Open Source Is Choice

If you haven’t been following along with the licensing kerfuffle surrounding the open-source Audacity audio editing software, take a sec to read Tom Nardi’s piece and get up to speed. The short version is that a for-profit company has bought the trademark and the software, has announced plans to introduce telemetry where there was none, made ominous changes to the privacy policy that preclude people under the age of consent from using the software, and requested that all previous developers acquiesce to a change in the open-source license under which it is published. All the while, the company, Muse, says that it will keep the software open, and has walked back and forth on the telemetry issue.

What will happen to “Audacity”? Who knows. But also, who cares? At least one fork of the codebase has been made, with the telemetry removed and the old open licenses in place. The nicest thing about open source is that I don’t care one bit if my software is named Audacity or Tenacity, and this is software I use every week for production of our podcast. But because I haven’t paid any license fees, it costs me absolutely nothing to download the same software, minus some anti-features, under a different name. If the development community moves over to Tenacity, it’ll all be fine.

Tom thinks that the Audacity brand is too big to fail, and that Muse will have a hit on their hands. Especially if they start implementing new, must-have features, they could justify whatever plans they have in store, even if they’re only available as a “freemium” Audacity Pro, with telemetry, under a more restrictive license. When that does happen, I’ll have to make the choice between those features and the costs, but I won’t be left out in the cold as long as the Tenacity fork gets enough eyes on it. So that’s just more choice for the end-user, right? That’s cool.

Compare this with closed source software. There, when the owner makes an unpopular decision, you simply have to take it or make the leap to an entirely different software package. This can be costly if you’ve gotten good at using that software, and between licenses and learning, there’s a lot of disincentive to switching. Not so in this case. If I don’t want to be tracked while editing audio offline, I don’t have to be. Woot.

The elephant in the room is of course the development and debugging community, and it’s way too early to be making predictions there. However, the same rules apply for devs and users: switching between two virtually identical codebases is as easy as git remote add origin or apt get install tenacity. (Unpaid) developers are free to choose among forks because they like the terms and conditions, because one group of people is more pleasant to work with, or because they like the color of one logo more than the other. Users are just as free to choose.

Time will tell if Audacity ends up like the zombie OpenOffice, which is downloaded in spite of the much superior LibreOffice just because of the former’s name recognition. I know this split riles some people up, especially in the LibreOffice development community, and it does seem unfair that the better software somehow enjoys less reputation. But for those of us in the know, it’s just more choice. And that’s good, right?

79 thoughts on “Open Source Is Choice

    1. More in-line with the article, the FSF and certain factions in the Free Software community would certainly argue that more freedom isn’t always better. It can be downright evil, and I’m not referring to permissive licences.

      Walking through the rather extensive [1], there are more than a few instances where they’ve worked to curtail what people are Free to do with their software and hardware.

      There’s the usual removing of microcode/firmware loading functionality so that the user can’t be tempted by Sin, but then there’s stripping warnings of miroarchitectural vulnerabilities that user really ought to know about. See no evil. Hear no evil. Speak no evil.

      Then there’s Respects Your Freedom only certifying hardware if the ability to modify (or even examine) firmware is crippled. Loading a binary from /lib/firmware? Sinful. Baking it into the hardware so the user can never replace it? The word of God himself. Product contains optional components that would require loading a blob to be functional? Cripple them or burn in hell.

      Purism tied themselves in knots on this one, forever sealing a blob into their Librem phones that has direct access to all of system RAM during operation, just to qualify for RYF. If Raptor has done the same with their network interface, they wouldn’t be able to ship an open source replacement now.

      The icing on the cake was a point-bank an accusation from around 2015 (reproduced in LWN) that Stallman himself was curtailing the Freedom of Emacs users to make their editor the most powerful tool it could be. This was out of fear that proprietary compilers would leech off the GCC AST (instead of using libclang, which is what they actually do).

      A lot of this policy is seriously out of tune with the interests of freedom.

      [1] https://twitter.com/marcan42/status/1377899929209774081

      1. To be VERY clear, I don’t think this is typical of all of “free software” or even of a large portion of the community.

