Exactly, this has nothing to do with free software. Most people are so messed up mentally and emotionally these days that they react like injured animals to any kind of outside stimuli. In the end I'm pretty sure they hate themselves more than anything. Expressing anger over the obvious injustice in this shit world is one thing, flaming any one who threatens to rock your boat another.
There is this joke that there are two kinds of people, those that divide the world into two kinds and those that don't. Is this some kind of elaborate riff on that theme?
He's talking about people complaining that Canonical are developing their own display server Mir, rather than using standards like Wayland. But people have a reason to be suspicious when a company doesn't want to play ball with others. Yes, they might have a great idea that nobody else has thought of, but rather more likely they have some plan for vendor lock-in.
At the time Mir was started, Wayland was not the standard (and lets be honest, still isn't - X is still the standard).
"Use the standard" doesn't ring true when there is none - and having multiple projects do the same thing is not a bad thing - they can learn from each other and try different approaches.
> rather more likely they have some plan for vendor lock-in
I think given Canonical's track record, this is a really unfair accusation. Yes, they did not always invest all the effort to make things like Unity work on other distros as well, but the reason for that was always prioritisation of their time (no one else took up and kept up the porting task either), never of intending lock-in.
Canonical (and Mark Shuttleworth) have given a lot to the free software community, and these are exactly the disingenuous accusations that have clearly and understandably affected Mark Shuttleworth personally.
Wayland it not even finished YET so really who can say it?
Wanting standards defined after some competition exist and we know the best approach could be positive, but not before based on dogma on what should be better on paper but on reality.
A reason to be suspicious? Personally I am suspicious of standards by committee in which anybody at a table have to be happy with unrealistic expectations.
This is "software as politics". Most people don't even know what the best approach is until a least they code sever tests on the real world and work some time on the problem.
Software has also some part of exploration on it. To have small focused teams exploring those ideas makes total sense.
For example the original software for the Iphone was done with a small team of 10-20 people. And it took them some iterations(years of work) to really know what people wanted or needed.
To same something is the standard because we say so is highly presumptuous.
The "grab some popcorn" mentality is offensive. It's actually a symptom of this problem - people aren't taking critical discussion seriously. Everyone suffers because of this.
Privately finding a public spat amusing does not hurt anyone. It's a very human response. The problem is thinking that you are above it and looking down on people who are more in touch with their emotional responses. People who are not in touch with the unsavory parts of themselves set themselves up for depression and worse.
When canonical announced Mir, i was excited and thought let's try how this display server differs to X window system, in case if it didn't work to my taste, you can always switch back or try other ubuntu derivatives. Why need to hate it? It's free, you can try it or leave it.
When it comes to Open Source Software, hate is not the option, you use it or take the Source Code to change according to your needs, but if you are a common user just change the distro or change your opinion.
This is why you should Love Linux or any Open Source/Free Software : https://goo.gl/eP6sy
Mark & the gang thought they could build something better and so they went for it. Their ideas may be wrong, but they took a stab. Most of those criticizing Mir never go beyond mere talk. That's the main difference to me.
That's a nice change of history. Canonical was behind Wayland. They secretly (for 6-9 months) started working on Mir. This while they said they'd do various amount of work on Wayland. Up to that point they publicly said Wayland is the next thing.
After 6-9 they revealed what they worked upon and made loads of incorrect statements about Wayland. Obviously this resulted in pretty harsh response.
Wayland is quite usable in Fedora as well as various other distributions, so your "never go beyond talk": eh? This while Canonical delayed Wayland by not announcing they wouldn't help out with Wayland anymore.
The Qt-based phone OSes (Sailfish, LuneOS, Tizen, Plasma Mobile) all seem to have adopted Wayland. So the claim that Wayland was unsuitable for mobile seems hollow 4 years on.
From my POV, Mark said Mir was better than Wayland without saying how and without delivering. I don't feel like I have anything to criticize except the fact that he wasted his time and set us back in a way. It's not up to me to criticize Mir, since I'm never going to compete against this software. The real question is not: "why is everyone against Mir", it is : "why was Cannonical against Wayland in the first place". The antagonist is not the community, it has always been Cannonical.
Mir, like Unity, was a political play. Canonical keeps trying (and failing) to establish itself as the owner of some core, critical desktop software.
Anyone who knows open-source hackers should know that their bullshit detectors are highly attuned to typical political malfeasance, and the transparent overlays of conventional corporate politics are not going to work on them (other approaches would work supremely well; hackers are not invincible).
