact when we go to this address, we can found all information to build Alfresco via Maven. Documentation is not necessarily clear at the first glance but after a while, you begin to understand all the ins and outs.

Let's start an example

I want to create an AMP to add my custom data model and my custom web client config.

In my toolbox, I have one maven plugin : maven-amp-plugin and one maven archetype maven-alfresco-amp-archetype.

For now I will only use the plugin. I want to start a little bit "From scratch". It's more shaping.

In pre-requisite, I installed Eclipse 3.5 and Maven 2.2.
I set the environment variables MAVEN_HOME and I configured my Eclipse workspace with this command in my eclipse workspace folder.


mvn archetype:create -DgroupId=fr.opensourceecm -DartifactId=alf-amp-osecm -Dpackagename=fr.opensourceecm.alf.amp.osecm

How to Create my AMP ?

1. Create Maven 2 project
Via a command line in my Eclipse Workspace Folder :


mvn archetype:create -DgroupId=fr.opensourceecm -DartifactId=alf-amp-osecm -Dpackagename=fr.opensourceecm.alf.amp.osecm

I create a complete folder tree of Maven 2 project (type jar).

2. Clean generated sample
I delete the subfolders of src/main/java and src/test/java to remove default class generated.

3. Add my custom model
In the src/main/config/model, I add my data model: openSourceEcmModele.xml

NOTES If you have no model, you can take "customExample.xml" from an Alfresco extension folder ( tomcat\shared\classes\alfresco\extension)

4. Add my web-client-config
In the src/main/config/ui, add the client configuration web openSourceEcm-web-client-config.xml

NOTES
If you have no model, you can take "web-client-config-custom.xml.sample" from an Alfresco extension folder ( tomcat\shared\classes\alfresco\extension)
NOTES
Previously, we have verified that the two above files are well-formatted and valid using the principles of dynamic models: http://wiki.alfresco.com/wiki/Dynamic_Models

5. Add module-context.xml
In the src/main/config, I created the file module-context.xml. Through this file, I defined spring beans to identify my model and web client configuration.



   
       
           
               alfresco/module/${groupId}.${artifactId}/model/openSourceEcmModele.xml
           
       
   

   
       
           
               classpath:alfresco/module/${groupId}.${artifactId}/ui/openSourceEcm-web-client-config.xml

6. Add module.properties
This file is needed to identify the module and its details when it was installed by the ModuleService. It incorporates the values defined in the pom.xml of the project.


module.id=${pom.groupId}.${pom.artifactId}
module.title=${pom.name}
module.description=${pom.description}
module.version=${pom.version}
module.repo.version.min=3.3

7. Change pom.xml
This file is the unit of work for Maven. This is an XML file that contains information and configuration of your project. Maven used it to configure, generate and build what you may except of your project. It also contains the default values of the projects.



4.0.0
fr.opensourceecm
alf-amp-osecm
amp
0.0.1
alf-amp-osecm
Open Source ECM - Extension



alfresco-public-snapshots
http://maven.alfresco.com/nexus/content/groups/public-snapshots

true
daily



alfresco-public
http://maven.alfresco.com/nexus/content/groups/public




alfresco-public
http://maven.alfresco.com/nexus/content/groups/public


alfresco-public-snapshots
http://repository.sourcesense.com/nexus/content/groups/public-snapshots

true
daily







true
src/main/resources


true
src/main/config
alfresco/module/${groupId}.${artifactId}





org.alfresco.maven.plugin
maven-amp-plugin
3.0.1-SNAPSHOT
true


false

We notice the presence of maven-amp-plugin as dependency. It allows the packaging of the project to AMP (see tags <packaging> amp </packaging>)

To find this dependency, we add the tag <>. It authorize Maven to search and get various dependencies from this place.

We also note the presence of tags <resource> & <filtering> for updating ${...} values by the values defined in the pom.xml.
For example, in our case the value of = $ (pom.version module.version) in the module.properties will be replaced by the value <version> 0.0.1 </version> in the pom.xml. Useful for defining a single variable values.

8. Compilation
Now we have everything to create our module. We just have to execute one command line in our module folder :

Now you can find in the target directory of our project, the file alf-amp-osecm-0.0.1.amp. It's now possible to integrate it in an Alfresco.war.

9. Integration
Copy/paste the file alf-amp-osecm-0.0.1.amp to the amps directory in your Alfresco installation. Then run the installation script : apply_amps.bat in your root Alfresco installation folder. Follow instructions, wait few seconds and after that you have a new alfresco.war in tomcat/webapps folder.

For more information : http://wiki.alfresco.com/wiki/AMP_Files

10. Starting Alfresco
Now you can run Alfresco and verify the integration of your custom module.

Good Job !

For me, this is the simplest example of creating an AMP into Alfresco.

In the next post, we will explore how to manage dependencies and inheritance.

PS : Sources of this project, called alf-amp-osecm, can be downloaded at:
http://www.box.net/shared/tvkc3olink

My news...

2010-04-21T00:44:00.000-07:00

Hello World !

As you have seen, it's been a while that i didn't post anything on this blog ?
Why...? I'm a french guy so I have a lot of excuses ! :o)

Seriously, I have 3 excuses ...

1. Merry Christmas !
What's the link between my production of posts and Santa Claus ...? Bill's gaming console !
Since christmas, I passed most of my free time

searching Cortana and save humanity in my green armour suite...
fighting in the street...
becoming the hero of bowerstone...
traveling through Pulse and Cocoon...
dropping bombs everywhere to explose funny character...
etc...

If you recognized and played all of those game, you have spend a lot of time like me.
Welcome in my gaming world ! :o)

2. Project Time !
Since the beginning of january, my Alfresco project for "French Postal Company" started a new phase. Naturally, I had a lot of work to do. So I spend my working time developping new features, improving my code and trying new pratices... I had no time to blog during this time...

3. Winter break
Finally, it's good to have a break :) Like bear or marmot, a good hibernation is always a good thing.

And now, what's the plan ?

Since last week, I re-start blogging on my french blog about the integration between Alfresco and Maven. I used it a lot in my work and I didn't know if my approach is the best approach. So I'm going to explain it to try improve it. It's time to share my experience.

So rendez-vous très bientôt ! Stay tuned!

Nuxeo in 2009

2009-12-02T00:02:00.000-08:00

It's been a while that I didn't share some news about Nuxeo on this blog. I'll try to make up for it...

Let's go back to the future... In 2009.

Nuxeo 5.3
During October, Nuxeo announced availability of its new version of Nuxeo DM: the 5.3.
This version is the most complete Nuxeo that we've seen. Functionally richer, it covers more complex and extensive needs.

Among the list of new features, I will mention two of them

Open Social: When we met "clients", one recurrent need is the famous "portal". I put out the word "portal" between quotes because, as the term ECM, portal are not a friendly term and can explain anything at the same time. Here I talk about Widget/Gadget Portal like Google or Netvibes. We can arrange on a single page a set of small square to form a mosaic that meet the functional requirements of the customer. In other words how, in my application, can I aggregate content and applications from different sources ? (and if it looks like Google it's even better!)

To satisfy this need, Nuxeo has integrated http://www.opensocial.org/. This is THE reference (de facto standard) in terms of creating composite application. Nuxeo DM becomes both Open Social Gadgets client and server . Think about it when you customer want to add the famous Weather box. ;o)

Federated publishing : This need is more technical oriented and really important for integrator or any large company. To understand the usefulness, consider the example of a company which has 2 remote sites (geographically and functionally). One site is the head office while the other site is a production site. At head office, product team develop and creates all of new products documentation (product schema, manufacturing instructions, user guide...) and stores / shares then in ECM Repository. The production team uses this documentation everyday. A lack of documentation can generate the malfunction of production site. To avoid such unforeseen, we will create an 'master-slave "solution between the main ECM repository and a local one only for production. Between these two sites, we can create a link to share and distribute information. This practice guarantee and ensure high data availability.

You can find the announcement at http://www.nuxeo.com/en/about/news/nuxeo-dm-53

If you prefer video, you can find a webinar presenting the capabilities of Nuxeo at:

http://www.nuxeo.com/en/about/events/webinar_nuxeo_dm_53

Nuxeo and marketing
For over a year and to support its international expansion, Nuxeo has considerably improved its marketing presence. Concretely this means more events both on the web via webinars, videos, presentations, participating in social networks or concrete meetings.

Reminder : Nuxeo World http://www.nuxeo.com/fr/societe/actualites/nuxeo-world-2010

One of the highlights of this strategy was the appointment of Cheryl McKinnon as CMO (Chief Marketing Officer in English). You will say "yes, what's else?" I will answer that she was Director of Program Management, Enterprise 2.0 from Open Text (proprietary editor solutions content management world and a leader in this market ... and Nuxeo competitor) and secondly she is known and recognized in the international ECM community. If you want further evidence of the rise of open source document management, I think you have a new ;-)

Nb: If you want to follow Cheryl McKinnon

Blog: http://candyandaspirin.blogspot.com/
Nuxeo Blog : http://blogs.nuxeo.com/cmckinnon/
Twitter : http://twitter.com/CherylMcKinnon

Three axes of conquest
During 2009, we discover that Nuxeo is moving to 3 directions (which I consider important)

Semantic area
Apache Chemistry and CMIS
Nuxeo Cloud

Semantic area
To be honest, I'm not an expert in this area but I know how important it is to provide intelligence into data stored in content management system .

To get an idea of integrating this component in Nuxeo, I'll refer you to the various available resources . They are mainly from Olivier Grisel work at Nuxeo and Yerbabuena Software which has developed the first module Nuxeo in this area.

Article

Videos

Nb: If you want to follow Olivier Grisel: http://twitter.com/ogrisel or Yerbabuena http://twitter.com/yerbabuenasoft

Apache Chemistry and CMIS
Nuxeo has launched for several months now the Chemistry initiative : http://incubator.apache.org/chemistry/. This initiative seeks to become the open source implementation reference of CMIS standard. Due to the approaching end of CMIS public review, we can suspect that this part will be increasingly important in the coming months and will contribute to Nuxeo reputation.

NB : Have a look to jonontech.com blog :
http://jonontech.com/2009/11/26/cmis-jcr-and-osgi-for-idiots/

Nuxeo Cloud
You want an ECM Solution without worry of technical infrastructure ? You want a distributed system which can support easily high availability ? So take Nuxeo cloud

And tomorrow?
This post by Laurence Hart on his blog wordofpie.com inform us about Nuxeo main development axes. We can find Record Management and a catalog of application services (mail management, DAM, Cloud ...).

Voila Ladies and gentlemen, all I can summarize about Nuxeo. I remain at your disposal for further information and to share various news about Nuxeo.

