Here’s a quick post on how we are creating subsites using the client side object model:

using (ClientContext ctx = new ClientContext("http://yoursite/"))
{
                WebCreationInformation wci = new WebCreationInformation();
                wci.Url = "YourSite"; // Relative URL
                wci.Title = "YourSite";
                wci.Description = "Your nice description.";
                wci.UseSamePermissionsAsParentSite = true;
                wci.WebTemplate = "STS#0";
                wci.Language = 1033; //LCID
 
                Web w = ctx.Site.RootWeb.Webs.Add(wci);
                ctx.ExecuteQuery();
}

And that’s it :)

The WebTemplate might be a bit odd, check this PowerShell script to to retrieve the correct values you can enter. Also note, custom web templates are possible and are also retrieved using this PowerShell:

PS C:\Users\ruarco> $s = Get-SPSite("http://yoursite/");
PS C:\Users\ruarco> $s.GetWebtemplates(1033) | select Name, Title

Note the LCID in the GetWebtemplates. The select is handy because custom web templates get a long guid and you’ll only want to select Name at those things.

Here are the main enhancements (copied from the release notes):

1. New support for:
• Windows Server 2008, including R2
• SQL Server 2008
• Visual Studio 2008
2. Non- AD Installation
3. SSL Support
4. Inline Functions
5. Reports
• New filters and parameters
• Using SmartObject direct execution functionality
6. Performance enhancements to SmartObjects and the K2 Workflow Server
7. In MOSS/WSS K2 configuration Features are deployed as solutions
8. K2 Designer for SharePoint enhancements include:
• Inline Functions
• SmartObject Filters
• Custom Work Item Notification
• Process Data Fields
• New Wizards
9. Internationalization of the codebase
10. Numerous documentation updates
11. Inclusion of all Updates

For more info, here are some links:
K2 Underground 4.5 RC Forum

Registration form for What’s new to K2 4.5

Check out the K2 Event Calendar for other webinars and events.

Jey’s report on the release!

PeteStilGoe.com’s post on 4.5

The new K2.com website showing 4.5 info!

More URL’s coming…

Why are inline functions so important?

To explain why inline functions are so important for K2, it is needed to provide some background. Way back before the South African development team started working on K2 blackpearl, a vision was created on how to design processes from a business perspective. The current K2 architecture has a lot of features that can be mapped to that vision. An example of this is SmartObjects. The vision shows that (enterprise) data should not be included/part of the process and should be referred from within the process. SmartObjects do just that, making live a lot easier when working with the K2 platform.

Another big part of the vision is to allow you to design processes without the need to write code. A lot of processes can be implemented without code, but I’ve had lots of occasions where I still needed to write small pieces of code. Within the new K2 designer for SharePoint and K2 studio, you’re unable to write code and you’re stuck going back to Visual Studio. This isn’t how K2’s vision was created and so a solution was needed to reduce the use of code and allow you to call (simple) functions.

What are inline functions?

Every wizard within K2 uses input fields where you can drag and drop items from the Object Browser to the input fields. Inline functions allow you to call simple functions inside those input fields. The functions are categorized in groups, so all the Date and Time functions are together and all the Text manipulation functions are together. A complete list of inline functions will be listed in the product documentation which will also be up on the knowledge base site. For now, you can check this blog post.

So, why is this so good?  Well, as stated in the previous paragraph, this will allow you to do more with the designers that do not support the use of code.

How do they work?

This youtube video will hopefully explain how you use inline functions while designing a process.

From a technical perspective, inline functions are executed within the workflow server.

[UPDATE] I forgot to share the video for the public, it should be better now :-)

But, the function I need isn’t there!

Ok, K2 can’t provide you with every function thinkable on the planet, but they always provide you with ways to extend the K2 platform. This is also the case with inline function! In the end, it all comes down to a simple .NET assembly with some class and method attributes. The developer reference will have an article/how-to on this.

Conclusion

I hope this blog post makes you understand how valuable inline functions can be when designing a process. For myself, it’s another way of doing things and allows me to create processes more easy and (in a lot of occasions) without code!

