This is the mail archive of the cygwin mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: setup.exe needs package name selection filter

Is there a ticketing/tracking system for Cygwin where one can submit
feature requests?  I think having it documented in the system as
something more than someone's wishful email would be helpful.

As far as improvements to setup.exe, I think the Cygwin team could
potentially save a lot of effort in the long term by porting and
adopting one of the Linux package managers in its place - apt or yum
or whatever.

On 6/20/08, Hugh Sasse <> wrote:
> On Fri, 20 Jun 2008, Dave Korn wrote:
>> reikred wrote on 20 June 2008 02:08:
>> > Christopher Faylor wrote:
>> >>
>> >> Yes.  It's called
>> >>
>> "make-setup.exe-demands-and-then-sit-back-and-wait-for-people-to-implement-y
>> >>
>> >> cgf
>> >>
>         [...]
>> >
>> > Please read it in the spirit of a suggestion of a feature
>> > that I think could be VERY useful to a lot of people, not
>> > just the 1% of cygwin users that know how to implement it.
>> >
>> > rr
>>   It's definitely a good suggestion, quite possibly a great suggestion.
>         [...]
>>   I think the point that cgf is getting at is that there is no shortage of
>> suggestions or good ideas or enhancement requests or wish lists nor any
>> difficulty in coming up with new ones.  But without a bit of
>> getting-it-done
>> elbow grease, a suggestion is really very little use on its own.
> This is a recurring problem for the Free Software movement.  People
> want Free Software to spread, and there are efforts, such as the
> attempt to get a record number of downloads for Firefox 3, as
> promotional activities.  But a constant theme is that suggestions
> from users are not welcomed; instead they provoke various forms of
> the response: "If you want it, send a patch".  This fails to
> recognise that one reason programmers don't like maintenance
> programming is that reading code is more difficult than writing it.
> This is a pre-requisite for changing the code.  It also fails to
> recognise that a user for one project who is completely unfamiliar
> with the code base, may be busy contributing to other projects, and
> that some suggestions may be very much easier for someone familiar
> with the code than someone who is not.  Also, if users' suggestions
> are (effectively) dismissed in this way, it will prevent the use of
> Free Software by non-programmers, which runs counter to the desire
> for it to spread.  Most of the potential users of Free Software are
> non-programmers.  [Yes, this is less so for Cygwin.]
> Clearly decisions are not made in the same way as for a software
> business, where finance is fundamental, but I'd suggest that unless
> the needs of users are given greater status, then Free Software
> advocacy will be somewhat hobbled.  But as this is a cultural problem,
> and there is no immediately obvious technical solution, I'm unsure
> how one might practically improve the situation.  I do think more
> thought should be given to it, though.
>>     cheers,
>>       DaveK
>         Hugh
> --
> Unsubscribe info:
> Problem reports:
> Documentation:
> FAQ:         

Sent from Gmail for mobile |

Mark J. Reed <>

Unsubscribe info:
Problem reports:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]