See you soon!

Open Source ECM V2

2009-11-27T02:00:00.000-08:00

As you can see, http://www.open-source-ecm.com/ has changed its look!

Lighter, larger, whiter, more simple, morecollaborative ... in other words more better!

So What's New ?

I changed the blue template for a white template. It's more ergonomic and allow an easy-reading. I drop "full-post" design for a preview one. Indeed on the homepage it is now possible to see the latest articles published on the blog (and not the long last one)... To go further, just click on "Read More" to see the whole article.

A menu bar appeared! Cruel missing in my previous blog, this bar will allow me to create links quickly and easily to a particular aspect of my blog. The 3 subject are currently "About", "My Interviews", "My Presentations".

To the right of the menu bar (this time properly integrated), we find a field of "quick search". Useful is not it?

A new mascot is also appearing! Tuxitecte (in english equivalent as Architect Tux) arrived from http://tux.crystalxp.net/fr.id.13668-tuxitecte.html site and was created by kami23 under creative commons license. You can find his works on the site http://jonkie.ch/index.html. Merci kami23!

In 2.0 section, you can find all my sharing space and "social networking" in My Universe Space. In this same category please subscribe to the RSS feed as a feed or as email.

The categories changed into "cloud" for better readability. We realize very quickly what are my interests. ^^

Each section is now "retweet". On my old version I tried Dzone and Digg indicators but I soon realized that these indicators (for the content of my blog) were not necessarily very convincing ...

The creation of comment is now simpler and clearer. By doing a little inventory of my old blog, I think the lack of simplicity was probably a brake ... I hope now that this mistake will now be corrected. Stay tuned...

Each post now has the famous "5 star ratings". I don't know you but for me I do it more and more. It become a reflex. If this works, these little stars will guide me a little setting up my articles.

Finally, each post also has links to various "favorites" to social tools & networks.

Now it's your turn, enjoy !

How to choose a *CM solution ?

2009-11-25T00:06:00.000-08:00

Hello everyone,

The topic of the day is devoted to an old problem:

How to choose *CM solution (Content Management)?

(* represents a letter of your choice, but commonly an E or W)

This question implies many issues, include many methodologies and is unique to each project context. In other words it is not easy (or impossible) to standardize a single approach to the problem. Nevertheless it is possible to avoid major pitfalls and use best pratices. This is what we'll try to see later through a series of rules.

Note: This rules are from different feedback which you can fin as references at the end of this post. It engages only me and can't be THE good road to choose its solution ...

Rule No. 1: What are my needs ?
Need : I remind you that this is the basis ... Without a LITERAL definition of need, without use cases DIAGRAM described by LITERAL explanations , without a HISTORY of functional requirements it is necessary to start a project that will result in inadequatechoice ?
This point is the famous WHAT for "What is the future scope of my work?"

In document management project, you must identify documents type that the system must manage. What are their format, their average size, frequency of creation or modification, deletion. Who is responsible for this data ...?

Rule No. 2: What are my technical requirements?

Following the recommendation of my Enterpise Architect I must be able to retrieve information from an application. Is there a connector with this application or what's the feature which can help to retrieve this information?
From my architects, my future solution must be based on an EJB3 architecture and must be installed on Ubuntu Linux.
The operating service needs to monitor the future application via JMX.

When such constraints is known, it is easier (for both client and supplier) to choose an appropriate solution. These constraints could create a list of MUST-HAVE. It can be useful to discriminate when comparing offers. It's up to you to place your cursor on the importance of criteria (don't forget of course that few solutions satisfying all criteria)

Rule No. 3: Small streams make large rivers (translation of french proverb.)
Based on this principle or the famous KISS (Keep It Simple Stupid), it's best to start simply, quickly and humbly then verify, validate and begin to see in a second time how to extend the functionality and features offered. With this methodology, you can have quick feedbacks from your operational and this is priceless!

In the area of development this is called Agile Development :o)

Rule No. 4: To choose a solution, the rating scale/matrix you must ban!
To have performed the exercise as a consultant and as user grids are useful in fact... for nothing... Except to prove and say what you want!

For those who had never seen, here's a little description:

This is usually an Excel file with a list of criteria grouped into families, which themselves are clustered in groups, which themselves are grouped by category. Each critera has a weighting. Sometimes it's also considered for families, groups and categories ... This list represents the Y axis (from top to bottom).
Then on the X axis (left to right), we'll add different solutions. They have to pass with success or not the treatment of the matrix.
The intersection of critera line and solution column must be filled by a combination of score between 0 and 5.
Finally, in the bottom of the Excel sheet, there is the final grade. Resulting in average weights between products and notes, this note is responsible for the choice of solution. This "scientific" approach allow to find right choice without doubt!

Simple isn't it ? :o)

The main criticism of this practice comes from the "subjectivity" in the criteria selection, weighting and rating.

Example:

My need
The technical staff (designer of our products) need to share their good practices in a collaborative way.

Possible criteria
Wiki Integration in my CM System

Possible responses

Response from a consultant: For the solution X, there's a possibility to integrate a Wiki solution via the API.
Response from an editor: In my solution, One of our features can manage a wiki.
Answer from an integrator: With a little code, you can create a simple wiki system that you can connect to the solution Z DM system.

Now It's your turn. Post a score from 0 to 4.

In fact, do you ask what is the syntax of the engine? Do your colleagues want to learned a new syntax ? or do they simply want a simple (or rich) text editor, with the integration of media such as images ...?
Then you have to think how these contents are stored? How they can be reused, shared and indexed? and so on ...

Now you would put the same note?

This essentially limits the Matrix system ...

Rule No. 5: In a RFP don't require a matrix list of critera !
Corollary to the previous point, if you do not want wrong and subjective answers, don't ask to complete a grid. Simple...

Rule No. 6: With whom I want to work?
It's a simple rule but it's not always applied. Is not it easier to work with people we know and with whom we have create trust link, respect, honesty and efficiency?

Do not hesitate to meet different people who are involved both in choice phase, design phase or production. Those are the people who are foremost guaranteeing the success of your project.

Rule No. 7: You know your budget? Then don't hide it !
I know that this choice may be considered as crazy, but it will save your time! On the one hand you will know which suppliers are really interested in the project and on the other hand, it allows you to choose the product range wich best match your needs.

Note: As a fervent open-source addict, I want to remember that Open Source Solution is NOT a synonym of FREE!

Rule No. 8: Open Source or Proprietary same fight or in other language there's no prejudice you should have!
Although I am a strong supporter of open source, you can't exclude the choice of a solution due to this Proprietary approach(and vice versa!). In choosing a solution only your needs should count (and your budget of course ^^...)!
If the need is well covered by a proprietary solution and you can buy licences with your budget, why deprive yourselves?

Rule No. 9: A prototype you have to test !
Sometimes in the selection criteria, we find the words: "2.0 user Experience", or "Enriched Usability and Experience". If you have never installed or used the product how can you trust a note or a description. It is in these times that we see the usefulness of a prototype.

Small reminder: A prototype is not the final solution. It's a few days work and not your several months project result.

Prototype goal is to demonstrate the strengths and weaknesses of a product on a particular use case. In addition, you will see solution degree of mastery by an integrator, or product degree of adaptability by an editor. This is not the Supplier(Integrator or Editor) to show what the product can do but the product to show what he can do in your context. And if the prototype crashes it does not matter, it shows instead what it should really do.

Rule No. 10: I don't understand ...
You have trouble understanding the principles of content management project. Simply, call for help !

That's all folks. Don't hesitate to post a comment.

This entire article wouldn't have been possible without precious post and articles on the subject. Here are some selections:

[FR] BPMBulletin :

http://www.bpmbulletin.com/2009/11/23/5-facteurs-cles-de-succes-pour-mettre-en-oeuvre-une-application-ecm-bpm/

[ENG] BIG MEN ON CONTENT

http://bigmenoncontent.com/2009/11/03/8-dimensions-of-cms-technical-evaluation/

[ENG] BLEND INTERACTIVE

http://blendinteractive.com/blog/business/getting-good-rfp-response/

[ENG] CMS WATCH

[ENG] CONTENT HERE (My favourite!)

[ENG] JON ON TECH (MUST READ!)

http://jonontech.com/2009/04/12/which-comes-first-the-crew-or-the-cms/

Wave & ECM : My feelings and opinions

2009-11-23T13:57:00.000-08:00

Hello everyone!
If you don't live on another planet, you've probably heard about GOOGLE WAVE phenomenon. Having obtained the precious sesame since october, I've tested the tool and I'm now ready to explain my opinion about it.

I'm not going to get an n-th presentation of Google Wave. The resources on Internet (and google of course) are not lacking. So today I just want to share with you my recent thoughts on the relationship between ECM or CM in general with Google Wave ... Does Google Wave will crush ECM or marry to it ? Is it a white elephant or technological breakthrough?

Collaboration and Communication
First Wave is like Google's definition says:

Google Wave is an online tool for real-time communication and collaboration. A wave can be both a conversation and a document where people can discuss and work together using richly formatted text, photos, videos, maps, and more.

a tool to exchange information (in other words communicate...) between one or more persons synchronously (interactive) or asynchronous (Static) ...

Let's do a little inventory. In the category of collaboration tools and communication exist:

E-Mail : is a method of exchanging digital messages.
Forums: Exchange Place where users ask or answer a given question. The various contributions, accessible to everyone, from forming a thread (thread in english). Each forum addresses a specific theme.
Instant Messaging: allows the instant exchange of text messages between computers connected to the same computer network, and most commonly the Internet. Unlike email, this medium is characterized by the fact that messages are displayed in near-real time and enable an interactive dialogue.
Mailling List: is a specific use of email that allows the mailing of information to users who are enrolled.
Podcast: podcasting allows users to listen or download the automatic audio and video broadcasts for personal music players for an immediate or later viewing.
Wiki: This is a website that offers users the chance to make editing pages.
Collaborative Document Edition: Set of tools to create jointly in a group one or more documents.

Now put everything into a single application, add a strong dose of Mashup-based widgets, shake it all and you get a Collaborative Google Shake ie Google Wave!

Google Wave: A Genetically Modified Collaboration Tool .
That is what appears when you use Google Wave. This UFO is impossible to store in a box and that is what is surprising and disturbing at the same time most users. Deja vu for some, a revolution for others. And you, how you can see this UFO?

Google Wave: One ring to rule them all?
Compare that with the phone. 10 years ago, the phone was used to ... Call! Then it allowed us to send SMS, MMS, E-Mails ... It turned into portable music player, photo album, and movie theater ... Today it serves to locate us, to browse Internet or just playing ... Yes there's an app for that!

