Re: Someone using COMPAT_SVR4(_32) ?
Maxime Villard <[email protected]>
| Newsgroups | gmane.os.netbsd.ports.sparc64 |
|---|---|
| Message-ID | <[email protected]> |
Le 01/08/2017 à 21:20, Dave McGuire a écrit : > Actually, in the long term, yes it is. If you take the time to > actually read what I wrote, and at least make some effort to understand > the point I was trying to make, this might be more clear. Take your > current activity relating to COMPAT_SVR4 and extend it to the rest of > the code base. We'd have no OS left if everyone just took the easy > route and chopped out anything that has a bug in it. It appears that you too would be well served to make some effort to understand the point I was trying to make. Following your argument, we would never remove anything, and we would keep stockpiling dead code. That's nonsense. Several compat options got removed in the past, some got added, some will be removed in the future. That's a cycle, and there is no reason for svr4 to be any different. >> Again, if it "simply" needs to be fixed, why didn't you fix it? > > Ah, finally the predictable response. I was beginning to worry. > Yeh; sorry, but I'm indeed one of the ones that spend time improving code quality, and yes, when someone comes in and wants to tell me what to do and how to do it, I'm automatically tempted to ask them why they don't do it themselves. >> Finally, removing this piece of code has little to do with shiny stuff. It >> has to do with making the code base clearer, in such a way that several of >> our ports are easier to maintain, and as a result, more functional. > > You misunderstand. I wasn't accusing YOU of only wanting to work on > shiny stuff. Rather, I was describing why COMPAT_SVR4 remains unfixed, > and other areas of the code become stagnant, while new features get added. Yes, I understood that. What I was implicitly telling you, is that as a result of removing dead code, the code quality of the functional parts is increased, in such a way that even if a potential feature is gone, there is a benefit for the users and developers. There is no "letting people down to work on more shiny stuff" - people benefit from clean code. But that's out of the point. I'll remind you that my initial question was whether someone could show a use case. Maxime