PHP alternative for Ruby on Rails?

[Luke Redpath]

I'm still confused as to why there are all these attempts to port Rails to other languages...just learn Ruby and be done with it. It's not hard. If Rails was better off being written in a different language, it would have been in the first place.

I love Ruby On Rails, and see myself using PHP less and less in the future. Ruby is just so elegant and the Rails framework is excellent, even at such an early stage. I think people will be making a mistake to write Rails off...I think its going to be the biggest thing in web development for years.

And maybe in 2 years from now, Ruby will be quite attractive for business. But right now it's no

Tell that to Epson, Bank of America and all these people

[33degrees]

Ok, I don't want to upset anybody. I am thinking more on commercial side of things. We all have to eat . How many projects are made in Rails ? How many people choose Rails over Java or PHP or Python or Perl (I know there are some, but not enough). Of course, it has gathered a wonderfull comunity around allready. And that's very good. And maybe in 2 years from now, Ruby will be quite attractive for business. But right now it's not. And that's imaturity (unless proven wrong a language is considered imature from the start).

Ok, I agree with what you're saying, but I think calling the language imature because of this is misleading. It's the market for ruby development that's imature, not the language itself. But I can remember a time when the same was said true PHP, and look where it's gotten now.

[Dr Livingston]

Umm...

ionami has been working with EPSON for over 3 years. While we created the original EPSON Developers site in Java, they were surprisingly open to a switch. We sold them on maintainability, speed, and the lower overall cost. We took the opportunity to do a redesign, and move the original, public site to Rails for ease of management.

This is a bit of a surprise for me, as was the other link. The unexpected uptake of Ruby by the number of people and companies (via above links) if it's to be believed is something that isn't to be sniffed at.

If the uptake is there, then this could be a direct threat not only to PHP but also Java in the medium term (3 years+). I'm still not convinced as a developer to take an interest in Ruby at the moment, but it has stirred something inside of me, and I don't like the feeling...

What I mean is that Ruby has recently got a lot of good press and the hype machines are still going around the clock, promoting a relatively unknown technology. And you have all these people taking it up and yet as far as web development goes, it's (still) untested in a great number of areas.

I'm still wary, and so I would say not to drop PHP so fast, that you take up Ruby and then have to backtrack later. The impression I get is that Ruby is said to be perfection, and as we all know, there isn't such a thing in this world.

[meryn]

just learn Ruby and be done with it.

It isn't that simple, Luke. There's only a small chance that a developer's current application deployment host supports Ruby, or will be supporting it in the near future. Not all developers get to choose their hosting themselves...

Also a company might very well have a policy regarding which platforms to use. Many companies still have a Java-only policy, and approving even just one 'scripting language' platform would be a huge step forward. Big companies have just gathered themselves behind PHP.

If Ruby (and Rails) is 'the future' (because of it's technical advantages), many developers will still be 'stuck' with PHP for a long time. The market doesn't move that fast.

Personally, I think it's clear that Ruby wins hands down on language features. But what about the available extensions? I think that has always been the strong point for PHP. I just read an article about how hard it can be to get image resizing working (can't find it, it was on a blog of a new Ruby programmer). That's almost a default for PHP.

The application platform one ends up with depends on lots of things. If it would only be language features and the availability of a powerful framework, I think there would be a major shift to Ruby right now. (I don't mean early adopters, but the early majority)

I think its going to be the biggest thing in web development for years.

That could very well be, though I'm not sure how it will play out in the market.
A good Rails clone on an other platform could influence the future drastically, I think.

Now, the Ruby platform is extra attractive because of Rails, but what if you don't have to go to Ruby to get the 'power of rails'? Then you would switch only on language features. I think this changes the equation quite a bit.

In my view of the market, PHP has currently the best industry acceptance among dynamic languages (followed by Perl and Python, and then (currently) WAY behind: Ruby). I'm not sure where ASP.NET is positioned, I think the Microsoft world is kind of isolated from the rest, it's sort of a monoculture. A rails clone on PHP could secure its lead for the future.

Funny thing is that I don't know if there will be a good rails clone in PHP. It seems Ruby has attracted much of the innovators / early adopters and highly motivated developers. Ruby currently has the most powerful developer community behind it (in the form of the Rails community). The PHP community is very fragmented in comparison...

