Re: Gtk-OSX (was: Website proposal for usability)
On Mon, 30 Aug 2010, Paul Davis wrote: > On Mon, Aug 30, 2010 at 9:47 AM, Allin Cottrell wrote: > > On Mon, 30 Aug 2010, Tor Lillqvist wrote: > > > >> > It has certainly been explained that that is the situation on > >> > Windows, and I fully accept it. It's less clear that it should be > >> > the situation on OS X, with its *nix-type substructure. > >> > >> You have it backwards. It was from the GTK-on-OS-X people (well, at > >> least those that I have heard from) that this convention originated. > >> Only a bit later did the GTK+-on-Windows people (well, many of us, not > >> all) realize the same. > > > > I was interested in the logic rather than the history, > > The logic is that despite its "*nix-type" infrastructure, this is how > Apple has intended ISV's to distribute software, and as a result, its > what users expect... OK, you make a good case for this. Allin Cottrell ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Mon, Aug 30, 2010 at 9:47 AM, Allin Cottrell wrote: > On Mon, 30 Aug 2010, Tor Lillqvist wrote: > >> > It has certainly been explained that that is the situation on >> > Windows, and I fully accept it. It's less clear that it should be >> > the situation on OS X, with its *nix-type substructure. >> >> You have it backwards. It was from the GTK-on-OS-X people (well, at >> least those that I have heard from) that this convention originated. >> Only a bit later did the GTK+-on-Windows people (well, many of us, not >> all) realize the same. > > I was interested in the logic rather than the history, The logic is that despite its "*nix-type" infrastructure, this is how Apple has intended ISV's to distribute software, and as a result, its what users expect. You will rarely (if ever) see an OS X application that has a list of prequisites other than a particular version of OS X and perhaps some hardware. The notion that "too use this app, you also need to have the FOO framework installed" just isn't something that exists in the OS X user culture. People shipping apps for OS X package up everything their app needs that isn't part of OS X itself. And yes, this causes all kinds of potential security issues and all the rest that Linux distributions hate about things like AppImage, but for better or for worse, that is the way Apple wants things to be. It is hard for it be otherwise without the kind of centralized repositories that most linux distributions use, and faced with DLL hell as the alternative, I guess Apple felt that the all-in-one package was the best option. Regardless of whether it is or isn't, its what people have come to expect. --p ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Mon, 30 Aug 2010, Tor Lillqvist wrote: > > It has certainly been explained that that is the situation on > > Windows, and I fully accept it. It's less clear that it should be > > the situation on OS X, with its *nix-type substructure. > > You have it backwards. It was from the GTK-on-OS-X people (well, at > least those that I have heard from) that this convention originated. > Only a bit later did the GTK+-on-Windows people (well, many of us, not > all) realize the same. I was interested in the logic rather than the history, and I don't think I had that backwards. But anyway, I'm OK with assuming that a common user-runtime for GTK is not on. > > I don't think that invalidates the idea that it would be very > > useful for app developers to have a GTK runtime package > > available, as we do for Windows. > > As usual, people seem to be constantly jumping between talking about > "packages" for developers, and "packages" for end-users. There is no > "officially sanctioned" GTK end-user runtime package for Windows > available, in the sense that it would be something that could/should > be installed as such on end-user systems. It's the developer and/or > packager that is expected to pick out those files his application > actually needs at run-time from the run-time zip archives on > ftp.gnome.org (or from the "bundle" which just combines all the > run-time and developer zip archives for the GTK+ stack). This is not > the same set of files for all applications. Maybe "runtime package" was the wrong choice of words, but I didn't intend anything thst contradicts what you're saying. As an app developer I'm happy to select what I need from the run-time zip archives for Windows. You have done us a major service by making those archives available. I'd like to be able to do the same for OS X (if a common runtime package for users is deemed infeasible or undesirable). Allin Cottrell ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
