[ameister14]

Why are there so few rails devs? Its not as if Ruby is super hard to learn or the rails framework is incredibly complex.

I think I would hire someone for 40k for their first year, train them in RoR and then pay them more their second year. You get 6 months of good work out of them after they're totally trained and at a lower price.

[patio11]

That's an attractive option iff you're training non-programmers and, when you say "pay them more", you are prepared to more than double their wages after they reach journeyman proficiency in Rails. If you expect people to have a CS degree prior to year 1 or react to a 10% raise in year 2 as generous, you would not find the Silicon Valley hiring market offering you as many candidates as you desired positions.

[ameister14]

I know that; Let's say it takes 6 months of training to be fairly proficient. you have 6 months after that at a proficiency worth 42k, but you've only paid 40k for the entire year. Then when you raise the salary to 84, you have someone you know works in your company, who is happy with the huge raise, and you've gotten a bargain out of it.

The problem is you have to think a year ahead, but I don't think that's such an issue.

[mechanical_fish]

How are you going to keep this hypothetical honor student down on the farm? [1]

In other words, once the employee has the skills to earn $85k, why should she wait six months for her giant raise when she could just switch employers and get the giant raise right away?

Not that this is the reason why the fantasy scenario doesn't work - if it were reliably possible to turn a raw recruit into DHH with six months of training, the fact that we'd have to pay market rates the entire time would be the least of our problems. Indeed, it would not be a problem at all.

A real reason why more companies don't try training up raw recruits is that (a) running a school is a specialized business and (b) the yield is far less than 100%, so it only works at scale. You need to admit - optimistically! - 10 or 20 bright and motivated nonprogrammers in order to graduate one person who, after six to twelve months of training, can be expected to successfully attack problems like:

"A customer has called; he has an obscure problem with his web site. Help him debug this problem over the phone. You have no access to the code or the server."

or

"Here's a legacy codebase that spans 267 files, two major versions of Rails, and three generations of programmers. Improve it. Don't break it, though, because our revenue depends on it."

or

"Here's a collection of 175 cloud instances running in Amazon. Build a system that reliably backs them all up once a day, with no downtime, and that can verify on demand that those backups exist."

or even something as "simple" as

"Here is a Wordpress site with a handful of specialty plugins installed. Here is an empty Git repository. Fill this Git repository with Rails code that implements a site that looks and acts exactly like the Wordpress site."

[1] http://www.azlyrics.com/lyrics/andrewbird/howyougonnakeepemd... - this song is now 95 years old, so I guess I better footnote it!

[bambax]

I may have missed a step in your reasoning but why are you assuming they have to start with a "raw recruit"?

Why couldn't companies hire actual developers with no previous experience in {some_tech} or {some_framework} and train them in said tech/framework?

It seems companies, just like recruiters, feel safer when they hire a "rails dev" for a "rails position" when in fact a good programmer with zero rails experience would be much preferable to a mediocre programmer who has some rails experience.

PS: the song is excellent but I couldn't find any recent cover; this old guy with a Banjo is killing it: http://www.youtube.com/watch?v=FlUExq2hNBU

[mechanical_fish]

I think hiring existing developers and training them up is a fine plan. And a lot of companies do just that.

But it works best if you have an existing team, trained in the technology, of sufficient size that it can afford to spend a little time mentoring, and with a long-term outlook.

There are teams like that who use Rails. But, as it happens, Rails is also a favorite tool of brand-new startup teams who are already critically overworked and who probably won't be around in six months. And it's a favorite tool of one- and two-person teams: When a one-person team loses a developer, the new hire has no mentor available. So there's plenty of openings for pre-educated Rails developers.

Mind you, there's also the cynical explanation: It's more "efficient" to advertise for "rails devs" because you can screen the resumes so much faster. Those who can't answer trivia questions about ActiveRecord syntax are out! This would, of course, be dysfunctional. But the whole job-ad system is already so dysfunctional that this particular bit of dysfunction may be lost in the noise.

[rmc]

But developers know that they can learn on the job. So an offer of "half regular salary, but we'll train you!" is not too appealing. Many companies will just pay regular wages. So we're back to where we started.

[bambax]

You're right. What I meant is that companies trying in vain to find rails developers should hire generalists, regardless of rails experience (at a competitive, normal salary according to the skills/experience of the recruit).

It seems companies only advertise for specific languages or frameworks; maybe they think they need someone capable of hitting the ground running, but if they spend a year finding them... it's pointless.

[ameister14]

I agree, the yield would be a major problem.

In order to solve the problems you mentioned, it's confidence and a particular mindset you need, not simply skill. I had overlooked that.

I think you could actually keep employees, though, with the right management. There are lots of reasons she would wait six months before the giant raise. Money isn't actually the main motivator for a lot of people, and with the knowledge that she is appreciated and people depend on her as well as the fact that she will be better paid upon contract negotiation at end of term, she's more likely to stay.

Oddly enough, lots of people want to have a career. They just don't expect to get loyalty from a company anymore and so don't give a company theirs.

If training is specialized, why not pay for people to go through some external training program in exchange for contract? I know some companies do this now for Rails and it's common for other skills.

Master craftsmen have apprentices. It's a model that works, that's all I'm saying.

[rmc]

So you're suggesting that the company change how it does employment to have more old fashioned, long term employment. But how do you convince someone of that? Everyone is saying "We're different, we'll look after you!".

You could do it with contracts & law. Offer 10 year salaries-index-linked-at-a-minimum employment contracts. Employment contracts that specifically state the employer won't do things like outsource, (or if you're in USA won't do 'at will'). Employment contract that'll give 5 times the legal minimum of redundancy leave. Require that upper level management/directors don't get bonuses if there's been any pay freezes or pay cuts. Lots and lots of little things like that might be a way.

This will (obviously) restrict the business, and limit its resaleability. ("Come buy us out, you just have to pay 50 people for the next 8 years").

[ameister14]

To be honest, a lot of it is about the character and credibility of management. If you work for a guy who is openly against outsourcing, you can generally expect them not to outsource. If you work for someone who was hired by the board to cut costs, you can expect them to do that in any way possible, even if it harms the company long-term.

Yes, contracts play a role in this and it would definitely limit a business' sale price. But the thing is, if you're looking long-term, why would your exit strategy be acquisition?

Money is the main motivator in financial services. People don't really have any other metric to use for their performance, so they are paid a lot. In other industries, though, you have many alternatives to pure financial reward, and people are actually happier.

I don't know if this would entirely work or not. I'm just saying that it makes sense to me and that many people value stability and appreciation for their skill a great deal.

[jdminhbg]

But now you have to hire a Rails trainer as well as a Rails dev!