I agree re: bags of bits of code. e.g there is no single binary that can try all permutations of various techniques on a single data set and pick the best model. instead there seems to be N different binaries for every (classifier) technique at least circa 0.5
Is there no presubmit checkin hook for svn or git? On Sun, Jun 10, 2012 at 8:24 AM, Sean Owen <[email protected]> wrote: > My point was rather that it shouldn't even reach review, and was more > suggesting I'm giving up on trying to fix it that continuing. > > I do think a review process is not realistic to expect, which is all > the more reason that if individual committers are going to dump code > in directly, that it at least pass basic checks available in IDEs. > This isn't happening here, and despite some discussion, hasn't > changed. > > Ted I think you made an offhand comment that you encountered an > independent review of the code base and it was viewed as fair-to-poor > quality. I encounter the same, and I agree. It's a real barrier to the > project going forward; if you don't 'trust' the code, can't easily > change or understand the code, it discourages using or building on it. > > I am not sure it's truly fixable at this point; such is life. This is > mostly a bag of bits of people's code. This is why, for example, I'm > doing my own new development in a separate project. But still worth > striving for quality, for those that want to continue to work on this > project. > > > On Sun, Jun 10, 2012 at 12:38 AM, Robin Anil <[email protected]> wrote: > > +1 > > > > Sean you can always keep cleaning this but the root cause is many, and > > issues like this will slowly creep in over time. > > > > One of the root cause is that reviewing is hard in apache at the moment. > I > > am really hoping we move to a set of tools that allows any contributor > just > > to make changes -> send to review -> (get responses from jenkins) -> > rework > > -> resend patch -> get approval from a committer -> submit. > > > > If you ask me to inspect a diff submitted on jira with absolutely no time > > available in day time. I will just look for general correctness. I would > > like to see static analysis output directly from jenkins, when someone > > checks in a patch. > > > > Robin > > ------ > > Robin Anil > -- Yee Yang Li Hector <https://plus.google.com/106746796711269457249> Professional Profile <http://www.linkedin.com/in/yeehector> http://hectorgon.blogspot.com/ (tech + travel) http://hectorgon.com (book reviews)