> It has certainly been explained that that is the situation on > Windows, and I fully accept it. It's less clear that it should be > the situation on OS X, with its *nix-type substructure. You have it backwards. It was from the GTK-on-OS-X people (well, at least those that I have heard from) that this convention originated. Only a bit later did the GTK+-on-Windows people (well, many of us, not all) realize the same. > I don't think that > invalidates the idea that it would be very useful for app > developers to have a GTK runtime package available, as we do for > Windows. As usual, people seem to be constantly jumping between talking about "packages" for developers, and "packages" for end-users. There is no "officially sanctioned" GTK end-user runtime package for Windows available, in the sense that it would be something that could/should be installed as such on end-user systems. It's the developer and/or packager that is expected to pick out those files his application actually needs at run-time from the run-time zip archives on ftp.gnome.org (or from the "bundle" which just combines all the run-time and developer zip archives for the GTK+ stack). This is not the same set of files for all applications. --tml ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Sun, 29 Aug 2010, Paul Davis wrote:
> On Fri, Aug 27, 2010 at 4:17 PM, Allin Cottrell wrote:
> > On Fri, 27 Aug 2010, Kristian Rietveld wrote:
> >
> >> For a GTK+ runtime package ("GTK+ Framework"), I think you should
> >> check out what has been done in the past. It is by no means an easy
> >> task. The latest code and instructions for this are at the GTK-OSX
> >> website if I am not mistaken.
> >
> > I'm aware it's not an easy task. That's why I'm requesting that
> > such a runtime package be made available as a download via gtk.org
> > (and offering to help in building one to the extent I'm able).
>
> its been explained quite a few times before that applications that use
> GTK and ship a bundle for OS X will almost certainly include GTK
> *inside* the bundle. providing standalone GTK frameworks is almost
> useless for users.
It has certainly been explained that that is the situation on
Windows, and I fully accept it. It's less clear that it should be
the situation on OS X, with its *nix-type substructure.
But even if we suppose that the best way to distribute a GTK app
for OS X is with GTK bundled internally, I don't think that
invalidates the idea that it would be very useful for app
developers to have a GTK runtime package available, as we do for
Windows. As things stand, rolling your own GTK for OS X is a
significant hurdle, and one that doesn't really have to be there.
Allin Cottrell
___
gtk-devel-list mailing list
gtk-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Mon, Aug 30, 2010 at 5:43 AM, Alberto Ruiz wrote: > Hello Paul, > > 2010/8/30 Paul Davis : >>> As long as the people working on GTK-OSX do it with a us-vs-them >>> attitude (like you display here by talking about the GTK developers in >>> third person), things are not going to change. If you start >>> considering yourself part of the team and actively engage, things can >>> and will change. >> >> this is pretty obnoxious. > >> i don't know how tor manages to keep his temper with the windows port, > > This is a resources issue, most of the existing developers are focused > on Linux, and they have no resources/time to focus on Windows > development. > I have contributed several improvements to improve the Windows support > and I have felt more than welcome to do so. I wasn't disputing the resource issue. But there are (at least) two approaches to the resource issue. One is to say "well, we don't have enough (or the right) person-hours to implement this for all backends, but we will go ahead and do it anyway; the backends will catch up when someone does the work there". Another would be to say "well, we don't have enough (or the right) person-hours to implement this for all backends, and because its going to break/change the semantics/stop compilation on some of them, and because GTK is trying to be a cross-platform toolkit, we won't actually push this change until we can figure out how to get it in place for all the backends". Now there is clearly a perfectly good rationale for choosing the first approach - most GTK developers and most GTK users are on X11 platforms, and yes, viewed from that perspective, it doesn't make sense to hold up the development of GTK because of a lack of human resources for other backends. But then if that is the decision, it also doesn't make sense to claim that the sense of "other backends don't really count" is so clearly wrong. Let me give you a recent example, although I am wary of doing so because I don't want to make it appear that I'm being critical of the design decisions of process that was involved. There is work going on for GTK+3 to add a GApplication object to Glib and probably a GtkApplication object to GTK. I've had quite a few discussions on IRC about what this might need to look like to be useful on OS X, and I think that because of this, the result will not be too hard to "port" to OS X (and probably Windows too). But ... the clearly overriding goal for this object has been to provide some functionality for a GNOME/X11 platform, functionality that is already present on OS X and Windows c/o the OS. So although other platforms are loosely taken into account by the design process of this potentially central part of