Zacker.org

I think we can probably all agree that RoR is just getting started, and that it's way too early to say that it's going to take the world by storm. But there are indicators that for me are quite interesting, and should be noted.

• It *has* enjoyed rapid growth and high praise in the last year.
• It has an awesome architecture and design philosophy (Don't Repeat Yourself and Convention Over Configuration).
• It's attracted the attention of a lot of large players. One example: I just noticed that salesforce.com has released RoR integration tools. That's the kind of thing that indicates the development of a larger ecology around the framework.
• ROR is ahead of the curve in having a wide array of built-in AJAX tools that make the replacement of desktop apps not only viable but easy. The [[http://demo.script.aculo.us/ajax/autocompleter|autocompletion]] feature alone makes me drool as a custom application developer. The gradual replacement of desktops apps with web-based ones is huge, and it's just begun.

I'd be *very* surprised if RoR were to displace Java or PHP in the marketplace (it's an apples to oranges comparison anyways). "Enterprise" people will still want slow and clean. Legions of PHP programmers will, for reasons both legitimate and questionable, resist learning a new framework and will want to continue to do things Their Own Way rather than adopt someone else's approach wholesale. Certainly, many developers will redesign their own tools to adopt some of Rail's architecture and philosophy.

But I do predict that RoR's rapid growth is going to continue for next five years, and that within its niche, it's going to gain a very loyal following as a way to quickly build featureful and well designed database-backed web apps. The web framework market is stunningly huge now, and there is plenty of room for something like Rails to thrive while not enjoying market dominance a la PHP.

It would be interesting to compare the growth of RoR with the growth patterns of other technologies that have preceded it. PHP certainly started from humble beginnings.

I have to agree that probably ror will not become mainstream.
It does have the potential for it.
I don't think the problem lies on IDE or things like that. Although a nice ide would be nice.
BTW, I have being using radrails from the day that it came out and I love it. It gets better every release.

I see the problem being the learning curve.

1. Ruby is a beautyful language that brings many different/new/advanced concept that are foreign to most programmers. ROR shines because of ruby. These new concepts are not for everybody.
2. Ruby is totally OOP. OOP again is not for everybody.
3. ROR is for people that want to get dirty and into the code. Most people don't want to get dirty, they just want to get it done and over with it. PHP does that. (no configuration, no convention. Now that's quick!)
4. ROR and Ruby are not the easiest to configure and maintain, a lot of mistakes have been made during releases. We/they are learning.
5. ROR is not very well documented and documentation is not well organized, altought in the last release the api site got much better. Again not good enough for newbies.
6. Switchtower/Capistrano not very easy to setup and deployment tools are still in infancy. Again huge Learning curve here. You need to understand: svn, ssh, unix commands, rake and ruby and maybe even lighttpd. Most people would give up on it.
7. Most web project don't need the power or complexity of ROR. It's overkill for more than 50% of the sites on the net. No need 4 ROR when all you need 5 pages, or just a forum/cms site. (phpbb/postnuke take care of that in 5 minutes flat from download to deployment.)
8. Not enough prebuilt plug and play components. Compare calendar/forums/cms/pic gallery and other components that are freely available on the net. Almost everything need to be built from scratch. (Compare the # of open source projects like phpnuke, postnuke, phpbb, mambo, etc... .net component out of the box, grids and so on with what's available for ROR, give it time you say?, ok I say, just document it, make easy to find the answer.)
9. even .net is easier to configure, deploy, debug. Especially since 2.0. 2.0 is the response to many many requests from developers and clients. 2 thumbs up for 2.0 (can definetely get better, maybe with iron_ruby??? :)

What ROR is really contributing is to a switch of views and methodology. ROR shook the world of web developement (thanks to great marketing from DHH) and every other framework took notice.

ROR has a lot of pluses, some great hacks and brings a lot of finesse to the table. It does come with a huge learning curve.

That's why there are so many ROR clones, there are a few for each language. My guess is that it was a sign of times, other people were working on different new ways of doing the same thing.

Diego

I personally am also a huge fan of seaside and secondarily of Django which came out about the same time with different philosophies.