I'm actually worried about Canonical after their repeated political failures. Despite their best efforts, Red Hat continues to stomp all over them, with systemd winning over upstart and now Wayland over Mir.
They need to re-evaluate their strategy for land grabs. Do they not have the technical skill on staff to conceptualize technologies that get people on board? Is Red Hat intentionally undermining them (hard sell since most of these projects existed before Ubuntu decided to try to copy and win), and if so, how can they compete?
I believe that jbmorgado is making a more specific comparison than that, Lennart Poettering having in the past expressed similar sentiments to Mark Shuttleworth's here.
He's probably just angry that so many people went from Ubuntu to Linux Mint and his daydreams about becoming a kind of open source Bill Gates didn't really turn out as planned, even though Ubuntu is doing fine as distro.
He needs to chill out and listen more to the people who do the real work.
Hate is a strong word. People can be very critical at times, myself included, but this isn't the same thing. People often vent frustrations and justify their own personal choices by criticising. It might often be a tad anti-social but I don't believe it is generally "hate". Personally I am sad to see less competition for the GNOME technology stack.
For example I (German) have to deal with east European mathematicians in my job (they are actually really nice people). The kind of strictness that they treat people who present mathematical bullshit (they are typically right in their mathematical reasoning) is sometimes on the borderline of what is considered as impolite in Germany (and in the US probably even more).
I also went climbing with former Russians. I just say: What in their culture is considered as normal language is in Germany already considered as sexist (and US citizens would be deeply horrified :-) ).
TLDR: What is perfectly acceptable behavior can be deeply different in cultures.
Both cultural (and not just by country), and certainly a personality type thing. In one company I worked at, I once had a conversation with someone while we were waiting for someone else, and this person out of the blue told me he used to think I was a massive jerk. I was flabbergasted - I had no idea why he'd think so. I liked him, and though I generally treated him well.
Turns out that he had the impression that whenever we were in a meeting, and he presented an idea, I'd pounce on it and tear it apart. It took him a long time to recognise that I was generally presenting valid criticism, and that it was not a personal attack.
But my reason for being surprised was that my impression of it was that he was someone I respected for often coming up with good ideas.
Good ideas are worth taking the time to figure out how to improve. Bad ideas you shut down in the least amount of time possible - they're not worth spending time on.
To me, the fact that I spent time going through his ideas and finding lots of smaller flaws in them was evidence they were good ideas. All ideas have holes at first, but good ideas can be turned into great plans.
To me, investing time in his ideas was praise. But he at first saw only the criticism.
It taught me to take care to praise explicitly as well, because I could not expect people to understand my motive for spending time criticising their ideas. But I might never have realised if he wasn't the type to be willing to be abrasive enough to start that conversation, as this was how I was used to people communicating in my circles.
Agreed, I have to remember myself that the hugging and kissing and close personal contact that is normal in Spain is considered offensive, sexual approach in north Europe or USA normally after some awkward situation.
This culture disagreements combines with only text communications in which people can't see the body language and tone of voice of others.
You add to it that people don't care about what they say on Internet like they care on reality. Specially after "text only" communication would require orders of magnitude more effort to express yourself without misunderstandings and most people only have limited time for it.
> You add to it that people don't care about what they say on Internet like they care on reality.
My impression about this is different.
My personal hypothesis (which might also be wrong) rather is that the internet is a mouthpiece for people who have strong opinions about some topic (which does not imply that they are wrong here) that few people in the "real world" care about. Since few people in the real world will be interested in this topic, these people often appear as nice people in the "real world", since hardly any discussions about the "heated" topic will accidentally happen there. The same happens if these people discuss about different topics in the internet. The main difference is in the internet there are enough people that care about this exotic topic deeply enough to give contra (whether they have good points or not - at least they have a similar level of opionatedness). This leads to the impression that these people are toxic, etc.
TLDR: I believe the difference is not about how you say things in internet vs. real life, but about how you talk about your "opinionated topic" vs. something different both in internet and real life.
There are ways to criticise that don't mean being an asshole about it. The problem is that people have a tendency to act like an asshole when giving criticism, which is abusive to the person/people on the other end of the criticism.
I blame (in part) the linux popularization approach that turned to drama to create a more compelling narrative. The now defunked linux action show for example. Having an opinion is cheap. People who have never contributed a line of code or donated a cent to a project think that by being super opinionated about systemd vs upstart or Mir vs wayland, that they can virtue signal how "l33t" they are. Then they justify their strong opinions with arguments about "fracturing and redundancy" but never anything technical because they actually know nothing. Telling those people to GTFO is both cathartic and better for the FOSS community. Right on Mark!
