'GNUstep doesn't really leap forward into the present time' -- WHAT DOES THIS EVEN MEAN!?

[gcasa]

Could you PLEASE explain what this means? I see that you did LiveStep and that was looking just fine... but without a word to the list you abandon it! SMDH, this is a PRIME example of EVERYTHING that is wrong with the open source community. No communication? Did you ever even think that we might react positively to whatever you wanted?

Also, I really must say giving the "history" on your wiki as if this project is in any way some kind of successor to OPENSTEP/COCOA or anything like that is ridiculous. GC

[kettle-7]

I think the main misunderstand here is the GNUstep has not many active development that can be seen by many people. The releases are getting less frequent because people working on the project have their own lives and other projects to keep up with, and so as a result the project seems like it's not changing much. Also in terms of appearance it closer matches the original NeXTSTEP than the newer early 2000s OS X that this project tries to imitate.

[gcasa]

As you can see from looking at the previous project created by the lead of THIS project, (LIVEStep) GNUstep can be themed to look very much like macOSX. It surprises me that developers can't look past the SURFACE of something. SMDH. What you mention is not lost on me, but it strikes me of a great deal of ignorance. I am the lead of GNUstep and I have posted about these developments on twitter, I have pushed and pushed and pushed and people persist in the "It's NeXTSTEP" claim just because of how it looks IN SPITE OF THE FACT THAT IT CAN LOOK EXACTLY LIKE macOS. Learn to look past your nose.

[gcasa]

I apologize for being upset. The fault is not entirely on helloSystem... the GNUstep project has failed to make its capabilities known for a long time. Looking at the website objectively, it is not showing some of the nicer things.

[probonopd]

Please don't be offended by my wording in the wiki, I changed the sentence from the title of this thread to something that is hopefully more acceptable. My intention never was to discredit GNUstep and I apologize if it came across as such.

[gcasa]

https://github.com/gnustep/plugins-themes-nesedahrik

Here is an example of how themable GNUstep is, If people would bother to look at the repos.

[gcasa]

But... waste your time redeveloping all of our work yourselves. You're free to crib off of project which is ACTUALLY rooted in the technology you seek to imitate.

[probonopd]

Hello @gcasa, thanks for checking in!

Before I will answer your questions, let me give some background: As a longtime Mac and Linux user, I increasingly saw shortcomings both in Apple's latest OSes (especially ever-increasing lock-in and lock-down) and with the LInux Desktop platform (or lack thereof, really). Hence I started investigating BSD + GUI options. Of course, GNUstep was a natural path for investigation, and indeed I created a Live ISO based on FreeBSD and GNUstep, as part of various experiments that ultimately led to the creation of helloSystem: https://github.com/probonopd/LIVEstep

There are many aspects about it that I really, really like about it.

This being said,

'GNUstep doesn't really leap forward into the present time'
Could you PLEASE explain what this means?

Let me try to explain what I was thinking when I wrote this. But let me also say that I have updated the wiki with hopefully less misleading wording.

Out of the box, the look and feel at the time of me writing this was stuck in the 80s. Grey rectangular boxes like in the NeXT days. Scroll bars on the left side. Applications that could be minimized to free-floating icons on the desktop. ObjC, no Swift (that I could see).

I was speculating that this could mean two things (or both):

1. It's made for users who prefer the NeXTSTEP-like look and feel to the (early) Mac-like look and feel (e.g., NeXT style menus by default rather than a Mac-like global menu)
2. Users would like a more Mac-like look and feel but developers have not found the time to make the switch, 20 years after the NeXT to Mac OS X transition happened

You seem to be upset that people always see "NeXTSTEP" in GNUstep, yet also you are not making changes that would make it more "Cocoa" out of the box (e.g., by changing some defaults and use the projects you have linked above). I am curious why that is. My (wrong?) conclusion was that 1. (GNUstep people prefer the NeXTSTEP-like look and feel) is the cause.

Now, don't get me wrong: I don't think that the latest and greatest version is always necessarily the best. In fact, I think the Mac has peaked quite some time ago (somewhere around 10.4). But I am looking to put together a system that is welcoming to switchers from the Mac, not so much to switchers from NeXT.