Well you can compare Google Wave Revolution to mobile phone Evolution . It will allow us to make a lot of stuff most brilliant (or not. It depends on the ideas ...) between you and me (why not!) Or between you and us (I have friends!) Or between you and us (because you also have friends) or within a community now, tomorrow or even later.

In content modeling we call this a primary or generic content (Does Google put a node into genetic heritage of a Wave ... Mystery ...) able to adapt and adjust.

Google Wave: ECM Killer?
And well, given its genetic heritage ... My answer is NO! (After I'm possibly go wrong ...)
Rather it is a very promising PLUS (meaning +) to make content management easier (?). Indeed the interest of a wave is that

Is not replicated on each client (as in every mailbox in the case of e-mails)
Is easily shareable (you can easily tell who has right or not being in a wave)
Is classifiable and therefore indexable and thus discoverable (it's still the job of google to find after all...)
Can be changed at will! (everybody can put his personal touch)
Can be synchronous or asynchronous (you're not there ... too bad I'll leave a message. You're here so we can go faster!)
Manage versions (Playback. .. I'd like to see who added what and when)
Is deployable at both in supplier and in my network safely behind my firewall!

Google Wave has taken some basic principles of content management (versioning, categorization, collaborative editing, user management ...) but can not be a "tool for content management" (as I said earlier we can't put in a box ...). Its raw material is the content! Remember : Content is everywhere and is everything! Wave can create and disseminate information easily and simply.

Google Wave in Business?
Google Wave will, of course, land in business like most Google tools. Who has never used a Gmail, a Google map, a Google Reader or Google for his professional usage? (Yes, you can excludes those without Internet at work...) But this will not immediately .. Firstly because it is a beta (a real one), partly because it requires that everyone starts to tame the monster to be a faithful companion and finally because we must invent new practices based (or not) on the ancient best-practice.

Who will benefit?
First, the small will still lucky for one simple reason.
Why do we invest in each tool like wiki + instant messaging + e-mail + collaboration tool + Forum ... One tool is clearly enough for my needs. Furthermore I have no infrastructure and I don't want! An SMEs / PMI offers should be available soon as is already the case for Google Apps.

Then Apps Creators
As for Iphone, Facebook and other means of communication in the wind, we'll see (and is already the case) full of gadgets, widgets and other add-ons to increase and enrich our user experience. It will be probably the same with some shiny and smarty User Interface.

After Third integrators
A good integration will be Archiving Application which can long-term preservation of waves ie how to archive information exchange between one or more persons. This will be less dependent on storage volume but probably more complex due to integrating external data sources.

Other integrations are those that we see recursively : office desktop tools, CM or ECM apps, business applications, security system...

And then WE... Because we are at the root of everything.

Finally
Once Windows providing all services through one operating system and softwares.

Now Google offers all services through Internet and Apps.

This becomes increasingly visible and palpable. Impressive for an integrator like me, normal for lambda users ... That is also the magic of computers!

Thank you for reading and see you soon.

PS: I still have 2 invitations Google Wave ... The first 2 to post a comment will be lucky! That is also the magic of the waves. The hunt for the invitation ...

Install and Test My Open Source ECM Family

2009-10-29T01:07:00.000-07:00

Hello Folks,

Do you know one main advantage of OpenSource ECM Solution ? You can test it !

As ECM consultant / Integrator, I have to install, test, configure, use, modify many different solutions in differnt environment. But I have a BIG problem... I just have my XP dev machine and no test Server... Of course I can install all of this things in my computer... But I'm very limited by my hard disk space... And I prefer to dedicate my machine for dev only! I fed up sparing time... Futhermore I can't share my installation and best pratice configuration and my colleague can't help me and can't share their experience.

So what's the solution ? Virtualization of course!

That's why, since a few moment now, I start a collection of virtual OpenSource ECM Solution.

How ?

First, I have to find "Ready" VMWare Images. Indeed, I don't always have time to download and burn ISO images of operating systems and then create a VM image. It could be very long. So I was looking for a way to get VMWare images and then use them directly without having a VMWare Workstation installed on my computer. So I opted to retrieve images on sites like Http://www.thoughtpolice.co.uk/vmware/ I have an external 320GB hard drive and a good ADSL internet connection.

Then to use them, I download VMWare (FREE) Player http://www.vmware.com/products/player/

I have all I want to start my collection. All of this for 0 € free of taxes and legally!

Now what's the menu boss ? SHARING !

Below, you can find presentation (or simple HOW-TO) to install my collection of OpenSource CM Solution. You will also find some important links.
PS : This installation is only for test purpose!

Configure Your Virtual Server

Ubuntu 9.04

Simple Installation

Alfresco
EzPublish (Work in progress)
Drupal (Work in progress)
GateIn
Jahia
KnowledgeTree
Nuxeo

Advanced Installation

Alfresco (Work in progress)
Nuxeo (Work in progress)

In a near future(If all goes right...), I will share other presentations on specific aspect...(LDAP Connexion, Database configuration, continuous integration, clustering...)

With all my CM family, I can create my private network! Test combination, configuration, new version or just create prototype and share it.

Of course, I can't start all of this virtual machine at the same time. But it's enough to me :-)

Hope it will help you!

Interview : Daniel Tellez, CTO & Owner of Yerbabuena Software

2009-10-14T12:46:00.000-07:00

Buenos dias a todo el mundo !

This time we make a travel to Spain.

Today, I have the honor and pleasure to interview Daniel Tellez, CTO & Owner of Yerbabuena Software.

Hello Daniel,

Hello Jean Marie. I'm very pleased to answer all your questions and to appear in your blog. I've been following your blog for some time because of your interesting posts about ECM, specially in the opensource world.

To begin this interview, I wanted to thank you for giving me a little of your time to answer this interview and to share your knowledge of open source ECM and specially Nuxeo.

First Daniel, what's the purpose and objectives of Yerbabuena Software ? What kind of services do you offer?

We, Yerbabuena Software, are a very young company, and we are very proud to count with a staff of young professionals who want to do very big things in their career. I tell you that because this is one of the main reasons because we have embarked in many interesting and great projects, like the one that will let us annotate semmantically and classify contents in Nuxeo, or access to its repositories through a smartphone.

Our mission and goals fit perfectly with all of these, and can be summarized in the next sentence: to be a leading and international company, reference in the field of development, research and innovation based on opensource enterprise content management solutions. This mission has not been chosen randomly, but has been the result of a hard but funny process or rediscovering of our ideas and goals. You can find a picture here, in the form of a roadmap that drives our actions to that goal:
http://translate.google.com/translate?client=tmpg&hl=en&u=http%3A%2F%2Fblog.yerbabuena.es%2F2009%2F04%2Fmapa-estrategico.html&langpair=es|en

Sorry, we still have to translate it. ;-)

About our services portfolio, I have to say that our company provides mainly consulting, training and development services over opensource ecm software solutions. It is in this field where we can add more value to our ecosystem, and where we can deploy more freely our ideas and innovation, to apply them to our customers.

As Owner and CTO can you describe your work in Yerbabuena Software ? What's your role and your work day after day ?

In Yerbabuena Software we have a very horizontal organization. We are all folks, in and out of the office, in the most of times. So, my daily work is mainly to help my colleagues in the three development teams we have, to care for the methodologies used (always agile) and to provide a seamless connection between development and comercial sides.

Why did you choose CM ecosystem ? Is there some opportunities ?

Well, as I said before, we have a small company that knows perfectly the problems of small companies. In the earlier days of Yerbabuena Software, we had to face with all these problems, decisions and investments like all the companies do. How easy it looks when you are in college !

We thought it will be a good way to earn money to sell ECM solutions to SMEs, because we are convinced that this is not a field of software that could only fit in great organizations. All the companies have to manage papers and documents. We have seen very little companies with whole rooms full of boxes full of papers. Yes, very hard.

Web content management and opensource is a good entry point to begin to work with SMEs. Basically because of the simplicity of the WCM concepts and the cheaper costs of opensource. It can begin publishing news, events and descriptive web pages, and sooner the need to share and generate documentation comes through this tool, as so does the need to deliver it with other departments, with companies, with customers. Very soon the need of applying workflows, auditing, etcetera comes.

Actually we are jumping from SMEs to bigger organizations with large amounts of documents and users, like Diputación de Malaga (regional government of Malaga) or La Salle colleges.

Definitively, in Yerbabuena we are always researching these gaps not covered for common ECM providers, from earlier days by offering ECM to SMEs, until today, when we develop solutions for problems like accesibility and usability in reading documents ( http://translate.google.com/translate?client=tmpg&hl=en&u=http%3A%2F%2Fblog.yerbabuena.es%2F2009%2F05%2Fprimera-beta-de-easyreader-desarrollada.html&langpair=es|en ), or increasing ubiquity in the access to the repository ( http://translate.google.com/translate?client=tmpg&hl=en&u=http%3A%2F%2Fblog.yerbabuena.es%2F2009%2F07%2Fprimera-beta-de-yerbabuena-ercp.html&langpair=es|en).

What are the main trends in ECM in Spain?

Well, in the last months we are hearing very much noise around ECM like some time before, at least in Spain. This is good. The market of ECM is becoming bigger in Spain. And the companies that did know ECM before, are replanning its resources to change to a more opensource scenario. In these crisis days, harder in our country, the companies are realizing that it is not easy to justify licenses costs.

Now, we can see how many companies knows about the leading ECM providers in the open source side, like Nuxeo and Alfresco. The first mainly because of us ;-)

Now, let's talk about Nuxeo. Why did you choose to use Nuxeo for CM problematics?

When we began to plan our business strategy around document management, more or less in 2006, Alfresco was a little toy, while a french company called Nuxeo few years working with a very cool document management and web content management tool. We loved it. And we did many projects with that foundation. And we continue doing it. (Nuxeo CPS)

After, Nuxeo decided to switch to JavaEE technologies. This was a full reimplementation of the previous tool, maintaining all they learned before about ECM.

But we are an independent consultancy company, so we didn't take the JavaEE option given by Nuxeo 5, but we decided to take a look at the tools around ECM in this year. I will not go into much detail, but we couldn't find no one platform that fits so well than Nuxeo: we loved its architecture, we loved the real opensource policy, we loved that all of the CPSLovers like me didn't have to face many troubles to understand Nuxeo 5 concepts.

On the technical side, what are the advantages and disadvantages of this solution ?

