Archive for April, 2009

MEF Preview 5

Just upgraded to MEF preview 5 and thought I would just post a gotcha that had me puzzled for a few minutes. After upgrading none of my exported parts were being picked up after composing an instance of the composition container class. I was exporting my types as per the example below:


    1 [Export("DataTransformation.Task")]

    2 public class Task : DataTransformationTask

    3 {

    4     // ….

    5 }



With MEF preview 5 you have to explicitly pass the type you are exporting in the Export attribute constructor as per below:


    1 [Export("DataTransformation.Task", typeof(DataTransformationTask))]

    2 public class Task : DataTransformationTask

    3 {

    4     // ….

    5 }



If you would like to find out what else has changed I would recommend taking a read through Glenn Block’s blog review here: http://codebetter.com/blogs/glenn.block/archive/2009/04/11/mef-preview-5-changes-and-enhancements.aspx

TinyURL and SEO

So, I have finally decided to setup a Twitter profile to find out how it can change my life. With Twitterberry installed on my phone, my profile picture uploaded its all systems go!

One of the first things you will notice once you are up and running is the prolific usage of URL shortening services such as TinyURL (due to the 140 character limit imposed on one’s tweets). This got me thinking about what the the impact of these services have on SEO. No doubt they add value (they take up less space) but obviously contain less metadata about the target of the link. Google’s famous PageRank algorithm utilises a link as a vote towards a target page’s score while the actual URL path is also utilised in some way to identify the relevance of key terms on the target page. So, the question is, do search engines such as Google unpack shortened URLs and uncover the source URL for indexing purposes? Food for thought…

I must admit that although there is value in shortening URLs (such as the example posted on TinyURL’s homepage), they remind me of the kind of links you find in email spam messages. Perhaps that is why I find myself thinking twice before clicking them. I don’t like the fact that I don’t know where I am being navigated to. Although its tough to judge the content of a page just by looking at the URL, I believe many of us have become very good at it. No doubt many people (including myself) have become very adept at rapidly trawling search engine results and picking links they wish to explore, central to this is the URL and the path it exposes. In some cases it may be useful if authors of shortened URL’s exposed a tooltip with the full path to the original URL although I understand that this is not always possible and, to an extent, defeats the purpose.