        I like to think most people recognize how utterly self-defeating policies like these are, much users and developers alike hate it when you lock things down.

        But these people are the *identified* figureheads of free software/open source, and that’s a big deal.

      2. These are typical over-reactions that humans tend to make when trying to right a perceived or real threat to their freedoms. The identity left are pushing well past the left and are looping around to the right in a lot of ways. The authoritarian right is pushing well past conservatism and are finding themselves over on the left. The ironic thing is that neither of them understand that they’re doing it. The same seems to be happening to free software activists by your report. It’s frustrating and disappointing, but seems to be normal for certain groups of people (draw your own venn diagrams).

    1. I find it hard to track “late stage capitalism” with software that’s essentially free, and a movement (free software) that didn’t exist 30 years ago.

      Can you elaborate on how “late stage capitalism” applies to free, modern software?

      Always willing to learn something new about economics…

      1. Idk if “late stage” capitalism applies, but capitalism for sure.

        There is still no good financing system, besides for very large projects, though they as well, depend more on companies with all the influence it entails.

        Independent projects often don’t get the same level of quality than supported ones, or if they do, then at the expense of individuals that sacrifice.

        1. If you can call it quality. The presumption that Commercial == Quality is as silly as the other. You call all the virus and security ridden stuff that is Windows, “quality”? I don’t.

          I pick and choose…and quality varies per program, not the developer’s finances, etc.

          1. >> >> You call all the virus and security ridden stuff that is Linux, “quality”? I don’t.

            >> Fixed that for you.

            You think Linux is virus ridden compared to Windows? Your judgement is questionable. Certainly Linux’s problems are legion, but let’s stick to the facts.

    2. @Jul13 said: ‘“anti-features” Yes! Such a perfect term for so much of late stage capitalism.’

      What does that mean? Maybe you prefer Marxism, where all software and communications are State controlled, with telemetry and back doors?

      1. You clearly do not understand Marxism, so maybe don’t throw it around like an insult. I believe you are referring to Communism with the big C. Repeating capitalist propaganda doesn’t make you look intelligent, it makes you look gullible.

        1. @John said to Drone: “You clearly do not understand Marxism, so maybe don’t throw it around like an insult. I believe you are referring to Communism with the big C. Repeating capitalist propaganda doesn’t make you look intelligent, it makes you look gullible.”

          Sigh… The beliefs and teachings of Karl Marx and Communism are like two peas in a pod. Go to Wikipedia’s page on Karl Marx then search for all instances of the words Communist and Communism. Whew that’s a lot of hits, isn’t it John! Next be sure to read the text surrounding those words so you fully understand the context. At the end of this simple exercise if you don’t see the equipollence of Marxism and Communism, then congratulations John you have been fully indoctrinated :-(

  1. I think I’ll just stick with Audacity V.2.3.3. It does everything I need it to and doesn’t track my activity. It also doesn’t auto-update, so no worries about being thrust unwillingly into the corrupt, greedy hands of the idiots at Muse.

    1. That will go well for a year, maybe two. Then you’ll want to upgrade *something* in your system that Audacity interfaces with. And then you’ll hope someone developed the fork.

      1. That’s a Linux problem. Under Windows I only update drivers (better performance) and security (it’s a must in Windows lands as it’s the most popular OS family). My old scanner used software developed in 1997 for Windows 95, and it worked under Win10 last year. Them my daughter took the box with the scanner and banged it a few times against furniture and walls, and scanner mechanism broke. OTOH my Creative webcam won’t work under anything newer than Windows XP x86. It forces obsolescence on driver level on purpose…

        1. Agreed that this is a problem on systems with centralized library management rather than a focus on single-installation shared libraries. Arguably, also, Audacity is relatively relaxed there – it haven’t made significant use of any of the progress in UI libraries of the last ca 20 years.

          Still, this can go wrong when your OS feels forced to update some interface, for example the audio interface (audio is an interesting thing, as on modern devices, you want to have iOS/android style granularity of audio access rights, lest you want to enable your mouse vendor’s tray icon to send sound samples for ML-based marketing analysis to alibaba cloud services)

          1. Well if you REALLY needed it, you could just have a non-networked computer run only the software you need. Never install anything else, never update, never change, and it will never break (software wise). If you’re adept at repair work, it will run your favorite software forever.