The majority of Ruby web programmers works with Rails , but not even 1% of PHP developers knows about CakePHP... Or so I think.

It'll be interesting times in web development.

[meryn]

To strengthen my point:
The fourth hit on Google for "Ruby" is Ruby on Rails.
The fourth hit on Google for "PHP" is .... PHP-Nuke.

For all things, PHP-Nuke!!

[Luke Redpath]

Well, its my opinion that if you snub RoR now, you'll regret it in the future.

I don't buy the hosting argument - for the majority of people this shouldn't be an issue. If your current host (assuming shared hosting) has no plans to implement support for Rails, then move hosts. Don't lock yourself into a host that has no flexibility. Textdrive already seems to be proving itself as a good shared hosting option for Rails.

And then there is dedicated hosting...if you, or your client are running off a dedicated Linux box, then again, support isn't an issue.

[Luke Redpath]

As for Ruby's modularity, there are plenty of things available on RubyForge.

In fact, only yesterday, I was working with some hefty XML data, and I thought...wouldn't it be nice if there was some way of automatically/semi-automatically mapping XML to native Ruby objects to save the hassle of dealing with XML parsers (such as REXML) directly. I figured this would offer much of a similar advantage to working with XML that things like ActiveRecord offer when working with databases.

A quick google, and I found this:

http://xml-mapping.rubyforge.org/

Image processing? Use RMagick, Ruby's binding to the ImageMagick.

I'm already finding Ruby a joy to work with and this is outside of the realms of Rails.

Finally, whilst I see myself using Rails more and more in the future, learning Ruby and using RoR doesn't mean you have to stop using PHP. In fact, I imagine I will be using PHP(5) for a good while yet as we are just about to deploy a huge PHP-based kiosk project...it's already been through several rewrites (ASP, then a straight port to PHP, then finally a huge refactoring job and move to PHP5/OOP (from the horrible procedural code it started off as). As much as I like Rails, I feel no desire to rewrite from scratch again.

[meryn]

then move hosts.

Sure, if you can choose your hosting yourself. I was talking about situations where that isn't the case. I'm convinced that many 'independent' developers will switch to Ruby very soon, as many already have.

OTOH, a client or employer who restricts a developer shouldn't complain about longer development time / higher costs... That would mean there isn't a big market for phpCake, because all the smart developers who have the 'freedom to be productive' will switch to Ruby no matter what.

phpCake would then only serve a niche: Smart developers who want to increase their productivity, but are somehow dependent on - or have significant investments in - the PHP platform; (custom) libraries or extensions that aren't available in Ruby.

[meryn]

learning Ruby and using RoR doesn't mean you have to stop using PHP.

A pragmatic programmer indeed.

[1PhpNut]

I'm still confused as to why there are all these attempts to port Rails to other languages...just learn Ruby and be done with it. It's not hard.

I am not porting Rails, as I said before the concepts are being used.

Cake is a MVC Framework that makes it easier for people whouse PHP to create applications fast.
I am doing it because people are using it, and communicating with me things they want to see.
This is what motivates me.

Am I doing it for money, haha, I have spent more money on this project than any other I have worked on.

Each developer that has joined the core team has received a copy of the Zend IDE to use while working with me.
We have a dedicated server with the Zend Platform also. All of these things to improve the coding and code.

I have more people contacting me about CakePHP also.
Our mail list is growing.

If Rails was better off being written in a different language, it would have been in the first place.

CakePHP is not RoR
It is a MVC/RAD Framework. People who use it and like it will continue to do so.
Do some people feel that this little project is going to put RoR out if business or something?
Every comment that is posted seems to be negative about this.
I use both Ruby and PHP, and others languages also.
I like what we are doing, and it is damn fun!!!

[Luke Redpath]

Do some people feel that this little project is going to put RoR out if business or something?

Not a chance, I can't see it ever getting as I think Rails will. But I'm not looking to criticise any particular port of Rails in any language - I just think between the choice of porting something to a different language, and learning a new language, I favour the latter, if the language is as good as Ruby.

