On 2013.03.14 05:28, Xiaofan Chen wrote: > Thinking a bit more into this, do we really need 2.0 so badly if it > is only for a few "nice to have" API changes? I personally > do not think so. It seems to me that this could be a testing > branch instead.
I don't particularly care if it's flagged as testing or experimental. But I'd like to have some flexibility with regards to where we take libusbx next, especially if we're going to handle hotplug, without having people like Greg KH complaining about the smallest of changes, or without being forced to devote a lot of time trying to find another trick to get a new square peg fitting into an old round hole. With Hans meant to pick the 1.x branch after the 1.0.15 release, I don't exactly see an issue with this. > Pete mentioned the following in the other reply. > "What's more, if we are only going to consider priority, and index it on > what our users request, we should probably drop everything else > (including releasing 1.0.15) and devote all of our time to hotplug. This > has been our #1 feature request for years (that we got another request > for that today is no exactly a coincidence)." This was only meant to be interpreted as an example of an extremist view we could take, if we went purely with what people want to have us prioritizing first. This was done to highlight the danger of trying to be too rigid with regards to rules. Of course we don't want to do the above. > I think we should release 1.0.15 and then focus on hotplug. I'm planning do work on hotplug _as part_ of 2.x, on account that I don't want to be constrained by our existing API if I find we can integrate hotplug better by reworking it. But I'm also not going to prevent anyone working on 1.x hotplug if they want, and I'm ready to help them backport any hotplug code from 2.x into 1.x. It's of course very possible that the changes between 1.x and 2.x will remain minimal, but it's also possible that they may not. Plus, I still have some hope to work on a proposal for a process based enum sometime this year, to remove this annoying need we have to constantly reenumerate. Even if we somehow agree this is what we want, I doubt it's going to sit down to well with the existing 1.x branch... > We can drop 2.0 and rename 2.1 to 1.0.16 if it is agreed > upon. You're not going to get my vote on that, sorry. Also, if your concern is about trying to speed up our release process, I don't really see how having or not having a 2.0 is going to matter one bit. Regards, /Pete ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_mar _______________________________________________ libusbx-devel mailing list libusbx-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libusbx-devel
