I don't see why Google couldn't do that today.
I'm not saying they are morally obligated to do so. At the end of the day they are a for profit company and they don't have to do anything at all about this problem. And, sure, the donations they are making instead are probably better than nothing. But as I mentioned, and as top19 mentions on what is as of this writing the top post on this article, the track record for these job retraining programs going back decades isn't very good.
What I'm saying is that if Google were to go to Pittsburgh, Youngstown, or Detroit, hire some bright unemployed people, pay them a decent but by no means exorbitant salary while it taught them how to code and then, as you point out, how to be a productive engineer at Google, at the end of the process those people would likely have very marketable skills to either continue moving up the ladder at Google or elsewhere. Of course this would cost Google something -- not only in the salaries while people were learning and not contributing but also in the salaries of people that were training and mentoring them. But Google is planning on spending a billion dollars anyway, so here's another way they could do that. A way that I think would be more effective.
--
Congrats on getting what sounds like a great job.
Rather than hire a traditional Stanford CS graduate, we find people who have self-taught to some basic extent and give them the chance to work under an apprenticeship. We started with a class of about 25 with decent results and will tune it more based on what we learn
This made sense back then because most of the tools, languages and techniques were either proprietary or niche. IBM wouldn't hire an ATT employee as FB hires a Google employee today, and back then people saw jobs as "until retirement", with a very different culture.
If you invested a lot of money training someone, you could expect them to stick around for a while and "pay back" your investment.
Today that isn't worth it. You could hire someone out of high school, spend a lot of resources training them just to have them poached by another company right after they are "ready".
That time is gone.
But here Google is looking to donate $1 billion as a charitable contribution to address a specific social problem. I'm suggesting an alternative way of "spending" that money. From that standpoint the fact that such a training program is sub-optimal in a bottom line sort of way isn't as important -- indeed it is kind of the point. The difference between the optimal business focused process and such a hiring and training program would be in lieu of the donation that Google is planning on making.
It's my contention that this way of spending money would yield more benefits than the current plan. I could well be wrong, but just criticizing the idea from a business efficiency angle misses the point.
That does make sense and I think that would be super cool if companies could do that today.
I admit I don't know much about the history but isn't building software today more complicated though not necessarily harder than it was before?
My impression is that before we would just build programs that ran on computers. So in that sense, if you knew how to program and had a compiler to use, that was sufficient to do what companies wanted to do. However, now we have people building other things like services, which is requires additional complexity and tooling on top of what was previously necessary.
The point I'm trying to make is that I suspect nowadays there's more prerequisite knowledge for most jobs than before.
> But Google is planning on spending a billion dollars anyway, so here's another way they could do that. A way that I think would be more effective.
That's a really good point. I imagine even if it takes a very long time to train people, with a billion dollars you have the time to do that..
The optimistic side of me wants to say it's because handling the logistics for that would be a nightmare and it's really outside of our core expertise.
The realistic side of me thinks that it probably is too expensive to give the same benefits to those employees that we give to current full time employees.
The pessimistic side of me also thinks that there's a certain level of prestige associated with working at Google that has been, at least to some level, successfully marketed both inwards and outwards, and hiring people who don't meet whatever "bar" would undermine that.
> Congrats on getting what sounds like a great job.
Thanks!
A more reasonable path is HelpDesk / Hardware Tech -> operations software / SysOp -> software engineering
I work for Google and I find there's already a very large amount of on the job training required due to all the powerful but complicated internal tools. My manager has told me that I should not feel pressured to contribute at all for my first 6 months and I should feel free to just focus on learning as much as I can.
I've been here for almost a year now and feel like I still know nothing so the training is still certainly not done.
I didn't mean to jump to the defense of my employer as I'm obviously biased but do think we are far from perfect (although I am very happy here :). I just wonder what the right balance is. It seems like if people can't program at all then it's not really on the job training because they wouldn't be working right?
At least my own experience has been that I came in with some college experience but no college degree and not really knowing anything besides the bare minimum to contribute (being able to program, having some grasp of CS fundamentals) and have learned a ton on the job and still have much much more to learn. I think there is an expectation (and pressure!) for all SWEs to hit the senior level, L5, so in a way everyone who is hired under that level is doing on the job training no?
Sorry for the long post!