[BerislavLopac]

Well, reading all those stuff about Ruby and RoR popped a few things into my mind. I'll share them, i no particular order:

1. Comparing PHP with Ruby, Python or Java seems just a bit odd. Basically, PHP is a Web development platform, which happens to have a very powerful programming language to implement the logic in, and that language is available as a scripting tool in some environments. On the other hand, the mentioned languages are fully fledged general purpose languages. It might not be comparing apples to oranges, but perhaps oranges to grapefruits.

2. There are only two things that helped make PHP so popular as it is today. One is the fact that it easily provides scripting abilities to Apache server, thanks to the fact that it is primarily a Web development platform; high-volume WSPs could give their customers a server-side scripting abilities for next to no additional cost. The other reason is the fact that it is very simple to use: it was the first platform that I know of which enabled one to mix logic code into the HTML, with which simple sites and apps could be done extremely fast. It took some time until similar platforms, like ASP or JSP, were developed for other languages, and they often needed special setups (Tomcat can sometimes be quirky, thank you very much). PHP also got more complicated and advanced over the years, but its original simplicity is still there. (And it might even be an obstacle to its advancement: things like the dropping of namespaces in PHP5 or the recent "controversy" over the references bug might be the first signs of it becoming to large for its skin...)

At the time of PHP's ascent there was another similar language widely available on most Linux hosts -- Perl -- but it was nowhere so easy to program and deploy as PHP, which is why most newcomers chose the latter. Today we have more similar languages (yes, I know that some are nearly as old as PHP, but have only recently became increasingly popular), from which it would seem there are two that are getting a significant following: Ruby and Python. But they are, unlike PHP and like Perl, general purpose languages, which for the long time have had no dominant Web application platform. (Yes, there is Zope, but even the Python buffs admit that it's a beast of its own kind.) If it is really so great, RoR might prove to be the killer app for Ruby, and we might easily see more of it in the future.

3. Ruby or not Ruby, that is the question. Or, rather, Ruby or Python? I've had my eyes on those two languages for some time, and still can't decide which should I try first. Whichever I choose, the odds are that I'll like the first so much that I won't even feel compelled trying the second, or that I'll find the experience so bad that I'll curse myself for losing so much time and refuse to try the second fearing another time-waster.

Note that I wouldn't like this to become a Ruby vs Python (vs PHP) deathmatch, as I have see to many of those. I guess I'm waiting for a miraculous solution which will tell me whether should I point myself towards LARR or LARP...

[Ryan Wray]

Not a chance, I can't see it ever getting as I think Rails will. But I'm not looking to criticise any particular port of Rails in any language - I just think between the choice of porting something to a different language, and learning a new language, I favour the latter, if the language is as good as Ruby.

Which is fair enough to say, a different thing to do. If I wanted to start coding in Ruby tommorow I may have to (general, certain items might not apply to me):

• Scrape, port or find alternatives to libraries I have written my self.
• Find hosts that have Ruby as an option (big problem if you are licensing you're code out, not many hosts support Ruby).
• Spend a while learning the languages features, it behaviour, etc, then more time to figure out it's quirks (which every language will have).
• Train developers working with me in Ruby
• Etc, etc, etc

Ruby might not suit everyone, while the Rails framework might. Like I would write in Perl just to use DBI (before PHP5 had PDO), doesn't mean I wouldn't like that feature or something like it.

[Luke Redpath]

I might agree if you are talking about the odd component here or there, but Rails is a full stack so the point is you shouldn't need to start porting over all of your own libraries - but like I said already, there is plenty of stuff out there if you look for it. Don't tie yourself so strongly to one language. As for not many hosts...there are hosts, and at the end of the day, you only need one reliable host right?

How hard should it be for an experienced programmer to pick up a language as relatively simple as Ruby?

I'm sure people were saying all the same things about PHP when it first arrived on the scene too. Don't be so resistant to change. Embrace it.

[bonefry]

If Rails was better off being written in a different language, it would have been in the first place.

You basically say there aren't new things left to invent How much experience do you have with Rails ?

This is a bit of a surprise for me, as was the other link. The unexpected uptake of Ruby by the number of people and companies (via above links) if it's to be believed is something that isn't to be sniffed at.

