> On Jan 17, 2017, at 3:01 PM, Robinson, Paul <paul.robinson at sony.com> wrote: > > - As Medhi says, according to surveys and discussions in forums like the LLVM Dev Meeting BoF, most people who care are in favor of mono-repo. > > From the online surveys, I think the split was roughly 50:50. > > I don’t know on what data you’re basis this on. I looked very closely and here are two questions that contradicts your view. > > Question: “If we could go back in time and restart the project with today's technologies, which repository scheme would be best for the LLVM project?” > -> 55 to 36 in favor of the mono-repo > > So, more like 60:40. Clearly "more" people are in favor, but I'm not so sure it qualifies as "most.” <>To clarify, *I* didn’t write “most”, I wrote earlier "My reading of the survey is that the monorepo has a significant lead.” So I guess your targeting Chris sentence, but my reading of it above is that it doesn't focus exclusively on the raw survey results, he added to this end 1) "and discussions in forums like the LLVM Dev” and 2) "people *who care*”. That’s subjective obviously, but just to say that the survey is one data source, that the BoF made some people attitude towards monorepo change, and the “binary" nature of the survey does not allow to capture nuances, i.e. how many people consider “I’m in favor of A but I’ll be fine with B” vs “I’m totally against B, it’s has to be A”, which discussions helped to catch. — Mehdi -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170117/2e1767ba/attachment.html>
I'd think that the survey respondent population would tend to be heavily weighted towards people who care (enough to spend time filling out the survey) as opposed to the likely smaller population of people willing to speak up in a contentious email thread. We surely did not get anywhere near so many voices in the thread as in the survey. So, in terms of gauging community sentiment, I'd lean more on the survey and less on the discussions. The discussions are useful for raising technical feasibility questions and workflow descriptions, but will tend to be misleading about community-wide thinking. (In most email discussions, not so many people are that interested, and people who aren't interested commonly aren't affected either, like in the optnone discussion. This topic really will affect basically everyone, which is why I think favoring the survey results over online debates is the right way to go.) --paulr From: mehdi.amini at apple.com [mailto:mehdi.amini at apple.com] Sent: Tuesday, January 17, 2017 3:12 PM To: Robinson, Paul Cc: David Chisnall; Keane, Erich; Chris Lattner; llvm-dev at lists.llvm.org Subject: Re: [llvm-dev] Git Transition status? On Jan 17, 2017, at 3:01 PM, Robinson, Paul <paul.robinson at sony.com<mailto:paul.robinson at sony.com>> wrote: - As Medhi says, according to surveys and discussions in forums like the LLVM Dev Meeting BoF, most people who care are in favor of mono-repo. From the online surveys, I think the split was roughly 50:50. I don’t know on what data you’re basis this on. I looked very closely and here are two questions that contradicts your view. Question: “If we could go back in time and restart the project with today's technologies, which repository scheme would be best for the LLVM project?” -> 55 to 36 in favor of the mono-repo So, more like 60:40. Clearly "more" people are in favor, but I'm not so sure it qualifies as "most.” To clarify, *I* didn’t write “most”, I wrote earlier "My reading of the survey is that the monorepo has a significant lead.” So I guess your targeting Chris sentence, but my reading of it above is that it doesn't focus exclusively on the raw survey results, he added to this end 1) "and discussions in forums like the LLVM Dev” and 2) "people *who care*”. That’s subjective obviously, but just to say that the survey is one data source, that the BoF made some people attitude towards monorepo change, and the “binary" nature of the survey does not allow to capture nuances, i.e. how many people consider “I’m in favor of A but I’ll be fine with B” vs “I’m totally against B, it’s has to be A”, which discussions helped to catch. — Mehdi -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170117/f7687054/attachment-0001.html>
> On Jan 17, 2017, at 3:41 PM, Robinson, Paul <paul.robinson at sony.com> wrote: > > I'd think that the survey respondent population would tend to be heavily weighted towards people who care (enough to spend time filling out the survey) as opposed to the likely smaller population of people willing to speak up in a contentious email thread. We surely did not get anywhere near so many voices in the thread as in the survey. <>I was not talking about the email thread, but the many *in person* discussion I had, and my experience is what I wrote before, i.e. people *have* to answer A or B in the survey, but it does not capture *why* they do so, or how strongly they are in favor or one, or opposed to the other. — Mehdi> <> > So, in terms of gauging community sentiment, I'd lean more on the survey and less on the discussions. The discussions are useful for raising technical feasibility questions and workflow descriptions, but will tend to be misleading about community-wide thinking. > > (In most email discussions, not so many people are that interested, and people who aren't interested commonly aren't affected either, like in the optnone discussion. This topic really will affect basically everyone, which is why I think favoring the survey results over online debates is the right way to go.) > --paulr > > From: mehdi.amini at apple.com [mailto:mehdi.amini at apple.com] > Sent: Tuesday, January 17, 2017 3:12 PM > To: Robinson, Paul > Cc: David Chisnall; Keane, Erich; Chris Lattner; llvm-dev at lists.llvm.org > Subject: Re: [llvm-dev] Git Transition status? > > > On Jan 17, 2017, at 3:01 PM, Robinson, Paul <paul.robinson at sony.com <mailto:paul.robinson at sony.com>> wrote: > > - As Medhi says, according to surveys and discussions in forums like the LLVM Dev Meeting BoF, most people who care are in favor of mono-repo. > > From the online surveys, I think the split was roughly 50:50. > > I don’t know on what data you’re basis this on. I looked very closely and here are two questions that contradicts your view. > > Question: “If we could go back in time and restart the project with today's technologies, which repository scheme would be best for the LLVM project?” > -> 55 to 36 in favor of the mono-repo > > So, more like 60:40. Clearly "more" people are in favor, but I'm not so sure it qualifies as "most.” > > To clarify, *I* didn’t write “most”, I wrote earlier "My reading of the survey is that the monorepo has a significant lead.” > > So I guess your targeting Chris sentence, but my reading of it above is that it doesn't focus exclusively on the raw survey results, he added to this end 1) "and discussions in forums like the LLVM Dev” and 2) "people *who care*”. That’s subjective obviously, but just to say that the survey is one data source, that the BoF made some people attitude towards monorepo change, and the “binary" nature of the survey does not allow to capture nuances, i.e. how many people consider “I’m in favor of A but I’ll be fine with B” vs “I’m totally against B, it’s has to be A”, which discussions helped to catch. > > — > Mehdi-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170117/8ea6951b/attachment.html>