On 6/1/12 3:27 PM, Richard Fateman wrote:
> On 6/1/2012 11:53 AM, Rupert Swarbrick wrote:
>> Ok, well I guess we disagree here. For me, a file containing source code
>> is not required to be a detailed "biography" of the history of the
>> project.
> I feel it should contain all that is necessary to read and understand
> the program. For
> a program like Maxima which has parts that were written in 1965 and have
> been
> unchanged for 45 years, I think it is relevant to include info as to
> why, after 45 years
> someone decided that it should be changed, and who that person is.
I can sympathize with Richard on this. Sometimes it's just really nice
to have the whole function to look at instead of a patch. Less work if
you just want to try out something. Other times a patch is better
because you can tell what changed without having to have the old version
around.
But like Rupert, I don't really care who wrote it because git can tell
me[1], but Richard makes a really good point that if something that
hasn't changed in decades is now changed, it would be really nice to
have a comment in the code about why and not in some checkin message
that often times is wholly inadequate to explain why.
Or if the author is unclear about exactly how something works, it's
awful nice to know who wrote the comment so you can ask instead of
digging through version control to find out when that was added and by whom.
Ray
[1] This isn't entirely true. If the really important change was done
5 years ago, but in the meantime the code has been reindented many times
(for other valid reasons), it becomes really hard to know what the
relevant change was and who made it. Perhaps some advanced usage of git
blame can tell you, but I'm not knowledgeable enough to know.