Hacker News Reader

Comment by inklesspen | Hacker News Reader

Web2py is a joke. The guy wrote his own ORM, rather than working with the best-of-breed Python ORM SQLAlchemy, and he touts this as a feature. He was also quick to point out that Web2py won a rumble, but fails to mention that this was because none of the other Python frameworks had even heard of it.

If you want a good Python framework that doesn't unduly dictate your app structure, go to Pylons.

View on HN

The Flourish conference was sponsored by Google. The creators of Django, CackePHP and Rails were on the mailing list and some even replied to it. I agree that if nobody else participated it is not a victory. I would not say web2py won, I would say web2py was not afraid to participate and delivered a working project in 24 hours that met the specs. The organizers said web2py won. Ask the organizers. Ian Biking was on the jury.

You did in fact say web2py won:

"web2py is the only framework that participated to the www.flourishconf.com rumble (develop a web app in 24 hours) and won because all the other frameworks chickened out."

http://spyced.blogspot.com/2008/04/m-half-baked-thoughts-on-...

Three other people came to pick up the rules and some more were on the mailing list, were supposed to and didn't. I spent 24 hours coding and felt like a fool when nobody else delivered. You are taking a quote I made to my users' group out of context. Why do you turn a discussion about some work of mine into a personal thing?

No, I'm taking a quote you posted on someone else's blog in an attempt to promote web2py. I'm not turning this into a personal thing; you are.

inklesspen, don't all framework/orm whatever come out when a guy is frustrated with what's there and starts it's own? Doesn't Django have its own ORM?

You can say that his ORM solution is crappy, or doesn't work well, but you can't dissmiss something just b/c it is not one of the old packages. While I may not want to use his framework for anything mission critical, it seem to be a good choice for a small and app or for quick prototype development.

His main points in his framework is simplicity and ease of starting.

Yes, and Django's ORM is also inferior to SQLAlchemy.

If his ORM were better than SQLAlchemy, it would be more widely used. Since it is not widely used, I see no need to look into it before dismissing it.

SQLAlchemy is better than the web2py ORM in dealing with legacy databases. web2py has restrictions in this case. But the web2py ORM is much better integrated with the rest of the framework than SQLAlchemy is integrated with, for example, Pylons or TurboGears. If you do not use the entire web2py framework you are better off with SQLAlchemy. If you do use web2py, you are better off with its own ORM. Here are some preliminary benchmarks (take them with grain of salt) http://mdp.cti.depaul.edu/AlterEgo/default/show/76

Reply on news.ycombinator.com