On Mon, 8 Oct 2018 at 16:30, Ian Lance Taylor <[email protected]> wrote:
> On Mon, Oct 8, 2018 at 4:50 AM, Scott Cotton <[email protected]> wrote: > Your real concern > has to do with running audio code without interruption. That concern, > too, is abstract. We don't know how well it works today and we don't > know how well it will work in the future. Let's focus on real > problems. The lack of cooperative preemption is giving us real > problems in real Go programs today. That is why we are trying to fix > it. If the fix causes problems for audio code, then we'll try to fix > that. But let's not get ahead of ourselves; let's first see if there > is a problem. > > Ian > There is documentation about the "reality" of the problem is on the issue tracker. If there is disagreement about that I would find it more productive to address why in light of the arguments and documentation present on the issue tracker, rather than simply to assume that it is not real here (and well after the fact). In any event, I take my experience, research, and interactions with others as sufficient to assess they are real. I also find that all assessments directly to the contrary or responses which ignore the problem to come from people with far less experience in the matter (at least as far as I know) than those other sources I have consulted and probably also myself. That said, I too took your stance about a year ago, said to myself, let's wait and see if it's a real problem. It is quite reasonable and really the only practical entry point into tackling the problem anyway. But asserting a priori that well established guidelines from the field are not to be treated as "real" without backing up that position is in my opinion a risk with long term consequences. -- Scott Cotton http://www.iri-labs.com -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
