keekerdc.com

essays and pithy thoughts

On coding boot camps

in notepad

Today’s economy for code writing talent, particularly in the web segment, is driven primarily by simple supply and demand. Now that ‘webmaster’ has become a rather antiquated job title, and organizations understand the value that someone with a solid grasp of programming fundamentals can bring, there’s too few coders with verifiable chops to fill all the open jobs.

So as any economist would expect, prices for code writing talent have gone up. Way up. Up to the point where NPR last week published a piece titled ‘Twelve Weeks To A 6-Figure Job’. And hey, for those of us in the industry, that sounds like pretty good news; demand is so high that, as long as you don’t torch bridges along the way, you are almost guaranteed to be able to find new work on-demand.

For someone who was an educator in a past professional life, there was a vaguely troubling part of that NPR piece: ‘Twelve Weeks’. That trouble gained greater clarity after this great post from my former colleague at LivingSocial, Travis Valentine, came across my timeline the same day. He arrived at his position through LS’s Hungry Academy, which was a half year program.

Throughout the post, the one thing that kept coming up was time. How much time and effort it takes to simply feel like you have core mastery of the fundamentals of the job is one matter; how much time it takes to get over the ‘imposter syndrome’ and not feel like you’re taking the piss every time you sit down at your desk is another. That timeline definitely doesn’t square with twelve weeks. From Travis’ perspective, it doesn’t even square with 26 weeks.

The place of humility and clear-headed realism that Travis writes from is a far cry from the hubris and opportunism that has typified the other side of the code writer supply problem. The very notion at the center of sales pitches for these heavily accelerated courses — that a student can go from zero to mid-level developer pulling a six-figure income in roughly 400 highly-consecutive hours of instruction or less (!) — is at best highly misleading, and at worse highly predatory. While there are exceptions (this link being probably the largest), the vast majority of marketing around bootcamp courses promises that a ridiculously well-paying job is waiting for you after the season turns. Oh, also after $10,000.

Even assuming the best of intentions, the glut of companies that have sprung up to capitalize on this market imbalance will essentially eat themselves in their rush to push students through, and leave the profession worse off. Flooding the market with flocks of green coders, folks who must land work after dumping their previous gig for an unreasonably short immersion program, will only bring on a sharp and early correction to this imbalance and bring starting salaries back down through reasonable levels, into a territory where it would no longer make sense for people to take the plunge.

Telling eager new developers they can “start a career as a front-end developer” in just twenty three-hour sessions spread over ten weeks saddles them with an imposter syndrome that could take years to get over, and simultaneously saddles organizations that hire them, on the backs of certificates from unaccredited for-profit institutions, with junior teammates that can’t possibly be ready to ship code at the quality and pace expected.

Everyone eventually loses at the current pace of things.

Not that I’m particularly sympathetic to the fate of the vast majority of bootcamp upstarts when the overall message they send to the business world at large is this: experience in the field is practically worthless. And judging from the vat of ice-chilled beer that was made available at the single course I ran at General Assembly in DC, they’re fine with amplifying the worst stereotypes of the industry’s culture to fill their classrooms.

Why should an organization pay a premium for someone who has honed their skills over the better part of a decade when they can buy a kegerator and hire two bootcamp grads, who we’re assured will have the same skill-set and won’t balk at longer hours, for the same cost? That sort of organizational calculus has always existed in this field, sure, but now there’s a $100-million industry whose existence more or less rests on validating that equation.

I suppose what I’m trying to say is that my sympathy rests with the new developers getting squeezed through these machines, who have a really difficult lot set for them that nobody’s being up-front about. This new educational apparatus is not serving anybody well, and is especially failing new devs.

I’m not arguing that the academic world has kept up with and is serving the development industry well. It hasn’t, and it isn’t. Even ten years ago the gap was evident and widening. But it’s irresponsible to fill that gap with courses that are shorter than most professional projects; and equating, implicitly or otherwise, a twelve-week course with a four-year degree.

I’m also not arguing that we should be discouraging people from joining our industry. We desperately need people like Travis. A return to normalcy in our labor market will make it a hell of a lot easier for everyone, from startups to large companies, to fill vacancies with a reasonable amount of effort and at reasonable salaries.

But we also must demand that the programs established to help fill these jobs run closer to twelve months than twelve weeks. It’s a pedagogical imperative. Basic competence in this field takes time. It takes space between instructional sessions to decompress and process information; to reflect on mistakes and corners-painted-into. The larger technological revolution that allows us to reinvent entire industries will not allow us to reinvent the way or the speed at which the mind absorbs new concepts and skills.

We need to call these ridiculously short ‘kickstart’ courses what they are: cash grabs. We need truly serious, vocational-style training programs, not the coding equivalent of Dr. Oz’ latest fad diet. We need things like Turing School. We can probably do without the rest.