Among all the desktop distros, Ubuntu/Unity came closest to making me switch away from macOS. While I'm a programmer, my interests in the desktop are that of a simple user: I want stuff to work without me having to spend time actively keeping it alive, and I want some amount of polish and UX.
I also firmly believe in order to accomplish something, you sometimes have to be a pioneer and charge ahead in a direction the majority doesn't deem worthwhile, even if that means making mistakes and incurring aggressive dissent. Not everything can advance by committee or mass consensus.
Ubuntu de facto cancelling their desktop program to focus on the cloud sounds to me like the premature death of something that worked pretty well, at least for my use cases. I really hope that decision was not entirely based on petty open source community squabbling. I love Brian Lunduke (he has a pretty good open source podcast by the way), but I really find it hard to get behind his long-standing criticisms of Mir and Unity.
Among the arguments against Mir and Unity, I find the one based on duplication of effort the hardest to swallow. Innovation is an exploratory process. Sometimes that means many people coming up with many solutions. In other areas
of open source development this doesn't seem to be perceived as a problem. In fact, it's celebrated as a marker of freedom for giving the user a lot of choices.
But somehow, Ubuntu lost all of its public support, for no reason I can (as an outsider) readily identify. If this is symptomatic of the future of open source, it's a bad precedent.
> But somehow, Ubuntu lost all of its public support […]
That seems hyperbolic. Ubuntu is popular and pretty mainstream for a Linux distribution. That will always attract hate and virulent nay-saying, as well as valid and constructive criticism.
I disagree. While it's probably impossible to come to an objective consensus about any matter of public discussion, at least the vocal people all fall over themselves celebrating the death of Mir and Unity. It also seems to me the hatred toward these software packages prompted Ubuntu to abandon them in the first place, and mind you these were integral pieces that set the Ubuntu desktop apart from every other distro.
So, no, I don't think it's hyperbolic at all to come to this conclusion since we're talking about losing public support to the point of Canonical outright canceling two of their most distinguished programs.
at least the vocal people all fall over themselves celebrating the death of Mir and Unity
I'm close to celebrating the death of Unity but won't because:
1. the damage is done. Ubuntu and Linux already lost a whole lot of momentum. We had a really nice desktop. We were selling it everywhere, looking forward to be able to justify paid support agreements.
2. Some people actually seems to like Unity and while I really cannot see myself using it I think it would be offensive toward those if I celebrated loudly
> I find the one based on duplication of effort the hardest to swallow
Having a display stack specific to one distribution makes things quite a lot more difficult than just another desktop environment. That was heavily criticised.
Mir supporters often mention that the toolkit will easily handle this; but that's not the case. The Wayland version of Firefox still has quite a few issues. It took a huge amount of effort to have X11 applications work nicely under Wayland using Xwayland (copy/pasting, dragging, etc). There's still various things left.
Various things in Wayland are still a bit iffy. E.g. some Wayland solutions are specific to one desktop. Fixing all of this and getting an agreement across desktop environments is a huge amount of effort.
Then this has to be doubled because hey 'Mir'. Obviously that's heavily criticized. Though various developers just said "No" (e.g. Kwin developer).
Yes, I'm usually not one to whine about "fragmented" project spaces and "wasted talent", but Mir is a special case because the Linux desktop is necessarily dependent on cooperation from hardware vendors. Historically, they've seen Linux as a source of confusion, and avoided wading too deep because they couldn't really make anyone happy, and because it was hard to tell if their investment would pay off or if they'd have to redo all the work when the "community" decided to shift again.
With Wayland, we had the opportunity to present a unified, modern desktop environment to them and say "Here, support this and we'll be happy". Then Mir showed up with absolutely no substantiation or value proposition and retriggered the prospect of nightmarish half-support across a fragmented ecosystem, which tires everyone out, not just vendors.
It was a great illustration that open-source and Linux haven't changed: they're still difficult to work with, and you still don't know if the engineering effort you're dedicating is going to be flushed away due to distro infighting. NVIDIA had already said they were playing a wait and see game before committing more resources to the next-gen Linux desktop, and Canonical only proved their decision right.
>Sometimes that means many people coming up with many solutions. In other areas of open source development this doesn't seem to be perceived as a problem. In fact, it's celebrated as a marker of freedom for giving the user a lot of choices.