The more important technical advantage of Nuxeo over any other document management system is its transparency and easyer way to fit it in any existing information system. This can be checked in many aspects, like:
- It is totally and really open source, like no other ECM application in the market.
- The active participation in CMIS technical committee of specification, in which the other big players in the ECM world participates also, like EMC, Microsoft, Alfresco, etcetera.
Nuxeo participates also in the development of Apache Chemistry, in partnership with Day Software.
- Nuxeo 5 provides different types of APIs, making easier the integration with third parties software.
- The use of standars and wide used technologies ensure the interoperability with other systems, the easy of access to the system to software professionals, the low chance to get into a vender lock-in strategy, etcetera.

Is very difficult for me to find a technical disadvantage over other content management systems. Maybe we could go into the typical discussion about the systems needed by JavaEE based technologies, more expensive than in others like Python or PHP.
But JavaEE has shown that it is a more reliable framework, resolving common problems for enterprise software.

Did you make some contribution ?

We started to contribute to Nuxeo 5 in the typical starting way: reporting issues and providing translations to the system, like iberian languages: castilian, gallego, catalá, euskera, portuguese and from others places like vietnamese.

Today, we have a very interesting project, commented before. We want to build two great components to Nuxeo 5: one rich interface to access document management repositories and most functions, and another component to annotate semmantically the contents and to automatically classify docs in Nuxeo 5. This is being the most important contribution for us in the project.

Like a non-technical contrib, we have to say that Yerbabuena have been, and still is, the main driver in the spanish-speaking community for Nuxeo 5 and one leader in evangelism of open source content management in this community and also in others like Viet Nam. To do it, we have participated in events and programs like Open Source World Conference last summer in Málaga, FOSS Bridge program or Open World Forum in Paris.

Personally, are you part of another open source community? Do you have examples of participation (Animation, articles, forums ...)?

Me and my colleagues have participated actively mostly in Nuxeo 5 and Nuxeo CPS projects. The main proofs of it can be viewed in the mailing lists of both projects.
Although Nuxeo CPS has lost great part of its strength as a community in favor of its elder brother Nuxeo 5, it continues to be an interesting project and a very powerful option for ECM and WCM, competing greatly with more popular platforms. So I participate yet in this project all the times I can by reporting new incidences or patches.

Our contributions feed also our blog posts, so I recommend you to visit frecquently ;-)

In the other hand, we have an own product, Bluesearch, that can be viewed as totally independent of ECM business line. This is a project oriented to marketing, which relates user profiles of mobile phones, so if they match in preferences or likes, it lets them connect to talk or so. This product is also totally open and our other great contribution to open source community.

To conclude this interview, what's your favorite blogs and links on IT in general?

Well, I follow many ECM related blogs, although in the latests days I prefer to use twitter to be in touch with the latest technologies and news. I know you also like twitter ;-) Here I can link you and interesting resource to follow the list of the main ECM gurus:
http://jonontech.com/2009/07/21/follow-forty-twitter-cms-gurus-in-three-clicks/

... BTW, Jean Marie... you are also here ;-D

In the blogsphere, I follow specially and recommend the Nuxeo one, the John Mancini's blog ( http://aiim.typepad.com/aiim_blog ), and the Word of pie ( http://wordofpie.com/ ), the three highly recommended, and of course yours ;-D

One last word before leaving?

Don't forget to visit our blog. We are starting publishing our only english blog, in the way you do it with http://www.open-source-ecm.com/ , and where we publish very frecquently our contributions to the community.

It's in http://blog.yerbabuenasoftware.com (english)
http://blog.yerbabuena.es (spanish)

And, of course, that the world of opensource ECM continue doing great!!

Thank you again Daniel for this interview and we wish you a nice and exciting journey on Open Source ECM Road!

Ps: As usual, you can download this interview at this url : http://www.scribd.com/doc/21070811/Open-Source-ECM-Interview-ENG-Daniel-Tellez-Yerbabuena

Interview Laurence Hart, Senior Manager/Information Management Solution Lead at Washington Consulting

2009-09-29T23:55:00.000-07:00

Hello Everybody !

For the next interview, we will go to U.S.A.

Today, I have the great privilege to interview Laurence Hart, Senior Manager/Information Management Solution Lead at Washington Consulting, Documentum Expert, blogger of www.wordofpie.com, ECM philosopher and twitter junkie.

Hello Laurence !

First of all as usual, let me thank you for the time you take to share your knowledge about ECM.

So Laurence, where content management experience has begun for you ? What was your first CM project and which solutions did you take?

Coincidently enough, I recently wrote about this very topic on my blog. My first system was a correspondence tracking system for the U.S. Air Force. I was brought onto the project as the database expert and was tossed the content management piece. It was DOCS Open from PC DOCS (later acquired by Hummingbird and then Open Text). We used a custom interface, but all of the content and search was powered by DOCS Open. cf. http://wordofpie.com/2009/09/08/my-first-content-management-application/

Nowadays, Is there an evolution of CM needs or is it the same as your first experience ?

The needs are the same, but the context and scale are changing. Users still need to track and manage their content, but now we are dealing with images and web content in addition to the scanned images and Word documents. Before, we were juggling between a network share and the new repository, now we are juggling SharePoint, multiple legacy repositories, and content stored in non-content focused business applications. I used to just worry about securing a document, but now I have to consider records management and eDiscovery when determining how to handle some content.

Sometimes I miss the old days.

Let's back to you for a moment, Can you tell us more about your position? What's your role and what are you doing day after day?

I run the Information Management Solution for Washington Consulting, Inc. I'm the leading subject matter expert for our company and I provide guidance to multiple projects and clients on how to address their needs. On good days, I'm sitting with clients and talking about how to solve their problems and designing solutions for them. Quite often, I am writing proposals and finding good people to work with on delivering solutions. I still keep my hands dirty by playing with the latest releases of software in the ECM space. I say it is to keep current, but I really do it because I find it relaxing and fun.

Could you present Washington Consulting ? What's the purpose and objectives? What kind of services do you offer?

We are a Management and Technology consulting firm located in Washington, DC. Information Management is a big piece of what we do, and that ranges from developing strategic roadmaps to "simple" Documentum and SharePoint work. However, we also have ERP implementations, Project Management Office (PMO) support, Organizational Change Management, and Business Process Reengineering (BPR) practices. They joy of my piece is that I get to work with all the different groups because everybody is trying to solve the content "problem".

You have worked with Documentum, since 2000. For my readers, who essentially love Open Source :o) , could you present this historical CM Solution (its history, its functionality and its architecture)?

No problem. Founded in 1991, Documentum has been one of the "big-three" ECM vendors for years, even as that membership has evolved. It is immensely scalable (I have a system with 40TB of content) and is free of proprietary languages or interfaces. It is very strong in core content management, RM, and BPM. It's collaboration efforts had been withering for a few years as eRoom grew old, but their new CenterStage product shows some promise. While they deliver on WCM and Digital Asset Management, those offerings are stronger when viewed in combination with the rest of their suite.

I think one area that shows great promise is their work with XML content. Their acquisition of X-Hive has really helped them here and I think that could become a big differentiators between the big ECM vendors, assuming that SharePoint doesn't wipe them all out.

Now if only they could simplify their license model.

Can you tell us what are the strengths and weaknesses of this solution from your point of view?

I think I covered that to some extent, but I think Documentum has two weaknesses, the user interface and the complexity. My most successful projects do not utilize the standard user interface. We use it as a content platform. This is also where it gets complex. I keep learning things about Documentum everyday. A Document expert is someone that knows enough to get through any client meeting and can then quickly look-up detailed answers to the things that they had to bluff their way through. I, and others in the community like Scott Roth and Johnny Gee, have forgotten more about Documentum than many with 3-4 years experience.

As for strengths, it is strong. I can throw users and content at it without fears that it will scale. I think that it's object-oriented approach to modeling content is flexible and powerful. They've extended this to allow custom behaviors on actions, and with version 6.x, they support aspects. Their is very little that the Documentum platform cannot do, as long as you understand the complexity.

Now let's start some philosophy :o), during this summer (and last years), you have post on your blog some information about your vision of ECM ? Could you resume your vision ? Why have you your personal ECM definition?

My vision for ECM is more than the tools. That is a problem with the AIIM definition, it talks about tools. Content is everywhere. We need it accessible from everywhere without emailing it to other companies or having to take it out of one system and placing it in another. ECM is a platform that provides content to users in the business context where they need it most. Users shouldn't know about Nuxeo, Alfresco, or Documentum. They should only know that if I scan this document, I can grab it and the presentation I made yesterday and share it with my business partner without worrying about the "how".

The technology isn't there yet, but we should be aiming for that target. When ECM was first coined 10 years ago, no vendor could state that they met the definition.

During this year, you also worked on CMIS. Why are you enthusiastic whith this future standard and Is it a revolution ?

I'm enthusiastic because the CMIS standard is the first step to making my vision for ECM complete. Applications need to access content without worrying about the specifics of the repository in which it is contained. It will allow legacy systems to stay in place longer and help separate the business application decision from the platform decision.

Is CMIS a revolution? That is a tough question. It is still early to tell as it is just approaching the Public Review stage. It is already starting to gain traction with vendors and customers, so I would say that it has the makings of a revolution. If Microsoft continues to support it, and incorporates it into SharePoint, then I see a revolution as being entirely possible.

What's your feelings about Open Source solution and open source move in CM ecosystem ?

I'm glad that Open Source is gaining ground. I think it will force the ECM "leaders" to focus more on simplicity and responding to users. I already see the vendors starting to be more responsive to user feedback than they had in the past. To be honest, I think the Open Source movement is a blessing for small to medium organizations that needed content management but couldn't afford the license fees. I am also beginning to strongly believe that the Open Source approach is the only way to deliver WCM systems that can keep up with the rapidly changing face of the Web.

For you what was the worst and best idea on content management area?

The worst idea is simple, moving everyone to a web interface in the 90s. Users had a tightly integrated desktop environment until IT management fell in love with using web interfaces for content management. Bad move. We are just now getting to the point where the integration is as tight with the desktop applications, and that seriously hurt user adoption across the board. The browsers and web applications just weren't advanced enough, but everyone fell in love with the sexy new kid on the block.

The best ideas are a toss-up. I think standards like ODMA and CMIS are some of the better ideas. XML has also been a great idea for changing the way content exists, but the adoption has been incredibly slow. The best idea may boil down to going electronic and forgetting about all of that microfiche.

Why have you started wordofpie.com ?

I started the blog back in 2007 because I wanted to rant about a recent Documentum conference. Luckily I started engaging in dialog with other bloggers and turned it into something far more constructive. Now I use it to expand on opinions initially shared on Twitter and to share news and other insights. I wish I had more time to write for it, but my day job and family keep me very busy.

Finally, can you recommend us weblinks or blogs about ECM or IT in general ?

