Unless I've misunderstood this article, I think this is a bad idea. When working for a company your work should ideally stay at work. Come 5pm, if you're privliged enough to work a 9-5, your day should be done.
This growing push for work to be extended into the family home really erodes the necessary work life distinction that allows for a happy life. Thats not to say you shouldn't have a lab at home to train yourself on things that interest you in your own free time, but if staying at a job means investing unpaid hours at home then one should probably consider whether the job is worth it.
After all, if a company expects you to do unpaid training in your own personal time then what will it expect from you in the future?
Furthermore, these articles constantly push the message that more work is better for your career than less. My father always told me its not the quantity of time you spend at a task but the quality.
It's less about what your company wants you to do, and more about what you need to do to develop your career. If you want to stay on top and keep being employable, you'll have to invest in skills, and I see no reason why your employer would provide that for free, especially if those skills aren't directly useful to them.
Company's job isn't to help you find a work-life balance. Its goal is to find the long term plan that extracts the most work out of you. It can involve giving you some cool perks to keep you healthy on the long run, or it can involve churning through new hires all the time and overworking them while they last. Depends on their strategy.
This mentality towards training people is common, but is a textbook case of stupid management. Companies under-invest in training, then pay enormous recruiting costs in getting people with the "right skills" as well as paying opportunity costs when projects can't go forward because they can't find the "right people." It's a bit of brain-dead management that's common in some other fields, but acutely common in software/IT.
> there is only one thing worse than training (or growing)
> your staff and having them leave, and that is not training
> or developing them and having them stay.
Yes, I think this is the spot on point. You don't need a home lab to do your current job, but you are going to need a home lab to prepare for your next job. Because your employer certainly won't be shelling out so that you can leave (well, probably won't anyway).
My current job involves me writing code for robots. This sounded interesting to me, until I realized that I am working with some ancient software here - I'm talking .Net 1.1 on Windows 2000! Not to mention that writing software for robots isn't really all that challenging.
If it weren't for my after-hours exploration of open-source technologies, my skill-set would be severely outdated, and my career would effectively be dead here.
I work with something very specific as well, and completely missing out on the web and what's hot. I have to do something in my spare time to ensure I'm employable. But I'm not going to do job-related training at home, screw that.
Yep, because no one wants to change the world like Steve Jobs did. Nope, everyone should have your values and value family. If they don't something is wrong with them.
That's not at all what I said, nor what I was implying. Further it's asinine for you to claim that anyone who wants to have cut off time is the antithesis of someone who wants to change the world.
> unpaid training in your own personal time then what will it expect from you in the future
...it's not unpaid training, it's lifelong learning, and you do it because it's fun! The purpose shouldn't be to learn about what your employer wants you to or what your employer might need from you. You shouldn't need to justify having a home lab to your employer or family or whatever. The article's author got it all wrong indeed. Fuck your employer, you have your "lab" 'cause it's fun!
Companies that won't invest in training or research are not worth working for. If they want you (as an employee) to pay for that out of your own pocket and you do, then you're a sucker, unless you've negotiated a fat salary and a 4 day week.
Many companies will pay for training to enhance your value to them but may not always pay for training which enhances your value to you, in part because occasionally (good! ethical!) companies have a plan for your career which is not the optimum plan for you. In this case, it makes sense to pay for training personally when the day job won't subsidize it, or otherwise invest in one's own professional development. (Via side projects, OSS contribution, yadda yadda whatever capital-building exercise you can licitly do outside the confines of 9 to 5.)
For example, my old day job happily dropped $10k to fly me to JavaOne so that I would more effectively map XML files and JavaBeans together, but "Most efficient cog in an Enterprise Java consulting shop ever!" was not a long-term career aspiration of mine. They were pretty understanding about providing for career growth during work hours but when they didn't I did it on my own time. That eventually cost them my services but, hey, capitalism happens.
Generic career advice for everyone: You are the captain of your own ship. You will have mentors, employers, and other folks who you are to some degree reliant on, but at the end of the day you have both responsibility for creating and authority to mandate forward progress.
True enough. And employees who wait to be "trained" on interesting new technology instead of learning it for themselves aren't worth hiring.
Obviously neither statement is (completely) true. At the "rockstar" level, all learning is self directed. And rockstar employers expect and rely on that. But for everyone else I think it's reasonable that there be some level of support for education from employers, and some level of self motivation from good employees. Neither is going to make up for a total shortcoming in the other.
> And employees who wait to be "trained" on interesting new technology instead of learning it for themselves aren't worth hiring.
> At the "rockstar" level, all learning is self directed.
This is a bit of brain damage that is, as far as I can tell, limited to programming, for various weird cultural reasons.
Go to Goldman Sachs and ask them how many of their "rockstar" traders learned the craft at home on their free time. Or go to a top research hospital and ask them how many of their surgeons did the same. Or go to a top architecture firm and ask them how many of their architects learned skyscraper design on the weekends. They will probably all laugh at you.
Rockstars can be created through institutional processes by identifying, nurturing, and systematically training talent. All industries have their rockstars, but ones more mature than the software industry have saner and more efficient ways of taking raw talent and nurturing them into rockstars.
