For The Beantin Index that I launched earlier this week, there are posts that contain ratings and then there are blog posts. The main focus of the site are the ratings, so I didn’t want blog posts appearing on the home page, and I wanted two separate RSS feeds.

Here’s what I did…

This article from Brightscape was really useful, but it needed some extra tinkering to work it to work with Pagelines.

Exclude categories from home page

here’s the function added to functions.php – i’ve put cat=-24 in order to exlude my “blog” category. You need to find out your own category id.

Elsewhere in functions.php i added the following Pagelines hook:

Removing from recent posts widget

This doesn’t stop the pages appear in the “recent posts” widget, so I installed the WP PHP widget plugin so I could put some PHP code into a wordpress widget (the default text widget doesn’t run PHP)

Into the PHP widget I put the following code to make my own “recent posts” widget:

Excluding from RSS feed

Finally, I needed to exclude the category from the main RSS feed, and add a feed of it’s own – the info at the end of the Brightscape blog post was enough to sort out that combined with the following hook and code to make sure the 2nd RSS feed appeared in the header:

Posted on Beantin Beta Blog

Google has rushed the implementation of Google+ Business Pages without thinking through how pages are too be owned and managed over time.

Right now, it would seems you can’t create a G+ company page without first having a public profile (for the same account). Plus, you can’t have multiple admins for a page, or transfer admin rights either.

This means that you will be creating a page that is tied to your G+ profile. If you change job, the company page goes with you, with no way of moving it over to anyone else.

If you create a “company google account” (non-personal, which many already have for administrating such things as Google Analytics, Feedburner, etc) you will still be forced to create a public profile before a Page.

If you use your Google Apps account (if your company uses that) then you yet again will be forced to create a public profile – perhaps not something you want to do if you already have a G+ profile from your own account. And this still doesn’t get round the problem of not being able to add additional admins.

As they are, Pages are going to create a number of duplicates and “scrap” profiles – something which Google has (so far with G+) tried really hard to avoid.

Clearly very early stages for Company Pages on Google+, get the feeling they’ve pushed this out as early as they possibly could – but I think it’s foolish of them to have overlooked (or down-prioritised) the account architecture side of Pages.

Such a decision can do nothing other than create headaches in the future.

Update 2011-12-21

Google have made a welcome update to Google+ pages. Not only can you have up to 50 administrators for a page, you can also transfer ownership of a page between accounts. Read Write Web has written a quick guide for these new features.

Posted on Beantin Beta Blog

I’ve been storing my backups “in the cloud” for a number of years now. For the past 3 or so years I’ve been using Amazon S3 and various linux-based solutions to store my backups of absolutely can’t afford to lose things.

During that time I’ve used Jungledisk and S3FS, but now I’ve moved onto S3Backer. So far so good.

Here is a quick guide to what I did to set up s3Backer on my Ubuntu 10.04 LTS server. The instructions on the project site are a little limited, and I didn’t find any useful “how to” blog posts.

1. Compile and install s3Backer and dependencies
2. Create your mount points. You will need one mount point for s3backer (which is used for the files stored on s3) and then one mount point for the file system (which S3backer looks after and saves to your S3 mount)
3. Start s3backer. The command I’m using is s3backer --blockSize=1024k --size=250g --listBlocks --vhost --ssl --timeout=250 --blockCacheThreads=3 --maxUploadSpeed=35k --accessFile=/etc/passwd-s3backer --prefix=s3 buggles /s3
   The blocksize option is good to have at 1024 when you are using it for backups; good balance of file system cache and put requests to S3.
   
   The vhost option is needed because this is an EU bucket.
   
   The timeout option I need as the server sits on the end of my ADSL, with a limited upstream. s3backer runs 20 simultaneous threads by default, so I’ve limited the upload speed to 35k and dropped the number of threads to 3, which means it will take at least 90 seconds for 3mb to transfer (blocksize times number of threads divided by bandwidth cap) – hence a safe timeout of 250.
   
   I’ve put my password file in /etc (rather than in Root’s home directory which is the default) as you can see by the accessFile option.
4. Make your file system. I went for ext4. There were some warnings on the project site that certain file systems would create too many “empty” files in your s3 account (and therefore cost you money even if they weren’t in use by the virtual disk). On creation, a 250 gig ext4 disk used 17mb of storage in my s3 account. mke2fs -t ext4 /s3/file
5. Finally, time to mount your cloud-based disk. This is the one you actually use and backup to. mount -t ext4 -o loop /s3/file /backups
6. Create some cronjob scripts that rsync your files across to your virtual disk. Here’s an example of the kind of rsync command I use: rsync -rptgoDv --del --inplace /stuff/ /backups/stuff

I currently store about 160 gig of backups in my Amazon S3 account for a costs of approximately 30 USD a month.

