3 ms·
But Peter's method, as he himself said, isn't good enough for software that would be released by a company for people to use: > Thoman says the difference betw
by pavel_lishin 13y ago
But Peter's method, as he himself said, isn't good enough for software that would be released by a company for people to use:
> Thoman says the difference between his fixes and official patches are primarily an issue of quality assurance. “When I write and program something, I make sure I didn’t make any obvious errors, and then I just release it,” Thoman told WIRED. “If it works for 90 percent of people, I already consider that a success.”
> From there, Thoman starts working for that other 10 percent, but he says that there’s probably one percent of people that he will never be able to help. If he were in charge of a commercial product rather than a fan-made patch, he’d have a responsibility to make sure whatever he released worked for everyone and didn’t break anything.
- MichaelGG 13y agoWhat's wrong with having having a checkbox to "Enable experimental renderer" and let people tweak the settings there? Certainly that's better than restricting your entire customer base because some of them don't have the right setup.
- pavel_lishin 13y agoFrom a product management perspective: it adds complexity, and it adds other things for people to complain about your product. People might complain that your car's top speed is 60mph, and that the acceleration is crap, and that is bad for your car and company's reputation. It would be much worse if there was a switch a driver could flip that would double the speed and acceleration, but would soimetimes eject the windshield, exhaust pipe, and door handles.