3 ms·

It's not my first time seeing this, but I'm struck by some disappointment in the legacy of OSS here. The author clearly understood what was happening enough to

by hirsin 2y ago

It's not my first time seeing this, but I'm struck by some disappointment in the legacy of OSS here. The author clearly understood what was happening enough to post this to the unix SO - but not enough to find the `man` code and grep for the string that had to be present? I'll admit I don't immediately know where to go to look for the source code, but I can't imagine it's some tightly held secret address to a file share somewhere. Alternatively - maybe it wasn't the difficulty of finding it but the (perceived?) difficulty in requesting that it be changed? E - maybe the likely alternative is the "the customer has nukes so please fix it" entitlement? https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95644#c4 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95644#c4

gavindean90 2y ago
What does this have to do with the link?
jdnbndhxb 2y ago
It's a nice anecdote, don't think too much about it
- MiguelX413 2y ago
  No.
strken 2y ago
I don't see Bill as doing anything wrong here. Given that he works for Cray on their own Fortran compiler, it seems reasonable that he would be allowed to report bugs in the competing product gfortran but not allowed to actually fix fix them, and that he would ask a question knowing full well that the answer would be "fuck you, pay me" just to be able to truthfully report that back up the chain. The nuclear-armed client seems entitled, but I don't think he necessarily is.
- eloisant 2y ago
  Nukes or not, telling open source developers "the customer is asking for it" twice is either entitled or very ignorant.
  - coldpie 2y ago
    I spent years working on open source and I disagree. Politely nudging bugs is fine. Sometimes things fall through the cracks, and a quick prod will get someone's attention to take the time to go fix it. We do actually want to fix bugs for our users, after all. It is very easy to nudge bugs impolitely, so do this with extreme caution, but I don't think Bill stepped over the line in that issue.

by hirsin 2y ago

gavindean90 2y ago

What does this have to do with the link?

jdnbndhxb 2y ago

It's a nice anecdote, don't think too much about it

MiguelX413 2y ago
No.

strken 2y ago

I don't see Bill as doing anything wrong here. Given that he works for Cray on their own Fortran compiler, it seems reasonable that he would be allowed to report bugs in the competing product gfortran but not allowed to actually fix fix them, and that he would ask a question knowing full well that the answer would be "fuck you, pay me" just to be able to truthfully report that back up the chain. The nuclear-armed client seems entitled, but I don't think he necessarily is.

eloisant 2y ago
Nukes or not, telling open source developers "the customer is asking for it" twice is either entitled or very ignorant.
- coldpie 2y ago
  I spent years working on open source and I disagree. Politely nudging bugs is fine. Sometimes things fall through the cracks, and a quick prod will get someone's attention to take the time to go fix it. We do actually want to fix bugs for our users, after all. It is very easy to nudge bugs impolitely, so do this with extreme caution, but I don't think Bill stepped over the line in that issue.