1.7.1 release date?

Corinna Vinschen corinna-cygwin@cygwin.com
Tue Dec 8 10:33:00 GMT 2009

On Dec  7 10:42, Christopher Faylor wrote:
> On Mon, Dec 07, 2009 at 10:58:19AM +0100, Corinna Vinschen wrote:
> >On Dec  6 14:43, Christopher Faylor wrote:
> >>Maybe you've never been involved in a "what do we call it" discussion
> >>before but, in my experience, they are never short.
> >
> >Then, why did you bring it up in the first place?
> >http://cygwin.com/ml/cygwin-developers/2009-12/msg00042.html We didn't
> >have to touch the directory names, nor the setup.ini files.
> >http://cygwin.com/ml/cygwin-developers/2009-12/msg00044.html This
> >entire discussion could have been avoided.  You were the one starting
> >it and now we are the ones blamed for it.
> I do have the right to raise a concern.

Yes, of course.  But the problem is that the directory names aren't a
concern at all, at least not from the user perspective.  Raising the
concerns started after the proposed renaming and were not taken
seriously from the start.

> I was talking about not changing the directory name for a non-beta
> release.

Sorry, but I don't understand this.  You proposed that the directory
gets renamed from release-2 to release, which would collide with the
old release directory name.  How is that talking about "not" renaming
a directory?!?

> Your adamant assertion to the contrary, I raised it because I
> thought this was going to be the plan all along.  That wasn't entirely
> right but some archive speleunking showed that the issue was still up in
> the air.
> I've raised what you should consider to be a real issue given your
> concern about end-users being bitten by the upgrade.  The issue would
> not be solved by giving a directory a cute animal name.

See below.

> Rather than cutting down my email so that you can make ad hominem
> responses, I was hoping to see a discussion on the points I made.

Sorry if it came over as ad hominem response.  I guess it was just a

What points exactly havn't been discussed?  Let's see.

- Shall I add the getopt patch?

  No, I don't think you should.


- I think setup-legacy makes more sense.


- Doesn't calling it "win-9x" almost guarantee the "What about Windows ME"

  It's just the name of a html file.  The actual text is going to take
  care of that.  So, ok.  Hopefully.

- I think it is moderately bad idea to have a "release-2" directory
  permanently associated with Cygwin 1.7.

  Why?  It's just a directory name and it only expresses an iteration
  of the release.  It's not connected to the release number of Cygwin.
  This has been raised by three people, but turned down as bike-scheding.

- Unnecessary 10 Gigs update on all mirrors.

  You don't care.  I go along with this.

- Mirrors will be in an inconsistent and potentially quite disastrously
  unusable state during the duration of the changeover.

  Ignored.  I go along with it.

- If users might be surprised by the 1.7 update, we have to do something.

  The upgrade from 1.5 to 1.7 will have a certain bumpyness for the
  users which is neither solved by renaming, nor by not renaming a
  directory.  You don't solve the user mount points problem by it, and
  you certainly don't solve the ANSI CP -> UTF-8 problems by it.
  These problems are major changes, inherent in Cygwin itself, not in
  the release.  You can't fix them up for the users in setup.exe, you
  can only band-aid them.

  So what *are* the choices we have at this point?

  - Just release 1.7.1 and see how bumpy it really gets.

  - Or, change setup.exe to check for an existing 1.5.25 release and
    warn the user that the upgrade might become tricky.

  - Or, change setup.exe to refuse an upgrade from 1.5.25 entirely.

  And besides

  - Keep the release area in a shape that users of an old setup don't
    get forcefully updated to 1.7.


Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

More information about the Cygwin-developers mailing list