        2. Is not a real Linux problem in this specific case either. Personally I feel like I have to walk back a few negative comments I made about audacity only because I finally upgraded to version 3 and a couple bugs that stopped me from using it in some projects were fixed.

          1. I’m on 2.3.0 and had no problems with it for the things I do from time to time. If I update to newest version, I’d block it in firewall. If it goes to cloud, I’d grab a (probably pirated, depending on price) copy of GoldWave or look for freeware or FOSS alternative. I try not to use paid software and often switch to FOSS or freeware because I just can’t afford it and I don’t like piracy anymore. There are too many (pejorative term describing the most hated and despised person), who turn cracks and keygens into Ethereum miners…

        3. Same here. I still play games I bought in the 90’s and use windows software that is no longer maintained. I have licensed copies of old software I later cracked because the licensing servers went down – so they still work – which would not be the case if I had to play along with the Linux ecosystem where everything is in a repository and the people who keep the repository would decide whether to support it or not (they wouldn’t, for the sheer amount of work involved).

          1. You use old stuff that is long out of support, not officially available etc and say that is better than the Linux model, in which you can choose to do the exact same damn thing if you so desire, but almost certainly won’t have to, as M$ obsoletes stuff much more readily than Linux. You can always ask for the older package if you so desire too. At least up to a point old versions are available just for asking to install version x…

          2. Foldi-One, what if there is no alternative software? There might be software so specific, no one made a fork or alternative version of it, and there is no way around it but to keep an entire computer setup obsolete on purpose just to run it.

            On the other hand too many forks lead to not enough people to develop one good version of software. How many Linuxes are no longer developed or supported because they were one-man forks? This is also the reason why Linux won’t be a valid alternative for Windows – there are too many major distros in development. No major software developer would support them all. Besides, a commercial software is hated by FOSS people, even if their offerings are very poor substitutes…

          3. >as M$ obsoletes stuff much more readily than Linux.

            Quite the opposite. The software I’m talking about is not provided by Microsoft, who in turn has provided me a system that still runs this software as it came out of the box. I’d say Microsoft has done a bang up job NOT obsoleting the stuff I paid good money for.

          4. >You can always ask for the older package if you so desire too.

            It’s another matter if anyone will bother to provide it to you… whereas under Windows I don’t need to, because I still have the CD that the original software came on. I don’t need to ask anyone to put it into a repository and maintain it, which they won’t do anyways because maintaining millions of old software titles is more work than it’s worth.

            The common Linux repository model is broken, because even maintaining all the new software you could have is too much work for anyone, so the repositories contain only a subset of the highest profile packages. It was a mechanism designed for system admins to deploy a limited number of pre-approved software packages – not a platform where everyone is free to distribute software without middle men.

            The reason why app-stores on Android and Windows devices work well is because all they do is distribute the installer or apk that you could download independently from anywhere else. It’s not particular to that phone or PC like with Linux distributions, where a package for OS version 18.04 is not directly compatible with 19.06 and vice versa. There’s no need to “port” software packages between revisions and versions of the same operating system because the OS provides reliable standard interfaces.

          5. @Moryc I’d love to see some evidence such software exists… And even when it does its not hard to run older software on very up-todate Linux… Maybe you have to do a tiny bit more work yourself, but then running ancient crap on M$ OS’s is no different, it can take some hoop jumping to make it work – and if M$ has cut support for some required feature you 100% must keep that obsolete machine if you want to use it – where in Linux you can just bring the required bits in and get it to run..

            But its all largely a moot point software that runs on Linux generally speaking doesn’t die, even when it probably should… Software running on windon’t however is probably dead before you even install it more often than not – they got you to pay up for it at version x, that is all you get, bugs and all – there is no interest in fixing or improving anything…

            Also really doesn’t matter what distro you run, the software developer only needs to support whichever bits they interact with – those particular library, and that is not at all different to M$ there. Doesn’t matter if x distro uses a different x,y,z, unless its a required part for your program, and as in many cases the external interfaces are the same either way, your program doesn’t need to care then either… You also really don’t need to worry about supporting anything other that the big ones Ubuntu/Debian etc. And if it works on Debian it almost certainly works exactly the same on all the derivatives. In reality for giving great support you only need to worry about the supporting the root of the two main families of Linux, the strange Linux guru folk that like running stuff like Arch don’t care, they will find a way to make it work if they want it.

