Format for JDK 6/7 changeset comments?

[Mark Reinhold]

> Date: Fri, 09 Nov 2007 11:33:39 -0800
> From: andreas.sterbenz at sun.com

> Mark Reinhold wrote:
>> Why does anyone who doesn't know how to navigate the Mercurial web
>> interface need to find this information?
>> 
>> In what sense is the location of the bug database better known than that
>> of the Mercurial repositories?
> 
> Just thinking of Sun employees for now: management, support and services
> people, developers for other products that use the JDK, e.g. App Server or
> Access Manager.

We need to think of a wider set of people than those who work for Sun.

> Any of them may need to get in touch with the code reviewers in case the
> primary author of the fix is unavailable. All of them know where to find
> the bug database but probably won't know the magic URL for the correct JDK
> repository.

We're already planning to put changeset pointers into bug reports,
automatically.  So I don't think this is really much of an issue.

> ...
> 
> Anyway, the main issue is that there is a lot of value in having all
> relevant information about a bug stored in ONE central place, which is the
> bug database. If you feel differently, then we'll have to agree to disagree.

Perhaps so.

There's an important non-technical aspect of this issue that hasn't
really been discussed here: Giving credit where credit is due.

Reviewing code -- or at least doing it well -- is hard work.  If I were a
non-Sun OpenJDK contributor then I think I'd prefer to receive credit for
having done a code review in the form of having my name in the changeset
comment rather than listed somewhere in a bug database.  The changeset
comment will travel with the code, wherever it goes.  Many people will
see the code but rarely, if ever, look at the bug database, which exists
in just one place.

The names of non-author contributors also belong in changeset comments,
for the same reason.

It'd be interesting to hear some non-Sun perspectives on this.

- Mark