Glib/GTK, the actual reality is that it will provide almost no functionality for apps on other platforms, and will instead be a way to get some specific (very useful) things done within the context of GNOME and DBus. To stress again, there is NOTHING WRONG WITH THIS. I don't have any disagreement with the way things have happened, but I also know that it reinforces *my* personal feeling that for most of the core GTK development team, GTK is an X11/GNOME toolkit that happens to run on other platfoms to some extent, rather than a cross-platform toolkit that happens to have some specific support for GNOME. Its therefore no suprise that John and perhaps some others should feel a little "edge-dweller-ish" in their efforts to get GTK fully implemented on OS X. > So yeah, I totally support Matthias here, if you want a better > situation, feel free to JFDI. Everytime I've needed something in the OS X backend, I've had to JFDI and have. I've made numerous improvements to the OS X backend, and will continue to do so on as-needed basis because I already have a full time job as a developer of an app that is actually *using* GTK. ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
Hello Paul, 2010/8/30 Paul Davis : >> As long as the people working on GTK-OSX do it with a us-vs-them >> attitude (like you display here by talking about the GTK developers in >> third person), things are not going to change. If you start >> considering yourself part of the team and actively engage, things can >> and will change. > > this is pretty obnoxious. > i don't know how tor manages to keep his temper with the windows port, This is a resources issue, most of the existing developers are focused on Linux, and they have no resources/time to focus on Windows development. I have contributed several improvements to improve the Windows support and I have felt more than welcome to do so. So this is not obnoxious at all, if the people with the focus and time actually helped, the situation would be a lot better, what you can't ask is that the development of Gtk+ to be stalled until someone shows up and helps with the Windows or Mac OS X port to make a given change. So yeah, I totally support Matthias here, if you want a better situation, feel free to JFDI. -- Cheers, Alberto Ruiz ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Fri, Aug 27, 2010 at 12:48 PM, Matthias Clasen wrote: > On Fri, Aug 27, 2010 at 12:35 AM, John Ralls wrote: > >> You might not like the warnings about the quality of Gtk+ Quartz, but when I >> wrote them a year ago, no one had touched the quartz backend for 8 months. >> Since then, one developer (Kristian Reitveld) has fixed many of the >> outstanding bugs, and some of the other Gtk devs have become a lot more >> receptive to minor patches... but the general attitude remains that it's OK >> to implement (or rewrite) features in Linux, and if it breaks Win32 and >> Quartz, oh well. There's a list of features that aren't yet implemented, or >> aren't implemented completely, at http://live.gnome.org/GTK%2B/OSX/. > > As long as the people working on GTK-OSX do it with a us-vs-them > attitude (like you display here by talking about the GTK developers in > third person), things are not going to change. If you start > considering yourself part of the team and actively engage, things can > and will change. this is pretty obnoxious. i don't know how tor manages to keep his temper with the windows port, but the truth is as john stated it: the core GTK development team has *consistently* demonstrated that its considered fine to implement something/change something for the linux port with the expectation that other backends will just follow along. i have lots of IRC quotes to support this claim, quite apart from the history provided by git. the people working on gtk-osx (which at this point is pretty much kristian plus a few occasional patch providers) do not consider *themselves* to be in an us-vs-them situation IMHO. instead, the attitude consistently displayed on IRC and this channel supports a view of the world in which the core parts of GTK are developed for the linux/gnome platform and then the rest of the world might, at some point, follow along. the very idea of trying to implement major new functionality on all supported backends before its committed to git (or certainly before its released) seems like an anathema to the development process. ***which of course is fine*** as long as you then don't go and get upset when people whose major focus is a backend other than the the linux/X11 one feel distinctly edge-dweller-like when yet another change to GTK is done without much consideration of their chosen platform. ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Fri, Aug 27, 2010 at 4:17 PM, Allin Cottrell wrote:
> On Fri, 27 Aug 2010, Kristian Rietveld wrote:
>
>> For a GTK+ runtime package ("GTK+ Framework"), I think you should
>> check out what has been done in the past. It is by no means an easy
>> task. The latest code and instructions for this are at the GTK-OSX
>> website if I am not mistaken.
