>>>>> "CF" == Christopher Faylor <[EMAIL PROTECTED]> writes:

..
>> 
>> | On Fri, 8 Apr 2005, Brian Dessent wrote:
>> | ....
>> | > So, you're changing the system-wide cygdrive prefix but since you have a
>> | > user-defined cygdrive prefix it is taking precedent.  I believe this was
>> | > a recent change which is why you are seeing it now.
>> | 
>> | and Pavel Toskov said:
>> | 
>> | PT> Yes! There was a change in October 2004 which allowed the system 
>> prefix to
>> | PT> be used no matter that a user prefix was specified. This is not the
>> | PT> expected behaviour and it was changed in Cygwin 1.5.14 to work as it 
>> used
>> | PT> to be in the past.

CF> There is nothing like that in that message.  You apparently ignored the
CF> part that would actually have helped you, went on to read the rest of
CF> the thread when reinforced Brian's advice and then ignored the
CF> advice.

You're correct, it's not in that message; Pavel's response is in the
followup to that message:
http://cygwin.com/ml/cygwin/2005-04/msg00355.html

>> <annoyed> Excuse me Brian, but this was not user error; the system
>> behavior changed from one version to the next -- and then was changed
>> back.  And this is the precise issue that pushed me to join the list.

CF> I have no idea why you would be annoyed at someone's attempt to help,
CF> especially when they weren't even trying to help you.

Lin Q started this thread; I had not asked for help, as I had found a
unlovely but sufficient workaround some time ago.  I see that I
encountered the bug in 1.5.13 -- in fact, it looks like that was also
the issue in the thread that Brian referred to.  

And yes, I prefer Brian's fix to hacking the registry, it's always a
good idea to use the official interface.

CF> Just to clarify: There was a *bug* which was *fixed* in 1.5.14.  The
CF> *bug* was that the user mount of /cygdrive was ignored.

Quite so.  This is the bug which I encountered, and which was
referenced in Pavel's response.  

>> Nothing quite like having my mountpoints and paths inexplicably change
>> from one upgrade to the next, with /cygdrive suddenly appearing from
>> nowhere, breaking my nice, comfortable useful setup that had been
>> running smoothly for the last six years. 

CF> There's nothing quite like indignant email from users who don't really
CF> quite understand what is going on but nonetheless feel obliged to bite
CF> the hand that is trying to help.

Excuse me -- I didn't intend to bite the helper -- nor was I the
person being helped -- but as I said, I don't like my filesystem
structure changing out from under me, and both Brian and Pavel's posts
seemed to imply that the change to mount had been intentional though
mistaken.

As someone who has worked as user, developer, designer, and tech
support, I also prefer clarity about what is a bug, what is design
error, and what is user error.

I gather it was an accidentally introduced bug?  My inference that it
was an intended change, which unexpectedly changed the visible
filesystem structure (added to the time I spent wading through
archives, doc, and system changes) is a good bit of what ticked me
off.

Anyway, I think the mount doc needs clarification.  Having different
mount points for the user and for the system is unexpected, and it's
also not clear from the doc that in cygwin, the user mount points
take precedence over the system mount points.

I'd be happy to write a paragraph or so, if one of the usual tech
writers would be willing to pull it in if they find it reasonable
quality.  Anyone?

-- 
Patricia J. Hawkins
Hawkins Internet Applications
www.hawkinsia.com


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Reply via email to