That's because in other areas more choice is always a good thing as it doesn't equal fragmentation. Yet another choice for parsing JSON? Yet another email program? Great, why not? But operating systems are platforms, not just libraries or tools that you can combine ad libitum.
For an OS to be an attractive target for developers it needs users. If you split the user base into several smaller user bases, the OS becomes that much less attractive for developers and as result it becomes less attractive for users as well.
A big player like Canonical should understand that and consider consensus building as much part of its role as creating new software.
Here is all of the stuff, filtered through Wikipedia's editorial process that is intended to eliminate unfounded personal opinion, stuff from unidentifiable sources, nonsense, and bias. Look, it isn't a hate parade; therefore there is no hate parade as M. Shuttleworth claimed. is a rather flawed argument, note.
> Here is all of the stuff, filtered through Wikipedia's editorial process that is intended to eliminate unfounded personal opinion, stuff from unidentifiable sources, nonsense, and bias. Look, it isn't a hate parade; therefore there is no hate parade as M. Shuttleworth claimed.
Actually, that Wikipedia link is not meant as evidence against Shutterworth's comments or in defense of my own. It's provided merely as context to those who are just now hearing about this situation.
No, my comment regarding the lack of a hate parade is supported only by common sense. A position easily taken once one digs a bit deeper into this subject. You should click through some of the actual references mentioned in (and outside of) that Wikipedia article:
Then, tell me if these opinions or the community comments they generated actually seem hateful.
In fact, the only substantial hate generated by this debacle was really in response to Canonical's unfounded swipe at Wayland, which was later retracted, and Shutterworth branding of opponents as "The Open Source Tea Party". And even that level of grandstanding only generated a modicum of hate from the community in comparison to things like systemd or SELinux.
The problem with Mir, Unity and all the convergence ideas was that they were poorly communicated.
They didn't really engage with the community and kept all the discussions and development behind closed doors, releasing dumps of code once in a while once they reached a milestone that they were satisfied with.
That's what Google does with Android, for example, and that would be okay... but only if they were able to take on such ambitious plans.
You can't expect the community to welcome you when you blindly follow your ideas and you can't even produce anything in quality. And when I mean in quality, I mean something that justifies going your own way, not just another "GNOME clone" without any added features.
I'm pretty sure they had amazing ideias in mind and that they had reasons on why they avoided going to Wayland but... is anyone able to point out any public mailing list, or public blog discussions, where they discuss all of this? I really tried to follow all Ubuntu/Mir/Unity/Phone/Convergence project but all the information I found was poor and outdated. Even working Ubuntu Phone images for the Nexus 5 was hard to find.
> Another problem with Unity was that it was totally not like the thing that made many of us like Ubuntu in the first place.
Well not so much a "problem" per se, but this was always something that amazed me: Unity was the last thing I was/am looking for.
I get that choices for the DE are great. Unity isn't my cup of tea, but having the option is nice. However, my main quibbles with Ubuntu lied - and still lie - totally elsewhere.
For example, I have never done a dist-upgrade with Ubuntu that didn't break everything and force me to do a new installation. This confuses me a bit, since Ubuntu is a great go-to distro for newcomers to Linux. And from that perspective I get that some people might be upset by the "NIH" of some parts of Ubuntu while others still could use more improvements overall.
That said, I like what Canonical does very much. Ubuntu is - amongst other things - a great live-system. I always make sure to have an Ubuntu Live flash drive around in case I have to save a friend's files because their Windows system kicked the bucket. And I suspect Canonical wouldn't have ended up where they are now if they didn't have the guts to go their own way on some things.
> You can't expect the community to welcome you when you blindly follow your ideas and you can't even produce anything in quality.
So you'd rather destroy a promising program just because you weren't asked about your opinion? Doesn't this effectively mean that we stopped caring about open source, the only thing that actually matters now is the perception of broad community support? It doesn't seem to be about users anymore either, it's all about the approval of gatekeepers now.
It used to be anyone, a guy in a garage, or even a huge company, could make something, share the source code, and we'd be happy for their contribution. Now that's apparently shifted to "don't start anything new, just fall in line with the existing stuff". What company or individual would want to publish anything new in this kind of environment?
The same way you can build anything you want on your own, I also can say it's crap and I don't want it. And that's fine, you don't need my permission and I don't need to be your public target.
But it's a bit odd that once you realize that you can't build your own vision without asking for my opinion, you start throwing a tantrum saying I'm a negativist and that I don't embrace the community spirit.