I see that you did LiveStep and that was looking just fine... but without a word to the list you abandon it! SMDH

Well, I would not really call it "abandoned", as the code is still there and I would be more than happy to hand LIVEstep over to caring hands (yours?). But indeed my focus since has shifted on helloSystem, which is using Qt as the native GUI framework. This was done after careful consideration, mainly for the following reasons:

• The hope to attract more developers since Qt and PyQt are more commonly used than GNUstep/ObjC
• A significantly larger selection of "large", cross-platform productivity applications written in the system-native toolkit (e.g., Scribus, Krita, FreeCAD, etc)
• Compared to ObjC (with manual memory management!) Python feels much more accessible to me (read: I would not even have known how to write the Installer in ObjC)

There were also a few things that confused me:

• WindowMaker not being ObjC and not being part of GNUstep - in fact most people seem to use GNUstep with WindowMaker but WindowMaker contains functionality that partly overlaps with GNUstep (at least this is how I vaguely remember it now). At one time I tried uroswm but I couldn't get it to compile. It would be helpful if the window manager and the session manager came natively with GNUstep
• No apparent plan what to do with Etoile and its components. Is it part of the GNUstep roadmap?

this is a PRIME example of EVERYTHING that is wrong with the open source community.

I am not a company with a roadmap, just a regular person with a hobby.

No communication? Did you ever even think that we might react positively to whatever you wanted?

Indeed I tried to reach out a couple of times to members of the GNUstep community (mainly through GitHub issues as I generally avoid mailing lists for privacy reasons), and during that time I got the impression that this community is very small. I won't blame anyone for not responding immediately, as I am also just a volunteer who gets overwhelmed with tasks occasionally, but the response to the tickets I opened was not exactly making me feel enthusiastic about the level of activity in GNUstep related projects.

Also, I really must say giving the "history" on your wiki as if this project is in any way some kind of successor to OPENSTEP/COCOA or anything like that is ridiculous.

Sorry if it came across as such. That is not what I intended to say. What I wanted to say is that helloSystem owes a lot to the conceptual thinking behind the Mac, NeXTSTEP, and GNUStep, and tries to implement those general concepts (e.g., application bundles) using languages and frameworks (e.g., Qt) that are more commonly used today.

As you can see from looking at the previous project created by the lead of THIS project, (LIVEStep) GNUstep can be themed to look very much like macOSX.

It surprises me that developers can't look past the SURFACE of something. SMDH. What you mention is not lost on me, but it strikes me of a great deal of ignorance. I am the lead of GNUstep and I have posted about these developments on twitter, I have pushed and pushed and pushed

Design, to me, is not just how it looks, but how it works. Here are some examples I had trouble with getting configured in a way that would be appealing to switchers from the Mac:

• How to set language, keyboard layout, and mouse speed?
• Entirely remove the miniaturized window icons in the bottom left of the screen. Users don't understand what they are for, and they seem entirely redundant if there is a Dock. Make the icons in the Dock do what those extra icons do
• Single-click on Dock icons rather than double-click
• Make it possible to keep items in the Dock by Ctrl-click/right-click -> Keep in Dock
• Put scrollbars on right side rather than left side by default
• In Icon view, remove the browser toolbars at the top
• Move the bookmarks from the top of the window to a left sidebar and use miniaturized icons (like in Xfce, Gnome, KDE,...)
• In GWorkspace, Alt-Backspace to push things into Trash
• In GWorkspace, make cut, copy and paste of files and directories work
• In GWorkspace windows, make typing the first letter(s) of a file activate it
• In GWorkspace windows, make Alt-DownArrow do the same as Alt-O
• In GWorkspace windows, make Alt-UpArrow go one folder up in the hierarchy
• Make GWorkspace browser column width changeable from the browser itself rather than Preferences
• What happened to Spatial mode? It is described in the documentation but seems to be missing
• Make Ctrl-click/right-click work when clicking on icons in the GWorkspace file manager
• Make Alt-I bring up "Information", a combination of all Inspectors in one window
• Make Alt-Shift-G bring up a text field where one can enter a path to a folder, which then will be shown
• Make Alt-Shift-3 take a screenshot of the whole screen, Alt-Shift-4 of the window, etc.
• Dock on the left rather than right side by default
• Dock shows an arrow or dot for the launched apps, not three dots for the non-launched apps
• Write all shortcuts in the menu in uppercase letters, e.g., Alt-Q instead of #q (could check whether an Apple keyboard layout is selected and then call it Apple-Q) and Alt-Shift-O instead of #O
• Combine Preferences.app and SystemPreferences.app
• Hide .app extension in GWorkspace by default (configurable by GWorkspace -> Info -> Preferences -> Advanced -> "Show all filename extensions" checkbox and defaults write NSGlobalDomain GSShowAllExtensions -bool true)
• Make GWorkspace request thumbnails from thumbnailers (or make its own), e.g., for pngs
• Dock can go to the bottom
• Rename "Recycler" to "Trash"
• Rename "Finder" to "Search"
• Make "Use Terminal Service" the default if Terminal.app is there
• Make the color for selections (e.g., font panel) the same as the color in menus (I prefer the one in the menus)
• Rename " -> Info -> Info Panel..." to " -> About "
• Rename " -> Info -> Preferences..." to " -> Preferences..."