>
> I'm aware it's not an easy task. That's why I'm requesting that
> such a runtime package be made available as a download via gtk.org
> (and offering to help in building one to the extent I'm able).
its been explained quite a few times before that applications that use
GTK and ship a bundle for OS X will almost certainly include GTK
*inside* the bundle. providing standalone GTK frameworks is almost
useless for users.
> GTK-OSX is focused on a native Quartz build but I'm talking about
> an X11 build (taking at face value the statement on GTK-OSX that
> the Quartz port is still immature).
the extent to which is it immature has been overstated in this thread.
___
gtk-devel-list mailing list
gtk-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Fri, 27 Aug 2010, Kristian Rietveld wrote:
> For a GTK+ runtime package ("GTK+ Framework"), I think you should
> check out what has been done in the past. It is by no means an easy
> task. The latest code and instructions for this are at the GTK-OSX
> website if I am not mistaken.
I'm aware it's not an easy task. That's why I'm requesting that
such a runtime package be made available as a download via gtk.org
(and offering to help in building one to the extent I'm able).
GTK-OSX is focused on a native Quartz build but I'm talking about
an X11 build (taking at face value the statement on GTK-OSX that
the Quartz port is still immature). Building GTK+ for Windows is
not an easy task either, and it's _very_ helpful that app
developers are able to download a well-built runtime.
Allin Cottrell
___
gtk-devel-list mailing list
gtk-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Thu, 26 Aug 2010, John Ralls wrote: [ in response to http://mail.gnome.org/archives/gtk-devel-list/2010-August/msg00244.html ] > Gtk-OSX is *not* part of Gtk+. Well, perhaps that's the problem right there. I've got nothing "personal" against Gtk-OSX but it seems that the main GTK site should not be sending people who are interested in GTK on OS X directly to http://gtk-osx.sourceforge.net/ . There should be a link, obviously, but it should be subsidiary to a more general account of GTK on OS X, which -- as things stand right now -- is probably going to mean GTK-X11 for most uses. > Building Gtk-Xll on OSX isn't Gtk-OSX's department. OK, I understand that, but I think it should be GTK's department. > Try http://www.macports.org/ and http://www.finkproject.org/ for > that. You can install GTK via Fink, but that's not what I'm talking about. It's much too complicated for most users. I'm talking about a pre-built GTK Framework bundle that you can install by simple point-and-click. As I mentioned, the R project people have such an installer, and IMO the same should be available from the main GTK site. > My experience with Gnucash is that there are few Mac users who > even know what X11 is, and even fewer who want to deal with it. Since Leopard, at least, Mac users don't have to know anything about X11 to use it as a means of running GTK apps. They just install two bundles, the GTK one and the app one (or they could even be combined). The bundled app contains the "magic" for starting X11 on demand. The only downside is that the GTK app is a little "foreign" in some ways, but if the app is useful enough people can get over that pretty quickly. I've used my GTK statistical app with my students, both PC and Mac users, and the Mac users don't seem to have any special difficulty with using it. Allin Cottrell ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Fri, Aug 27, 2010 at 12:35 AM, John Ralls wrote: > You might not like the warnings about the quality of Gtk+ Quartz, but when I > wrote them a year ago, no one had touched the quartz backend for 8 months. > Since then, one developer (Kristian Reitveld) has fixed many of the > outstanding bugs, and some of the other Gtk devs have become a lot more > receptive to minor patches... but the general attitude remains that it's OK > to implement (or rewrite) features in Linux, and if it breaks Win32 and > Quartz, oh well. There's a list of features that aren't yet implemented, or > aren't implemented completely, at http://live.gnome.org/GTK%2B/OSX/. As long as the people working on GTK-OSX do it with a us-vs-them attitude (like you display here by talking about the GTK developers in third person), things are not going to change. If you start considering yourself part of the team and actively engage, things can and will change. ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
RE: Gtk-OSX (was: Website proposal for usability)
I am just a humble user of GTK, I have yet to develop the skills to contribute,
but half way through the development of our app, we moved to OS X, from linux.