No one forced Mark to shut down the project, he made the decision on his own.
Also, community and open source is all about contributions and public discussions, not _just_ code dumps.
> Also, community and open source is all about contributions and public discussions, not _just_ code dumps.
Personally, I think software should be the central point, not populism. Ideally, an open source project would surround itself with a support community of people who actually use the software instead of wasting so much time fighting.
If I build something that 10 people like, I might not care that 3 billion people choose to ignore it. But I would probably care a great deal if 5 people spend a lot of effort and air time campaigning against it.
> No one forced Mark to shut down the project, he made the decision on his own.
I'm disappointed in that as well, in case that wasn't clear. But as a vocal critic you also don't get to wash your hands completely if something you hate goes down the drain.
> The same way you can build anything you want on your own, I also can say it's crap and I don't want it.
Sure, I just don't like how much enthusiastic assent you can get out of the implication that the main reason for your dislike is simply that I built something on my own.
There's a difference between "destroying" (active) and "not supporting" (passive). If you want to garner support, you need to be open, communicative, and, yes, elicit opinions.
You know, being "validated" opens access to a lot of opportunities and most underdogs want just that. Open source is a fancy way to refer to unpaid work nowadays, can't see any revolution here, all being marketable if optimised for production.
People get paid to develop Mir. They're a limited resource in open source - paid developers are the ones who can drive projects, fix tricky bugs and take on big features. They also wind up being gatekeepers to contributions.
At the time Mir was announced what I heard and what a lot of others heard was "Ubuntu's paid developer resources with relevant expertise are being diverted from Wayland".
Replacing X is sorely needed in the Linux desktop space, and it's a huge project. So big that unless it's frightfully mismanaged then it's extremely dubious what a split in development effort is going to accomplish. And what did it accomplish? I have absolutely no idea what Mir actually shipped that was usable. I have no idea what they were doing substantially differently and why that was good (something about convergence by using bionic so it ran on Android too I think?) And I do remember Canonical at the time arguing they would beat Wayland to something usable... But here we are, I'm still using X, my video payback still tears, and I can't for the life of me think how the Linux desktop has benefited.
I suppose Cinnamon got created and I now run that on top of others vanilla Ubuntu?
Wayland is actually fairly usable at this point, though some programs are buggy. With aliases to run problematic programs in X, it'd probably work quite well.
> So you'd rather destroy a promising program just because you weren't asked about your opinion?
Canonical made blunt claims about deficiencies in Wayland which did not exist, they even made claims that others like KDE would happily adopt Mir but those did not know about their luck and did not plan to do this. Oh and add some CLA requirement for contributions to the mix. This understandably provoked a very harsh response from the affected projects.
> What company or individual would want to publish anything new in this kind of environment?
Those who don't insult other people and their work I would say. Then everybody will either be happy about your contribution or at least don't care if your contribution is practically unusable because it's incompatible with everything.
> Doesn't this effectively mean that we stopped caring about open source, the only thing that actually matters now is the perception of broad community support?
Oh it's been that for... I dunno, 15 years now? Maybe more to be honest. It's just got worse recently.
Not that the other side doesn't exist too of course, but acceptance is hard - even for better ideas - if the community doesn't buy in.
Ubuntu was my first distro, and it was a great way to learn. I'm still enough of a n00b that I mostly use Debian stable. But when I need the latest versions of packages, I generally go for Ubuntu over Debian testing. However, I use Ubuntu server, and add my own desktop environment. Unity is just too much of a resource hog. I really can't afford to give more than 1GB RAM to a VM, because I use so many VMs.
I assume Canonical wants to have some influence with Gnome and Wayland going forward. The nature of open source, and just working with groups of people for that matter, is that you have to build consensus before trying to push changes, especially big and visible changes.
Why would a volunteer coordinator (or paid contributor at a different company) even look at a contribution from another team that just called him anti-social if the code isn't perfectly in line with his goals.
Even if Mark is 100% right, his 'anti-social' shots are not a good way to build in-roads with the projects Canonical will be relying on.
The Unity UI is more minimal and looks better, has great keyboard shortcuts to the point that you don't need to use the mouse at all, a great interface to search programs and files. Unity took the best approaches from other DEs and put them together.
Gnome, on the other hand, is a collection of things that don't make sense. From the people who
* initially removed the shutdown function, because it was too destructive [1]
* almost bankrupted the project by funding SJW bullshit [2]
You can't possibly expect these people to come up with a good UX.
