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: FLTK versions in Cygwin [was: Re: units: update, FHS compliance]


Yaakov (Cygwin/X) wrote:
> On 25/01/2010 04:57, Albrecht Schlosser wrote:
>> Point taken. But unfortunately the FLTK community decided it the
>> other way with a majority of 74% (this must have been in or before
>> 2003):
>>
>> http://www.fltk.org/poll.php?r1
>>
>> Thus this is not likely to be changed. Sorry.
> 
> Since when are matters such as this decided by opinion polls?  I bet
> most of the voters didn't use Cygwin or even know what it is.  Did you
> have a poll to decide how FLTK should be built on Linux?  Didn't think
> so.  I am NOT impressed.

Meh.  Unlike linux, there is a significant portion of the cygwin user
base that treats cygwin simply as a "build environment" -- but use a
compiler for native win32 $hosts.

I don't much like it, but that's reality. (...why did Cygnus fund the
early development, in the first place?  To have a windows-hosted build
environment for xxx-target compilers: in this case, xxx = native-win32)

So, the FLTK development and user community thinks of Cygwin that way,
by a 3-to-1 majority.  Maybe we ought to try to convince them to treat
cygwin as a full-class platform, rather than as an afterthought to
native win32 support?  I suspect a better approach to that end is more
honey, less vinegar.

I know, I know -- that's pretty rich, coming from me.  But...there it is.

> Like any other platform, the only people that should have a say in any
> package's behaviour on Cygwin are the Cygwin managers and maintainers.
> We have made Cygwin a *NIX/X11 platform, and we have decided that gcc
> will not support -mno-cygwin (which was broken anyway).  *That* is the
> ONLY opinion that should matter in this discussion.

Well, by removing -mno-cygwin, we've more or less assured that this will
happen eventually. People are going to get tired of being stuck with
'gcc-3 -mno-cygwin' eventually, and are going to want to use
i686-mingw32-gcc(4) at some point.  Then, they'll have to switch over to
an actual cross-compile configury for their cygwin-hosted native builds.

But that'll take a while.

> Unfortunately this attitude is by no means limited to FLTK, and any
> frustration with this topic is based more on the frequency of this sort
> of thinking than with FLTK in particular.

Yep, but...that's the nature of (our) beast.

--
Chuck


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


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