          6. A lot of the Windows compatibility issues that Microsoft gets blamed on are actually caused by them trying to act MORE like Linux.

            Like when Microsoft started pushing for driver updates through the Windows Update service. Since Microsoft started rejecting drivers that are not signed and/or distributed by them, all the old drivers for Windows Vista, 7 and 8, quit working in Windows 10, which meant that companies like Hewlett-Packard could choose not to add support for old printers to the new drivers. As a result, all the HP customers had to buy new printers or start hacking their OS to make the printer work every time you get a system update.

            That illustrates the exact problem why a central software repository is a bad idea. The people who control the repository can and will play games with you.

          7. MS don’t tend to break backwards compatibility. It would wreck their ecosystem, and people would walk. That’s why you’ve got IE in Win10.

            Linux OTOH gives you “oh, grab the old source and build it yourself”, which is great if (a) you know how, and (b) you can find a combination of source, libraries, compiler etc which will work together. It’s quite hit and miss.

          8. Dude > Windows I don’t need to, because I still have the CD that the original software…

            You can do the exact same damn thing in Linux, if you have the entire program, which you can have, you can run it. But why you would ever want to do that when software in the opensource world is forever getting better, not stuck in past….

            Also I’ve seen many websites now offering a windows like download and run installer for linux – which is doing nothing at all different to the repo model, its just adds their software to the package manger’s list. Or sometimes is effectively an “install.exe” who’s only tie to the package manager is that the actual install is handled by it – so it can then remove it properly and be sure any dependencies are present..

            There is no damn middle men required – you want to bundle software for Linux you can, in any one of a number of methods – just pick one you like, a PPA is a common choice but its not the only one. Nothing in Linux is specific to that device either, the Linux packaged repository model is basically the same damn thing as the Android App store, works in basically an identical fashion – Android is basically Linux with a little Google thrown in, so they can take a cut of all software used on their platform, and gather yet more data on everyone…

            You can also just distribute a flatpak of your app – and that will run on any Linux box unless it is set up to prevent such things, probably till the end of time, or at least close enough it makes no matter…

            I’d also point out that WINE is the only way to run many many older windows native applications – as M$ has dropped support for its older stuff – so many of your old programs might only run on obsolete Windoze or via Wine on Linux…

          9. Dan, there is this great piece of FOSS software for musically inclined called Ardour. It’s available for Linux for free. For some time they had Windows version that was behind paywall. Their attitude is this:
            “Want to get it for free, you dirty Windows Scum? Compile from sources. Good luck, as we only know one or two people who managed that, worldwide.”
            Unless you are a programmer or software developer, compiling something from sources is a complex, arcane process, a chore with low success rate…

            Also Microsoft makes most of its profits on corporations and governments that run Windows. So they made it as user-friendly as possible. As backwards-compatible, as possible. They know that corporations would rather not upgrade to newest Windows than to be forced to upgrade or ditch all their special software and tools. Linux and FOSS won’t win at this because there are too many distros, forks, versions, repositories and standards that are incompatible, inconsistent and poorly designed. Windows has a GUI. Linux has a dozen, and each is incompatible with the other. 99% of tasks done in Windows is done via GUI. In Linux that’s 20% at best. You can use bash or any other of dozens command line interpreters to solve problems under Linux that Windows user never experienced.

            LibreOffice is great because it mimics MS Office.
            GIMP is great because it mimics Photoshop.
            Blender became great when they decided that they need to be user-friendly, not user-hateful.
            I can’t stand KiCAD because it mimics software developed 30+ years ago for engineers. I can’t stand FreeCAD because it has terrible user interface based on Object-Oriented Idiocy.
            LMMS, another great piece of software, mimics FLStudio (formerly FruityLoops, don’t confuse it with cereal). It lacks some critical features, like a sampler, audio recorder and good VSTi support.
            You see the pattern?