Reason: Faster, better looking interface, much better looking chassis.
You tell'm John, and thanks for all the hard work to you and Kris.
I was using GTK on Linux, because there is nothing else. really. I have
mentioned this before, GTK is a linux GUI, with SOME cross platform abilities
(not for the faint of heart).
It only looks good native on Linux, because it is designed to be used with
GNOME. It does not look good on Windows or OS X. I have sat days with users,
applying all kinds of themes (half of which don't work), only to get the same
block-ish look, on Windows and OS X.
There is a HUGE gap, between what is being touted on GTK's website, versus
reality.
"GTK+, or the GIMP Toolkit,
is a multi-platform toolkit for creating graphical user interfaces.
Offering a complete set of widgets, GTK+ is suitable for projects
ranging from small one-off tools to complete application suites."
http://eboyjr.homelinux.org:8081/
The only platform that has a complete implementation is Linux. Windows has a
lot of issues, and in OS X most of the imp are TODO TODO TODO. Without the help
of GTK-OSX, I don't even know where I would be!
So I would certainly be supporting the work, not chiding them for telling the
truth. If the point is indeed to be multi-platform offering a complete set of
widgets, I would certainly be spending a great deal of effort looking at Quartz
and some of the functionality it touts, and applying them backwardly to
windowing systems that don't. I would also spend a good amount of time on
another project that needs to be brought into the fold. GTKGLExt. Most of the
OS X interface uses OpenGL now.
I would also make sure to point user to some place where they can quickly see a
chart of widgets and functions, and what platforms they are complete, in
progress, or TODO. This way, (as the archives will show) hapless programmers
like me, will not leap to far before they get the truth.
WidgetOS X WindowsLinux
GtkWindow60%99% 100%
GtkButton
...
...
> Date: Fri, 27 Aug 2010 10:46:55 +0200
> Subject: Re: Gtk-OSX (was: Website proposal for usability)
> From: k...@gtk.org
> To: jra...@ceridwen.us
> CC: gtk-devel-list@gnome.org; cottr...@wfu.edu
>
> On Fri, Aug 27, 2010 at 6:35 AM, John Ralls wrote:
> >> I don't know how many people share these views, but if I'm not
> >> totally out on a limb I would be willing to draft a page along the
> >> lines I'm talking about (recruiting help from those who are more
> >> knowledgeable). I'd also be willing to try making a runtime
> >> package if I can get some time on OS X -- though I suspect others
> >> are better qualified than I for that job. The R guys have
> >> some packages at http://r.research.att.com/libs/ and maybe one
> >> of them would be willing to do an "official" build.
>
> For a GTK+ runtime package ("GTK+ Framework"), I think you should
> check out what has been done in the past. It is by no means an easy
> task. The latest code and instructions for this are at the GTK-OSX
> website if I am not mistaken.
