[EMAIL PROTECTED] wrote: > On Sat, Jul 19, 2008 at 10:03:55AM -0700, David Brown wrote: >> Really. They are probably the single most obnoxious attribute of our use >> of P4 at work, especially given P4's blurring of branch and directory >> names. Just branching a file causes the version stamp to change, making >> merges or even just diffs obnoxious. > > How about just using the file name and data keywords then? But foregoing the > version keyword? > >> Another is a "ChangeLog" file. The need to write history information into >> a file suggests that the history in the revision control system either >> isn't very good, or isn't accessible. > > Yea I've never written nor read a ChangeLog file. The fact they may be > obsolete therefore makes me feel better. :) > > Wow I didn't know ChageLogs predated VCS.
I was surprised not too long ago when a project changed it's procedures to eliminate the ChangeLog. My first reaction was to believe I'd miss it and that commit logs would never adequately substitute, because they were invariably terse and often cryptic. After some reflection, it dawned on me that my reaction was unreasonable. To be effective, ChangeLog entries had to be written prior to the commit --by the committer! So if there was any chance for a decent CL entry, it could just as well (better, in fact) go in the commit log. The end result would generally be less effort by the programmer, and would more likely not be forgotten. A further refinement that was suggested, also makes sense to me. Since some tabulated listings display only the leading N characters of the CL, it was suggested that the first line of a commit log entry should be intentionally made a concise summary (containing an issue id, if relevant), and following lines should be used for detail, where necessary. Regards, ..jim -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