Wow, a couple companies want to try Rails and the whole world is allready amazed. I think in a week from now I will loose my job to a Rails developer. But seriously now. Big and smart companies never rely on a single platform to avoid platform lock-in. That's why there is room for both NET and Java, that's why there is room for PHP on the presentation tier, and that's why people are experimenting with Rails.
BTW: if you look at the macromedia.com site you will find that ColdFusion is used by 75% of the Fortune 100 companies. But I for one, don't see many ColdFusion applications around.

If the uptake is there, then this could be a direct threat not only to PHP but also Java in the medium term (3 years+). I'm still not convinced as a developer to take an interest in Ruby at the moment, but it has stirred something inside of me, and I don't like the feeling...

No, not really, remember the initial .NET hype back in 2001. Java doom days have comed and passed a couple of times allready Multiple languages they said ? I think I had the same feeling of "I don't believe it's worth it".

I don't buy the hosting argument - for the majority of people this shouldn't be an issue.

Really ? I had trouble finding a good Python hosting. I had trouble finding a cheap Tomcat hosting.

...I'm convinced that many 'independent' developers will switch to Ruby very soon, as many already have.... OTOH, a client or employer who restricts a developer shouldn't complain about longer development time / higher costs... That would mean there isn't a big market for phpCake, because all the smart developers who have the 'freedom to be productive' will switch to Ruby no matter what.

75% of all developers have switched to Rails allready and we are still waiting for reports on this weekend . Microsoft is bankrupted and Sun has shoved Java up their as*. IBM as ussual switched sides and has rewriten it's entire Websphere line for Ruby. An IBM official said that Ruby in the long run should provide better web services support and it's free candies for everyone who joins IBM vs RBoss debate entitled "were are we actually going"

Let me tell you something. Smart developers have always had the freadom to be productive (words like Python, Eiffel, Lips come in my mind). And they always did it at home. And at work, managers always made the best decisions for their companies (or for them) imposing to programmers languages like COBOL, or C++, or Java. Because you know why ? Smart developers aren't always interested in details, team work, finalization, i18n, bugs and others things like that. Smart developers are interested in doind "smart things". And the same analogy goes for stupid programmers. They simply do stupid things. And it's a manager's job to combine the "smart things" along with the "stupid things" to make something usefull .

Oh, and BTW: If Ruby on Rails succeeds, it will be over this hype created around it by programmers who bash other technologies just because it's fun to say "my new technology looks cleaner than yours" without any real arguments. That means it won't deserve it, and it will end up being hated.

The arguments on this thread are getting more and more religious. Someone should move this post from the PHP section. PHP is no Ruby.

[bonefry]

Not a chance, I can't see it ever getting as I think Rails will. But I'm not looking to criticise any particular port of Rails in any language

Please tell me something. You saw how people around here actually use a pattern called "unit testing" with PHPUnit and SimpleTest. They are quite efficient in NET too and in any languages so far. But it all started with JUnit from Java. Yes, JUnit is the daddy. Does it make it worst on all other platforms ?

Actually it would be a serious blow to Rails developers pride to see their praised technology being used on other platforms without the efort of learning of yet another language.

[BerislavLopac]

Yes, JUnit is the daddy.

Actually, AFAIK, sUnit was first.

[bonefry]

really ? cool

[Luke Redpath]

Please tell me something. You saw how people around here actually use a pattern called "unit testing" with PHPUnit and SimpleTest. They are quite efficient in NET too and in any languages so far. But it all started with JUnit from Java. Yes, JUnit is the daddy. Does it make it worst on all other platforms ?

Actually it would be a serious blow to Rails developers pride to see their praised technology being used on other platforms without the efort of learning of yet another language.

It's a very strange comparison that compares a full stack MVC framework to a unit testing utility.

[sweatje]

Stumpled across PHP on Trax, which also proports to be a rails port.

[33degrees]

There are only two things that helped make PHP so popular as it is today. One is the fact that it easily provides scripting abilities to Apache server, thanks to the fact that it is primarily a Web development platform; high-volume WSPs could give their customers a server-side scripting abilities for next to no additional cost. The other reason is the fact that it is very simple to use: it was the first platform that I know of which enabled one to mix logic code into the HTML, with which simple sites and apps could be done extremely fast. It took some time until similar platforms, like ASP or JSP, were developed for other languages, and they often needed special setups (Tomcat can sometimes be quirky, thank you very much).