>
> > You might not like the warnings about the quality of Gtk+ Quartz, but when
> > I wrote them a year ago, no one had touched the quartz backend for 8
> > months. Since then, one developer (Kristian Reitveld) has fixed many of the
> > outstanding bugs, and some of the other Gtk devs have become a lot more
> > receptive to minor patches... but the general attitude remains that it's OK
> > to implement (or rewrite) features in Linux, and if it breaks Win32 and
> > Quartz, oh well. There's a list of features that aren't yet implemented, or
> > aren't implemented completely, at http://live.gnome.org/GTK%2B/OSX/.
>
> I would say the quality has been slowly increasing, though there's
> enough left to do. I do try to track the latest developments in GTK+
> master and adapt the Quartz backend wherever necessary so it does not
> break. This is also pretty time consuming, but did result in a Quartz
> backend that continued to work when the XI2 and rendering-cleanup
> branches where merged into the master branch. There's some more
> backend work planned I think, that will hopefully affect the Quartz
> backend to a lesser extent. In the meantime I will continue with
> reviewing patches/implementing missing features to end up with a
> feature-complete backend some day :)
>
>
> regards,
>
> -kris.
> ___
> gtk-devel-list mailing list
> gtk-devel-list@gnome.org
> http://mail.gnome.org/mailman/listinfo/gtk-devel-list
___
gtk-devel-list mailing list
gtk-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Fri, Aug 27, 2010 at 6:35 AM, John Ralls wrote:
>> I don't know how many people share these views, but if I'm not
>> totally out on a limb I would be willing to draft a page along the
>> lines I'm talking about (recruiting help from those who are more
>> knowledgeable). I'd also be willing to try making a runtime
>> package if I can get some time on OS X -- though I suspect others
>> are better qualified than I for that job. The R guys have
>> some packages at http://r.research.att.com/libs/ and maybe one
>> of them would be willing to do an "official" build.
For a GTK+ runtime package ("GTK+ Framework"), I think you should
check out what has been done in the past. It is by no means an easy
task. The latest code and instructions for this are at the GTK-OSX
website if I am not mistaken.
> You might not like the warnings about the quality of Gtk+ Quartz, but when I
> wrote them a year ago, no one had touched the quartz backend for 8 months.
> Since then, one developer (Kristian Reitveld) has fixed many of the
> outstanding bugs, and some of the other Gtk devs have become a lot more
> receptive to minor patches... but the general attitude remains that it's OK
> to implement (or rewrite) features in Linux, and if it breaks Win32 and
> Quartz, oh well. There's a list of features that aren't yet implemented, or
> aren't implemented completely, at http://live.gnome.org/GTK%2B/OSX/.
I would say the quality has been slowly increasing, though there's
enough left to do. I do try to track the latest developments in GTK+
master and adapt the Quartz backend wherever necessary so it does not
break. This is also pretty time consuming, but did result in a Quartz
backend that continued to work when the XI2 and rendering-cleanup
branches where merged into the master branch. There's some more
backend work planned I think, that will hopefully affect the Quartz
backend to a lesser extent. In the meantime I will continue with
reviewing patches/implementing missing features to end up with a
feature-complete backend some day :)
regards,
-kris.