probonopd/LIVEstep#9

So, I really tried, but I never got the scroll bars to the right side (where Mac, Linux, and Windows users expect them), for example.

people persist in the "It's NeXTSTEP" claim just because of how it looks IN SPITE OF THE FACT THAT IT CAN LOOK EXACTLY LIKE macOS. Learn to look past your nose.

Just my 2 cents, maybe it would help to make https://github.com/AlessandroSangiuliano/rik.theme and https://github.com/gnustep/plugins-themes-nesedahrik the default in GNUstep, then make a major release, and communicate a lot about it.

@gcasa thank you very much for the great work you have been doing over many years. If I had to choose between GNOME, many other desktop environments, and GNUstep I would take GNUstep without a second of hesitation. At the same time we should not see open source as an either-or playing field. It's good that there is choice, and GNUstep application bundles should run just fine on helloSystem. There is now also Airyx which is much more ambitious in terms of compatibility - my hope is that we can all work together.

Do let me know if you would be interested in taking over maintainership of LIVEstep. I really would like to see it thrive and succeed, and probably the chances would be highest if someone deeply rooted in the GNUstep project would steer it.

[gcasa]

Hello @gcasa, thanks for checking in!

Before I will answer your questions, let me give some background: As a longtime Mac and Linux user, I increasingly saw shortcomings both in Apple's latest OSes (especially ever-increasing lock-in and lock-down) and with the LInux Desktop platform (or lack thereof, really). Hence I started investigating BSD + GUI options. Of course, GNUstep was a natural path for investigation, and indeed I created a Live ISO based on FreeBSD and GNUstep: probonopd/LIVEstep

I was very happy to see LIVEstep.

There are many aspects about it that I really, really like about it.

That's good to know. Thanks. :)

This being said,

'GNUstep doesn't really leap forward into the present time'
Could you PLEASE explain what this means?

The look and feel is stuck in the 80s. Grey rectangular boxes like in the NeXT days. Scroll bars on the left side. Applications that can be minimized to free-floating icons.

You know as well as I do that modern themes are possible. You used one in LIVEStep for goodness sake. Every aspect of GNUstep can be themed. Application Icons can be turned off or modified. And those things that were missing weren't even discussed before you put aside your efforts on LIVEstep.

ObjC, no Swift (that I could see).

I forked swift a while back. My efforts have been more focused on getting the API to the point where we can build 10.15 apps without much effort. I also wrote libs-xcode and buildtool to make working with xcodeproj files easy without the use of GNUmakefiles.

I have not had time to devote to working on a swift bridge. I am one of the only people on the team who is familiar enough with swift to try this. Also, as a side not, I did once try to write a transpiler (it is not complete) you can find it at https://github.com/gcasa/phoenix

This can mean two things (or both):

• It's made for users who prefer the NeXTSTEP-like look and feel to the (early) Mac-like look and feel (e.g., NeXT style menus by default rather than a Mac-like global menu)
• Users would like a more Mac-like look and feel but developers have not found the time to make the switch, 20 years after the NeXT to Mac OS X transition happened

Given the theming already mentioned I think you are CATEGORICALLY mistaken about the above. GNUstep is made to have as wide an appeal as possible. If we ONLY wanted to appeal to people who wanted square and grey we wouldn't have...

This

Or this...

Or even this...

And a myriad of others. Including a GTK theme.