This blogpost is part of a post on inline functions and only for reference!

SharePoint Connection NL 2010

MSOL06: Nintex Workflow 2010: Extending 2010-Era SharePoint Workflow to End User and Enterprise Needs Alike by Mike Fitzmaurize

As a K2 insider it seemed weird going to the Nintex session, but I went to see what they have to offer as it is always good to keep an open mind. I’ve already seen Nintex a while ago but now they showed its integration in SharePoint 2010. Mike is a nice speaker and listening to him was good, I also think nobody noticed that the workflow he build couldn’t actually start and he didn’t return to fix it. But then again, he was working on an alpha build of Nintex. Nintex still runs on the SharePoint WF, you can either see this as an advantage or a disadvantage. Also, the integration with backend systems is done via wizards. Nintex has a lot of those which makes the out of the box package product pretty complete. If I look at K2 from that perspective, it’s able to integrate with SmartObjects and more generic solutions which allow for more flexibility.

MSC11: SharePoint 2010 Search by Richard Taylor

Richard Taylor about make people smile!

Yesterday I really enjoyed listening to Richard, when I saw he did a functional session on Search, I was curious and went for it. The session had a lot of information on what search actually is. And that search is not actually what you want to do, you want to FIND things. This al leaded into explaining what Findability is and what you need to do for it. Interestingly, it wasn’t SharePoint that provided the solution according to Richard. He says if you have a mess that using SharePoint will not solve the problem, you need to structure your data and an Information Architect is crucial to get Search working correctly and make things findable. This was a valuable lesson and enjoyable to listen too, the hour flew by and everybody left with things to think about in the next SharePoint project.

MSC27: Claims-based identity in SharePoint 2010 by Spencer Harbar

I expected a lot from this session, but it didn’t really deliver. Partly because there was no demo but only slides and a lot of theoretical stuff on the claims based identity framework (Geneva).

The session did hold a lot of good content which is way too much to sum up here. It was important to note that with claims based authentication you’re not always able to eliminate Kerberos, this is because the backend system needs to be claims based aware, which some systems are simply not (yet). When you want to use Kerberos, you’ll need to use claims based authentication within SharePoint, which (from my understanding) will be able to create a windows identity and authenticate via Kerberos.

I really like the claims based authentication and really want to start working wit hit and try some different scenarios.

Spencer also noted that claims based identity (currently) only works with MSIE8. I presume that’s for the claims based identity to work as if it’s integrated authentication. So for older browsers we’ll probably see forms based authentication that uses claims based identification.

UPDATE: Spencer commented to this article about the browser requirements. He also provided some tweets:

http://twitter.com/harbars/status/8318105648

http://twitter.com/harbars/status/8318666635

http://twitter.com/harbars/status/8318676296

http://twitter.com/harbars/status/8318716190

Ask the experts area

MSC31: SharePoint UI Customizations by Marianne van Wanrooij

This session was a ‘redeploy’ of the session already held in Las Vegas. Marianne made her own demo’s, which was good to see.

JavaScript, Status bar and Notifications

SharePoint has a nice feature called JavaScript on demand, which only gets the JavaScript if needed. This allows for a small size of the basic page that requests the correct JavaScript when you need it.
After this, Marianne showed how the status bar and the notification area can be manipulated using JavaScript and the server client object model.

Ribbon

The ribbon is also extendable in SharePoint 2010, it allows you to add and remove buttons using a SharePoint feature which gives you very tight UI integration with SharePoint. The ribbon is actually one big image which is cut up by CSS. This is a very efficient technique that eliminating callbacks to the server. In the demo, a button was added to the ribbon and some simple code was run.

Dialog framework

The dialog framework was also part of the demonstration. There is not much to tell about it (which is a good thing!). You can open a dialog and provide it a URL to use as a page. The dialog framework hides all the html elements with the ‘s4-notdlg’ css class. This allows you to use one page for dialogs and for normal pages!

MSC30: Advanced Web Part Development by Jan Tielens

Jan Tielens ready to rock!

