On Fri, Dec 1, 2017 at 3:44 PM, <matthewju...@gmail.com> wrote: > > Here’s an explanation behind my assumptions about Go: > > Recently I encountered a crash in the latest stable version of Go that > blocked my development and was root caused to a mistake in how pointers are > handled as map keys by the runtime. The fix was beyond my short-term ability > and leaves me with 1.8.5 until the next release; as a Go application > developer I am dependent on the language implementation experts who, for > quick support via the github issue tracker and for fix acceptance via the > code review process, seem to mostly be Google employees. I am not aware of > any Go forks.
It's worth noting that the key step in the analysis of that issue (#22781) was from Alberto Donizetti, who does not work at Google. It's true that the fix was made by a Google employee. > While I recommend Go as a general purpose language (an improved C) I also > think that convincing management of a commercial company that this > relatively young language will always have support will be a tough battle, > and more groups of people need to commit to contribution for the language to > grow past this barrier. I apologize if my assumptions are incorrect, but my > understanding is that Go would not exist if Google did not pay for it > initially, and if Google removed support (such as by privatizing Go 2) the > public would have a compiler and specification that while great are not yet > as mature as C++ or Python and would no longer have daily support and does > not have a clear organization or set of organizations to inherit it. > > My assumption is that Google management will look out for the company's best > interests even if that means going against what its employees want, that > Google has the capability to own the language on their own, and that the > critical path of Go is primarily developed today by and for Google. Open > source does not mean perpetually supported and updated in reasonable time > for commercial use, although Go seems to have support available for an > indefinite future. > > For wider adoption perhaps a blog post series showing who is contributing > would be helpful to break these assumptions? There are quite a few people who do not work at Google who have commit access to the Go repositories. For the current release cycle more than half of the changes were made by people who are neither on the core Go team nor work at Google. You're right that if Google suddenly reassigned all the Go team members to some other project that it would not be clear who should carry Go forward. Personally I've occasionally pondered whether a Go Foundation should be created at some point, along the lines of the Linux Foundation or the Python Software Foundation or similar organizations. But I don't think it is necessary or useful at this point in the language's development. (And I have to say that it is extremely unlikely that Google would suddenly reassign all the Go team members; Go is a small project, and at the level where such things would be decided it seems to me that Google is barely aware that Go even exists.) Ian -- 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. For more options, visit https://groups.google.com/d/optout.