Posted on Beantin Beta Blog

I’d love to see the research and testing that went into Facebook’s latest changes to the news feed. I’m making the presumption here that they *have* researched and *have* tested it before launch.

They’ve moved the real-time feed over to the right column, which includes “actions” by your friends (and pages) such as liking or commenting – but further interaction with this items is an extra step away – which will result in many of them never receiving further interaction.

The main news stream is a odd jumble of “top” news and “recent” news “since your last visit”. This is a very abstract collection and will be difficult for people to understand what it is they are seeing (and why they aren’t seeing some things)

We all know that a key factor to success is simplicity, and given the complexity of the new news stream on Facebook, i’d say it’s s step away from success.

…but then we always complain about changes to Facebook, but nevertheless we stay and put up with it.

(this post was originally posted to Google+)

Posted on Beantin Beta Blog

For a few websites I’ve used my Google Profile URL as a OpenID identifier – allowing me to log in using my Google Account. That identifier changes format when you sign up for Google Plus. Google hasn’t managed the change of URL correctly and it stops you from logging in.

Bad news

This naturally is terrible news for anyone using (or running) a web service that makes use of OpenID. The handling of multiple personas is something that OpenID copes with perfectly OK with – providing the account provider (in this case Google) does their stuff.

Google have changed the format of OpenID identifiers based on profiles before. Originally (in 2009) they were of the format http://www.google.com/profiles/[username] which migrated to http://profiles.google.com/[username] (in March 2011). Both of those work and are authoritive.

Changed profile URL

Once you sign up to Google+ your profile URL changes to the format http://plus.google.com/[number]/posts. The problem at the moment is that Google isn’t answering in an authoritive way for the two previous incarnations of the profile URL, breaking OpenID for you.

You can read a discussion on the OpenID mailing list. Some sites are still working, but they shouldn’t (if they are, they’ve got a security flaw!). Let’s hope Google get’s their act together before too many people and companies are affected.

Posted on Beantin Beta Blog

I’ve just migrated all my contact from my old Sony Ericsson C905 into my shiny and new Samsung Galaxy Ace. It took a bit of time, but I’ve just managed to do it pretty much perfectly. Here’s how…

1. Use MyPhoneExplorer to Export contact from C905 to a CSV file
2. Fill in one contact completely (with all the fields you expect to use) in Google Contacts
3. Export that contact as a comma separated file (CSV format). This gives you a template with the current columns being used by Google Contacts (they vary reasonably often as i’ve understood)
4. Copy your data column by column to the correct column in your 1-contact template file. It’s essential that you keep leave the names of the column headers untouched as well as the order of the columns.
5. Fill in the correct type in the corresponding type column for each value column. This applies to telephone numbers, email addresses and each set of address fields. Values are: home, work, other (plus numerous others for telephone)
6. Save your completed “template” CSV and then use the import function in Google Contacts.
7. Hey presto, you should now have lots of nicely formatted contacts, with all the right data in the right fields.

At the time of writing, the columns and the column order for your CSV to import into Google Contacts is as follows:

Name	
Given Name	
Additional Name	
Family Name	
Yomi Name	
Given Name Yomi	
Additional Name Yomi	
Family Name Yomi	
Name Prefix	
Name Suffix	
Initials	
Nickname	
Short Name	
Maiden Name	
Birthday	
Gender	
Location	
Billing Information	
Directory Server	
Mileage	
Occupation	
Hobby	
Sensitivity	
Priority	
Subject	
Notes	
Group Membership	
E-mail 1 - Type	
E-mail 1 - Value	
E-mail 2 - Type	
E-mail 2 - Value	
E-mail 3 - Type	
E-mail 3 - Value	
IM 1 - Type	
IM 1 - Service	
IM 1 - Value	
Phone 1 - Type	
Phone 1 - Value	
Phone 2 - Type	
Phone 2 - Value	
Phone 3 - Type	
Phone 3 - Value	
Address 1 - Type	
Address 1 - Formatted	
Address 1 - Street	
Address 1 - City	
Address 1 - PO Box	
Address 1 - Region	
Address 1 - Postal Code	
Address 1 - Country	
Address 1 - Extended Address	
Organization 1 - Type	
Organization 1 - Name	
Organization 1 - Yomi Name	
Organization 1 - Title	
Organization 1 - Department	
Organization 1 - Symbol	
Organization 1 - Location	
Organization 1 - Job Description	
Website 1 - Type	
Website 1 - Value

Posted on Beantin Beta Blog

If you are running a Facebook page for your business or organisation, then you’ll want to make sure you get you design your profile picture exactly right so that you make optimal use of the space available and at the same time making sure the thumbnail picture is spot on too.

The maximum size of a Facebook profile picture at the time of writing (May 2011) is 180×540 pixels.