I can’t help saying it again, Jan Tielens his presentations and demo are super to see! They simply work and are so clear. He makes it look way to easy!
The session covered topics on web part development in SharePoint 2010. There are some new things, like XSS Safeguards which are important to know about as a developer. A big improvement is the Virtual Web Part. For me, that’s nice to see, but I don’t understand why it was so important to drag and drop UI elements.  Apart from that, Jan showed how you can use the visual web part to create connect web parts. After that he went on and made two connected web parts update with Ajax.

Conclusion

I enjoyed the SharePoint Connections conference, although a lot of it was already known from Las Vegas, it still added knowledge for me. For me, that’s the most important part of going to a conference. Apart from that, it was nice to see all the SharePoint people again and shake some hands!

Keynote by Matthijs Hoekstra and Mike Fitzmaurice

The keynote started with the nice promotional video that was also shown at the SharePoint Conference in Las Vegas back in October. It’s a nice movie with lots of fast moving images and music, good to wake up and get you in a good mood. Matthijs Hoekstra kicked of the keynote by introducing Mike Fitzmaurice.  Mike is always a good speaker and he told a story how SharePoint 2010 came to be as it is now. Starting in the pre-2001 age, Mike probably knows most about the revolution SharePoint has made. Although the story was nice to hear, it wasn’t what I expected from the keynote, as I expected a bit more ‘power-talk’. But then again, do we still need that?

MSC34: Implementing Multi-Lingual Solutions on SharePoint 2010 by Spencer HarBar

I’ve never done anything with Multi-lingual sites in SharePoint 2007. Spencer gave a good overview of what it is and when/how variations come into play when doing multi-lingual sites. The important message was to plan up front because turning on variations has a large impact on your SharePoint site, site collection and even farm.

Language packs

Language packs provide the SharePoint chrome in a different language, this makes the ribbon and menus (site actions, etc) show in a different language. These language packs basically provide site templates in different languages. After the installation of a language pack, you must rerun the configuration wizard. If you install multiple language packs, you only have to run it at the end of the process once.

Variations

Variations are there to provide your content in a different way. This also means that site targeting mobile devices are actually a variation too. It’s just shown in a different way. The same principle applies to languages. It’s the same site provided in a different language. There is always a Variation Source, which is where you place the initial and original content. Posting or editing content can kick of a process to provision that content to the variation targets. The variation target gets a draft version so the translators can translate that document into the language required for that variation.
I’m pretty sure this content isn’t new for SharePoint 2010, but still nice to see when you’ve never done variations. Spencer did mention some new features in SharePoint 2010 like:

• View changes editing the draft document shows you the original and new version which makes editing easy.
• The process that copies the documents from Variation Source to Variation Target has changed and is now more configurable.
• Sites can have an alternative language, which makes it possible to have the chrome of SharePoint shown in a different language without having to create a variation of it.

MSC33: Understanding the Service Application Architecture of SharePoint 2010 by Richard Taylor

This talk was about the new Service Application infrastructure which has completely been changed in SharePoint 2010. The room was packed so I was happy to get a seat, near a wall socket even :-)

In SharePoint 2007 there is one SSP that hosted services which sites could use. This is a monolithic design meaning that every service you need is hosted within that same SSP. In SharePoint 2010, this has all been changed and you can now run Services separately and only enable what you really need.

This changes a lot, it makes these services more secure because they can be run under different users and application pools, but it also makes things way more complex. For instance, every services uses its own database, this means that database management has become more complex. But also in terms of architecture, which site needs what service? Also note that some service application can be shared between farms, and some can’t be shared.

For every service application a proxy is created, so all the clients (features, webparts, etc) that need to request information from a service do that using a proxy.

I enjoyed Richard’s session a lot because it made me laugh when we got to the point about upgrading from SharePoint 2007 to SharePoint 2010. He basically says you should not do that unless you have a very simple site. The way this was told was pretty funny. It also made lots of people unhappy, but I think they didn’t notice until they walked out, as the session was funny, entertaining and informational!