Now, don't get me wrong: I don't think that the latest and greatest version is always necessarily the best. In fact, I think the Mac has peaked quite some time ago (somewhere around 10.4). But I am looking to put together a system that is welcoming to switchers from the Mac, not so much to switchers from NeXT.

We are not stuck int he past. Period, as demonstrated above. What more do you need to see other than the efforts of myself and others.

I see that you did LiveStep and that was looking just fine... but without a word to the list you abandon it! SMDH

Well, I would not really call it "abandoned", as the code is still there and I would be more than happy to hand LIVEstep over to caring hands (yours?). But indeed my focus since has shifted on helloSystem, which is using Qt as the native GUI framework. This was done after careful consideration, mainly for the following reasons:

I was thinking of forking it since one of the things weighing on my mind is to create a "reference" distro.

• The hope to attract more developers since Qt and PyQt are more commonly used than GNUstep/ObjC

I get it. Qt and such have a larger user base on Linux and other free OS's. But you also have to consider that, since GNUstep is following Cocoa, it also has a large potential base. There have been a few companies and individuals that have ported their stuff over. Keysight/Eggplant, Apportable (a company that used GNUstep Foundation as part of their UIKit -- now PocketGems), Algoriddim (djay software for making music)... and a number of others. The userbase on the GNUstep side is NOT small.

• A significantly larger selection of "large", cross-platform productivity applications written in the system-native toolkit (e.g., Scribus, Krita, FreeCAD, etc)

There were also a few things that confused me:

• WindowMaker not being ObjC and not being part of GNUstep - in fact most people seem to use GNUstep with WindowMaker but WindowMaker contains functionality that partly overlaps with GNUstep (at least this is how I vaguely remember it now)

WindowMaker works with GNUstep apps fairly nicely... that's pretty much where that ends. I have been working on a Dock replacement. GNUstep actually doesn't even need WindowMaker or any window manager at all... the backend can draw it's own decorations.

• No apparent plan what to do with Etoile and its components

Paying attention goes a long way towards helping to understand what happened here. By that I mean... you had to be watching the repos and reading the list. libobjc2 was incorporated into GNUstep as the clang runtime and theming was refactored from the experimental stuff into GNUstep's GUI library so that it works with the existing code instead of overriding it in a messy way (like Etoile did).

Some of the things proposed by Etoile were a little too far reaching, but it is certainly NOT forgotten.

this is a PRIME example of EVERYTHING that is wrong with the open source community.

I am not a company with a roadmap, just a regular person with a hobby.

I understand. So am I, but I admit being very hurt by the simple fact that you put us on the "unacceptable tech" list and spouted unverified facts without so much as a word! When I say "a prime example" what I mean is someone forming an opinion without actually talking to or consulting those involved. Did you ever stop to think that if you had engaged us directly that any one of the team would have been HAPPY to work with you?! I certainly know I would have. The issue is that simply taking a "blow up the world" approach, as you did, does NOTHING to help, but... oh well.

No communication? Did you ever even think that we might react positively to whatever you wanted?

Indeed I tried to reach out a couple of times to members of the GNUstep community (mainly through GitHub issues as I generally avoid mailing lists for privacy reasons), and during that time I got the impression that this community is very small.

I am active on the list. Also it is WELL KNOWN that I am the lead. You could have reached out to me, Fred Kiefer or Richard Frith-MacDonald.

I won't blame anyone for not responding immediately, as I am also just a volunteer who gets overwhelmed with tasks occasionally, but the response to the tickets I opened was not exactly making me feel enthusiastic about the level of activity in GNUstep related projects.

I will look at the tickets you submitted to verify this. It would have helped had you ACTUALLY engaged someone involved with the project. That would have had a much more positive impact.

Also, I really must say giving the "history" on your wiki as if this project is in any way some kind of successor to OPENSTEP/COCOA or anything like that is ridiculous.

Sorry if it came across as such. That is not what I intended to say. What I wanted to say is that helloSystem owes a lot to the conceptual thinking behind the Mac, NeXTSTEP, and GNUStep, and tries to implement those general concepts (e.g., application bundles) using languages and frameworks (e.g., Qt) that are more commonly used today.

Okay, I understand. As I said in a previous post... the issue is also that GNUstep has never done a great job at "marketing" if you will. So much so that people just don't get that... (I am going to capitalize)...