          10. Linux GUI’s are not half as incompatible as you are suggesting Moryc, in most cases while it might not work optimally software will just work in any of the desktop environments even if it was ‘meant’ for KDE say.

            Also don’t knock FreeCad it is absolutely fantastic software – just because it doesn’t work exactly like the software you already know doesn’t negate that. If you choose not to learn a new bit of software that is on you. Its also fine to like a different logic to the programs operations, as in some other software – but neither make the software you dislike bad. It might be bad for other reasons, but FreeCad really isn’t, its damn good already and developing at quite remarkable pace.

            As for compiling from source, for most things it really isn’t hard at all – just get the project and dependencies, do a config pass so make knows what it will be doing and make (or which other ‘make’ tool the project prefers – its easier not to swim against the devs though you could…). Some projects are more complex, its certainly not always trivial, but often it really is. At least to compile for Linux…

          11. Oh and in many more mainstream disro’s now you really can do as much, probably even more config than windows allows from the GUI if that makes you comfortable…

            Still probably only 50% of the total configuration options linux provides, maybe even less – but it covers on most mainstream desktop oriented distros 100% of what Windoze does, has a less horrible hunt for the setting – is it control panel? start menu Settings? types bollocks Win 10 is infected with and just works.

            Might well have a been a valid complaint in the 90’s… Maybe even into the 2000’s, but it really isn’t now unless you go looking for a disto that chooses not to bother with a silly GUI settings – config files work and are vastly more versatile – so there can be good reason to drop lots of common GUI elements – even the GUI entirely… But that is definitely a distro you shouldn’t choose – its meant for real nuts and bolts techies to implement stuff windows can’t do – like the standalone networking gear, or high performance x on baby processor, RAM, storage or power budgets…

          12. Foldi-One, I actually used FreeCAD as my first CAD software. I even finished a small project in it. The software itself is quite awesome. The UI is horrific. And if you are a bit visually impaired, as I am, simple solution of inverting colors for contrast improvement doesn’t always work. I could adjust many colors in newest version, but that’s tedious and doesn’t fix the main problem: UI is bad and user experience is worse. So I switched to Fusion 360, very complex piece of software with many bugs, but both UI and UX is much better. I’d rather use it despite many bugs and shortcomings than try again with FreeCad. Blender had shown that the UI can be fixed. And that’s the only thing FreeCad needs – a better UI and user experience…

            A task for you: grab Ardour sources and compile it for Windows 10 x64 under Linux. All compilers are cross-platform, so it shouldn’t be a problem. I don’t need the software, just report here, how many days it took…

          13. How long ago was your FreeCad use?
            As I would say its user interface is just fine, and has been for quite some time. Its not what I’d call pretty, but its got everything you need, and in groupings that make sense, kind of reminds me of Gimp, lots and lots of complexity you can get lost in, but the core stuff is there.

            But I don’t have any visual impairment, so if it can’t be as easily configured to suit you I can understand why you would think otherwise – I’ve seen what that it is like, and don’t envy you – though how you can run windoze at all I don’t know, trying to get windon’t 10 to do the right colour profiles and not glitch out, or have vast amounts of now blinding brilliant white spots where it didn’t take just on M$ programs.. What my aquaintance needed looks horrible to me at the best of times, but its even worse when it just won’t work right, very adamant they wanted windoze though, and couldn’t stick with the one that was already working, so suffer I did…

            I’m not going to try compiling anything for Windon’t… Its never worth the hassle, as that means I have to actually run Windoze to test it… And I only do that these days for games that have anti-cheat that forces me to (in a VM)..

          14. @Foldi-one: The subset of settings Linux exposes to the GUI is nearly 100% of what Windows does? Great, find me a way to stop my Ubuntu 18.04 laptop from going to sleep when I close the lid.

        4. The security aspects of this (and flatpack, for that matter!) tend to be overlooked. If you use the same libraries that the software was distributed with, the software will always work, but this also means that security vulnerabilities in those libraries won’t be fixed, either.

