Mart Muller's Weblog

Benjamin and Joost-Wim, two interns here at Tam Tam, have been working at a very cool solution in MOSS 2007. Their goal was to create a knowledge management solution in MOSS 2007. Of course there are some tools out there that support rating and tagging and cloud tags in SharePoint. This solution however offers a rich user experience of those mechanisms, personalized (this is very cool) and full integration with SharePoint search and Ontolica search!

I've put some screen shots below, but I'm sure the movie they created tells the whole story in 2 minutes.

 
Tagging, and personalized rating integrated in SharePoint list, sort the lists by rating

Add new or existing tags.

 
Rate the items, both personalized and global

 
Integrated in search

Or just watch the video here

I'm going to the SharePoint Conference 2008 in March! Today, the sessions became available here. Seems like it will be a great conference. If you're also going and want to meet, please drop me a line.

With the release of the AKS, the discussion about accessibility in SharePoint has really started now. If you're interested in the discussions going on there are two places you should go to:

• The SharePoint Accessibility discussion on MS Forums
• The AKS Mailing list

Especially in the last one, you can learn a lot from the experiences from MVP's and the community members!

I just received an email from Ian Campbell about his Silverlight Browser web part for SharePoint (a SharePoint Silverlight web part which displays SharePoint documents and pictures in a rich Silverlight flow control). A cool example about how to upgrade the default SharePoint UI! You can download the beta bits here.

See SharePoint Silverlight Browser in action Silverlight Browser Video or read more on Ian's Blog

Recently I got a question about in what cases should you split site collection and content databases in a 'site/portal/your configuration here'. The question came up after reading my post about MOSS limitations. Although it's very hard to get to those limitations, it could be the case that would like to create 2000 sub sites... right?

Anyway, the choice of splitting it into different content databases, and therefor also different site collections, depends on other things as well. If you won’t expect to exceed 2000 sub sites, it wouldn’t be a reason to split up. But there are some other things to take into consideration:

• The integration level – if you split, you cannot use content query web parts to query all the collections. The site collections are separated in content and security with the advantages, or disadvantages, that come with it.
• Security – splitting site collections is easier to control when it comes to security. Of course it is no problem to secure different sub sites, lists or list items, but the chance of mistakes is always a risk. Different site collections are much more controllable and easier to maintain when it comes to security.
• Storage – if you are intending to use huge amounts of data in a collection you can use different databases to be more flexible. It's easier to migrate databases or to control the load of the SQL server.
• Site collection quota's. Its easier to manage and control site collections with quota's.
• Backup and restore. In most cases we see that if it comes to a portal divided into departments, one site collection is used. When also using WSS sites for projects/communities etc, in the same interface, it might be a good idea to split those into other content databases.

Basically, there are no hard guidelines in which cases you should split up or not, but it's always good to plan before you try. So, also check out the planning section for MOSS on TechNet.

Check out our new vacancy site and get a virtual introduction to Tam Tam!

It's in Dutch but I think you can find your way anyway :) Enjoy!

It's still hard to explain to developers why the Content Type ID's are as they are: "one-dimensional hierarchical partly-hexadecimal unique identifiers"...

The structure is documented at MSDN here: http://msdn2.microsoft.com/en-us/library/aa543822.aspx

"Content type IDs uniquely identify the content type and are designed to be recursive. The content type ID encapsulates that content type's lineage, or the line of parent content types from which the content type inherits. Each content type ID contains the ID of the parent content type, which in turn contains the ID of that content type's parent, and so on, ultimately back to and including the System content type ID. By parsing the content type ID, you can determine which content types the content type inherits, and how two content types are related.

Windows SharePoint Services uses this information to determine the relationship between content types and for push-down operations.

You can construct a valid content type ID using one of two convention:

• Parent content type ID + two hexadecimal values (the two hexadecimal values cannot be "00")

• Parent content type ID + "00" + hexadecimal GUID

One special case, the System content type, has the content type ID of "0x". The System content type is the sealed content type from which all other content types ultimately inherit.

For all other content types, you must use one of the previous methods to construct a valid content type ID. "

But why doesn't a content type just have a parent property with a consistent GUID (like al the other ID's)? Basically I don't know, but the article give the 'push-down operations' as a reason, but I'm not really convinced that is the only option to achieve that. Maybe it's got to do something with performance as well.. to find out the parent will require a recursive function. Anyway, it works.

Also check out Andrew May's article with more info

Just received some lines of code for my colleague Wouter about setting audience targeting on a web part programmatically. I thought sharing would be useful because there is very little documentation out there...

SPLimitedWebPartManager mgr = webMS.GetLimitedWebPartManager("default.aspx", System.Web.UI.WebControls.WebParts.PersonalizationScope.Shared);

SPLimitedWebPartCollection col = mgr.WebParts;

webparts[0].AuthorizationFilter = ";;;;" + "ROLE 1" + "," + "ROLE 2";

mgr.SaveChanges(webparts[0]);

In this case the first web part gets the role1 and role2 audiences. Checking the MSDN library it says: "Gets or sets an arbitrary string to determine whether a Web Part is authorized to be added to a page". Hmm can someone explain this a little more?

...and the best wishes for 2008!