Liam On Fri, Jul 26, 2024 at 9:42 AM Liam Proven <[email protected]> wrote:
> Your formatted-text email makes it a bit hard to pick out quotes. I hope > I didn't miss any. > > > On 25/07/2024 2:37 pm, Gregory Casamento wrote: > > > I can only speak for myself, but I wasn't in the greatest of moods > > yesterday at all. I apologize if I came off as abrasive. > > No problem. It was justified, in context. > As lead I try to do so by example. It wasn't the best example of how to deal with a different point of view. > > If that's so it's best not to make assertions based on assumptions about > > where this project is going. We have had a wayland backend for about 3 > > years. Granted it was never posted under news because it is still > > being worked on. One thing I can tell you from my own experience is > > that announcing something too soon can sometimes be detrimental as > > people will try it and immediately get turned off. > > Fair points. Conceded. > In the future ask... we are a welcoming bunch. :). You could also come to the bi-monthly meetings. You would see that there is plenty going on. I will send out the link in another email to the group soon as we have a meeting coming up in August. You and anyone who wants to attend is welcome. > > There are only about 6-8 active developers right now. This is not > > counting the people at Algoriddim or Keysight who do occasionally submit > > PRs. It's difficult to know when, also, to put someone else's work up > > as news as they may not be ready for it to be reviewed. In other > > words... we are vastly under manned. :) > > Everybody is, unless they work on RHEL or Ubuntu. > GNOME... and their Foundation? And their 1.5million EURO donation from the STF, are they undermanned? > > Very true. If you have noticed, things are a bit weird here in the > > states on that spectrum, but we won't discuss that here. I should keep > > the politics out of the mix. I have questions about using X/Twitter in > > generate since was bought because it doesn't seem that people are really > > listening there anymore. > > Yup. I agree with Lars: switch to Mastodon, and maybe just use > Buffer.app (a free account will do, it's all I have) to schedule posts > to both. > Not a bad idea... X/Twitter (will never just call it X) is very political these days. Many developers are leaving it, so I don't think it provides a platform that's good for the project. > I had one post that was a little bit harsh against a recent client due > > to their misunderstanding of "Free Software" as the person who ran the > > company thought that because I was their "liaison" to GNUstep that the > > work done on GNUstep on their behalf was free. That's why that one was > > missing. > > OK. Then maybe delete the tweet as well? > Ah, yes, I will. > > You mean calling Probono out for what he actually did? > > Not just that. The whole paragraph and the dismissive comments about > overlapping projects, Ricardo's about Cocotron, and so on. > It was only meant to be dismissive about HelloSystem. Cocotron has not seen a significant number of commits in over 8 years and the projects using it have not updated the code written by the original project in a meaningful way. My assessment of cocotron's status (the class I pasted was only ONE example of unimplemented code) is based on my observations of that project. As they were once a competitor I kept an eye on what they were doing. My conclusions are not based on any sort of bias or being dismissive. The are fact. HelloSystem is not a competitor because it seeks to recreate the look of macOS, not the development environment. It does a good job at this and people looking for a macOS like environment may be happy there. RavynOS does seek to be macOS compatible on both the OS level and the API level as, from looking at their repo, I can see that they have made changes to the BSD layer relating to Mach-like calls. But it is also not a competitor because it doesn't seek to offer the APIs on a multitude of platforms, only it's own. Additionally, RavynOS uses the older Cocotron codebase which, as I said above, has not been updated in nearly a decade. These assessments are not dismissive, but based on fact. I am still keeping a close eye on RavynOS. Indeed, I have told the creator of the project that if she needs help enhancing the existing APIs that she can use GNUstep's implementations so long as she gives proper credit and doesn't strip copyright headers or licensing. Just as an aside, I still do and always have had doubts with respect to the originality of Cocotrons code. I watched their forums and saw many references and even cut-paste instances of GNUstep code, but I chose not to make a big deal about it since I had no desire to start a war. Cocotron ultimately defeated itself when Xcode's architecture changed so radically that they couldn't integrate. > > No, it's not. I'm the lead and I have the agreement of the core > > developers. It's a development environment and not a desktop because WE > > SAY SO. > > You continue to miss my point. > > You have built a desktop. As a demo that GS is remarkably complete and > powerful, this is a *huge* thing and you do the project vast harm by > denying it