I guess I am missing what this guy is on about. What seems to me to be the case is reading through the article, nothing prevents anyone using the Microsoft stuff in a "Lite" ie: "free as in beer" manner, the "proprietary free distribution" to code up anything they want and distribute it under GPL. Meaning, you can use all the proprietary language extensions you want under any Microsoft license you want, create code, and send it out there under GPL.
It's when you try to use the "Microsoft free stuff" to produce commercial code you want to keep closed source, and make money off redistributing binaries of the software, that you are screwed. In that case - you have to buy the official subscription from Microsoft. You have to make sure your code does not link into anything of theirs that has a license that won't allow you to include it. In other words, you have to hire a maid and code auditor and lawyer and do all the stuff a "real" software company does. And as long as you are doing the "release cycle" thing and continuing to produce new versions - you gotta keep paying the subscription. If you turn yourself inside out you MIGHT be able to use SOME of the "free" stuff to produce code that is proprietary where you can make money selling binaries. But it is going to be risky since you could have the rug yanked out from under you. Why is this a problem? FreeBSD has had a solution for this since it's inception. Ports. You download the source code and compile it ON YOUR OWN FREEBSD SYSTEM. If by compiling the code you are breaking some license - so what. Let them come after you and the 20,000 other guys doing this - assuming they can find you since there's no way for them to know you are doing it since you aren't redistributing the software. I think this is the sheer "laziness" factor of so many users in the Linux community. They want to pretend that their GPL/GNU system is exactly the same as a commercial Windows system. They want to obtain their copy of software by binary only, they want to pretend that there's no difference in the source code of those binaries as in the source code of Windows binaries. When they have a problem they want to complain about a particular binary that someone built and put in a repo to that person and shove reporting responsibility for it on to the maintainer of the package, they don't want to be bothered going to the actual developer who wrote the package. I view binary repos as a short convenience only. Maybe I'm installing on something with a 120GB disk in it I had lying around and I just need a router made up of the operating system and a few binaries. But if I had a problem with those binaries - I'd go to the developers who wrote the CODE, download the CODE onto a big brother Linux server, and COMPILE it myself. Then if the bug still existed I'd try to fix it myself and if it was trivial I'd report the fix back to the AUTHOR of the package. If downstream "maintainers" happened to benefit - great. But I'd expect the AUTHOR of the package to gladly take my fix and incorporate it so that everyone benefits. THAT is what the FSF and the GPL have always been all about. Richard Stallman, author of the GPL has repeated this over and over in talks and lectures. When someone tells you what their license is all about BELIEVE THEM The author here claims: "Open-source was created as a financial weapon to destroy proprietary on-prem software and to ensure file formats (eg. msword doc vs msword docx) remained open for mixing between different pieces of on-prem software. Open-source as a financial weapon is also why making money from open-source is so god damn hard." This is utter rubbish. Open Source was created for ONE reason - to insure the SOURCE CODE remains accessible so that USERS OF THE SOFTWARE can modify it FOR THEIR OWN PURPOSES. The license merely exists so those users can't pull a fast one and do a quick mod of the software, compile it, release the binary and charge people for copies of it. And when OSS was coined - the distinction between cloud and on prem wasn't even a thing. In addition on-prem software IS NOT proprietary. I have a 42 inch rack of servers in my basement that is full of "on prem" OSS software. And I never had a problem making money off it. I did it the old fashioned way - by charging people who weren’t interested in buying my book and learning how to setup their own mailserver using Open Source, for doing exactly this. The world is full of people who aren't interested in learning how to change their own motor oil and paying someone else 2 or 3x cost to do it. Spend a little time learning to change motor oil and you will never be out of bacon. And, I taught that to my kids as well - which is why this last summer my college daughter wasn't working at some fast food place for The Man and "protesting" on the weekend about why minimum wage jobs weren’t living wages - she was out there self-employed, posting flyers and making money cutting people's hedges and walking their dogs. This entire article is - to me - an example of someone whining that Microsoft closed the loophole that used to exist in Visual C++ version 6 - they made 1 build of it and released it and you could use it to build commercial software. Microsoft didn't support anything unless you paid for it - assuming anyone creating commercial software would want access to their tech support under a support contract if they ran into a bug that was a result of the compiler. But then tons of commercial entities took the free version, built their proprietary packages, sold them, without even bothering to buy a single copy of C++6 let alone a support contract. Now if someone can explain to me where I am wrong on all of this in my assumption - please, please do so. Otherwise this seems to me just greedy people pissed off the HOA is requiring them to pay someone to cut the grass now and then because they won't get out and do it themselves. Ted -----Original Message----- From: PLUG <plug-boun...@lists.pdxlinux.org> On Behalf Of Tomas Kuchta Sent: Monday, September 30, 2024 6:27 AM To: Portland Linux/Unix Group <plug@lists.pdxlinux.org> Subject: Re: [PLUG] Code editors and IDEs to use on headless machines Since this is Linux forum and there were nubber of suggestions for using Microsoft vscode product - I would like to recommend checking this blog post: https://ghuntley.com/fracture/ Especially the part when Microsoft describes its build process: "When we [Microsoft] build Visual Studio Code, we clone the vscode repository, lay down a customized product.json that has Microsoft specific functionality (telemetry, gallery, logo, etc.), and then produce a build that we release under our license." Best, T On Fri, Sep 27, 2024, 13:17 Tomas Kuchta <tomas.kuchta.li...@gmail.com> wrote: > I use jupyter for prototyping and documentation. It works great for > Python and bash (I run other language (Tcl, C, awk, SQL, ...) code > under bash) > > To put together the code I use whatever editor is available on > localhost - mostly kate, nedit, vim ... Most other developers around use > (x)emacs. > > Reaching to headless execution machine storage with nfs or sshfs. > Executing the code through LSF, other OSS schedulers or ssh. > > I hate people running vscode on shared machines - it opens millions of > browser and other processes sucking up all RAM and eventually CPU. > > You will probably need to find what works the best for you. It seems > that emacs/xemacs it the most popular around. I personally only know > fow to open file and exit from emacs. > > Tomas > > On Wed, Sep 25, 2024, 11:55 Vince Winter <thine.technoc...@gmail.com> > wrote: > >> I am wondering what people use for coding on headless machines. Both >> editors and IDEs. >> >> I am using vim and shellcheck and pytlint, for example. As projects >> are getting more complex, I am needing better tools. >> >> Bonus points for setups that don't require more then just >> installation, so I don't have to bring my configuration with me between >> systems. >> >