          Considering software from 1997, if they happened to use OpenSSL, there are hundreds of exploits out there that would apply. Maybe it’s image editing software from 2012 that’s exposed to dozens of libjpg exploits, or perhaps it’s the software that’s controlling your old 3d printer, that linked against Prusa’s slicer and can be exploited with something as simple as a 3d model.

          Sadly, both choices have huge flaws. For me, having my shared libraries in one place makes it much easier to keep up with these sorts of things, but it also means accepting that some software will break with alarming regularity. For others, the balance might absolutely tilt in the other direction, but there are very large tradeoffs that have to be weighed to make that decision.

          1. I generally just keep near but not on the cutting edge of Linux stuff so everything is updated for security, but tested enough to be that Linux level of Stable M$ can only dream of (though they have done alright with 10)… Staying at that point you rarely get any breakages, I actually don’t know if I’ve ever had one on one of those systems I didn’t cause myself… The cutting edge when you play on it is definitely double edged, all the newest toys, and all the new bugs.

            I do however have no trouble using old software if and when there is a reason to, which in FOSS doesn’t happen often. Just do it sensibly and know the risks – which as you quite rightly pointed out can be quite considerable.

    2. What is it about a company’s desire to make money, to stay in business, do you think is greedy? By the very definition of most companies, they exist to make a profit. If buying the ownership of a software package is cheaper than developing from scratch, I don’t consider that greedy but rather good business.

      In the case of Audacity, color me shocked that FOSS projects are available for out-right purchase. Money talks.

      1. That’s because you (purposefully/willfully?) don’t understand what people are finding greedy. The desire to make money to stay in business (notice I omitted your comma) is fine. The desire to create unending profit quarter after quarter, to pay insane amounts of money to people who are not doing insane amounts of work, to buy competitors, to restrict choice and repairability, to injure the consumer as a means of making such profits; these things and more are what people find to be greedy. Stop mischaracterizing others’ ideas and you’ll find them easier to understand.

    3. DainBramage said: “I think I’ll just stick with Audacity V.2.3.3. It does everything I need it to and doesn’t track my activity. It also doesn’t auto-update, so no worries about being thrust unwillingly into the corrupt, greedy hands of the idiots at Muse.”

      Anyone can download Audacity v2.3.3 and its accompanying manual from here:

      https://archive.org/details/audacity-2.3.3

      Archive.org should be a much safer source than those dodgy file download sites, but be sure to let something like VirusTotal.com have a look at your download first before installing (.exe file) or unzipping and running (.zip file – recommended). For my archive.org download VirusTotal says Audacity-2.3.3.zip is 0/59 (safe) and audacity-manual-2.3.3.zip is 1/60 (very likely safe too).

  2. Important point about LibreOffice.

    Reactions to Muse Group’s handling of Audacity vary quite a bit and become somewhat heated. Some of them end up dismissing Audacity itself, usually by comparing it to solutions which have little to do with it (e.g. Reaper, by Cockos).

    Something which still isn’t discussed that much is the way a project ties people together. (People often say “community” though “network” is more descriptive for this kind of social structure.)

    A project split is a wedge in that network. Yes, devs can all shift their efforts to one branch and many users might migrate to the same branch. There’s a whole group dynamic at stake. And strife isn’t out of the question. Taking quite a bit of energy out of the system. Since we’re talking about reactions from human beings, it’s difficult to predict the outcomes.

    It’s not too difficult to discuss which outcomes are preferable for whom.

    What would be the best outcome? “How Might We” get to that?

    1. LibreOffice is a major piece of software that easily can compete with MS Office. Audacity is a rather small program for a specific niche of users. That’s rather huge difference in scale and reach when comparing software…

    2. @Enkerli said: “It’s not too difficult to discuss which outcomes are preferable for whom. What would be the best outcome? “How Might We” get to that?”

      This never should have happened in the first place. It’s simple, for a FOSS Project and Community regardless of size, never spy on your users by default. If you feel spying on your users is beneficial (it may legitimately be good for everyone), explain your case and add a safe opt-in (not opt-out) method.

