On Sunday, 30 June 2019 02:36:46 UTC+2, andrey mirtchovski wrote: > > > I will let Andrey speak for himself. > > Since this is turning into a bit of fisticuffs I will quote my private > message to you for clarity, here it is: >
Like for Brexit, I think a new survey will be surprising: a lot of people have presumably realised that the consequences of tackling the verbosity of the current iteration of error handling comes with a big price tag. That was, I assume, not raised in the survey, at least not clearly enough. There is much that can be discussed here and Go has made it clear to me that for every correct outcome there may be far many more erroneous ones and that exceptions will continue to demand more code than the expected algorithm flow. I agree with Daniela that breaking the rules for exceptions is going to hurt eventually, if not immediately. The try pseudo function, even though it has precedents in new and make, looks wrong. My peek into the abyss says we're looking at the wrong problem, it is the *return* statement that needs to be enhanced and I have commented, perhaps not seriously enough, in that respect. Also, I have personally never used the *panic/recover* paradigm, but I suspect that if someone within the Go team were to write a clear exposition of how to use it in the rare case where it may serve as a *try/catch* construct, Robert (Engels) and others may well be able to offer even more guidance on where Go2 may be able to go to satisfy the clear necessity to summarise a set of errors into a single contingency handler good enough for the most challenged among us. Maybe that is not possible, maybe we need to iterate in that direction, rather than try to make a quantum leap. But if C++ and Java are going to wither away, them the message "if you want X you know where to look for it" is going to be no longer valid for exception handling. That will put pressure on Go and at that point and the outcome under pressure may be horrific. And, yes, I am not entirely serious about this :-). Go is not just simple, it is mostly "elegant". Go2 must not be allowed to become a second generation design, by democratic dictum. -- 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 golang-nuts+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/6ef60612-ee9b-4d34-b5b6-aade1f4db80a%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