when you ought to be celebrating it. > Are you referring to the LiveCD? That was not an official GNUstep project... even though it was called the GNUstep Live CD. It is undeniable that GS can be used to create a powerful desktop. But, you continue to miss MY point. Our cross platform compatibility is more important than our ability to make a pretty desktop. So, the LiveCD is another example of a third party creating something great out of our stuff and yes it should be celebrated. But it also demonstrates that the CORE project is better at making the APIs great... desktops are better left to others who are good at that. Compared to multiple other FOSS desktops it's amazing. Examples: > > * incomplete (Lumina) > * forks of old codebases (Trinity, MATE) > * the multiple half-done Win95 clones (Cinnamon, Fallback, ZorinOS) > * shallow skins around GNOME (Budgie) > * eternally unfinished and left behind by technology (Enlightenment) > > But you know what? *All* of these are first-class citizens and are > shipping in >1 distro or OS. > Yes, the POINT I am trying to make is that we have other projects READY AND WILLING to package us just like the above. The developers on GNUstep are helping them. The GNUstep project itself WILL remain like GTK is to GNOME. The framework that makes the environment POSSIBLE. GS is _better_. It should be up there with the others. > We are arguing over nuance about how to get there. I am not sure why you aren't absorbing that letting interested parties create their desktop WITH OUR HELP is better than the code developers doing it themselves. The difference is rather negligible. It in fact mirrors how other projects work. > Whether you intended to do it or not, _it is the best advert you have_ > for GNUstep. Stop disowning your abandoned child. It looks really bad. > Embrace it. You guys fathered it. Admit it and bring it into the family. > Oh... we own it. Everyone on the project thinks it's wonderful. It's a pure and simple demonstration of my point. :) > To your point... the desktop environments being done by the interested > > parties: GSDE, Gershwin, NEXTSPACE... are better than we can manage to > > do as the core developers. Those guys are FOCUSED on making desktops. > > That is _not_ the point. Not at all. > > > GNUstep does not have the advantage of being > > adopted by major distros as their default GUI. Believe it or not this > > was supposed to be the case. > > I know. It should be. Ask why not. Ask how you can fix that. > This question has been asked OVER and OVER again by myself and others. I don't know why distributions don't choose us. If it were based purely on merit, they might. But many tend ONLY to go with what is most popular. It's frustrating, but it's the way the world works sometimes. > > The FSF planned on using GNUstep as it's > > MAIN development and desktop environment, but when GNOME was introduced > > it stole our thunder. :) Long story which I won't get into here... but > > Miguel De Icaza was once a member of GS. I'll leave it there. > > Remarkable! Really? When? > If you search for his name in the mailing list archives you'll find it. > Also... what we do is analogous to what some companies that make cars > > do. Mitsubishi, for instance, provides the engines for MANY automakers, > > but they don't make the car. That's what we do. We make the ENGINE > > that makes things go. > > You make my point for me. *Mistubishi makes vehicles.* > True, but they also make engines for other companies. :) https://www.mitsubishicars.com/ > > This is in no way a conflicting goal with being a programming framework. > It is an _advertisement for it._ > > > I'm not so sure. > > My full time job is observing and commentating on the FOSS industry. I > am telling you. Be sure. > I do understand that. It might be useful for you to come on the list and ask what's going on or even to email some of us privately. I would very much LOVE to do an interview if you ever decided to do that. > https://www.apple.com/us/search/cocoa?src=globalnav <https:// > > www.apple.com/us/search/cocoa?src=globalnav> > > > > Note: the programming API is not there. > > > > > > You're delusional. > > I gave the *proof* in my post. Follow the link. It is right there. > > I gave the proof with the latest version of Xcode. When they stop putting Cocoa.framework in there we will stop referring to it as Cocoa. > > Also, it doesn't matter what THEY call it. Whatever > > Apple calls it doesn't matter. > > Yes it does. There is zero point in trying to explain what the project > is and does using names that are no longer used by the people who coined > the names. It is a huge waste of effort. > > See above. Xcode headers and dir. > It's also not helpful to refer to it as OPENSTEP as you did in an > > article a while back given that Apple no longer mentions that either. > > Then come up with something better. > See above. Xcode. > As is OPENSTEP and NeXTSTEP as you used in that poorly written article. > > *Angrily* Tell you what, Greg. If you try to stop insulting my work, I > won't insult yours. Deal? > Deal. See above. LATEST Xcode. *Very Frustrated, but politely* Please respect the wishes of people on this project and *STOP* referring