MSC01:  ECM for the Masses by Erica Toelle

I’ve been to ECM for the masses at the Las Vegas conference, so I didn’t expect a lot of new stuff. The session covert most of the same topics but more from a business perspective. It showed the new capabilities of the metadata services and the document management features SharePoint 2010 has to offer (Unique document ID, Document sets, etc).

In general I really like this presentation but not having a live environment and using movies to show wasn’t helping a lot. I think the presentation was good and Erica is nice to listen to but would rock if there was a live environment.

I was impressed by the last demo. It showed how you can change where SharePoint stores its blobs of data using configuration in SQL server. After that, a PowerShell script was used to import a directory structure into a SharePoint document library and create shortcuts on the FileSystem to keep the documents in place for the end user. SharePoint 2010 routed the document (based on metadata) to the correct folder in a SharePoint list. I was impressed!

Update: Erica commented on this post (thanks!). Her VM wasn’t working correctly so the video’s used where a backup!

MSC06: Client-Side Technologies in SharePoint 2010 by Jan Tielens

This session covered the 3 new client side object models available to the developer.

The Client OM

In the past, when coding a client side application you could only talk to the SharePoint web services which really wasn’t very nice to do. The new client side OM allows you to request and manipulate information using a normal .NET API on the client. You’ll use the ClientContext object to query SharePoint 2010 in a very efficient way. An example showed it getting information from a list. You can specify which columns you need and that makes the result set very small and efficient to transport over the network.

ECMAScript

This is the JavaScript version of the client side OM. It has all the benefits of the Client OM, but uses JavaScript. This will primarily help you create AJAX functionality that interacts with SharePoint data. The OM is also efficient using the same mechanism as the Client OM. I liked that Jan also dropped in some jQuery which is something more Developers should do and learn. But I guess that even counts for basic HTML and CSS skills.

Silverlight client OM

This is basically the same as the previous two, only in Silverlight. This Works very nice and very efficient.
Read you can, efficient the client OM’s are (joda talk)! Apart from the microphone problems this was a very nice session with a well prepared and worked out demos. Jan is a very good presenter and I enjoyed the session, especially because Jan Tielens was the first trainer I ever had on a SharePoint course so he sort of introduced me to the thing.
Jan also showed the SPVSX extension which allows ‘quick deploy’ for your handy SharePoint 2010 development work.

Conclusion

A nice first day at the SharePoint Connections 2010. A lot of people complain about the content being a bit poor. I do agree a bit on that, but that’s mainly because I feel the demos are not prepared or working very well. Which I think eventually is a responsibility of the presenter, but I can’t help blaming that SharePoint 2010 is still beta.

The other part that I’ve changed is the size of the borders on windows. To do that:

1. Go to Personalize by right clicking on your desktop.

2. In the Personalize screen, select Window Color

3. In the screen that pops up, select “Advanced appearance settings…”

4. Now you’ll get the normal screen you’re used to from XP and earlier. You select “Border Padding” to change the transparent border on the windows. You can also change the “Caption Buttons” to change the way the Minimize, Maximize and Close buttons look. Be careful with the last one. Your taskbar items will also be smaller.

5. Click “Ok” :-)

So, the reason i blogged about this? I wasn’t able to find this very quickly and i needed a reason to try out LiveWriter :-)

Apart from this new functionality, K2 will also support the current functionality in SP2010, which we saw a demo of at the conference. This makes K2 the only workflow engine i know that is able to work with SP2003, SP2007 and SP2010. To make it even more cool, you can combine those in one single workflow!

K2 Insider Conference

Last weekend was the K2 insider Conference in Las Vegas. They neatly planned it just before the SharePoint Conference to make it more easy for us to come over. This post will not cover a lot of info that’s been said at the conference, some of it is even under NDA, so I’m simply not allowed to blog about, which is a shame, as I can’t wait to get all those new bits!

Start of the conference

Chris Geier kicked of the conference by an introduction round and checking what everybody wants to get out of this conference, most of the insiders want to see the new bits and talk about issues that they had in their projects.