___
gtk-devel-list mailing list
gtk-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
Hi John, 2010/8/27 John Ralls : > > Gtk-OSX is *not* part of Gtk+. Maybe that's something we should fix? Resources around Gtk+ are already way too fragmented. -- Cheers, Alberto Ruiz ___ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
Re: Gtk-OSX (was: Website proposal for usability)
On Aug 26, 2010, at 8:08 PM, Allin Cottrell wrote: > On Thu, 26 Aug 2010, Devin Samarin wrote: > >> I moved the URL from >> >> http://eboyjr.homelinux.org:8080/gtk/ >> to >> http://eboyjr.homelinux.org:8081/ > > This looks good to me. But given that the website is getting a lot > of attention, I'd like to suggest one area where the content needs > to be changed: the material relating to GTK on OS X. > > The new draft site basically does what the old one does, namely > hands off to http://gtk-osx.sourceforge.net/ for everything to do > with OS X. But that site has some serious problems. It has an > old-fashioned, clunky look. Worse, the author(s) spell the name of > the target operating system incorrectly: Apple's OS X is > consistently referred to as "OSX". Much worse still, it gives an > overall impression of pessimism and disarray. To quote: > > "The native Quartz display is handled by libgdk-quartz and > libgtk-quartz... Unfortunately, these libraries are not yet > feature-complete. What's more, while most other Gnome > functionality can be made to work on OSX, few if any of the > developers have any interest [sic] cross-platform compatibility. > Developers considering GTK+ as a cross-platform environment for > new work are advised to evaluate other toolkits carefully before > committing to GTK if they consider OSX an important market." > > A nice advertisement -- not! > > It may be that the Quartz port of GTK is stalled. (It looks that > way, though I'm not an expert on the topic.) But if that's so, I > can think of a good reason why it might be: Apple ships a decent > X11 implementation with current OS X, and installs it by default, > so that GTK apps work well on the Mac without GtkQuartz. Sure, it > would be nice to have totally native GTK apps on the Mac, but > that's a luxury and I can imagine why working on it might not > motivate many people all that strongly. > > IMO, the site for GTK on OS X should "accentuate the positive" > (i.e. GTK-X11 works well) while also talking about the Quartz port > objectively and (if this makes sense) encouraging developers who > are fans of both GTK and OS X to contribute. > > One more thing: the site should offer a downloadable binary > runtime package for X11-based GTK on the Mac. Many GTK app > developers, I suspect, develop on Linux but wish to offer Windows > and Mac versions. We don't necessarily have time, opportunity or > interest to build the whole GTK stack for the target OS. > (Cross-compilation is of course not trivial.) We can download a > Windows GTK runtime to distribute, and that's great. It would be a > big step forward if we could also download a Mac runtime. > > I don't know how many people share these views, but if I'm not > totally out on a limb I would be willing to draft a page along the > lines I'm talking about (recruiting help from those who are more > knowledgeable). I'd also be willing to try making a runtime > package if I can get some time on OS X -- though I suspect others > are better qualified than I for that job. The R guys have > some packages at http://r.research.att.com/libs/ and maybe one > of them would be willing to do an "official" build. > Gtk-OSX is *not* part of Gtk+. Sorry you don't like the website. It isn't intended to be flashy or an advertisement. It's intended to show developers who want to port their Gtk+ applications to native Quartz and make a shippable bundle that Mac users will actually try out. You might not like the warnings about the quality of Gtk+ Quartz, but when I wrote them a year ago, no one had touched the quartz backend for 8 months. Since then, one developer (Kristian Reitveld) has fixed many of the outstanding bugs, and some of the other Gtk devs have become a lot more receptive to minor patches... but the general attitude remains that it's OK to implement (or rewrite) features in Linux, and if it breaks Win32 and Quartz, oh well. There's a list of features that aren't yet implemented, or aren't implemented completely, at http://live.gnome.org/GTK%2B/OSX/. Building Gtk-Xll on OSX isn't Gtk-OSX's department. Try http://www.macports.org/ and http://www.finkproject.org/ for that. My experience with Gnucash is that there are few Mac users who even know what X11 is, and even fewer who want to deal with it. For the most part they want a draggable app bundle, though they'll put up with an installer bundle if they have to. They certainly don't want to go through the pain of building Fink or Macports only to have the whole thing fail because the packager of x requires a newer version of foo than the packager of x's dependency y, and upgrading foo deletes y. (Yes, that happens. Often.) (In case you're wondering, I do this because there are a couple of Gtk+ applications that I want to use, and I don't want to have to deal with X11, or put up with the version-hell that plagues Fink and Macports.) Regards, John Ralls ___