    1. Seriously, since when is having a political mission a problem for FOSS?

      Specifically, I spend loads of time on FOSS development, because I believe in free access to resources to benevolent users all over the world.

      If you can’t join a project because it states its mission in its name, *maybe don’t join that project*. I know that sounds elitist at first, but *every* project is a conglomeration of people working on a base of a minimal common understanding on goals. That’s why projects have CoCs; that’s why there’s communities, not just commercial interest groups with mutually binding contracts forcing each other to work in the open.

      So, honestly, if “Tenacity” is too much of a political statement, think about how Linux is Linus’ UNIX that’s deliberatively not Minix. What about Mozilla? Mozilla.org currently has a **very** political title, “Internet for People, not profit”; does that really stop you from using Firefox?

      I think it’s a fostered misbelief that any aspect of human community needs to be free from “too much politics”. It’s not; the fact that it happens in community means people *need* to talk about, and settle on, what their goals are; otherwise, the big players just strongarm the rest. (This interestingly suggests that if you want to sell, say, most cars with ads on sports events, you are very interested in these not discussing the role of sponsorship money in deciding the sportive direction.)

        1. Hm, I’ve yet to encounter *one* place outside the core FSF (read: stallman himself) where that is true! And I’ve been the maintainer of project whose name starts with GNU for about three years.

    1. There would be some sort of authentication involved, surely.
      And the compile-from-source packages of audacity were not have the telemetry – seems Muse owns the copyrights and can compile a non-free version (or maybe audacity is not copyleft?)

      1. The Muse team think they have the copyright. But contributing coders have pointed out that there was no copyright transfer agreement, so every contributer still holds the right to their code.

        I think the Muse team issued a PR thing, claiming they will just buy out the main contributions, and “rewrite” every minor line they can’t buy. I’m anxiously awaiting another press release explaining how they plan to do that.

  3. You seem to completely ignore the actual downsides to the split? Like sure, all else being equal choice is good, but all else is not equal. When small communities split like this you don’t usually end up with two fully-functioning versions of the software, you end up with one being broken and the other leading the way. You also end up with the minority version stirring up additional trouble to try and gain users and developers, what they need to keep the project alive. Some people just quit the whole mess and never come back. That’s ignoring the entire ideological bent as well.

    1. While that can be true, you also end up with a hopefully friendly competition between the now split software, which can actually lead to new ideas, and more features in both – with less hands on deck for each its going to take longer to implement changes, but at the same time with less voices a good new idea gets heard easier.

      Both might not survive, but while they strive too you get some good development that ends up improving both.

      1. Maybe. There is no guarantee any of what you said will happen (especially in an unpaid environment), and honestly, there aren’t a lot of examples backing you up; for every one success story of a project splitting, improving, and the community repairing itself, there are hundreds of dead projects, where issues pile up, everyone realizes they’re not getting paid enough, and they stop. It’s very hopeful to assume the world will pick up the slack, but the threat/benefit of “friendly” competition has always existed in the form of someone else cloning/forking/merging, yet here we are. Nevermind all of the security issues behind splitting development a million, untrusted ways, the choice between a handful of budding repos all doing the same things is a huge gamble for anyone relying on them for long-term stability, without the technical skills to help themselves.

        1. I did say can, not will.

          Dead projects happen, both in FOSS and not. But most FOSS software doesn’t hit a complete dead end and die unless it has been replaced by something ‘better’. As the need for that software doesn’t go away, so while the pace of development can become much slower there are often still paid devs, and always devs who are also users that push the concept forward. It won’t ever be exactly the same again – big surprise there, may work out worse – though you would need a time machine to know as plenty of projects hit droughts without any drama, but it doesn’t lead to death of the whole tree – there will be a branch or two that carries on, at least until a healthier better tree or newer reworked branch replaces it entirely.

          1. Fair enough. I think it is a common misconception or assumption a lot of consumers-of-FOSS have; that FOSS WILL and MUST and ALWAYS DOES deliver at the speed and quality of for-profit software. Hopefully, the next GitHub feature will be AI-assisted time travel.

