(unable to comment to this without loosing my temper. So... no comment) On Mon, 10 May 2021 at 05:49, Ian Lance Taylor <i...@google.com> wrote:
> On Sun, May 9, 2021 at 8:33 AM abebeos <lazaridis.com+abeb...@gmail.com> > wrote: > > > > To me this sounds quite like an "disorganized mess, where bullies, > abusers and even IT-fascists can thrive". > > > > It is clear to me that some gcc project maintainers, the steering > committee and bountysource are crossing ethical (if not legal) boundaries. > > The GCC project maintainers and the steering committee are definitely > not crossing ethical or legal boundaries here. > > I don't know anything about Bountysource. Bountysource is completely > separate from GCC. It appears from your link that John Paul Adrian > Glaubitz posted a bounty for some GCC work. A number of people and > organizations supported the bounty, but the GCC project itself did > not. Although the work is for GCC, the GCC project has nothing to do > with that bounty. That is handled entirely by Bountysource. > > > > The Issue: > > > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92729 > > > > The Bounty (a bit higher than $7K) > > > > > https://www.bountysource.com/issues/84630749-avr-convert-the-backend-to-mode_cc-so-it-can-be-kept-in-future-releases > > > > The Complaint re Voting Process: > > > > https://github.com/bountysource/core/issues/1532 > > > > Bountysource may write whatever they want in their terms-of-service - > the relevant law is still above. And of course OSS-ethics, which are more > that a basic code-monkey-mentality of the kind "only code is work, only > patch authors are workers". > > > > * there is a bounty > > * I start work (working around a major gcc project deficit, which is a > missing CI, testing testing testing, concluding, "reviving" and existent > patch) > > * I claim 50% > > * a dispute starts, which is then aborted non-transparently by some > anonymous coward, without waiting for the major backers votes, and all gcc > participants simply keep silence. > > > > I am aware that the effort to fight for 50% of a $7K bounty is not worth > it - even the distraction for opening a discussion here is essentially not > worth the effort. > > > > I cared more or less only on what Microchip ($5K contibution to the > bounty) had to say, and the other two top backers. And I'm still curious > about this. > > > > If they too say "research, analysis and integration work is no work" and > "effort to validate abandoned patches and reuse of them is no work" - well, > then I guess I'll rest my case. > > > > But at least it gets "on file", so other people which struggle with gcc > (especially in combination with bountysource) have a point-of-reference. > > > > Very disappointing all this. > > > > I mean really? An OSS project which brute-force aborts a > voting-procedure (=IT-fascism)? Just to award a monetary value to an > gcc-project insider? > > > > And everyone keeps silence? > > If I'm reading the Bountysource page correctly, the bounty was awarded > to saaadhu, who I assume is Senthil Kumar Selvaraj. Senthil has been > a GCC contributor for a while with some 70 committed patches. But I > wouldn't describe them as a GCC project insider. They are not a GCC > maintainer or reviewer. > > > > > We all know that reproducing such things in order to have an informed > opinion/vote costs terribly high amounts of time. The simplified method > would be: > > > > * enter the issues here: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92729#c9 > > * then, just try to validate the until then available patch > > * => you'll fail, as no stable environment is available (major failure > of the steering comitee, which should insist "all targets need to have an > functioning CI" > > * => here you start integrating a dev/ci environment, try to find > reference points/versions etc., etc. > > > > @ Steering Committee > > > > A functioning CI across targets is a non-disputable must requirement in > todays IT landscape: > > > > * https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98574 > > > > Or at least reference-repos, like e.g. this one: > > > > * https://github.com/abebeos/avr-gnu > > > > We would not have this topic here, if the gcc-project had a decent CI, > or a build-setup used by all developers. > > I agree that having a good CI for GCC would be really great. It's > also really hard. GCC generates code for many different targets, and > the nature of GCC is such that running tests on a simulator rather > than real hardware, while helpful, is simply not good enough in > practice. So a good CI for GCC requires supporting a large variety of > hardware. That is very desirable. It's also very hard and very > expensive. The GCC project must rely on volunteers for this work. > You can see the available support at > https://gcc.gnu.org/wiki/CompileFarm and on the gcc-testresults > mailing list. Should the project have better CI? Yes, absolutely. > Who is going to put in the time, effort, and money to make that > happen? > > > In any case, this is not the right place to raise your concerns about > Bountysource. That is not part of the GCC project. > > Ian >
