Monday February 20, 2006

There's been recent much-ado about Krugle that I just don't get. The website isn't even open, yet folks are salivating as if they've never heard of such a thing as a code search engine. Not that I'm invested in anyway in Koders, Codase or Codefetch but they're providing a credible story already for that kind of specialized search. I'll be happy to try Krugle when they open their doors but the level of excitement in the posts about them seems really odd. Wired has some screenshots in their article about it. Let's see the web site live and then get all frothy about it, please! This just smells like more web 2.0 over-hype; people: control yourselves!

What I'd really find useful is code search built into proprietary projects. After reading Using Lucene to Search Java Source Code, I'm imagining building a Lucene full text index of a project at build time, as a maven plug-in or an ant task. The current tools out there to make source code and docs browsable would benefit so much from making them searchable; there's already a Jetty plugin, run your build, start your webserver and search the code base whenever you want to find a particular method call (or something. That would be cool!

krugle web 2.0 hype lucene koders codase codefetch ant maven