* Ricardo Signes <perl.p5p@rjbs.manxome.org> [2012-01-08 17:00]: > * Nicholas Clark <nick@ccl4.org> [2012-01-03T05:49:15] > > I don't see merit in keeping it. > > What is our path to deprecate SigRT if we choose to? > > My feeling is that while we won't see much contention on p5p, we may > get some from users, and it would be more polite to deprecate in > 5.17.0. > > Thoughts? Aye. At least if I’m reading the thread correctly, that it says this thing *could* be made to work by someone who read the source. Breaking working user code unnecessarily, however bad the code may have been, is bad form. Though if such code didn’t work in the first place, it would actually seem a service to users to break their code right away, as long as it is with a clear error message telling them that it never actually worked and how to fix it. I guess effectively that means a deprecation cycle either way. The difference is only whether the meat of the module is removed.Thread Previous | Thread Next