Why are there so many java web frameworks?
I try to keep up with the latest java technologies by subscribing to a number of different rss feeds, one of the underlying trends seems to be ‘hey look at my cool new java web framework’ and I keep asking myself why keep reinventing the wheel? Clearly I’m a big fan of wicket, but I’m not going to get
into a ‘wicket is the best ever’ rant.
Instead i’m going to say that regardless of your requirements, there surely must already be at least one framework that meets practically all of them. So given that what you want already exists, why go and build it yourself anyway? Or can they not use google?
But suppose that you do build your own framework, this means that out of all existing frameworks, none meet your requirements (or you can’t use google). From this we can deduce that you have quite specialized requirements (or you are less computer literate than a 10 year old), meaning that your requirements probably won’t be shared by many other people, so why oh why do you then take your poorly documented, highly specialized code and thrust it upon the world?
We aren’t interested that you can build an entire app that says hello world with one command, or that you have a slightly different method of hacking together java, html, ajax in a ‘really cool way’.
</rant>
November 21st, 2008 at 11:14 am
Hi Richard,
So you’d suggest to stay on the beaten tracks? If everybody would think so, we would probably still hack our webapps in JSP … or Perl … or C … quite scaring thought, isn’t it?
Regards
November 21st, 2008 at 4:39 pm
Stefan,
Thats not exactly what I meant. I was more ranting at the fact that every week I see a blog post for yet another new java web framework, and that many of them must be very similar or have similar goals.
I think that if the community worked together to produce a smaller number of really good frameworks then things would be better for everyone.
November 24th, 2008 at 8:16 pm
Perhaps another way of asking that question is “why do so many people find that improving an existing framework is harder than writing a new one?”.
November 24th, 2008 at 11:22 pm
@walter - yea thats probably a better way of putting it. I would guess that improving an existing framework means you have to put time into understanding it, whereas building your own you don’t.
December 3rd, 2008 at 6:19 pm
You mean something new has come out since Wicket left the incubator? I didn’t think anyone would bother competing!
Once a Web Framework Category is bloated enough (eg. action frameworks) then definitely yes, why bother adding another framework? Especially if the category’s obsolete. For newer categories like ’stateful component frameworks’ (Wicket, jsf, Tapestry), thick[er] html client GWT like frameworks (GWT,Echo), or RIAs then a little variety might work for a while to help framework authors learn what works and what doesn’t work.