"Go to Goldman Sachs and ask them how many of their "rockstar" traders learned the craft at home on their free time. Or go to a top research hospital and ask them how many of their surgeons did the same. Or go to a top architecture firm and ask them how many of their architects learned skyscraper design on the weekends. They will probably all laugh at you."
The difference in all these fields is that the practitioner, to learn, needs to things that cost significant amounts of money (trading/building skyscrapers), cause people to live or die (surgery) or go to jail or be executed (law), or need to do things that require significantly altering the real world to create an artifact(setting up skyscrapers). Hence the need for supervision, and systematic training with a lot of 'apprenticeship' feel.
In most programming, the real rockstars (say Carmack or Linus or whoever) are self taught, and did learn on their free time, because all you need is a laptop and the internet and you are ready to go and creating a significant artifact (a programming language, a game, an operating system) needs only knowledge (easily acquired from books and the internet) and time.
Not to take away from your larger (and valid) point that company training can be effective. But your analogies are a bit 'off' and that affects the power of your argument (imho, feel free to ignore).
The comment to which I was replying stated: "At the 'rockstar' level, all learning is self directed."
Parse this out logically. I read this as: "self-directed learning is a necessary condition for achieving rockstar level skill." My point is that in other fields, institutional training can create rockstars without self-directed learning. That is to say, self-directed learning is not a necessary condition.
Your point is that programming is different from those other fields, because you can become a rockstar learning on your own. I.e. that self-directed learning can be a sufficient condition for achieving rockstar level skill. But that does not contradict my point, which is that it is not a necessary one.
"The comment to which I was replying stated: "At the 'rockstar' level, all learning is self directed." "
In context, that quote sounds more nuanced.
"Obviously neither statement is (completely) true. At the "rockstar" level, all learning is self directed. And rockstar employers expect and rely on that. But for everyone else I think it's reasonable that there be some level of support for education from employers, and some level of self motivation from good employees. Neither is going to make up for a total shortcoming in the other."
I suspect if you add "(In programming)" before the sentences "At the rockstar level .. " etc it would make more sense,especially if those sentences are describing present reality than some imagined future.
And as for "the fact that it's easier to learn programming on your own than it is to learn other fields doesn't prove that", sure it doesn't prove anything.
But then if we are to descend to that level of hairsplitting, your quote about other industries training programs doesn't really prove the point that something similar would work in software/programming, and is equally 'neither here nor there'.
It may be true that you can create rockstars by company training, so you have a valid point in that we should probably explore the idea, look for supporting/opposing evidence and so on.
It still remains to be shown to be true,and ajross has a valid argument that "as of today company training programs haven't produced programming rockstars (vs competent devs)", so your comparisons to surgeons or whatever don't really counter ajross's statement.
ajross seems to be describing the present reality (and extropolating off that) and you seem to be saying that a different reality is possible (but unproven).
Not much of a conflict that I can see. (but hey ymmv, I freely admit I could be interpreting all this incorrectly)
And just fyi, I responded to (what I thought was ) excessive rhetorical flourishes like 'brain damaged thinking' in your reply. That phrase conveys no information beyond your annoyance and detracts from your main (and valid imho)argument. You can make a solid point without 'name calling' (note quotes), especially here on HN.
EDIT: rayiner changed his post from when I typed this reply(which is all right, I do that all the time), so this answer may not make as much sense as it should . Now his point is that "self-directed learning is not a necessary condition (to become a programming rockstar)".
It is probably true and certainly an idea worth exploring.
One possible answer to that, especially in a forum called Hacker News, is "Show us".
(to those fond of logic, to prove that a universally quantified argument, "All X s are blah" is wrong, you need to prove that "Not all X s are blah" and that in turn becomes "There exists a Z that not blah" and you have to show that there does indeed exist such a Z - aka an existence proof, what can I say, I'm a nerd).
And with that, I bow out of this thread. Have a nice day.
> But then if we are to descend to that level of hairsplitting, your quote about other industries training programs doesn't really prove the point that something similar would work in software/programming, and is equally 'neither here nor there'.
It's not hair splitting, it's the fundamental question of whether self-directed learning is necessary to achieve rockstar-level programming skills. This seems to be commonly assumed in the programming world, which I think is a bit of cultural brain damage. Indeed, I think the proposition that programming is somehow different from those other intellectually demanding fields in trainability is the exceptional proposition that has the burden of proof.
plinkplonk already pointed it out, but I'll say it again: this fact is commonly assumed in the programming world because it is empirically true. If that's "brain damage" then you have a strange definition for that term. We observe facts ("at the rockstar level, all learning is self directed") and make hypotheses to support a theory ("self-directed learning is necessary to achieve rockstar-level programming skills"). If you want to assert a theory counter to that, then I'm sorry but it is you that needs to provide some evidence to support it.
And seriously: there is none. There are no "trained" rock stars out there that I've seen. Can you find just one? Someone who through diligent classwork or apprenticeship reached a top-tier position in their field? I've never seen it.