For news and analysis, I keep close tabs on CMS Watch (www.cmswatch.com). I've met several of the analysts and they always give you their honest opinion. I enjoy reading the Big Men on Content (http://bigmenoncontent.com/) for their insightful view into events around the industry. They aren't as active since EMC hired them, but they always have some great reads. Billy Cripe's Fusion ECM blog (http://blogs.oracle.com/fusionecm/) and Cheryl McKinnon's Candy and Aspiring (http://candyandaspirin.blogspot.com/) are also favorites.

Oh, and I seem to recall enjoying your blog as well. :)

What would you say to conclude this interview?

Thanks for being patient with me in getting these responses and reading through them. If you've made it this far, it is fair to say that you have a passion for Content Management. Use that passion to help guide the industry towards your vision of ECM. You can sit by, watch things change, and point out issues, but don't stop there. Get involved. Offer solutions. Offer to help. It doesn't have to take much time. That is how I got involved with CMIS through AIIM. I saw a problem and realized that AIIM's iECM committee was chartered to find a solution. Go out and do the same thing so that our kids aren't solving the same content problems that we are solving today.

Many thanks, Laurence, for this interview. We wish you a nice and exciting journey on Open Source (or not) ECM Road! If you want to follow ideas and thinking of Laurence, you can follow this links :

Blog : http://wordofpie.com
Twitter : http://twitter.com/piewords
and his company : http://www.washingtonconsulting.com/

PS : you can download this interview at http://www.scribd.com/doc/20438655/Open-Source-ECM-Interview-ENG-Laurence-Hart-Word-Of-Pie-

Open Source ECM Interview : French to English

2009-08-27T11:53:00.000-07:00

Hello everybody !

As you know, I have two blogs : one in french (www.opensourceecm.fr) and another in english (www.open-source-ecm.com). I try to have same posts in both of them but due to time crisis it's really hard. Conclusion : I have some interesting post ONLY in French... I think it's really really sad for you my sympathic english readers... and futhermore I always made a link between my english to my french post but rarely in the other way... It's not acceptable!!

That's why, I decided to recap french interview here and, with my Google translator friend, I propose direct link to english translation.

08.09 : Julien Viet, Product Manager eXo SAS
06.09 : Benjamin Jean, legal advisor Linagora
06.09 : Nicolas De Loof, Architect Software Solution Capgemini
02.09 : Herve Quiroz, Lead Architect Componize
01.09 : Jean Marc Orliaguet, Contributor of Nuxeo Theme
11.08 : Jean Christophe Dichant, IBM collaborator & Blogger (bpmbulletin.com)
11.08 : Jean Philippe Clair, Commercial Leader Knowings
07.08 : Michael Harlaut, Solution Engineer Alfresco
07.08 : Christophe Bec, Senior Architect ECM Capgemini Sud
07.08 : Romain GUINOT (Atos Origin), Contributor of the month Alfresco
06.08 : Thomas CHOPPY, ECM Consultant Smile

Enjoy! :o)

Interview : Dr. Ulrich Kampffmeyer, Managing Director and Founder of PROJECT CONSULT

2009-08-13T04:20:00.000-07:00

„Guten Tag“ everybody !

Today we make a travel to Germany.

Today, I have the great honor to interview Dr. Ulrich Kampffmeyer, Managing Director and Founder of PROJECT CONSULT in Hamburg, Germany. He is as well a member of the Board of Managers of the DLM Network EEIG, the organization which administers the new records management standard MoReq2, published by the European Commission. He was also founder and chair oft he German association VOI and held different positions within the international association AIIM. In other words, he know the meaning of content and information.

Guten Tag Dr. Ulrich !

First of all, let me thank you for the time you take to make this interview reality and to share your knowledge about ECM.

So Ulrich, as usual I read information about my interviewee and I find a number : 30. You have been in the document, content and knowledge management industry now for over 30 years... For a young happy-go-lucky ecm consultant like me it's "Woooowww". Could you present yourself and main facts of your experience ?

Well JM, it is not „in the industry“ because we both know that our industry, described as document management or enterprise content management, is not that old. Since 1976 I have been working with software in the field of analysis and documentation. My first computer projects were in the archaeology and soil sciences domain. I developed software for cartography, automatic classification of objects, statistics, video acquisition, documentation, classification and reconstruction system for archaeological findings. So my first document management projects were for museums and research institutes.

Later on, I was working for scientific research institutions, document management vendors, universities and consulting companies. This led me to combine theoretical academic, programming and system design skills with project management, business process analysis, moderation and solution design advising. July 1st, 1992, I founded my own consulting company, which is working in Enterprise Content, Records, and Knowledge Management field for end user organizations only. Beside that I was active in associations like AIIM, VOI, DLM Forum and others, and wrote some books, studies, lots of articles and other publications like our PROJECT CONSULT Newsletter.

And now what's your work day after day ?

Part of the day – when I am not under a tone of our customers sites – is filled with administrative work. I have to perform as Managing Director. In regard to projects I have to control quality of final versions of documents where I have the role of quality manager in the project. A lot of time is taken by email correspondence and phoning with prospects, customers, staff, freelance consultants, journalists and so. I try to reserve blocks of hours for writing reports for my customers, research or just contemplation. Some hours a day, when I'm in the office, I go to the internet – communities like XING, Informationzen.org, facebook, bloglines, publishing sites, industry portals a.s.o. When I'm travelling for seminare, customer workshops, or congresses, most of this work is delegated to the traveling and waiting hours. But most important of my time is spent with my customers in workshops, discussions, or meetings. It's not only to give advice but also to learn each day as well.

You have found project-consult, in 1992. Could you present your firm ? What kinf of services do you offers ?

PROJECT CONSULT Unternehmensberatung GmbH works in the field of archive, document, content, records, business process, information lifecycle, and knowledge management only. With my colleagues at PROJECT CONSULT we have been running more than 200 projects designing information management strategies, carrying out analysis and design of organisation and processes, writing tenders, selecting products, solutions and system integrators, running projects as project managers or as quality control managers, conducting workshops, writing policies and legal expertises and so on...

We are small company combining hired staff with speciliazed free lance consulters. Our customers mainly come from Germany and Switzerland, but we have as well worked for supra-international organizations like UNEP or ESA, international companies in Europe and other clients from abroad. Clients come from all industries and our main task is to help them to solve organisational problems related to documents.

Beside projects work, we were engaged in the creation of standards like GEVER in Switzerland, the electronic archiving standards for the Federal Savings Banks Group, and the ECM standards for the Social Insurance Group in Germany, we engaged in MoReq2, ISO 15489 and ISO 19005, and more.

Let's back to future, I know it will be quite long but is it possible to summarize the computerization and dematerialization of information since you started your experience ?

In the last 30 years I faced a lot of changes from host computers with character-only display, the first PCs, early networks, evolving internet and now all this 2.0 stuff. I looked on all the short-sighted developments of the hardware and software industry in the early years.

Imaging was born because host and early PCs could not display and handle images.Document management was born due to the restrictions of hierarchical file manager.Workflow was born to support collaboration and to conncect the users.

Our industry in the past lived by the errors and niches created by the IT mainstream vendors. This time is gone and our industry is loosing its USPs, Unique Selling Propositions. Document and enterprise content management is becoming IT infrastructure, dissolving at the middleware layer, becoming services, being no longer visible at the user desktop. Modern clients and web interfaces are enabled with the capabilities to handle documents, processes and other information assets.

We are living in a digital world and I sometimes suffer as digital native. Especially the last ten years have changed the perception of how we handle and how we use information. With 2.0 it is beyond everything our industry had developed as assets in the past. Document related technologies become integral part of every software application.

For you what was the worst and best idea on content management area?

The worst idea in the CMS arena was to re-develop everything which had already been developed in the document management arena. The web guys often re-invented the wheel and created solutions, which didn't take care about compliance, documentation, consistency, authenticity, persistance and other issues. The web generated more problems for document management, records management and electronic archival than we had before and these problems are not solved today.

The best idea of content management was lately skipping the barriers between structured and unstructured information. The key is, to look at digital information as an integrative concept – data in an database is just information, content in an WCM is just another form of digital information, and records and documents are as well only different by format, legal or business value, it is all digital information, that may become - combined with logic, context and meaning - even knowledge. Modern information and content management solutions must and can handle any type and format of information. This is the important step to overcome the differences in systems and concepts of the past, which will lead to the development, that ECM and WCM will disappear inside the IT infrastructure. Their functionality will become part of standard software and operating systems, and will be a commodity in the near future

In functional domain is there the same or have you noticed an evolution? Is it possible to illustrate it with your own project experience?

Yes ! There have been some interesting developments in the past years. On the one hand vendors add more and more functionality to their products or their products suite. End users often get overwhelmed with un-needed functionality. The big shots are offering suites which include almost every function somebody wants. Smaller vendors have been moving towards integrated solutions which special applications for certain industries.

There's some kind of special markets like the capture industry, compliance software offerings, email-management or archiving solutions for certain standard ERP or CRM products, and others, which are common to most vendors. And there's another trend in functionality to cover the gap in big software vendors product offerings like Microsoft with Exchange and Sharepoint, IBM, SAP or others. Business process management developed to an industry of its own like output management always has been – both are major components of the overall ECM concept. Although Enterprise Content Management has something like an universal approach, a lot of vendors focus on „filling-the-gap“ of Microsoft. If we look on all of these developments – standalone products, complete suites, “Gap”-solutions and new functionality entering from the Web 2.0 world, a lot of mid size vendors are in the danger of overstretching their resources. And even the big players do not manage to get their product offerings consolidated.

Today the most important applications in Europe are
• electronic virtual folders,
• “in-bound” capture systems for unified acquisition and distribution of information,
• automated classification and categorization,
• electronic long term archival (call it digital preservation in the archivists realm),
• collaboration,
• Web-2.0-functionality inside traditional document management solutions,
• standardised records management for paper and electronic records combined,
• enhanced business process management including business intelligence,
• compliance,
• ediscovery and auditings solutions,
• portals with integrated document management and similar advanced ECM solutions.

Now let's talk about ECM, do you have your personal ECM definition?

Yes and no, I have a definition for ECM. I'm supporting the definition of AIIM international now for nearly a decade. It changed a little bit over time, but the message and the core ideas are still the same. You will find some books I wrote (for example the synchronized-three-lingual publication ECM Enterprise Content Management), lots of articles (even the entries in the German and English Wikipedia) lots of seminars and lots of presentation about this definition.

