Re: Bug Fixes Management

Ville Skyttä <ville.skytta@iki.fi> wrote:

>Perhaps the new dev box [when it's up] could run Bugzilla?

Well, it's unlikely to be ready in time to be any use in this beta test,
but otherwise it's a good idea.


>More on bug fixes *during* the beta test, should bugs be fixed "on the
>fly" or should the code be left alone during this period?  And where to
>commit, HEAD or to a branch/tag?

Changes are fine so long as they are "conservative". :8001 is subscribed to
validator-0_6_0-beta so changes on HEAD won't affect it until we move that
tag up with "cvs tag -F validator-0_6_0-beta".


>header.html has:
><link rev="made" href="mailto:gerald@w3.org" />
>....perhaps that should be changed to be consistent with the footer.

s/gerald/www-validator/


>There are various checks in the code like:
>if ($File->{Type} eq 'xml')
>....which should be using the &is_xml and friends functions.  Eg. the
>namespace for an X(HT)ML document won't be displayed because of this in
>the results.

This one takes some thinking I, uhm, "think" because it messes with the
somwhat hairy code that generates the table in the output, but if you think
it'll be low or zero-impact go ahead and commit a fix,


>The namespace table rows need a colspan="2" tweak.  That'll be a bug
>only when the above one is fixed, though :)

And lands smack dab in the middle of the frightening mess that is
"serialize_table" and friends. :-(


>/feedback.html, the "Usenet newsgroup" bit needs tuning; only the group
>name should be linked, and in monospace.

Knock yourself out.

-- 
I have lobbied for the update and improvement of SGML. I've done it for years.
I consider it the jewel for which XML is a setting.  It does deserve a bit or
polishing now and then.                                        -- Len Bullard

Received on Wednesday, 23 October 2002 16:53:45 UTC