First, even if it's empirically true that there are no trained rockstars in the programming field (which is a point I'm not going to concede), that doesn't mean that "at the rockstar level all learning is self-directed." There is an obvious difference between "all the rockstars I can see are self-directed learners" and "self-directed learning is necessary to become a rockstar." Asserting otherwise is a basic failure of logical reasoning.
My theory is that programming isn't any different in this regard from other intellectually demanding fields except culturally. I think the assertion that it is different is the one that requires proof, or at least an explanation for the mechanism by which it might be different.
As for self-directed learning, I'll point to Dennis Ritchie as an example. He's arguably somewhere in the middle between self-taught and trained. On one hand he started at Bell Labs concurrent with/immediately after Harvard. On the other hand, UNIX was something of a side project. Compare him to his father, who was a scientist at Bell Labs working on switching theory, and received his training in the AT&T engineering machine. What's different about the two men? Why is Dennis's field not trainable but Alistair's field trainable? My assertion is that there is no difference except the fact that there was an established framework for training engineers but not really one for training programmers.
As a 4th year medical student I can tell you almost all training happens on your own time. Its up to you to stay current. After the first 2 years of med school you are pretty much on your own to learn the rest of medicine. They'll test you along the way...but passing is far, far less than you actually need to know.
Interesting discussion here. I've always had a home 'lab' because I've always been interested in a variety of technologies. At college it was half my desk and a couple of carrying boxes, (one for tools, one for parts). Once I got married I had a desk in the spare bedroom which had other stuff as well. Eventually I got a room to myself in the house. Back in the dot.com days before it went tango-uniform I was going to build a basement (machine room, library, office) but alas that was not to be.
Since learning is my hobby and my passion it serves me well. But I could certainly understand it if you hated what you worked on and wanting to get away from it when you were at home.
His $1,362 per month spend (not including hardware, which it sounds like he gets for free) seems a bit spendy for just a hobby "lab". Since when was a rack of computers called a lab? Does a few servers in the closet count? I want to know if I'm in danger of losing my job.
The main problem with my rack is the noise - I live in a flat and there's nowhere but the living room to put it. I've seen a few noise cancelling designs around, but they're all too expensive.
I wonder if I could build a simple noise cancellation system with an FX-LMS algorithm?
I have six IBM 2U servers in my home rack (36U). They're really well-designed and the noise level is not too bad even in the room with the rack, but there is also a 1U SuperMicro server in that rack with two Tesla cards and it is so loud as to be quite annoying anywhere in the house. I keep it turned off unless I'm specifically working on it, which isn't how I'd like it to be. The lesson I learned is that bigger servers (4U would be even better) have bigger fans and run more quietly. Also IBM servers aren't as loud as Dell and Dell servers are quieter than SuperMicro. I'm thinking the next addition might be a cluster of Raspberry Pis.
If you can replace the dinky fans servers usually use with 120mm units, you can all but solve the noise problem. It would be totally custom, yes, but probably easier than sophisticated noise cancellation.
They're 1U and 2U servers on which I've installed ESXi. I love it - I can run a VM locally then transfer it across, then if I'm travelling I can chuck it onto VMWare Fusion on my laptop.
If an employee is going to have invest effort into a significant home lab, it'd make sense to go the extra mile and start using the lab to earn an income, reducing reliance on said employer.
Yep, having your own personal lab allows you to do what you want and can actually be a sort of pretest bed for applications and/or hardware you want to recommend for your employer.
To get true understanding of what you're working with you need to be able to trash the configuration of your gear at will. I learned a lot buying up some big old Cisco Catalysts and EMC Clariion SAN boxes back when I was doing data center work. Later on though, when I was in charge of buying new stuff I'd rack the stuff up first and go crazy with it before it was in production. Either way works. I don't miss having those fans running in my house though.
The really nice thing about labs these days is how much you can do virtually. Before I changed jobs, I had 2 NetApp filers doing replication, two EMC Clariion CX arrays, and two complete VMware datacenters doing DR failover, vMotion, and DRS. All on two PC servers in a downstairs closet.
"Girlfriend Impact" -- bleh. Yeah, I know, 99.99whatever9% of IT workers, esp in the ops role, are guys, but would it have taken so much extra thought to put "family impact"? Or even "balanced life penalty"?
And I say this as someone who has spent a ridiculous amount of money on tech books in the past year (probably four figures) and who is giving up 1.5-2 nights a week, on average, to tech meetups.
I'm pretty sure he was referring to his own personal "Girlfriend Impact". It was not a general statement about all IT workers, just the one who was talking about the lab.
This growing push for work to be extended into the family home really erodes the necessary work life distinction that allows for a happy life. Thats not to say you shouldn't have a lab at home to train yourself on things that interest you in your own free time, but if staying at a job means investing unpaid hours at home then one should probably consider whether the job is worth it.
After all, if a company expects you to do unpaid training in your own personal time then what will it expect from you in the future?
Furthermore, these articles constantly push the message that more work is better for your career than less. My father always told me its not the quantity of time you spend at a task but the quality.