I just finished up 2 days of Mashup University 4.
John Herren kicked things off Monday with his presentation on Cross Entropy Reduction for Image Analysis using Simulated Annealing and Fast Artificial Neural Networks. Unfortunately, all the supercomputers at the Computer History Museum (where Mashup Camp is being held) are unplugged, so we settled for an excellent Introduction to Mashup Development instead. This was a great choice to start thing off, since it set the stage for everything to come. The Mashup Pyramid is particularly useful as a reference to fit all the pieces together, especialy since mashups are evolving, and the tools and frameworks overlap in some areas, and leave gaps in others.
Over the two days, we had lots of very cool technology to look at.
For Mashup platforms, we had Zude, Google Mashup Editor, Windows Live, Microsoft’s Popfly, Bungee Labs, and IBM’s QED Wiki. (I missed the Bungee Labs presentation.)
For interesting technology other than platforms, we had Google Gears for offline apps, LignUp for communications services, and AOL’s AIM messaging , XDrive, and OpenAuth.
A couple of observations on all of this so far:
- The AOL ‘walled garden’ really is disappearing
- Authentication and Authorization are becoming increasingly important – OpenId‘s time may be here.
- Are Mashups becoming the best way to build situational applications ? IBM thinks so, I tend to agree.
- Is there really an enterprise mashup market emerging ? Probably, and I think the enterprise BI vendors will be the first to find out.
- We still need a wider variety of data sources to power all these mashups.
Things are evolving quickly in the mashup space – we’ll see what happens over the next couple of days.
Kudo’s to IBM for sponsoring the end of semester Pool Party