It is important that the industry sticks to one definition. A lot of vendors, analysts and consultants came up with definitions of their own – no better, no worse. But to help the market to identify itself with this complex software and technology environment, to position ECM besides other IT topics like ERP, CRM, Enterprise 2.0, and to help the end users finding the right solutions, to make them familiar with ECM, it is important, that we all tell the same story, have the same definition, and here it goes: „Enterprise Content Management (ECM) is the strategies, methods and tools used to capture, manage, store, preserve, and deliver content and documents related to organizational processes. ECM tools and strategies allow the management of an organization's unstructured information, wherever that information exists.“

I had the opportunity to read one of your report on ECM and I learned many good things about content management. Does your work participate to the elaboration of ECM Definition ? Do you work for a standardization of ECM with AIIM ?

I was fortunate to be one of the members of the AIIM Europe and the AIIM international boards of directors when AIIM developed ECM as new strategy for the future. So I had the chance to discuss issues of re-positioning the industry in 1999 (after the first internet wave) with AIIM officials and other industry peers. My first presentations about the new vision of the market where already in 2000 in London and Berlin. The roadmap was always developed by AIIM but it took the liberty to add some features and create my own image of the ECM message.

Today there are a lot of discussions if we should stick to ECM or if the community should move on the EIM Enterprise Information Management or Enterprise 2.0. The image of ECM is blurring at the edges as the big vendors like Oracle or Microsoft introduced their own interpretation of ECM (based on what their products do). So we might stick for a while to the current idea of ECM, but this will change within this decade and ECM will become history like microforms – still surving in niches, but no longer mainstream.

Do you make other contribution (Animation, articles, posts, forums ...) ?

Thank you for this question – the answer is YES! I wrote some books and studies – have a look on Amazon.de (German Amazon!). I was editor for a whitepaper series by AIIM with seven booklets. I have been member of the editorial board of MoReq2 standard. I am editor of the PROJECT CONSULT Newsletter, which covers the ECM industry in middle Europe since 1999. The newsletter is as well available as fill text database and on several social publishing platforms. There are lots of white papers I wrote for vendors like IBM, EMC, SER, HP and more.

And there lots of presentations including the slides, videos, multimedia and handouts. Most of the material can be accessed using our website http://www.PROJECT-CONSULT.com (in German), but I published as well quite a lot on www.scribd.com, www.slideshare.net, www.doxtop.com and other platforms (look for „DrUKff“ …).

I am running a discussion group about information & document management in four languages and with around 6.000 members on www.XING.com. Sorry, most of the material is only available in German language, but I have as well some whitepapers and presentations in English and French.

Do you have an explanation of ECM needs ? Is it normal the way of consolidation in content management (Document management, Web Content Management, Records Management, Information management...) into one system ?

It's not only about the integration of different components into one suite or system. The real challenge is, to make ECM known with the decision makers in the companies, who should by ECM solutions. I believe that quite a lot of components may be used as well stand alone in the future because ECM, as a whole, might be too complex for the users within a “normal” business organisation. And I already mentioned that ECM will become infrastructure/services. It will be important to avoid islands of information and to create universal federated repositories which can be used by every application, not only by ECM applications. To get a grip on information, to control information, to assure quality and correctness, to implement transparency, to support end-to-end processes, to bridge the existing applications and to help the organisation to be compliant – these are true ECM needs and ECM assets.

What's your feelings about Open Source solution and open source move in CM ecosystem ?

OpenSource has been playing a very important role in the WCM (Web Content Management) market from the beginning. It took some time before the first ECM open source products appeared. People find the interesting especially because there is a mix up of the terms freeware and opensource. Open source can be free ware, but it must. The market thinks of Open Source ECM products being cheap, being publicly available, being easy to adopt. This is not always true and I remember projects where the open source implementation was more expensive than a standard product solution!

In middle Europe most people think of Alfresco when considering an open source product. Other solutions like Knowledgetree etc. are not much known. I think open source will gain a little more importance in the ECM market in the near future, but the real thread for the traditional industry is SaaS, software as a service. We see the first ECM SaaS solutions on the market and more and more applications are moving into the cloud. Google is adding pressure and Microsoft is starting as well. So records management, long term digital preservation, document management, workflow, colloboration and other tools will be offered as well via the web. This will change the market in the next ten years because ECM installations at the customer site are known to be complex : you have to care for migration on your own, you hardly can afford a redundant backup solution, a.s.o. The same way project management and CRM customer relationship management is moving to web today, we will see all the ECM components and applications move into the cloud tomorrow.

In 2009, can you advise us on public meeting/ event or conference we must not miss ? Will you be present ?

The most important fair with a congress on site is the DMS EXPO in Cologne, Germany, September 15th to 17th, 2009. Nearly all vendors (IBM; Open Text, Oracle, SAP, Microsoft, Alfresco, Optimal Systems, Iron Mountain, ELO, Docuware, Ceyoniq, ITESOFT, COI, windream, Readsoft, Autonomy, Docuportal, Coremedia ...and integrators are present. ). There will be about 400 exhibitors and the KoelnMesse expects about 20.000 visitors. In the exhibition hall and as well in seminar rooms there are conferences and free presentations. The entrance and the catalogue is free of charge if you register in advance : www.DMSEXPO.com

Finally, can you recommend us weblinks or blogs about ECM or IT in general ?

First of all – internationally – there is www.AIIM.org. I am as well a member of the AIIM community www.informationzen.org. Most of my activities ist he the group on XING https://www.xing.com/net/informationlifecyclemanagement both communities are free.

There are a lot of blogs about ECM and related topics, so for consolidation and overview I use Bloglines and Delicious (you can subscribe to http://www.bloglines.com/public/DrUKff and http://delicious.com/DruKff).

If you are looking for sources in Germany and in German language, check on www.ECMGUIDE.de, www.ContentManager.de, or www.Documanager.de.

If you are looking for new records management standards and archiving practices, have a look at the DLM Forum www.dlm-network.org.

Groups on publishing sites with ECM related content are http://www.doxtop.com/groups/view/feb95ada/ecm-enterprise-content-management.aspx or http://www.slideshare.net/group/ecm.

And not to forget my own sites www.PROJECT-CONSULT.com, http://pcnewsletter.coextant.info/ and www.MoReq2.de.

What would you say to conclude this interview?

First of all – thank you for giving me the chance to participate in your series of interviews. To conclude this interview – the growth of information, the speed of development in the ITC industry and restrictions of human nature and brain capacity create together one of the most interesting mass experiments in mankinds existance. We moved in one generation from an analogue world of information management to a digital world of information management with new rules, new behaviours, new visions. I am glad living in this interesting era and being able to contribute a little bit to solve the big challenge to save the memory of the information society - in this early, still dark age.

Many thanks, Ulrich, for this interview. We wish you a nice and exciting journey on Open Source (or not) ECM Road!

To find out more about Project Consult : http://www.PROJECT-CONSULT.com
and if you want more information, you can follow this links :

If you want to download this interview, follow this link : http://www.scribd.com/doc/18551640/Open-Source-ECM-Interview-FR-Dr-Ulrich-Kampffmeyer-ProjectConsult-

CMIS SQL : Tutorial & Presentation

2009-08-12T09:29:00.000-07:00

Hello everyone !

Since CMIS was created, I'm interessted in CMIS-SQL. That's why I decided to search some documentation on it. Except CMIS Specification, I found very few documentation. I just want some CMIS SQL query sample...

Yes I know, it's normal there's no document.. It's a young idea! Nevertheless I decided to test it and to share my discovery with you.

So below, you can find my presentation about CMIS-SQL and how to test it with Alfresco ! As usual you can view, download, distribute and comment the presentation!

CMIS - CMIS SQL Search Tutorial - ENG

To be continued...

Nuxeo Tips : Installation and configuration

2009-07-29T05:18:00.000-07:00

Hello Readers and Followers.

Today I would like to share some information about how to install and configure Nuxeo 5.2.
Indeed, recently I help a customer to install, configure and customize this OpenSource ECM System in its Windows environment. So let's share :o)

Nuxeo Installation

How to install Nuxeo ?
In general manner, you should download and use Nuxeo 5.2 Windows Installer. You can find it at this url : http://www.nuxeo.com/en/downloads/

This installer is a good starting point to begin a Nuxeo Project. Moreover it also incorporates Nuxeo Shell (a beautiful gift for ECM Administrators). However it embed a java light database (Derby). This database is for demonstration , developments or experiments environments. It's NOT a production database! That's why we have to change Nuxeo settings.

How to configure Nuxeo and PostgreSQL ?

For step by step explanation, have a look at the :

FAQ : http://www.nuxeo.org/xwiki/bin/view/FAQ/

How to configure Nuxeo for a production environment?

For Nuxeo, the information is here : http://www.nuxeo.org/xwiki/bin/view/Main/NuxeoProdSetup/
For PostgreSQL, the information is there: http://www.nuxeo.org/xwiki/bin/view/FAQ/PostgreSQLSettings/

If you follow all advices and recommendations, you have a good foundation to start a production server. So it's time to consider the functionallity side, ie different features that we want appear or not.

Configure Nuxeo

Short Introduction

I usually say, in Open Source ECM context, there are 3 possible ways.

User approach : User only know how to click on buttons (This is not a criticism or an insult but it huge caricature...)
Administrator approach : Administrator know how to use and modify humanly comprehensible files (XML, Properties, txt)
Developer / Integrator approach : Developer is omniscient and omnipotent. He understands and knows everything in programming, configuration and utilization.

In our case we will look only at the administrator level. Indeed many features are configurable through XML files in Nuxeo.

To illustrate this point, I propose the following use case: I want to increase the number of documents displayed in a space/folder/workspace. By default this value is set at 10. If the space contains more than 10 documents, then the user interface will create an automatic 10 to 10 pagination. For ergonomic purpose, I wan to set this value at 25.

Let's explore what are the possible approaches to meet this need

First of all, Nuxeo has 3 IMPORTANT directories :

NuxeoEP5 \ NuxeoServer \ server \ default \ deploy \ nuxeo.ear \ System
This directory contains Nuxeo heart.
NuxeoEP5 \ NuxeoServer \ server \ default \ deploy \ nuxeo.ear \ plugins
This directory contains all the modules/ addons / plugins / extensions that are not part of the heart Nuxeo. In general this addons are your addons.
NuxeoEP5 \ NuxeoServer \ server \ default \ deploy \ nuxeo.ear \ config
This directory contains all Nuxeo configuration files.

These 3 directories will illustrate 3 ways to configure Nuxeo.

System directory
If you already explore Nuxeo files and folder, you've noticed that most files inside System folder are Jar. For those who are not familiar with this file format, take a look at wikipedia : http://en.wikipedia.org/wiki/JAR . To generalize, it is a ZIP file wich contains java-like files.
So if you have a zip utility like winZip or 7-zip... you can unzip a jar file! So... you can open it and therefore ... you can change the contents of the ZIP!