WE ARE NOT JUST A BUNCH OF GITS SITTING AROUND WRITING A NEXTSTEP CLONE THINKING THAT SQUARE AND GRAY GUIS ARE THE THING...

...

WE ARE A BUNCH OF PEOPLE WHO ARE PASSIONATE ABOUT COCOA AND HAVE TRIED TO MAKE GNUSTEP INTO SOMETHING THAT CAN MAKE AS MANY PEOPLE HAPPY AND LOVE IT AS MUCH AS WE DO.

One of my favorite poems that reminds me of GS comes from Tennyson... "We are not now that force which, in old days, moved both heaven and earth, but that which we are we are... one equal temper of herioc hearts, made weak by time and fate, but strong in will... to strive, to seek, to find, and NOT to yield."

GNUstep is a work of love for me.

Anyway... much aloha and happiness. Yours, GC

[gcasa]

Coming back to address some of your edits...

You seem to be upset that people always see "NeXTSTEP" in GNUstep, yet also you are not making changes that would make it more "Cocoa" out of the box (e.g., by changing some defaults and use the projects you have linked above). I am curious why that is. My (wrong?) conclusion was that 1. (GNUstep people prefer the NeXTSTEP-like look and feel) is the cause.

This is a fair assessment. GNUstep folks have gotten used to how it looks. A while ago the decision was made to not pursue making GNUstep into a desktop environment. The thought was, at the time, that the packagers would decide... that never happened.

One interesting side note: If you create a gui using a nib generated with Gorm or using code without setting the style you will get NeXT like controls. Just like how on Windows the "classic" theme is REALLY no theme at all, macOS's "default" is NeXT-ish... this can be addressed by creating a default theme which is installed with the framework. But it should be the users choice.

The best solution is to make it easy to choose and set the theme. I would argue that it doesn't matter what I or other GNUstep maintainers prefer.

Design, to me, is not just how it looks, but how it works. Here are some examples I had trouble with getting configured in a way that would be appealing to switchers from the Mac:

• How to set language, keyboard layout, and mouse speed?
• Entirely remove the miniaturized window icons in the bottom left of the screen. Users don't understand what they are for, and they seem entirely redundant if there is a Dock. Make the icons in the Dock do what those extra icons do
• Single-click on Dock icons rather than double-click
• Make it possible to keep items in the Dock by Ctrl-click/right-click -> Keep in Dock
• Put scrollbars on right side rather than left side by default
• In Icon view, remove the browser toolbars at the top
• Move the bookmarks from the top of the window to a left sidebar and use miniaturized icons (like in Xfce, Gnome, KDE,...)
• In GWorkspace, Alt-Backspace to push things into Trash
• In GWorkspace, make cut, copy and paste of files and directories work
• In GWorkspace windows, make typing the first letter(s) of a file activate it
• In GWorkspace windows, make Alt-DownArrow do the same as Alt-O
• In GWorkspace windows, make Alt-UpArrow go one folder up in the hierarchy
• Make GWorkspace browser column width changeable from the browser itself rather than Preferences
• What happened to Spatial mode? It is described in the documentation but seems to be missing
• Make Ctrl-click/right-click work when clicking on icons in the GWorkspace file manager
• Make Alt-I bring up "Information", a combination of all Inspectors in one window
• Make Alt-Shift-G bring up a text field where one can enter a path to a folder, which then will be shown
• Make Alt-Shift-3 take a screenshot of the whole screen, Alt-Shift-4 of the window, etc.
• Dock on the left rather than right side by default
• Dock shows an arrow or dot for the launched apps, not three dots for the non-launched apps
• Write all shortcuts in the menu in uppercase letters, e.g., Alt-Q instead of #q (could check whether an Apple keyboard layout is selected and then call it Apple-Q) and Alt-Shift-O instead of #O
• Combine Preferences.app and SystemPreferences.app
• Hide .app extension in GWorkspace by default (configurable by GWorkspace -> Info -> Preferences -> Advanced -> "Show all filename extensions" checkbox and defaults write NSGlobalDomain GSShowAllExtensions -bool true)
• Make GWorkspace request thumbnails from thumbnailers (or make its own), e.g., for pngs
• Dock can go to the bottom
• Rename "Recycler" to "Trash"
• Rename "Finder" to "Search"
• Make "Use Terminal Service" the default if Terminal.app is there
• Make the color for selections (e.g., font panel) the same as the color in menus (I prefer the one in the menus)
• Rename " -> Info -> Info Panel..." to " -> About "
• Rename " -> Info -> Preferences..." to " -> Preferences..."