This means you can have quite a tall profile picture and can use that extra space for branding, offers, bit of fun. Take your pick.

As far as the thumbnail is concerned. Facebook doesn’t let you use the full width of your profile picture. 16% of it is always impossible to select. This means, if you have a profile pic that’s the maximum 180 pixels, 29 pixels will be unselectable for your thumbnail – making 151px wide.

The thumbnails are always square, so this means the maximum area of your 180×540 profile picture is 151×151 – but you can of course which ever 151×151 square from the picture you want.

Which means you need to design your profile picture in a smart way so that it works in both situations – on the page itself, and then in all the places where a thumbnail is displayed.

Posted on Beantin Beta Blog

In firefox 2 and 3 it was possible to set the browser.tabs.tabMinWidth within about:config to zero so that tabs shrank in width the more tabs you had open until just the icon was left. In Firefox 4, that option has been removed, leaving you with fixed width tabs of 140 pixels.

Thankfully, although the option has been removed, the functionallity has just moved. In firefox 4 and 5 you can customise the Chrome of the browser by using CSS styles in the userChrome.css file.

This file is stored within your firefox profile. To get to that directory, click on the Help menu and then select Troubleshooting information.

On the tab that opens you’ll see that the first section is Application Basics and in that section is Profile directory with a button next to it saying Open containing folder. Click on the button.

Open the Chrome folder. In there you’ll find two example files, one of which is userChromeExample.css. Copy that file, renaming it to userChrome.css and paste in the following CSS style:

Save the file and restart the browser. Now you should have tabs that behave like you’re used to in Firefox 3.

Posted on Beantin Beta Blog

I’ve been baffled for several months as to why you couldn’t get at the mobile number via the Lotus Connections ATOM API – only the “work” telephone number.

In the summer I noticed that not all telephone numbers were visible in the XML returned via the ATOM API for Lotus Connections 2.5. Everywhere I checked behaved the same, including the Lotus Greenhouse. I’d studied the documentation, and there didn’t seem to be any way to get any more data out.

I even had a dialogue with IBM, as well as few Lotus people via Twitter, but it just didn’t appear possible. There was definitely a bug with vCards, and there seemed to be one with the API too.

I was on the verge of pushing for an alternative API to be quickly developed (rather than waiting another half a year for a fix-patch) when Jan Fjärnström discovered that there was a mistake in the Lotus Connections documentation.

On the Searching Profiles page, it says:

In reality the default value is infact lite.

So, if we’d have tested putting: &format=full at the end of API requests, it would have worked. Amazingly, I’d tested: &format=lite (the incorrectly stated “non default” value), but as full was default, I hadn’t tested to specifically add &format=full.

A more complete example would be:

Lesson? Never stop being curious! test everything…

Update, 2011-04-04

The documentation has been corrected for Lotus Connections 3.0 and as Michelle says in the comments below, a technote has been created highlighting the problem with the Connections 2.5 documentation.

Posted on Beantin Beta Blog

Tumblr can be surprisingly flexible using a bit of CSS and a few theme variables. When Tumblr added static pages (and redirects) in March 2010 it opened up a whole new range of possibilities. Unfortunately, Tumblr missed a few important features out – such as a block in the theme for controlling the display of static pages.

What this means in practice is that anything you have in your template for “permalink pages” – such as a retweet button, or a commenting system, will be visible on your static pages and, well broken.

What makes them even more broken is that static pages don’t have a {Permalink}, {PostID} or a working {ShortURL}. A whole load of variables return nothing or incorrect values when pages are displayed.

I’ve emailed Tumblr a few times, at regular intervals, and highlighted these problems and flaws in their system – but I usually get a thanks but then nothing. Guess everyone else is too busy uploading pictures of dogs to notice the bugs.

But, at last I’ve devised a way to identify static pages from your theme. This then gives you the opportunity to hide (or show) sections of your code. You can’t avoid things loading or scripts being fetched (such as retweet buttons or Facebook like buttons) but at least they will be hidden from sight.

The solution is thanks to the buggy {ShortURL} variable. On static pages {ShortURL} always returns http://tumblr.com/xpm0. So you just bracket the section you want to hide as follows…

<div id="{ShortURL}-hide">
	part of your theme you want to HIDE on static pages
</div>

…and then add bit of CSS to the style section of your custom theme…

div[id$="xpm0-hide"] {
	display: none;
}

If you want to add the ability to limit the visibility of something just to static pages, then you need adjust the style as follows…

div[id$="xpm0-hide"], .static-only {
	display: none;
}

div[id$="xpm0-show"] {
	display: inline;
}

…and then wrap the section you want to show like this:

<div class="static-only" id="{ShortURL}-show">
	part of your theme you want to SHOW on static pages
</div>

Posted on Beantin Beta Blog