After that it was Adriaan van Wyk to explain how they are handling the economical situation in the world. In general K2 is doing very well and handling the situation as they should be, which reassures me in the future of K2.
After this all, Adriaan gave us a overview of how they set out to build K2 blackpearl and how a lot of functionality is in the platform but not exposed via UI since they simply didn’t have the time and resource to do so, a good example of that is the Event Bus. It also shows why SmartObjects are so important and why they re-engineered a lot of the SmartObject code to make it perform better and really start using it as a way to surface and manage data around the processes.

Other good news is that K2 blackpoint and K2 blackpearl are now based on the same code-base and upgrades will be released at the same day.

Next K2 release

In the next release there’s been loads of work done to make installation easier. A lot of support tickets within K2 are based around infrastructure problems like Kerberos. With the new installer (which merges installation and configuration wizards), there’s an environment checker that checks your settings and sometimes automatically repairs them. The new installer also allows for unattended installations. Together this allows for super easy installation and checking if the Kerberos environment is configured correctly.

K2 designer for SharePoint

Also in the next release is a the new K2 designer of SharePoint. This basically is the Silverlight designer but is redesigned and made better, with one of the key features being inline functions and SmartObject filtering. Next to that, there’s new wizards available to allow you to do more with the designer.
The K2 designer for SharePoint has also been enhanced, concepts as loops are available, this might seem a simple item, but you’ll be amazed on how many workflow products do not support a loop/loopback.

Again, people will wonder it a Silverlight designer can bring enough functionality. If it does not, you can simply export the process and continue working on it in K2 blackpoint, and if that’s not enough, you can go to K2 blackpearl.

Inline Functions

Another new feature in the next release is Inline functions. It might not say a lot by the name, but it basically means that on every entry-field on a K2 wizard you’re able to use functions – all without code. What kind of functions? A lot. Think about math functions, expressions, string functions (like left, right, padding, etc), and much, much more. Now if you do think you can find a function that’s not available, you can simply create it yourself.
Inline Functions allow for less use of code within K2 processes, which is important for the K2 designer for SharePoint and K2 blackpoint, as there’s no code available.
To make it a bit more clear, here’s a screenshot:

Screenshot of an inline function (complex)

I’ll also try to create another blogpost on InLine functions to explain them better.

Performance and stability enhancements

Apart from new functionality, Koos Du Preez also had a talk on the performance enhancements made to the K2 server. In the past, there have been some problems with the Worklist performance, I can assure you, those days are over.

The performance of SmartObjects and the ADO.net provider has also been reviewed and improved. Some service brokers are now able to directly call the backend, eliminating a lot of overhead. Aside from that, a double buffer system is made so the SmartObject readers are way faster, sometimes even faster than calling the SQL Table directly.

Conclusion

The days flew by and where packed with info, the above article is just the tip of the K2 mountain. Can’t wait to go ahead and work with those new bits and see the product improve and improve.

The Event Bus/Scheduler itself is an impressive piece of architecture. You can read about it in chapter 21 of the Professional K2 BlackPearl book. The event bus is used to perform an action when a specific event occurs within K2. With the K2 workspace you’re able to subscribe to those events. This extension provides you the ability to add your own assembly with static method and call that based on the event that occurred within K2.

Example code

To make a small example, I’ve written a class with two static functions that use the Twitterizer API to communicate with Twitter. Although Jonathan King has got some more integration with Twitter, this seems to be a simple/good example for the Event Bus:

using System;
using System.Collections.Generic;
using System.Text;
using Twitterizer.Framework;
using System.Configuration;
using SourceCode.EventBus.Utils;
 
namespace TwitterCustomActionEvent
{
    public class SendTweets
    {
        private static Twitter CreateTwitter()
        {
            AssemblySettings settings = new AssemblySettings();
 
            string proxyUrl = settings["ProxyUri"] as string;
            string username = settings["TwitterUsername"] as string;
            string password = settings["TwitterPassword"] as string;
 
            if (string.IsNullOrEmpty(username) || string.IsNullOrEmpty(password)) {
                throw new ArgumentException("The username and password should be supplied by the Application Configuration file.");
            }
 