    2. If both projects remain opensource, they can take whatever updates the other makes and implement them. The ONLY difference could be that small ideological point. We could end up with Audacity v9 (now with xyz), and Tenacity v9 (now without xyz). The smart people will pick the package they want, and everyone else will do whatever luck decides.

  4. I don’t know why the hell there isn’t something in the GPL or whatever other open source license that says “this name cannot be transferred to a paid or closed source software or otherwise and remains the property of the public.” in legaleese of some sort to protect the original projects ubiquity. it’s literally beyond me in the age of branding and one word searches.

  5. also, tenacity already in my git repository on arch… most annoying part is now i’m going to have to correct everyone when we talk about audio editing software…..

  6. Splits and fights in Open Source Projects are always bad. Sometimes are worst than a dispute like this one.
    I remeber when the Ardour Project didn’t want to merge the funtion of importing projects from Protools, becouse the owner of the project didn’t like the guy that wrote the code, and we had to wait like 5 years or so to have that functionality. ANd when we had it it wasn’t that good either and still need work.
    So, basically we lost 5 years because some stupid fight between a dev and the owner.

    1. “Splits and fights in Open Source Projects are always bad.”

      I disagree. Sometimes fights and splits are essential to making better software.

      Case-in-point: LibreOffice

      The fight was getting Oracle to do something with OpenOffice. The split was when they ignored the community, so the community took a copy of the code and left. The result was an *explosion* of fixes and enhancements and the raging success that is LibreOffice.

      The spiteful move by Oracle of giving the OpenOffice imaginary property to The Apache Foundation instead of the Document Foundation was just icing on the cake confirming Oracle is pond scum.

  7. Everything was awesome here with the exception of stating Libre is superior. I suppose that’s a matter of opinion rather than fact. I’m not really down with these 20-30 year old type editors anyways but if I had to I would pick Apache OpenOffice over a LibreOffice any day.
    For some reason Apache has better features than Libre.
    I recently worked as volunteer for both Libre and Apache and found that open office has this one feature for example.. where if you insert an image into a circle it maintains the outer circle and places the image inside circle.

    Of course Libre office doesn’t do anything like this.
    If you take a look at the source code for Libre you will see open office links and directories everywhere like they didn’t even change it to their own which is kind of sloppy programming from a developer point of view for a fork.
    Libreoffice documentation is also in need of some improvement and when improvements were suggested these were met with resistance because one of the defectors from open office who was now at Libre refuses to take recommendations from an actual technical writer… Like maybe you shouldn’t have mac images featuring specific features not available in Windows or Linux and that sort of thing.

    All in all I think we’re going to move to less localized editors with 20-year-old features and move forward to better platforms and systems to make documentation better.

    I know that has little to do with audacity and tenacity but I am glad to see someone forked audacity when the telemetry came in.

    1. I used Open Office for years. Recommended it to friends & family, and they used it with success.
      Then came the OO update where myself and ALL of them had their windows broken by that update, that also nuked OO. Over the next year, various OO new install attempts also broke windows and the OO didn’t work.
      Uninstalled OO, windows issues gone. Installed LibreOffice, everthing worked.
      Haven’t looked back.

  8. I thought the BIG THING with the new Muse Audacity licencing, was that all new code contributions are free for Muse to use anywhere Muse wants to and under any licence Muse wants to use, be it in Muse Audiacity or any other Muse product, now or anytime in the future.

    1. I like that they banned the entire school from further commits. That was the right thing to do in my opinion. This isn’t a game for dumbass college kids to go screwing around with. If you want to make a point, don’t do it in a way that endangers a ton of people.

      1. Agreed, deliberate sabotage should get proper punishment.

        Though how long tainted code manages to stay undetected in FOSS vs close software to me means they are also missing the point with the article (assuming its like the one I read on this incident of similar title) – its not hard if you have the skills to get hired officially by a closed source software company, making it easy to slip such stuff in where nobody will ever get to audit it, and no new features are likely to break it… Not to mention such things can be slipped in through many other methods too…

  9. Now imagine an alternate reality where the USA decided software wasn’t protected under IP law. Not patentable or protected under copyright…. as it should be. The only push-back is from programmers who can’t imagine how they can monetize their software.

Leave a Reply to JohnCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.