Many of these can be done in GNUstep... but some of these are OS level things.

probonopd/LIVEstep#9

I see. If these issues had be filed in GNUstep they would have been visible to us.

Just my 2 cents, maybe it would help to make AlessandroSangiuliano/rik.theme and gnustep/plugins-themes-nesedahrik the default in GNUstep, then make a major release, and communicate a lot about it.

Also true. But you knew this with LiveStep as well.

@gcasa thank you very much for the great work you have been doing over many years. If I had to choose between GNOME, many other desktop environments, and GNUstep I would take GNUstep without a second of hesitation. At the same time we should not see open source as an either-or playing field. It's good that there is choice, and GNUstep application bundles should run just fine on helloSystem. There is now also Airyx which is much more ambitious in terms of compatibility - my hope is that we can all work together.

I agree.

Do let me know if you would be interested in taking over maintainership of LIVEstep. I really would like to see it thrive and succeed, and probably the chances would be highest if someone deeply rooted in the GNUstep project would steer it.

I will. GC

[probonopd]

Communicate we did:

https://twitter.com/bheron/status/681348123021717504
https://twitter.com/probonopd/status/1308101818166243334

Let's agree that communication was not optimal and I take my part in this. As I said, I am still fully committed to helping efforts for making a GNUstep-centric Live ISO, and in case you would like to take over the lead on LIVEstep you can be assured to have my support with everything I have learned about making FreeBSD based Live ISOs.

I was thinking of forking it since one of the things weighing on my mind is to create a "reference" distro.

So I take this for a "yes"?

[gcasa]

Communicate we did:

twitter.com/bheron/status/681348123021717504 twitter.com/probonopd/status/1308101818166243334

Oh dang! The points you brought up in that post are important. I suppose I was thinking more about email. For better or worse, posts on twitter tend to hold my attention for a moment or so... then I get busy. :/

Let's agree that communication was not optimal and I take my part in this.

Agreed. We both bear a bit of fault there.

As I said, I am still fully committed to helping efforts for making a GNUstep-centric Live ISO, and in case you would like to take over the lead on LIVEstep you can be assured to have my support with everything I have learned about making FreeBSD based Live ISOs.

Sweet

I was thinking of forking it since one of the things weighing on my mind is to create a "reference" distro.

So I take this for a "yes"?

It's a strong maybe. :) I am not sure if I am going to create a reference distro based on BSD or Linux. I will need consult with the other maintainers and get back to you. Your help would be greately appreciated.
GC

[darkoverlordofdata]

If I may interpose - I really like GNU step objective-c. But I find it works better on Linux. The FreeBSD version is stuck at 2.0, and on Linux I can install 2.1. And per the fresh ports page, there is no longer a maintainer for GNUstep on FreeBSD. Not to mention that when I ask for help with GNUstep in forums, the main answer I get is that their project is dead - to which I am sometimes uncertain. - getting help is difficult at best. I would say that the GNUstep team does have a lot of work to do to make it more usable. I view it as legacy on life support.. Too bad, I'd like to use it more.

[gcasa]

You CANNOT be serious about the "the response is that their project is dead". Even a cursory glance at the repo shows high activity.... also you asked once on the mailing list and MY response was:

We are most certainly still active. The website was shifted over because we lost our hosting since the person who maintained it passed away, so we are in a bit of a transition. Once the website is back on its feet, those tutorials should be updated.

GC

So I am not certain where you are getting the impression of non-activity. No one has said that to you on the list which is the only forum you should ask those kinds of questions on.

As far as getting help... Getting help is an email away. I don't understand why people ask for help on mailing lists, and then don't respond with any feedback. We can't anticipate any and all issues. We try to make the documentation.

I work for a company, namely, Keysight, that uses GNUstep to build their eggplant application... here: eggplant. It's available on Linux, macOS, and Windows. Windows is their main platform. It is 100% written on top of GNUstep using Objective-C. We are used by a number of other companies. So I am dreadfully sorry if you think we are inactive, but it is certainly the precise opposite.

So your "interposition" while appreciated appears to lack both accuracy and relevance.

