pre 1.0 issues in the database

I've looked over the issues tagged with release numbers.

Honestly, I think the list could be usefully trimmed by quite a bit.

----------

These seem to me like important, short-term issues (clearly pre 1.0):

894 eol-style problem (effects CRLF)

        I'm guessing this is easy to fix.

911 Build process enhancement

        Trivial task

900 Error message cleanup.

        Low priority but someone did send a patch.

880 Real bug

        Core dump with script.

-----------

These I'm not sure about:

896 output formatting issue(? "filter contents of log messages ...")

        I'm not sure what this is about.

689 Checksum feature

        This does sound like a plausible 1.0 safeguard.

        I would want to be really certain, though, that there
        wasn't a significant probability of corruption that would
        confuse both the server and the client in parallel -- making
        the checksums equal, but of equally bogus data.

        In other words -- if there isn't very high confidence that
        this safeguard addresses the most probably failure modes, then
        it is probably premature. But if there is very high
        confidence of that...then maybe.

-----------

These are things I wouldn't worry about too much for a 1.0:

851 eol-style problem (mixed line-ending files, svn:eol-style native)

        This looks both hard and relatively unimportant.

761 Usability issue (log message files)

        Not critical to back-end use. Fixing this certainly won't
        prevent usability complaints about 1.0 :-)

860 Performance (large files commits)

        Not critical to Wiki apps or to many development
        projects.

806 Bug fix clean-up

        But there's a work-around in place and
        discussion of the right solution isn't done.

840 Namespace clean-up

        There will probably be later namespace clean-ups,
        post 1.0. Putting this off harms nothing.

872 i18n issue (Windows only)

        This stands in the way of only a small number of possible
        uses and design work seems to be ongoing on for it.

913 Performance tuning (import over ra_local or ra_dav on a fast connection)

510 architectural/API change (invert txdelta interface)

        Design work sounds ongoing. This doesn't add user visible
        features or have any stated impact on performance and robustness.
        The issue appears to be an internal change in preparation for
        _future_ (post 1.0) enhancements.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun Oct 13 19:35:51 2002