Item — HN Reader

As much as I appreciate the momentum behind Rails, I cannot stop reminding me that if I go with the herd and do whatever most people are doing I will have to find my competitive advantages in places other than technology.

And I think that would be a huge loss for me.

3 comments

This seems to be a common thread for engineers building webapps.

Its not to say that you're incorrect, but if you're building a webapp that doesn't have huge special needs with regards to performance or some special technology sauce then your technology isn't a competitive advantage.

If you have something that can easily be built in Rails then I have a hard time thinking of a way to get a real competitive advantage working in something else. (On the flip side, Twitter was easily built in Rails and it was very much the wrong tool for them).

In fact working with well known, ubiquitous tools that get the job done well and fast is in itself an advantage. Being able to rev faster and easier then the shop down the block with their super cool custom built web framework that takes people 6 months to learn is an advantage.

Unless your technology IS your product, compete somewhere else.

It's not about doing stuff in platform A that can't be done in platform B. It's about delivering things faster than your competition, something that can't be done if you use the very same tools they use. You can, of course, be wrong and go with the wrong tool, but remember - it's a bet on a technology differential.

The Twitter problems stem not from the web framework they used but in flaws in the design of the data structures under it. When doing something like Twitter, one should never even consider using relational databases for more than a prototype.

Not only that, going with the herd would actually mean going with PHP and not Rails. There are dramatically more applications written in PHP and developers working in PHP than with Ruby on Rails.

I am not sure. Rails seems far better positioned than PHP for future growth. This means PHP will peak and Rails will surpass it eventually.

"Rails seems far better positioned than PHP for future growth. This means PHP will peak and Rails will surpass it eventually."

If this comes to pass I will be completely and utterly stunned. I don't expect to be stunned.

You do know that PHP usage has continued to increase, all throughout RoR's rise to prominence (and subsequent fade in prominence as Django, Erlang, Haskell, Scala, etc. shuffle in and out of the spotlight), right?

There are more PHP developers today than there were three years ago when Ruby started getting folks attention. Book sales are up for PHP while they're down for Ruby, etc. I'm not at all suggesting that RoR isn't awesome or that it isn't alive and well or anything of the sort. I just know the reality of the market. We have thousands of web developers using our software...and I can assure you that the vast majority are working in PHP, despite the fact that we went to great lengths to add a nice RoR environment a couple of years ago (we also added some Python niceties, but it aint exactly mainstream, either). In fact, I'd probably even go so far as to say that more new PHP developers came into existence during the rise of RoR than new Ruby developers. So, not only do I think Ruby on Rails won't surpass PHP eventually, I think it's been falling further behind PHP all along and will continue to do so, just like most languages in the web space.

I don't have to like PHP to recognize this. Believe me, I don't like PHP. But, I'm not going to bury my head in the sand and pretend like it is anything but a massive phenomenon. It's simply the unstoppable juggernaut of "worse is better", when it comes to web software development. Will PHP eventually die? Sure. But it's not going to be killed by Ruby on Rails.

PHP will eventually peak. Rails is almost certainly not the replacement. The replacement will have to be at least as easy to start with and as convenient as PHP, not less so.

I say that's a big maybe. For all it's strengths, Rails requires more programming skills than PHP... PHP will always be better for the "web designer turned scripter" set, just because it's easier to get results with if you aren't really a programmer by trade.

> PHP will always be better for the "web designer turned scripter" set

Funny, I think the same about Ruby/RoR, it just seems like a perfect crowd/match for "web designer turned scripter" mindset - looks like it is marketed to the same crowd too. PHP mostly marketed itself to "script kiddies", albeit not on purpose. It gained a huge following, and with it came lots of inexperienced kids that could "do stuff" easily. Mainly installing phpNukes, at the time, and modifying phpBBs etc. To say that Rails requires more programming skill than PHP is a moot point considering that developing web application, whatever the nature, could be done in both Ruby/Ror or PHP (with from scratch work or some framework) - thus it relates to the same problem set. Why would solving the same issues in one language/framework require a larger skill set then? It is as if you are saying RoR programmers are l33t, PHPs aren't - which doesn't make any sense, because I believe there are many great programmers and programs written with both.

"I think the same about Ruby/RoR, it just seems like a perfect crowd/match for "web designer turned scripter" mindset"

And that is probably one of the greatest strategic weaknesses of Rails - it will be used by those who shouldn't use a web framework, who will make every silly mistake and that could tarnish Rails reputation beyond repair.

Not necessarily. As you can witness, PHP attracted lots of "newbs" and guys that generally don't have a clue, or are just installing drupals, joomlas and whatnots while proclaiming or being proclaimed as PHP gurus. It did get a badge of being a shit language to php, like visual basic (which rightfully deserves so IMO - php does not), but it did fine. Culling of the masses that don't have a clue happens spontaneous, and usually follows them being hoarded into a separate camp - compare to the "normal" hard working php guys and joomla, drupal stuff.

Am I the only one completely underwhelmed by Drupal? The UI is awful and the demo doesn't even have a decent editor. Whomever revamped Wordpress needs to get in touch with the Drupal folks yesterday.

"PHP will always be better for the "web designer turned scripter" set"

It all depends on who you want to build your web applications, the guy who knows how to build them or the guy that knows how they should look.

Ideally, I'd want one of each.

Agree. I let the programmer write code and the designer create the looks. Hiring one to do both is a cost-cutting measure I am not ready to take

I guess that depends on what you're trying to accomplish. Often one wants to be first to market or add new features quickly. The author states that he was "getting a hell of a lot done in not much time at all." That means the front end of the business could be built quickly, accomplishing the main goal of satifying the customer as quickly as possible.

Now there may be another framework that does it better, if you can find that and use it, you will outrun your competition.

I chose PHP for my first startup, because it was a really fast way to get things done, warts and all, and it seems many feel that way. But I'm going to check out Rails this time because it has so many people getting things done with it.