Also, for clarity... I am the lead developer of the entire project. Thx GC

[probonopd]

Great to hear that GNUstep is alive and well, and that at least one high-profile company is using it to build a commercial application. (I'm tempted to see if we can get the Linux version to run on FreeBSD using the Linuxulator...)

Sometimes I wonder how the history of desktop computing would have looked if "Linux" had standardized on GNUstep early on, and had dropped Gtk/GNOME in favor of it. Somehow I have the feeling that either "GNUstep/Linux" would have taken over the world, or RH would have totally wrecked it by now.

[gcasa]

I apologize if my response was harsh but it really feels like people WANT us to be dead. Even misquotes like the one in the previous message. No one EVER EVER said the project was dead and I replied and told him the complete opposite and he still LIED!!!

[darkoverlordofdata]

I’m not trying to start a fight here. I did not mean to say you or your team told me that GNUstep was dead. My meaning is that’s what most people tell me when I ask about using it.
I’ve been trying to use GNUstep, and write objective-c for the past several years. It feels a lot like wearing a paper bag over my head at times. It’s not easy to get any information, and I have no corporate resources to do so, just whatever I can find online that has not 404’d - most of the links on https://gnustep.github.io/ are dead links. It’s frustrating.

[darkoverlordofdata]

In fact, here is the posting that earlier this year renewed my interest in objective-c: https://mclarenlabs.github.io/McLarenSynthKit/#introduction-to-the-mclaren-synth-kit

I learned more useful information from this article than all I had been able to glean prior.

[gcasa]

I’m not trying to start a fight here. I did not mean to say you or your team told me that GNUstep was dead. My meaning is that’s what most people tell me when I ask about using it.

I know you're not and my apologies for sounding harsh. What you said did sound like that's what people were saying on the mailing list but I get that others tell you that.

I’ve been trying to use GNUstep, and write objective-c for the past several years. It feels a lot like wearing a paper bag over my head at times. It’s not easy to get any information, and I have no corporate resources to do so, just whatever I can find online that has not 404’d - most of the links on https://gnustep.github.io/ are dead links. It’s frustrating.

I get and share your frustration. We have had one of our members working on getting the website up to snuff again but that work is going very slowly and it makes us look bad. I am going to make a concerted effort to improve things.

I feel sometimes like I am fighting the tide. Because no matter how hard I push or how hard I work on gnustep no one seems to care. And worse they remake our work either out of spite or ignorance.

GC

[darkoverlordofdata]

I did not mean to push your buttons. You have my heartfelt apology for that.
As far as the hatred - I’ve seen it. It seems mostly about the gcc/clang thing, and then they throw in the gui looking so old, which can actually be fixed with a modern looking theme - I use NesedahRik and disable the AppIcon, and it looks much newer.

The clang thing is hard - GNU purists hate it. I see hateful things said about the GNUstep name over that. But from what I can tell, it was gcc that dropped GNUstep, but there is little clarity there, and in the absence of information, people say terrible things.

Hopefully things will change - clang is starting to get some traction. I think what GNUstep needs is more positive socialization. And a newer more modern theme would also help.

Your passion is impressive. Don’t let the bastards get you down.

[gcasa]

I did not mean to push your buttons. You have my heartfelt apology for that.

Apology absolutely accepted. :). Again, I am sorry for the anger in my response.

As far as the hatred - I’ve seen it. It seems mostly about the gcc/clang thing,

Yes.

and then they throw in the gui looking so old, which can actually be fixed with a modern looking theme - I use NesedahRik and disable the AppIcon, and it looks much newer.

Absolutely... part of this is because GNUstep has been developed mainly as a framework. I trusted the packagers to do a much better job than they have. This is why I have been thinking about a reference distribution to showcase GNUstep at its best. This includes the idea of a new theme.

The clang thing is hard - GNU purists hate it. I see hateful things said about the GNUstep name over that. But from what I can tell, it was gcc that dropped GNUstep, but there is little clarity there, and in the absence of information, people say terrible things.

Indeed, this is true. GCC deems ObjC non-critical and has released at least once with a broken ObjC compiler. They also have not updated the compiler to handle newer syntax which is critical for acceptance by developers and the building of applications.

Hopefully things will change - clang is starting to get some traction. I think what GNUstep needs is more positive socialization. And a newer more modern theme would also help.

