Wednesday, September 30, 2009

Tools for The Current State of Web App Development

I've been doing this stuff for a long time. The technologies have progressed quite a bit over the years. If I had to start over today, these are the tools I'd choose:

App server: Resin 4
Web frameworks: RESTEasy, htmleasy and JSR299 (as part of Resin 4).
Persistence: Hibernate with ehcache for the second level / query cache.
Database: MySQL and Sequel Pro is awesome (try the nightly builds).
Presentation layer: Mix of JSP/JSTL with GWT controlling the forms and complex widgets. json and gwt async rpc for communicating with the backend.
Javascript helpers: jquery and possibly some YUI3.
IDE: Still sticking with Eclipse cause it is free and works well enough.

Other interesting technologies: HornetQ for messaging. Infinispan for data grid and possible map/reduce framework in the future.

I've recently spent some time with Stripes and I like it a lot, but the resteasy / htmleasy / jsr299 combo is so much more powerful and complete (while maintaing a fairly simple api), I just don't see the point in using Stripes.

Saturday, September 26, 2009

Twitter Valuation

How can a company that 'has yet to report any significant revenue' be valued at $1 billion?

Just because you have an audience doesn't mean they are going to be willing to spend any of their money. It could be the fact that their audience doesn't have to spend any money that gives them an audience to begin with. As soon as they start charging for services, someone else could easily come along and do it for free again and 'steal' their audience. Still, I just don't get the valuation.

That said, people pay for when OkCupid is free and arguably a more entertaining / useful service. But then again, maybe people pay for thinking they are actually getting a better product or maybe it helps weed out the people who aren't serious. Maybe that is what the investors are hoping for with Twitter.

Regardless, it will be interesting to see how all this plays out over the next few years.