This is really the main reason why I (and probably most people) use PHP; it's cheap, ubiquitous, and easy to learn. But I really have no love for the language itself, and the list of things that annoy me about it grows day by day; it's simply the language that best suits my current needs.

If it is really so great, RoR might prove to be the killer app for Ruby, and we might easily see more of it in the future.

I don't think the hype would be so strong if it wasn't at least partially deserved. RoR has definitely proven itself in some situations, some figures:

Basecamp - 400,000 dynamic requests a day without caching. "It’s currently handled by two web/application servers that each run 15 FastCGI processes and 50-100 Apache 1.3.x processes on dual 2.4GHz Xeons with 2GB of RAM. These machines usually sit between 0.5 and 1.5 in load." The DB server is shared with other 37S projects including BackPack and Tadalist

43things - 200,000 dynamic requests per day. Same three server set up but with 3GHz CPUS. "Load on the servers rarely exceeds 0.3 and CPU idle time is usually in excess of 80%."

And one you probably haven't heard of, rapidreporting.com - 300 requests/second though the system was tested with 3,000. The cluster has ten servers. The application uses PostgreSQL for the database, lighttpd on the web server, and around 10 FastCGIs per application server sitting behind a virtual server with IP tunneling"

Certainly powerfull enough for my needs! And powerfull enough for Justin Gehtland, who is the author of several Java Books; Here's an interesting post he made comparing the performance and size of an Java app he rewrote with Rails.

3. Ruby or not Ruby, that is the question. Or, rather, Ruby or Python? I've had my eyes on those two languages for some time, and still can't decide which should I try first.

The main difference between Ruby and Python is Ruby's everything-is-an-object approach; it's a pure OO language whose syntax allows it to look like a procedural language. Python is a procedural language with OO added in. IMO, this gives Ruby a big advantage over Python (and an even bigger one over PHP).

For more insight regarding this, you may be interested in reading this comparison.

[bonefry]

The main difference between Ruby and Python is Ruby's everything-is-an-object approach; it's a pure OO language whose syntax allows it to look like a procedural language. Python is a procedural language with OO added in. IMO, this gives Ruby a big advantage over Python (and an even bigger one over PHP).

You shouldn't believe anything you read. Literals in Python are objects. Functions in Python are objects. Classes in Python are objects. You can do metaprogramming in Python. Python has very good reflection. And if OOP was added later, then certainlly they did it right. What is that OOP purity you are talking about ? What features are missing from Python that are so important ? The only advantage of Ruby are those code-blocks. And that's it.

[bonefry]

It's a very strange comparison that compares a full stack MVC framework to a unit testing utility.

I was saying porting ideeas has always been good. Because it's the ideea that counts, not the implementation. It's the mistake you just made. Instead of thinking about it, you stumbled upon on what I was comparing.

[BerislavLopac]

The main difference between Ruby and Python is Ruby's everything-is-an-object approach; it's a pure OO language whose syntax allows it to look like a procedural language. Python is a procedural language with OO added in. IMO, this gives Ruby a big advantage over Python (and an even bigger one over PHP).

Yes, I have read quite a lot about both languages on the Web, and am aware of most of the differences.

However, what you emphasise doesn't seem like like such a crucial advantage -- there is nothing wrong about a language being both object and procedural, it's up to the programmer how it will be used. The fact that a language is OOP-only really doesn't prevent you from writing procedural spaghetti code -- I've seen a whole Java app written as a single procedure inside a single static code block. Don't ask.

The thing that is most on my mind is an apparent disparity between the ability to write code quickly (it's faster to use symbols, as Ruby tends to) and the clear and understandable code (unless you are very experienced in a language, it will be easier to understand verbose commands and keywords. I would like my code both to be written quickly and to be easily reviewed by someone who has extensive programming experience but is not necessarily an expert in the language at hand.

Off Topic:

Actually, if I really had a choice, I'd do all my programming in Javascript. But it's still lacking a good platform -- Whitebeam just doesn't strike me as the way to go...

[bonefry]

Off Topic:

Actually, if I really had a choice, I'd do all my programming in Javascript. But it's still lacking a good platform -- Whitebeam just doesn't strike me as the way to go...

You're in luck, because JavaScript will be the first scripting language integrated in Java in the Mustang release.