            Twitter twit;
 
            if (string.IsNullOrEmpty(proxyUrl)) {
                twit = new Twitter(username, password, "K2 Notifications");
            } else {
                twit = new Twitter(username, password, "K2 Notifications", proxyUrl);
            }
 
            return twit;
        }
 
        public static void SendTweet(string text)
        {
            Twitter twit = CreateTwitter();
            twit.Status.Update(text);
        }
 
        public static void DirectMessage(string toUser, string text)
        {
            Twitter twit = CreateTwitter();
            twit.DirectMessages.New(toUser, text);
        }
    }
}

As you can see, there are 3 methods. Only the Public static methods are seen by the K2 Custom Event Wizard. The private method instantiates the Twitter API and uses the AssemblySettings class to get some configuration settings. The two public methods simply send a Direct Message to a twitter user or send out a simple tweet (status update).

In the CreateTwitter() method, a AssemblySettings object gets initialized. This object can be found in the SourceCode.EventBus.Utils namespace and allows simple configuration files to be loaded. A sample of that config:

<?xml version="1.0" encoding="utf-8" ?>
<configuration>
	<appsettings>
		<TwitterUsername>username</TwitterUsername>
		<TwitterPassword>password</TwitterPassword>
		<ProxyUri/>
	</appsettings>
</configuration>

Although it might look like a normal application configuration file (.config), it is not! Normally you would see ‘<add>’ nodes in the appSettings list. The AssemblySettings class reads the the XML and parses it. Be sure to use ‘appsettings’ in lowercase, as the xml parsing is case sensitive.

The SendTweet and DirectMessage methods simply send calls the Twitterizer API.
The above class is put into a signed class library project.

Installing the Custom Event

After you have build this class, create a folder in C:\Program Files\K2 BlackPearl\Host Server\Bin, or whatever directory you’ve installed K2 into.
In my case, I’ve named the directory ‘twitter’ and placed 3 files:

• TwitterCustomActionEvent.DLL
• TwitterCustomActionEvent.DLL.config
• Twitterizer.Framework.dll

As you can see, also placed the referenced assemblies in the same folder, and create a config file that’s the same name as the DLL, with .config after it.

Now, open up the SourceCode.EventBus.Assemblies.config file located in the Host Server\bin folder. This file is a simple configuration file so the K2 server knows which assemblies it can use for Actions in the Event Notification. Simply add a assembly line, give it a Displayname and locate the Assemly that it sould use. Mine looks like this:

<?xml version="1.0" encoding="utf-8"?>
<configuration>
	<appsettings>
		<assemblies>
			<assembly displayname="EventAssemblies" fullname="c:\program files\k2 blackpearl\Host Server\Bin\SourceCode.EventBus.EventAssemblies.dll" />
			<assembly displayname="TweetAssembly" fullname="c:\program files\k2 blackpearl\Host Server\Bin\twitter\TwitterCustomActionEvent.dll" />
		</assemblies>
	</appsettings>
</configuration>

Configure the Action Subscription

After this, restart the K2 hostserver and go to the K2 workspace. Open up the Notification events menu and select. On the screen that opens,click New event in the bottom right hand corner.

Enter a name and description, select next.

Click to enlarge

Now, you get a screen where you can select the event you want to subscribe to, in this case, I’ve selected the OnProcessStarted event.

Click to Enlarge

Time to select the action, this is done by first selecting the assembly. Click search and get all of them, including the one you’ve just added.

Click to Enlarge

Click next and select the class you want to use. In this example, TwitterCustomActionEvent.SendTweets.

Click to Enlarge

Click next to select the Method you want to use, I’m going to do ‘SendTweet’, finally click finish.

Click to enlarge

As you can see, the screen is updated and you are now able to add Action Parameters. Simply select data from the left object browser and drag them into the Text field. You can add extra text if you want too.

Click to enlarge

After that, click finish and you’re done! Start the process and see your twitter happening!