Live data from Hacker News

John Carmack pushes out unlocked OS for defunct Oculus Go headset

arstechnica.com

121–130 of 316 posts

Re: John Carmack pushes out unlocked OS for defunct Oculus Go headset

#122
post #64

Earlier quoted context omitted.

> the job market favors engineers hopping around instead of staying at one place Does it? That's a fair description of the consulting space, but I don't think it's an accurate representation of the job market as a whole. When product companies hire software engineers, they tend to aim for the long haul.

You'd think. Most career advice I get is you need to jump ship to get an actual pay raise. Maybe stay 3 or 4 years. Not enough to build up deep knowledge, at least compared to older engineers imo

I don't think it's necessary to stay with the same company a very long time to develop "deep knowledge". on my last team there were two engineers with about 20 years of experience. one had spent all of it at that company, and the other had worked at five different companies in the same space. the former knew more about the company's code and was an invaluable resource for "why are things the way they are?" type questions. but the latter often proposed more novel (to us) solutions, drawing on his experience from previous roles. both were extremely effective engineers.

Re: John Carmack pushes out unlocked OS for defunct Oculus Go headset

#123

Earlier quoted context omitted.

You'd think. Most career advice I get is you need to jump ship to get an actual pay raise. Maybe stay 3 or 4 years. Not enough to build up deep knowledge, at least compared to older engineers imo

A buddy of mine just joined Palo Alto networks and they gave him a 30% pay increase, as well as a $100,000 signing bonus, and then put him on a team who sold well and he got a $40,000 bonus a couple months after joining.

One of my friends stays at his current company where he has stayed for years, despite only getting low single-digit pay increases each year.

Re: John Carmack pushes out unlocked OS for defunct Oculus Go headset

#125

> allowing for a randomly discovered shrink wrapped headset twenty years from now to be able to update to the final software version, long after over-the-air update servers have been shut down. This resonates so much with me. Each time I setup a new device that requires an Internet connection, I think about how we can enjoy booting 30 years old retro computers and how the next generation will not be able to do the sa…

Do you mean locked-down software?

Do you have examples of hardware that the next generation might not be able to boot 30 years later?

I'm curious and ignorant :D

Re: John Carmack pushes out unlocked OS for defunct Oculus Go headset

#126

Earlier quoted context omitted.

I don't think we need a law, we just need businesses to stay greedy. DRM licenses are kind of expensive and it's hard to publishers to justify the added cost of DRM after the initial release, when they've made 80% of the revenue for the game. So they often (though, not always) get patched out eventually.

Depends how the DRM license is written. It could easily be a perpetual license for a given title. Then there is no incentive to remove it.

It could be perpetually licensed. Sure.

But there's a business case to be made for patching out DRM very early in the product lifetime. The longer a product is protected by DRM, the more likely the DRM is to be cracked, and that this crack could be applied to a new releases. It doesn't make sense to put a new game release at risk just to keep protecting some five year old game from piracy.

If you're selling DRM, you want to disincentive publishers for using your protection for too long. And if the DRM is built in-house, you want it to keep functioning for a number of titles to amortize the cost of development.

Re: John Carmack pushes out unlocked OS for defunct Oculus Go headset

#127
post #39
post #5

Earlier quoted context omitted.

The same is true of some of the best work being done on the web, such as the NYTimes interactive infographics. Because they are so JS-heavy, and reliant on CI/CD pipelines for deployment, on custom CMSes, there is no way to archive them in the way that static pages containing just text and images can be archived on the Wayback Machine. Heck, even Flash projects from 15 years ago still run fine when compiled on Ruffle…

I don’t know why CI or a custom CMS would be the problem here, the output is still static HTML. IMO the problem compared to years ago is that the client side dependency trees are so much more complex, involving third party domains and such. But dependencies are still a problem even in Flash: it can load external JSON data for example, and it’s a lot more difficult to sniff all that out than it is with JS. Don’t get m…

Surely someone is archiving apks, just for the fun of it? I can imagine that someday we'd have a webarchive that embeds a simulation of the device where you can run that apk in your browser. Of course, if it's just an application shell dependent on a running backend, then yes it's not really an application so much as an "application shell" at this point, and you ALSO need to be able to move around the data corpus to systematically capture all reachable application state. This is a hard problem in general, but pretty easy I think when applied to specific cases.

Re: John Carmack pushes out unlocked OS for defunct Oculus Go headset

#129
post #96

Earlier quoted context omitted.

Unfortunately, a lot of new releases need a 0-day patch to be playable. It's basically an anti-piracy/anti-early release measure. So if you really want to preserve a game on disk, you pretty much have to pick up another copy stamped at a later date to include all of the patches.

GOTY editions rarely need 0-day patches.

Right, that would be the "copy stamped at a later date which includes all the patches" version I was mentioning that you'd want to pick up if you own a new release copy.
Post reply on HN