When you say “Rails doesn’t compare to J2EE”, what do you mean exactly? As has been discussed recently, the “doesn’t scale” argument is often fallacious.

Unless ActiveRecord has come a long way since I last looked, Hibernate is most definitely a more powerful OR mapper that can deal with many more situations. ActiveRecord is fine for new databases, but not so good for existing, possibly strangely organized ones.

I am looking for a new technology to cover that void. I stoped and looked at both Python and Ruby. Python is actually quite good. The frameworks are allready there. The problem is the Python’s community doesn’t realize ModPython and FastCGI are actually the way. Also, I had trouble finding documentation. And man, those oficial Python sites are ugly ! And maybe hackers take pride in them, but man, they need to change their attitude.

Ruby is still too young. And I don’t like the hype. It’s too fast too soon. Rails doesn’t compare to J2EE. It doesn’t even come close. It’s way back when compared to Python’s frameworks.

You are right and at the same time you seem to miss the point. David Heinemeier Hansson excels at this “marketing”, or genererating hype if you will. But that is not a bad thing, on the contrary. Rails is still a very young project. He has been able to single handedly build a thriving community around it, by no means a small feat. Through this community and the considerable amounts of interest he generates, Rails will continue to gather mindshare for the foreseeable future at a rate which no comparable Python framework can offer (except perhaps Plone).

> Python does have some advantages

You are right again, but these are not advantages most people will take into account or care about. They *will* look at the fact that you can get up to speed with Rails in only a few days and that it offers true integration without any need for DIY customisation of seperate components (basically what you did).

People will also look at the fact that within a few months the Rails community has created a truly great documentation package: videos, extensive Rubydocs, a wiki, how-to’s, active mailing lists and an active IRC channel (which are archived).

In the end anyone can experience that Rails *does* deliver on its promises. Just follow the excellent tutorials. Undoubtedly you might have the same experience with Python, but it will take much more work to get things going and it will lack the gloss or neatness of Rails.

For me Python is a nearly ideal language. But Ruby is not so much different (herecy here :-), it even has some features which I would like to have in Python itself (code blocks). In the end, the benefits of using Rails far outweigh any benefit the use of Python, The Language, might offer me.

The Plone members, I believe, have done a great job in marketing Plone as a CMS. I don’t think I’ve seen it marketed as a solution to the standard CRUD application needs.

You’re absolutely correct that many people don’t care about Python advantages that I’ve pointed out. Generic functions *could* be used to help create a web framework that is extensible in some great new ways that people would care about.

I don’t consider it heresy to point out the similarity between Ruby and Python. They are very alike in features, even if the syntax is a little different.

Subway can bring the integration that Rails has to Python. My point, though, was that the people working on Subway should not expect that creating something “similar to Rails” is going to create a similar draw to Python and Subway. The only way that it can do that is to have something worth talking about and to get people talking about it.

Funny that Java programmers are attracted to Rails; Java started out as 99% marketing as well.

Greg Jorgensen

my only frustration as a relatively new user is - surprise - documentation, and comprehensive examples that clearly tie all the components together in the new user’s mind. i can sympathise with the webware lead developers preferring coding to documenting, but i think this is an impediment to building a larger community around webware - which ruby on rails appears to be doing quite well.

on the whole it’s been a pretty easy transition for me (i’ve been using python for a couple months now). still discovering things here and there, though, that i really should have known about early on. complete, coherent documentation and a couple start-to-finish typical business-oriented database applications that lay bare the entire scheme would have taken care of this. i should note here that one of the lead developers of webware blogged his disdain for the word “enterprise”, which he considered to be the antithesis of creative software development. that may be, but it would be in webware’s interest to cultivate “enterprise” developers rather than snub them.

i hope i haven’t sounded too critical - the fact that i’m using java strictly for maintenance and doing all new web apps in webware should speak louder than anything i’ve said here that sounds so.

webware is gearing up for a 0.9 release, which i’m looking forward to.

Actually, it is Ruby itself that has the built-in Web server. Rails is Yet Another Ruby App that makes use of it.

In fact, despite the hype about Rails, it is mostly a clever application of intrinsic Ruby features, bundled up with user-friendly metprogramming and coding conventions.

Not to knock the greate work down by the Railsers, but it is more of a developer UI win than a feat of techincal programming. It abstracts the mundane, repetitive tasks, and uses introspection and reflection to make reasonable guesses based on small amounts of code, saving the developer the tedium of coding the obvious.

On the downside, the hype has gotten tiresome, and the engorged set of libraries and APIs means you spend far more time learning to do things The Rails Way than using Ruby directly.

LivePage’s Chatola was amazing, but I question how scalable it is because it maintains an open socket connection back to the server.