Don Dodge writes about several meme-tracking services popping up all over the internet (also see this article on TechCrunch), and what the cirtera will be for those that want to survive the shake-out in this arena.

I think it’s quite apparant that due to the flood of information that overwhelms us (I blaim the bloggers :-)), we need a good way of monitoring those things that interest us.

Services like those provided by memeorandum or technorati seem to be quite popular nowadays. But if you observe their userbase you notice that it mostly consists of
people that belong to the person2.0 profile (I’m talking about people that blog or read blogs quite often).

But what about people that aren’t that deep into the whole tech/blogosphere and simply want to stay up to date in their own little piece of the world?

Take for instance somebody that develops software for a living (somebody just like myself). To stay on top of the newest tools or techniques we need to either keep track of a lot of weblogs (my preferred way) or depend on knowledge passed on to us by our co-workers or any trade magazines. Fact is that a lot of people simply don’t have the time to read ~150 rss feeds a day.

Meme trackers could be an excellent tool for these kind of people to stay on top of the blogosphere without having to invest a lot of time in reading. But only if we happen to have meme trackers tracking this specific domain of software development (which in itself can be quite huge when looking at the vast majority of available platforms).
You can even take this a step further and combine this with social tagging to build a profile of the user in which our tracker can predict what the user might find interesting…

I’ll let you in on a little secret. For a while now I have been experimenting and building a prototype of a social tagging site doing exactly this. I talked about domain specific search engines before, imagine the disturbance this might cause with the big boys (Google, Yahoo, MSN) when a lot of these domain specific trackers slash social tagging containers slash search engines start popping up…

Update: a few minutes after posting this I read on John Batelle’s blog about Krugle (they have a blog, subscribed)… a domain specific search engine dedicated to finding code and everything related. It has begun :-)

Eweek is running an interesting article describing the architectural vision of Microsoft supporting the term Web2.0 in their future.

This is a good thing from the developers perspective because the acknowledgment by Microsoft gives us a rock-solid fundament for delivering ‘open’ solutions with a quicker time-to-market on their platform.

I was listening to a dotnetrocks show the other day in which Richard mentions a study confirming the power of Microsoft’s .NET platform in the middle-segment of the IT industry. This is exactly where the pain points lie in using the current web-service offerings.

The web-services paradigm that we have been given by vendors for enabling this ‘open-ness’ (I deliberatly avoid the term SOA) has its strong points but also its weak points. To name one of the biggest: the complexity XML and schema require in order to fully comprehend and use effectivly.

I do agree that the toolset helps by hiding these complexities… but think ‘open’ again. Try integrating a rails app with a dotnet webservice… it is very possible, but certainly not without weird hacks and quirks. The promisied holy grail of open web-services architecture (the technology formerly known Indigo) is getting there, but this is only Microsoft… a lot of other vendors have yet to catch up.

I was exploring the Yahoo Term Extraction API the other day and was able to integrate this in my application within minutes. No twiddling with wsdl-contracts or other funky stuff, just simple http GET requests and query parameters.

It is important, from the architects point of view, to understand when web-services are favored over the REST approach:

“The consumer edge is the peer-to-peer, Web 2.0 world and the enterprise edge is the SOA, ESB (enterprise service bus) model. In addition, the consumer edge is an asynchronous communications model based on the REST (Representational State Transfer) scheme, and the enterprise edge is based on the Simple Object Access Protocol scheme. REST is a dominant model on the consumer side, and SOAP is the model on the enterprise side.”

I do hope Microsoft adds more weight to these statements by actively supporting REST in their offerings in addittion to SOAP. But, until that is the case, for the time being you might want to have a look at this series of articles :-)