to GNUstep as an OpenStep or NeXTSTEP clone it is absolutely inaccurate. If you feel so strongly that MUST avoid calling it Cocoa at all costs (even though this is what most developers STILL refer to it as), then call it an "implementation of the macOS Programming APIs" whatever suits you better, but *PLEASE* don't write articles *MISREPRESENTING* us as OpenStep. This project WAAAAY SURPASSED OpenStep many years ago in terms of the APIs it offers. Saying we are either an OpenStep clone or a NeXTSTEP clone is very misleading to developers because the APIs follow the macOS APIs. GNUstep is a VAST superset of OpenStep. NOTE: I use the OpenStep capitalization as that was what NeXT used for the APIs. It was OpenStep = APIs and OPENSTEP for the OS... confusing. > So stop using those too when referring to GNUstep. Alright? > > No. > Good grief... *sigh*. So you are PUBLICLY telling the entire project, on the mailing list that you intend to continue to describe this project in terms that everyone here disagrees with? Be clear and be sure about this because this list is archived FOREVER by the FSF. While I am moderator on the list, I can't remove posts from the archive. I am *politely* asking you not to misrepresent us by using those (also dis-used/defunct) brand names. If you are someone who writes about FLOSS projects, then you need to have respect for the opinions or preferences of those who ACTUALLY WORK ON THEM. You need to do two things: 1) Ask a member of this project to fact check your articles 2) Respect the preferences/opinions of members of this project. As a journalist doing both of the above would help your articles be more accurate and useful to your readers and it would also help us reach more people. Additionally, whether Apple uses "Cocoa" is also entirely irrelevant. *Developers* STILL call it this, so that is the word we have been using because it is useful to them. To settle this I am going to ask the question to someone who works at Apple to see what they refer to it as. If I get a response I'll get back to you. As I said in the above paragraph... use "the macOS APIs" or something. If you're concerned about unused or disused terms then OpenStep/OPENSTEP and NeXTSTEP are even MORE disused/disowned than you claim Cocoa is. If you want to help, then help. Using defunct terms, no matter what they are, only makes us look outdated. When attempting to write about us... please ask the people most involved. Sometimes things look very different from the inside. > All of Apple's frameworks are still written in ObjC. > > Irrelevant. The question is not "are they going to rewrite the whole > thing?" because the answer is "of COURSE not!" > > The vastly more important question is: how much _new code_ is being > written in Obj-C? > It's a mix. There are some frameworks coming out in ObjC, but many are coming out in Swift. > > Also, you'll excuse me if I don't take the word of people on reddit as > > anything official. > > Sure. It was the first link I found, nothing more. > Ok. > I have my ear to the ground. I know what's happening. > > I think you are only hearing the whole story, because your worldview > here is radically different to mine. > What's wrong with hearing the whole story? How is your world-view different? > But, PLEASE, as a non- > > programmer, refrain from telling US (the people writing the actual code) > > what language we should use as the core part of our frameworks. > > And AGAIN you misread what I said. I did not say that. I am not even > hinting or implying that. You are putting words in my mouth I would > never say and that is rude and hostile and it is very hard not to be > angered by it. > I apologize. I guess I interpreted what you said as "you guys should stop using ObjC" rather than "you guys should integrate with Swift". The reason is I have heard so many things like this before. I agree with the latter interpretation and I accept that is likely what you meant. > I am not suggesting rewriting a single line. That is absurd. > Okay > *ALL* I am saying is that you need Swift bindings for GS so that Swift > code can be used with GS. That is all. > This is happening. It's important to note here that, when I posted to the swift list the ideal of GNUstep integration was not met with enthusiasm by them. Even Chris Lattner (the creator of Swift) mentioned he thought it would complicate things and that "GNu Step" (sic) should never be integrated. This was a few years ago before the syntax of swift settled down. Things seem a bit more stable now so I am starting, tentatively, to take another look. > -- > Liam Proven - Profile: https://about.me/liamproven > Email: [email protected] - Google Mail/Hangouts/Plus: [email protected] > Twitter/Facebook/Flickr: lproven - Skype/LinkedIn: liamproven > IoM: +44 7624 227612 ~ UK: +44 7939-087884 > ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053 > Yours, GC -- Gregory Casamento GNUstep Lead Developer / Black Lotus, Principal Consultant http://www.gnustep.org - http://heronsperch.blogspot.com https://www.patreon.com/bePatron?u=352392 - Become a Patron https://www.openhub.net/languages/objective_c https://www.gofundme.com/f/cacao-linux-a-gnustep-reference-implementation