I hope so too. I wake up every day with that hope. I agree with respect to positive socialization. I am doing everything I can. The modern theme is something I am thinking a great deal about. I don't know if the reference distribution will use one of the existing themes or something totally new, but it will have all of the available themes packaged.

By the way, that SynthKit link was awesome, I can't wait to download and check it out! :)

Your passion is impressive. Don’t let the bastards get you down.

Thanks, they have yet to succeed. This project is my life's work... it's more than just another Free Software project to me... it's my baby. Scott Christley, one of the early members of GNUstep once asked me 'If GNUstep died tomorrow, would you still be alive?" my response was "Yes, but a large part of my heart would disappear with it."

Yours, GC

[dszidi]

Do let me know if you would be interested in taking over maintainership of LIVEstep. I really would like to see it thrive and succeed, and probably the chances would be highest if someone deeply rooted in the GNUstep project would steer it.

This 100,000%
When Probono decided to go with Qt, he suggested I try and pick up LiveStep. I had a terrible time just trying to get basic things the way I wanted. In the end, the overhead of new language, new framework and uncertainty of compatibility with freebsd made it a non starter for me. This would have ended very differently with a GNUstep expert on board.

@gcasa
Please consider taking on LiveStep. You would have at least one enthusiastic contributor in me and probably many more in time. It would be priceless to have an expert to lean on. Not to mention the BSD community in general is much more Apple related tech friendly than the Linux side of the fence. I think you'd like it here :)

[dszidi]

Another interesting news item that may be relevant for cultivating interest in a return of LiveStep .
Looks like latest swift lang 5.9 now builds on FreeBSD 13.2

@gcasa is there still interest by the GNUstep project to make a UIkit for swift on non apple platforms? I've seen conflicting information on the internet about it

[dszidi]

Oops. just realized this thread was started in 2021. Apologies for reviving a dead thread

[gcasa]

Oops. just realized this thread was started in 2021. Apologies for reviving a dead thread

If you look at the recent replies you'll see it was very recently revived. :)

I'll reply to your other concerns on that thread.

[gcasa]

Another interesting news item that may be relevant for cultivating interest in a return of LiveStep .

Looks like latest swift lang 5.9 now builds on FreeBSD 13.2

@gcasa is there still interest by the GNUstep project to make a UIkit for swift on non apple platforms? I've seen conflicting information on the internet about it

There is. I have identified some frameworks that might fit the bill to be ported to gnustep for this purpose. I will post that link later.

I have looked into extending swift to support GNUstep but I believe that it will be easier to create bridge that will allow swift to call objc methods in gnustep without modifying swift.

[dszidi]

That's fantastic news. Sounds like there is much to look forward to!

[gcasa]

Do let me know if you would be interested in taking over maintainership of LIVEstep. I really would like to see it thrive and succeed, and probably the chances would be highest if someone deeply rooted in the GNUstep project would steer it.

This 100,000%

When Probono decided to go with Qt, he suggested I try and pick up LiveStep. I had a terrible time just trying to get basic things the way I wanted. In the end, the overhead of new language, new framework and uncertainty of compatibility with freebsd made it a non starter for me. This would have ended very differently with a GNUstep expert on board.

My only concern with basing any reference distribution on FreeBSD is exposure.

Linux is more widely used and it would likely see wider use but, as you said, bsd is closer to the roots of macOS. So it might be a more familiar and welcoming environment for it.

@gcasa

Please consider taking on LiveStep. You would have at least one enthusiastic contributor in me and probably many more in time. It would be priceless to have an expert to lean on. Not to mention the BSD community in general is much more Apple related tech friendly than the Linux side of the fence. I think you'd like it here :)

I may take on livestep but I also want to make my own distribution from scratch.

GC

[probonopd]

@gcasa I'd be more than happy to hand LiveSTEP over to you.

My only concern with basing any reference distribution on FreeBSD is exposure.

One thing to consider: Linux is "just" a kernel, and the rest of the stack is cobbled together by various distributions, making it very hard to support things. BSD, on the other hand, is a complete (minus the GUI) operating system engineered and released with the whole stack. For me this is part of why I choose to work on FreeBSD. My biggest gripe with it so far is that graphics drivers are not part of the core system, but apart from that I find it a very stable base to build a system from.