To complete our task, we have to focus on : Nuxeo-platform-webapp-core-5.2.0.jar
It is located in : NuxeoEP5 \ NuxeoServer \ server \ default \ deploy \ nuxeo.ear \
Open it (with 7-zip in my case) and locate the file OSGI-INF/querymodel-contrib.xml. Edit (or just extract) the file and replace the value 10 by 25 in the next block:

Next, save the file (or import it into the Jar).

You can now restart Nuxeo, importing more than 10 documents in a space and see that there is no pagination creation. We have to import 26 document to see this component.

Now to the second method:

Config directory
As you can see, this directory is composed primarily by configuration XML files. The main purpose of this directory is to configure the global parameters of Nuxeo application. So if we think 5 seconds, it's what we want : we want to set the number of items displayed in a space.

Before all, if you have tested the previous method, you have to remove your change. Then create a configuration-ui-config.xml (the name does not change anything except -config.xml part which should remain unchanged) in the config directory and add the block below:

The most important part is the tag REQUIRE. In this tag we will set the default component name value (which we saw in the previous method). This allows us to "overwrite" or "replace" the default component.

You can now restart Nuxeo, import more than 10 documents in a space and see that there is no default pagination creation.

You just learn a method that works for most of Nuxeo default component.

If we have to create a procedure (a recipe), it would be:

Open and explore jar files in system directory (with your zip utility: 7zip for me ...).
Search outXML files (especially files ending with-contrib.xml) in OSGI-INF directories.
Copy the block you want and remember the name of the component (in general it's the 2nd line of the -contrib.xml file).
Create your own config.xml file in the config directory .
Insert the block and change default parameters to match our requirements.
Remember : you have to respect XML validity and declare it as a XML file.
Restart your instance.
You can serve, it is ready! :o) : o)

Finally the last method
Plugins directory
This method (recommanded by Nuxeo) is more scalable, modular, but also the most complicated! (Everything is not allowed particulary in computer science...). I will not going deeper in details with this approach because it is well documented. Remember, if you're interested in the topic, I advise you to read the Nuxeo book draft at: http://www.nuxeo.org/static/dev-tutorial/

Everybody doesn't have time and skills to install and configure Eclipse, JDK, Maven etc ... It's one criticism of this method. It can be a complex task which can sometimes take quite some time ... and therefore money.

On the one hand : Is it advisable to install a development environment for our need ? The answer is NO of course. On the other hand : I want to add and modify many default Nuxeo components and functionality. Do I have to install Eclipse ? Answer is YES ! This environment is necessary.

As often: Everythin depends on your requirements and needs.

Conclusion
To simplify, this is my recommendations (to save time!)

Banish editing Jar files in system directory !! Sometimes this seems to be the easiest solution but it is the least profitable in the long term. And in the world of opensource change are everywhere (migration, upgrade ...)! The only possible case concerns feature removal. If you have identified a feature that does not interest you, it may be useful to remove associated jar (simplify your installation, gain of time when application start...)

If you want to change a default Nuxeo component parameters, choose the method based on the configuration directory. It will be in long term the best method.

If you need to change a component Nuxeo + add files (. Java, images, defining a workflow) or add a component, the preferred method is Plugins.

Finally, if I have an advice : Be curious, navigate and explore the depths of Nuxeo jars!

Have a good journey and see U

News of the week via twitter

2009-06-12T05:34:00.001-07:00

Hello everybody !

It's time to share some news finding on twitter. Did i mention Twitter is THE genial tools to find informations about ECM ? Yes ...? Ok! Let's begin.

eXo Portal & JBoss Portal : Happy Union !
First important news of this week is the union of eXo Portal and JBoss Portal project.

Business Wire : eXo Platform Merges Portal Software Development into JBoss Community
Reuters : eXo PlatformMerges Portal Software Development into JBoss Community
J.Boye : eXo Portal and JBoss Portal join forces
eXoPlatform Blog : Julien Viet Post
Red Hat : JBoss.org Community Grows
JBoss Portal Blog : Welcome eXo !
JBoss FAQ : http://www.jboss.org/community/wiki/JBossPortal-eXoPortalFAQ
CMSWatch : http://www.cmswatch.com/Trends/1615-eXo-JBoss-Merger

Twitter Source :

Tugdual Grall : http://twitter.com/tgrall
Julien Viet : http://twitter.com/julienviet
Benjamin Mestrallet : http://twitter.com/benjmestrallet

New version of KnowledgeTree
KnowledgeTree Releases New Version of Its Commercial Open Source Solution as well as a Microsoft® Office Add-In.

This release focuses on helping companies ease the burden of compliance, provides improved usability and ensures increased interoperability between KnowledgeTree and other productivity tools. Significant new features and enhancements include:

· Document Type Alerts make it possible to manage document retention policies, contracts and other time-based events with less effort by applying alerts to groups of documents.

· Configurable Electronic Signatures ensure accountability and compliance with regulations such as the US Food and Drug Administration’s (FDA) Title 21 Code of Federal Regulations (CFR) Part 11, Health Insurance Portability and Accountability Act (HIPAA) and others by requiring a user to re-authenticate when changing a document.

“KnowledgeTree can significantly ease the compliance burden for smaller companies,” said Daniel Chalef, CEO of KnowledgeTree. “The combination of electronic signatures with a complete audit trail fulfills regulatory requirements, while the affordable price appeals to decision-makers.”

Also featured prominently in this release is the Microsoft Office Add-in, which integrates KnowledgeTree seamlessly with the entire Office suite. Microsoft users can open, edit, save and email documents that reside in the KnowledgeTree repository from within Office applications, exactly as they would if that document was located on a shared drive. In addition, a powerful built-in search makes it even easier to find documents in the repository from within Office applications.

“The Microsoft Office Add-in allows users to enjoy the benefits of collaboration and version control afforded by KnowledgeTree without changing the way they work,” noted Philip Arkcoll, product manager at KnowledgeTree. “Customers using this feature report enhanced productivity as a result of the streamlined process.”

For more informations visit KT website : http://www.knowledgetree.com/

Other Links

CMSWire : KnowledgeTree Updates Open Source ECM, Adds Key Enterprise Features
Read Write Web : KnowledgeTree Makes Upgrades to its Open Source Document Manager

Twitter source :

KnowledgeTree : http://twitter.com/knowledgetreesw
Daniel Chalef : http://twitter.com/danielchalef

PS : If you are curious and want to know who are behind technology and sales in KnowledgeTree, let's take a look at this 2 posts :

New Alfresco Book
If you search a book to learn how to customize, use, and administer Alfresco, I think the last new book : "Alfresco 3 Enterprise Content Management Implementation" is the best one !

For further information visit Packtpub : http://www.packtpub.com/alfresco-3-enterprise-content-management-implementation/book

Twitter source :

DRB : http://twitter.com/DRB

Bunch of links

Twitter List : Capgemini & Sogeti

2009-06-12T01:48:00.000-07:00

After my ECM Twitter list, I'm going to create a Capgemini (official account) Twitter List :

Capgemini

Capgemini : http://twitter.com/Capgemini
Capgemini North America (Us/Canada) : http://twitter.com/CapgeminiNA
Capgemini Cloud, Global Center of Excellence : http://twitter.com/Capgemini_cloud
Capgemini Outsourcing : http://twitter.com/CapgeminiOS
Capgemini Espagna : http://twitter.com/capgemini_es
Capgemini Espagna : http://twitter.com/capgemini_rrhh
Capgemini OutSourcing UK : http://twitter.com/osukcto
Capgemini Norway : http://twitter.com/CapgeminiNorge
Capgemini FR ADC Toulouse : http://twitter.com/ADC_Toulouse
Capgemini UK Public Relation : http://twitter.com/CapgeminiUKPR
Capgemini Netherlands Global Services : http://twitter.com/CapgeminiGSCAPS
Capgemini RMG team : http://twitter.com/capgeminijobs

Capgemini Group
If you work @ capgemini, don't hesitate to join us at :

Sogeti

Sogeti France : http://twitter.com/sogeti_fr
Sogeti NetherLands : http://twitter.com/sogeti_nl
Sogeti NetherLands High Tech : http://twitter.com/sogeti_nl_ht
Sogeti USA : http://twitter.com/sogeti_usa
Sogeti Seattle Sogeti : http://twitter.com/SeattleSogeti
Sogeti Espagna : http://twitter.com/sogeti_es
Sogeti SAP Ireland : http://twitter.com/SAP_IN_IRELAND

Of course this listing is not thorough and extra hands are always welcome to complete this list!

Finally, if you want to follow me ?

http://twitter.com/jeanmariepascal

Happy One Year !

2009-06-10T04:46:00.001-07:00

Hello Readers !

Today it's a special day.
Today I celebrate the first year of this blog !

Happy One Year !

Licence Creative Commons (Attribution-Noncommercial 2.0 Generic) :
http://www.flickr.com/photos/smcgee/264384675/

For this event, I want to share some stats about this blog :

74 posts
Number of visits : > 9800
Number of unique visitors : > 6900
Number of viewed pages : >15000
Visits duration : 2 min (approx)
Country Stats and Repartition (USA, France India, Spain UK...)

For all of this result, I have just 2 words

Licence Creative Commons
http://www.flickr.com/photos/vernhart/1574355646/

Thank you for your visit !
Thank you for your share !
Thank you for everything !

And of course,
don't hesitate to come on the road to an Open Source ECM World !

Evolution of Content Management : Thoughts and Ideas

2009-05-19T01:11:00.000-07:00

Today I would like to share one of my recent philosophical... technical... functionnal... reflections in content management world. Particularly I want to speak about content creation ...

Today's theme (if you wish) is focus on the evolution of content creation best practices ... To be more explicit, I'll use a comparison with the world of software development (yes ... I will talk about code! ^^)

So if I had to ask a question:
to draw one's inspiration from.

How content production can draw inspiration from software code's production ?
Does content production should take the same path and adopt the same principles?

First, an observation
In non-structured information world (ie in the functionnal world generally...)

Users enjoy the use of desktop tools (MS Office, Open Office ...) to create the vast majority of content (Report, documentation, manual, form ...) inside the company.
They store information locally on their computer (hard drive).
They share content via network shares (shared hard drives in a network) or via an e-mail attachment.
Sometimes content will be grouped, merged with other content to form for example a multilingual advertising campaign, a response to prospect, a manual for mounting a furniture, a booklet, a catalog ...
One of the best practices in content creation is to create a template to re-use it. In general we want to re-use the style of the document (PowerPoint template), or content (word document template ...)
To share these models, we publish the file in a shared directory or website

In technical information world (eg software development)

Developers enjoy the use of development tools (Eclipse, Visual Studio ...) to create the vast majority of code file (.java, .c, .html ....).
They store files locally on their computer (hard drive)
They share code via a management system version (SVN, Hg, CVS ...).
With all code, developers create a project which will be compiled using automated environments (Maven, Continuous Integration...) to produce a program (software solution)
When code can be reused, developers include this code in a library(.NET, Java, Spring ...) This library can then be used in new code.
To share libraries, they publish it in a system directory or a website.

