Very important to highlight this; since joining a different big tech company I have quickly realized how much the internal processes (everything from expenses to security posture) are structured to put the maximum responsibility on the person furthest down the totem pole.
In small European companies I've worked for this would be considered a big red flag and a sign of broken company culture, but for US companies it is seemingly the norm.
> In small European companies I've worked for this would be considered a big red flag and a sign of broken company culture, but for US companies it is seemingly the norm.
Try working for a big European company, it’s no different.
Do you mean large EU companies are more similar to large US companies, or more similar to small EU companies? I presume the former, which would be a shame but not entirely unexpected since I imagine (in the tech industry at least) there is a lot of emulation.
As soon as you do business interbationally, especially in the US, you have to follow the same anti corruption laws, business conduct and so on. I guess that's one of the reasons big corps are so similar with regards to those trainings.
By your own argument, foreign companies can navigate the rules for selling into Europe just fine - so what is stopping a European company from doing the same?
If there is any remaining problem it is that the rules are not enforced strictly enough on tech giants.
An indie dev in New York does not care about the GDPR. The just build cool shit and put it online. Look at all the Show HNs here.
In the EU, the situation is very different. Indie devs are super afraid and work hard to make their stuff less useful to please the GDPR.
Now about larger players:
EU companies agonize their worldwide users with cookie banners. Because that is what the GDPR tells them to do.
Non EU companies dont do that. Because why should they? Will a lone Italian traveller in the USA sue them for using Google fonts? Probably not. And if they do - they can handle it. So they only agonize their EU users with cookie banners.
Unlike in UE, EU companies are not allowed to sell their customers data to random third parrty spammer, scammers, adtech companies and bounty hunters.
That is it - as a startup, GDPR is not a massive prohlem. You know what is a real problem? The fact that you can raise 10x more investment in the US with the same slide deck.
Small correction: the indie devs just building cools shit are fine under GDPR. The indie dev who wants to monetize his community while not caring about the externalities of possibly leaking their information has a headache.
If your business model depends on creating undesirable externalities for your "users" then you don't have my sympathy. The only shame is that we still need to enforce GDPR properly on large players, but that's a political and social thing, not per se a problem with the law itself.
And the oh so horrible cookie banners: the solution would be to not track people. If you aren't fully acting in the users interest, the cookie banner is easy to implement, or maybe not even required. So whenever you are annoyed by a cookie banner, it should be directed at the company, not the law.
Until they start saving PII that is not necessary for their core business relationship with the user.
If your app is literally about self quantification and the user pays you to collect that data and keep it private? You might not even need to state it anywhere, although the safe thing is of course to list all the ways you do or do not collect data.
If your app is about self quantification and you monetize by selling user data or its aggregates... GDPR. If you use a third party data provider instead of hosting the data yourself: GDPR etc. Because user data might not be important to you, but it is to your users, so you probably shouldn't be allowed to YOLO handling it
> you probably shouldn't be allowed to YOLO handling it
Exactly. If a company doesn't care enough about its users to even tell them what they are doing with their data (or in some cases even know what they are doing with it) then the user can't expect that company to secure it or to provide a valuable service with it.
Think of it this way: by starving the supply of that one bitcoin, you have contributed in some small way to the eventual loss of all bitcoins through similar events - speeding up the rate at which the world can move on from this silly fad.
> I do find it odd air-to-air source heat pumps are not considered more often in the UK
Could be due to housing density and the reputation for noise, though I'm not sure how well-deserved that really is.
Mostly I just think it is a mismatch between the age of housing stock (and the pre-existing heating solutions) with incentives.
If I were building a new house I wouldn't bother with a gas supply and would use a ground-source heat pump combined with underfloor heating - but retrofitting that to my 150 year old house is never going to be cost effective versus a boiler.
For residential developments, the government should just mandate no new domestic gas supplies and be done with it.
Underfloor heating has an 8 hour turn on/off lag time. It's nearly useless out here in the SF Bay area, since we go from short sleeve to jacket weather every day, almost year round.
Our air source heat pump is much quieter than old air conditioners. For efficiency, it has a giant, slow, variable speed fan. It's usually not running anywhere near full speed, and is still quiet when it does. Similarly, the blowers in the ducts are variable speed.
The only appropriate response here is for a corporate owner (Hashicorp preferably) to fork the repo just before the license change and to assert that their version is the default, and drive future development there.
The original author is entitled to their opinion and can restrict their own development to a repo with license conditions they define, but it is not reasonable to expect companies to accept non-standard and politically charged license terms. Getting shit done is hard enough and this change will not move the needle on Ukraine/Russia sentiment one iota.
You are asking a corporation to clone "a community-driven project" and maintain it. Ok. There are 2.7K of forks of just this module already if you want to use them.
I'm unsuccessfully trying to go the other way - but that's because I was never a developer to begin with.
Writing code (or even working with computers at all) is not something I ever aspired to. I put this down to the impact Microsoft software had on IT education in the UK in the 90's - computers were for miserable people with pallid faces who never smiled, poking around in Access databases.
Now, I work all day every day on a computer because a career as a scientist doesn't pay an acceptable amount. However, since I never studied computer science or did and significant programming early in my career, I've circled around the field doing sysadmin and project management work. I find myself too senior to take on the kind of programming tasks I can reasonably handle - there is no time allowance for me wanting to learn some of these skills in more depth. Doing personal projects and changing jobs could get me into writing code, but only with an intolerably large pay-cut.
The sad thing is, my field (scientific computing) needs exactly the skills I am aiming for - but since no-one will pay sensibly for that kind of work, people either steer around it like I did, or just jump sideways into DevOps or other "retail" IT.
Try it out on a local Linux machine first, if you have one. There are plenty of ML techniques outside of neural networks which train perfectly well on a CPU, so I'd start there.
Look at some kind of AutoML framework like AutoGluon, then dive deeper on the components it uses once you've got through the initial setup process. AutoGluon will let you train some basic models with all the data cleaning and normalisation steps handled for you.
My regular M1 MBP is comfortably managing 18 hours of normal use (non-video browser and terminal) or 12ish hours of more intensive use (video and Slack). Sounds like putting the M1 Max in a laptop form factor is a bit of a waste, as I've never even strained the base level processor.
Very important to highlight this; since joining a different big tech company I have quickly realized how much the internal processes (everything from expenses to security posture) are structured to put the maximum responsibility on the person furthest down the totem pole.
In small European companies I've worked for this would be considered a big red flag and a sign of broken company culture, but for US companies it is seemingly the norm.