Akelos Framework: too good to be true?

2006-06-10T09:26:00Z

Someone recently added a comment to my article about Rails-inspired PHP frameworks pointing out that I forgot another Rails-like framework, in my round-up. He obviously posted a link to this rather mysterious Rails port in PHP and spam or not, I’d like to thank this guy for letting me know of the existance of Akelos, a new PHP framework which seems simply too good to be true.

Let me just spend a few words more before writing more about it though. First off, it’s not available yet. OR at least it doesn’t seem to be: the author is planning to release his work to the Open Source community but… well, he’s a bit concerned about the current “PHP Framework War”: he wouldn’t like to end up like Subway or just be slagged off by those merciless reviewers who enjoy write round-ups and comparisons about frameworks. Bermi Ferrer is “just” a talented PHP developer who decided to create his own framework and he really enjoyed doing so, nothing more, nothing less.

“I considered other PHP ports of Ruby on Rails, but we could not find all we needed on them. One feature that I needed on the core was internationalization and Unicode support, so I decided to roll my own framework trying to keep most of the original rails interface so most of its documentation could work for it.”

Where did I hear that? Nothing new: it’s always the same story of frameworks not being as we want them to be etc., it’s human. And yes, it’s another attempt to port Ruby on Rails to PHP, and a damn good one — or so it seems.

Before proceeding any further, I’d like to write a short warning for a few people who may or may not want to pop in and start commenting about the Rails-is-better-than-any-PHP-clone issue: If I see a single comment slagging off this framework (or any other) only because it’s a port of Rails to PHP, it will be deleted, may it be David Heinemeier Hansson himself. Stop it, no seriously, I think it will be counter productive for Rails in the end: I really like RoR and I love the way it works, and yes, I think Ruby is definitely the best language to do that sort of things, EVERYBODY KNOWS THAT. Please, please, save us poor PHP developers the usual preaching.

Right, back to Akelos now. Curious as I am I immediately checked out the official page and all i found was a pretty long list of features which made me dribble, literally… ooops!

Let’s just quote the most juicy ones, shall we?

Active Record [“Model”]

Associations
Callbacks
Transactions
Finders [ $Project->findFirstBy('language AND start_year:greater', 'PHP', '2004'); ]
Versioning
…

Action Controller [“Controller”]

Filters
Pagination
Mime Type
Mime Response
Code Generation
Response handler
…

Action View [“View”]

Templates (using Sintags)
Web 2.0 javascript using prototype and script.aculo.us
Helpers
Partials
Template Compilers
…

And then more Akelos-only goodies:

Multilingual Models and Views
Locale alias integrated on URLS (example.com/spanish will load the es_ES locale)
Database migrations using DB Designer files
Pure PHP support for Unicode (no extensions required)
Unit Tested source code
Code Generators
Built in XHTML validator
Automated locale management
Clean separation from HTML and Javascript using CSS event selectors.
Ajax file uploads
AFLAX integration
Dojo Rich Text Editor
Format converters
File handling using SFTP for shared host running Apache as user nobody (as most CPanel server do)
Distributed sessions using databases

Impressed? I was, honest. And I’m talking as a CakePHP fanatic here, and I must say that if this framework can really offer all this and — and this is what really matters — is also as simple as Cake to learn and well performing… well, this is definitely going to be quite a promising player in the “PHP Framework War” (but is not a real war, is it?), although the author is quite worried about that:

“I’m also concerned about the PHP Framework war, I don’t want to play that game. Building this Framework was a great experience, it works great for me and it has helped me to become a better programmer so I don’t want to spend my time discussing about if this is better or not than other solutions. That’s the reason I’ll first look for great developers interested in the Framework to help me releasing it.”

Really, this will be an interesting project to check out, once it goes open source, and yes, I really wish Bermi all the best. Good luck, Akelos Framework.

Digg Effect - the day after

2006-05-05T01:59:00Z

…So it turns out that my last article appeared on Digg homepage.
This was quite a pleasant surprise: I didn’t expect that an article submitted to my own site could make it that far! I thought you’d need a relatively well-known website, mafia’s support, some divine intervention and a terrific amount of luck, but it seems that sometimes an interesting article about an interesting subject can be enough. I’ll probably write a more detailed report of what happened soon, in another article rather than a blog post, but for now I just wanted to post a short summary here.

Two days ago I decided to write a roundup of the six Rails-inspired PHP frameworks, CakePHP, Symfony, PHP on Trax, Code Igniter, Biscuit and Pipeline. The reason for this was that I couldn’t find anything comparing all of them and such comparison could have been useful for some new bakers_. OK, I confess, when I started writing the article I thought I’d submit it to Digg and see what happens: I saw that another roundup made it to the first page and people were quoting it everywhere on the net. It’s a nice article, but – in my humble opinion – not too exhaustive.
Then I read a comment by someone to the digg of the latest Cake release stating:

Yes, they are similar – both were inspired by Rails, but Cake has gone further to differentiate themselves. Here’s a decent (but not great) overview of some frameworks: http://www.phpit.net/article/ten-different-php-frameworks/

At that point, I thought that another round up, perhaps more Cake-centric, was in order. The other reason was that in one of my recent blog posts I tried to compare CakePHP and Symfony, but obviously my emotions got in the way and in the end I noticed I was kinda attacking Symfony. That was a blog post though, and that’s half-allowed, but I felt that I should have written a slightly more objective article mentioning also all the other competitors.

Anyhow, right when I went to submit my article to Digg, it turns out that another guy wrote a similar round up, which made it to Digg’s homepage. That was an annoying cohincidence, but in the end things didn’t go too bad: his roundup was more generic, while mine was more specific and detailed.

After submitting my article the reaction wasn’t instantaneous… 5, 7, 10, 13 diggs in the first two hours. Then shortly I made it to 30 and when the 40th visitor dugg it my article was moved to the first page!
I immediately noticed it when I refreshed my stats page: a minute before my girlfriend was here telling me “oh look, over 400 visitors… not too bad”. Then I refreshed the page and it said 539, I refreshed again and said 600-something… eeep… Digg effect!

A special praise goes to my new hosting company, BlueHost: the server didn’t go down and it managed the extra traffic fine! A good test for CakePHP as well, since I built this site with it.

So here I am… over 5000 visitors read my article, about 600 people dugg it, nearly 40 people commented it on digg.com and 20 directly on my site. And – except for the usual Rails-is-better-than-anything-else comments – they were generally positive. Over 250 people bookmarked on del.icio.us and many blogs mentioned it in many different countries.

Money? Didn’t make much with adsense at all: programmers don’t click on ads!
Bandwidth? About 1GB was gone in the first five hours, now is obviously slowing down: oh well, I still have another 398GB available till the end of the month :P

H3RALD - Tag 'web-development' (Atom Feed)

Akelos Framework: too good to be true?

Digg Effect - the day after