Now it's game time ! Let's spot the difference ! ... ie try to find the differences between the two worlds. You have 5 minutes!

Ok... let's forge ahead ...

My vision

I think best practices and principles acquired through development (or other) will be increasingly used in content creation.

It's time to explain ...

As you can see, there are differences in how to create / share / edit content between functional and technical domain.

In software development world, we assist in last few decades :

Local code
Shared code
Shared code in a version management system
Shared code in a standardized code management system

In content creation world, we assist :

Local file
Shared file
File in a content management system
File in a component management system

In other words, content management will know (and already knows... I have to be precise ! ) the same changes than development with programming had with object-oriented approach and continuous integration !

We try to apply to content the well-known principles :

Write Once, Deploy Everywhere
Stop Reinventing the Wheel

This new idea tries to create and use tools like :

A management tool for configuring multi-dimensional content (how to manage versioned content composed by other versioned content which are composed by other versioned content)
A generation platform
An editor (integrated in office tools or not)
A coherent system to rule them all !

What are benefits of this approach?

Take the example of "response to a call for tender" (Sorry if it's the wrong term...)

For those who are not familiar here is the principle.

A call for tender is a procedure by which a potential buyer requests different suppliers to make a commercial proposition to the detailed formulation (specification) of buyer's need (product or service). Of course all communication between the parties is usually done with documents in office format (printed or not). Response to a call for tenders is the technical proposal and commercial office format.

Let's explore the difference between a company called A which owns a component management tool and a company called B which have no tool.

Company A has made an habit of creating components. That's why it has a catalog of

Client references
Architecture and methodology
Proposals
Resources
Stylesheets

User, who are reponsible for the response, will simply assemble major components (up to 80% of the document) in a few clicks and work on the last 20% (the heart of the response...)

Company B does not have such a system. User must seek in his old one example the one that will serve as start model. It should then do a lot of copying and pasting, checking the different information, may update or ask to update ...
And usually, this person finds few hours before the deadline he had forgotten to take the right stylesheet or the last reference ...

To conclude, according to this quick example, the gains are many:

Gain of time
Gain of money (because time is money!)
Gain of modularity
Gain of efficiency ...

What's this system?

I do not know if there is still a true acronym, but the term CCMS for Component Content Management System (CCMS) is the one who is closest!
So don't hesitate to inquire about it! (And me too at the same time ....)

And future term may be ECCMS (Enterprise Component Content Management System) ...

Special Thanks to Componize.com (and staff cf. [FR] Interview of Herve Quiroz ) which help me to understand Component Content Management and don't hesitate to try their CCM solution !

Note
The only problem I find with CCM adoption would still the graphical interfaces and user experience. The man (and woman) still remain the key and most important variable...

Thanks to everyone and don't hesitate to post a comment !

Alfresco Meetup Paris : Review and Feedbacks

2009-05-06T02:11:00.000-07:00

Hello everyone!

On Thursday 30.04 I was in Paris at the Alfresco Meetup.

First the planning : the day was divided into 2 parts.
Part 1 : Alfresco "Crew" (Morning)
Part 2 : Alfresco Partners (afternoon).

Alfresco Morning
The first presentation session was devoted to ...........? Alfresco of course!

Submitted by Denis Dorval Vice President EMEA Alfresco, the session was about Alfresco, ECM and the ECM market in general.

I noted some interesting key points like :

A person who uses Alfresco is called: "Alfrescan(s)" (I didn't know I had this name too...)
Alfresco community represents 90 000 people worldwide.
France is the 2nd community after the USA.
Alfresco won 270 customers, 90 partners and 18 OEM 2008.
2009 is the year of open source acceptation.
2009 is a consolidation year for the software market.
More than a dozen clients manage more than 10 million documents with Alfresco.
Adobe (with acrobat.com) manages the largest number of documents with Alfresco: over 120 million documents.
Alfresco want to test its application to cover one billion of documents.

If you want more informations : Support Alfresco

Then I attended a case study entitled: "EADS (Astrium & EPS) paperless HR solution or UBIC project (Atol CD / Astrium)".
During this presentation, we discovered how it's possible to dematerialize all HR processes of a large company such Astrium (and EADS in the future).

Interesting points were :

2008 : Project Initiation
600,000 : Number of documents managed in a first step,
10 million : Number of documents managed in a near future.
Alfresco 2.2E, GWT, Red Hat Linux, Oracle 10g, JBPM, WebDav, Webscripts: Main components of architecture and implementation.

For more information, you can see a screencast of the application at this url : http://www.atolcd.com/ubic-demo-flash.html

At the end, UBIC will become a plug-in that can be added in Alfresco for HR document management. This project will also be supported by Astrium and Atol CD.

To learn more about UBIC project: http://www.atolcd.com/btn-ll/alfresco-ecm-ged/personnalisation-avancee.html

To conclude morning sessions, Mike Farman, Director of Product Management at Alfresco presented Alfresco Roadmap, Vision and Strategy.
Important points were :

Record Management: DOD5015.2 Certification in September 2009
Managing and archiving e-mails. (MUST-SEE!)
IMAP Implementation and creation of Virtualized Repository (MUST-SEE!)
Alfresco SURF means .... NOTHING! (always good to know ...)
Admin tools for Alfresco Enterprise subscribers (Enterprise Only Capabilities: clustering, monitoring ...)
Share Service Forms
Web Delivery Runtime
Alfresco Query Language (based on CMIS + Alfresco Extensions)
Index Consistency Check
Alfresco Mobile (Example:iPhone integration)

If I had to summarize, Alfresco continues to spread its perimeter on corporate documents life cycle. Next targets are e-mails and legal archiving. These 2 aspects are strongly expected by functional people. To support inevitable and unavoidable document spread, Enterprise customers will benefit tools for managing Alfresco clusters !
Did you say Alfresco continue its evolution ?

Partners Afternoon
The afternoon was devoted to technical presentations from Alfresco partners. Overview :

NB: "Alfresco Partner of the Year" to Sopra Group: http://www.sopragroup.fr/

Componize and XML document management solution.
As a neighbor(Marseille, France), I finally discovered Componize solution! It is based on components principle. Indeed when we create office or technical documents by default we create documents by components without realizing it.
A paragraph is a component of a chapter.
A chapter is a component of a book.
A book is a component of a library
etc ...

And often, we realize these components are the same in a lot of documents!
and what happend when we decide to modify one ? We have to go back over ALL the documents having this component! Lost time guarantee!
To optimize document management, we have to use a tool like Componize!

For more information: http://www.componize.com/overview/overview-home.html

Xenit and Castor integration .
Firstly if you do not know what is CAStor Caringo, let's have a look at : http://www.caringo.com/products_castor.html
Now try to make links with document management system like Alfresco. Next, use an addon developped by Xenit...
You have a solution that can easily manage millions of documents with high-scalability and high performance guaranteed!
Impressive!

For more information on Xenit: http://www.xenit.eu/

Openwide and data migration with ETL (Talend)
Generally, when we start an ECM project, there are always existing documents. So, we have to find a way to transfer this documentation to Alfresco.
How? With an ETL (Talend in our case) and a component developed by Openwide. This one allows you to create and integrate content within Alfresco through "jobs" (automated).

For more information on Openwide: http://www.openwide.fr/

Atol CD and SURF applications :
JSF is dead, long live REST! It's the main idea or slogan of this presentation which explains how to create a new user experience with Alfresco.

For more information about this presentation : http://www.atolcd.com/fileadmin/Publications/Alfresco_Meetup_RIA_et_SURF.pdf
and Atol CD: http://www.atolcd.com/

Workflow conception with BlueXML
How is it possible to generate 1200 lines of code to create and setup a workflow in Alfresco in a few minutes?
The answer is simple: You have to user BlueXMl Alfresco Generator !
This tool, based on Eclipse, allows via drag & drop (and forms) to create all necessary files for setting up a workflow Alfresco.
The dream of every developer!

For more information on BlueXML: http://www.bluexml.com/v2/alfresco-generator/

Finally

First as its logo suggests, Alfresco can be compared to a flower that has evolved in its ecosystem (content management).
In this ecosystem, it met

Bees (OEM) who have created their miels (solution) from the pollen of the flower (Alfresco. ..)
Florists (OEMs) who used the flower (Alfresco. ..) to create compositions (specific solutions)
Horticulturalists (Partners / Integrator / Community) who took the flower to make it evolve (Alfresco. ..) and make it even stronger.
Enthusiasts (Communitu) who think this flower is beautiful.

Then, if you want to live happy, you have to live hidden. This idea is THE fundamental trend of Alfresco, integrators and partners.
They are abandoning more and more history JSF interface to more responsive and more 2.0 interfaces. Goodbye Java, Hello Java... script! (I am exaggerating but the idea is here ...)

Finally Alfresco continues to integrate its application seamlessly with tools that the user knows and uses every day!
Yesterday CIFS for sharing documents in a Windows folder. Today Windows Office suite via Sharepoint Protocol integration. Tomorrow IMAP for sharing and managing emails in Microsoft Outlook ...

And me ?

For my first participation in an event Alfresco (my second for an ECM community event), I think I have to improve "networking" between sessions :o)
It is important that I improve my "commercial" side.

Then I totally confirms my interest for integration and information (content) systems.
Indeed, if you have read all of this article you can:

Manage your documents and e-mails (Alfresco ECM)
Collaborate (Alfresco Share)
Create and manage websites (Alfresco WCM)
Manage technical documentation and XML (Componize)
Divide your data in high availability (Castor - Xenit) with tools (Enterprise Capabilities Only)
Integrate existing data via ETL (Talend - Openwide)
Create rich interfaces (Alfresco SURF) for specific business needs
Create workflows in a simple and intuitive way (BlueXML)
Archive everything (Alfresco RM)

And who are the people with this kind of abilities ?

Integrators (like me ...)!!!

Finally, when I was young I loved playing with Lego ... I think I'll continue on this path ...

As you can see, the road is still long.

Thank you for everything. And see you soon!

PS: You can find some pictures I took during the event at: http://www.flickr.com/photos/36931301@N03/sets/72157617655796047/

Open Source ECM : This is my way...

2009-04-02T06:12:00.000-07:00

JM.PASCAL - This is my way...

View more presentations from Pascal Jean marie.