> * almost bankrupted the project by funding SJW bullshit [2]
The funding didn't cause this. It's the foundation handling money for organizations and them not paying on time. This caused a cashflow problem. Cashflow isn't the same as being profitable, though it is a problem.
Your source is much better. The fact that they spent almost half their funding on social justice programs is fucked up already and shows that their focus is not on developing good software.
It wasn't "their funding". The GNOME Foundation handled payroll for an internship program sponsored by the participating organizations. Net expenses on OPW for 2013 were about $25,000[1], which included paying several people to hack on GNOME.
And when Canonical went mainstream, it became the focus of irrational hatred too.
Never hated them. I do however think they made themselves a huge disfavour by abandoning a proven recipe that people loved. I mean even I used Ubuntu because of its almost perfect Gnome, and I was traditionally a KDE person.
Summarized: they built a userbase of people who loved their flavour of Gnome, then suddenly swapped that for something completely unproven that might be better but was very different - and expected us to continue to be fans just because they were Canonical.
Suddenly it turned out we weren't so much Canonical fans as fans of a polished free desktop that worked well.
The hatred for Ubuntu, Mir, Unity is increasingly irrational and difficult to understand.
Unity seems to be one of the more polished Linux desktops and there are many options, even if one doesn't like Unity or Mir there is zero compulsion to use them. Surely people investing their own time, effort and money to build free software and attract users is good thing?
But what then explains the hoarde of angry 'self appointed community spokespeople' singling out Ubuntu in every online discussion for NIH and duplication of effort.
By the same logic these community spokespeople should be accusing Flatpak of duplication of effort and NIH and haranguing Flatpak developers to join hands with the Snap project which precedes Flatpak. But all quiet on that end. This is a bit too hypocritical and seems to be more about some Redhat supporters seeking control than anything sincere.
How many Redhat, Gnome or Freedesktop projects are being targeted for NIH and duplication of effort the way Ubuntu's projects are.
Ubuntu has sizable market share in the Linux community. You can look no further than the Linux games available on Steam/GOG to see that many of them are offered specifically for Ubuntu, not Linux.
The fact that Canonical was switching to Mir was of great concern, particularly to gamers, who were worried that AMD or NVIDIA would start pushing Mir-only graphical/driver patches at the expense of Xorg/Wayland patches, resulting in gamers being "locked in" to Ubuntu.
I guess Shuttleworth deliberately wants to piss-off the Free Open Source project developers and users. Seems like a rather twisted, bipolar Trump like goal which if succeeds would server to Ghost Ubuntu. Perhaps mark Shuttleworth is having one of those Donald J. Trump 'bad hair' days. That doesn't however excuse such crass demonetization of a major Ubuntu userbase.
If there is one thing for which Canonical deserves some criticism is its insularity.
Many things Ubuntu end up being Ubuntu-only, sometimes for historical reasons, sometimes because a solution solves the problems Canonical has better and sometimes because things are done and few people know about them. Sometimes it's because Canonical is seen as an aggressive competitor. Launchpad is wonderful, but Github won. Bzr is not bad, but Git won. Unity is great for small screens, but Gnome 3 is an incredibly polished desktop environment and is clearly the future. Ubuntu phone is interesting, but Android clearly is unstoppable (and that has been clear for some time now).
It's really hard to compete with a whole industry by yourself.
Their software is good. They do awesome things but far too often they do it by themselves, without any help.
84 comments
[ 2.8 ms ] story [ 137 ms ] threadHere is a link: https://arstechnica.co.uk/information-technology/2017/04/ubu...
"Use the standard" doesn't ring true when there is none - and having multiple projects do the same thing is not a bad thing - they can learn from each other and try different approaches.
I think given Canonical's track record, this is a really unfair accusation. Yes, they did not always invest all the effort to make things like Unity work on other distros as well, but the reason for that was always prioritisation of their time (no one else took up and kept up the porting task either), never of intending lock-in.
Canonical (and Mark Shuttleworth) have given a lot to the free software community, and these are exactly the disingenuous accusations that have clearly and understandably affected Mark Shuttleworth personally.
Wayland it not even finished YET so really who can say it?
Wanting standards defined after some competition exist and we know the best approach could be positive, but not before based on dogma on what should be better on paper but on reality.
A reason to be suspicious? Personally I am suspicious of standards by committee in which anybody at a table have to be happy with unrealistic expectations.
This is "software as politics". Most people don't even know what the best approach is until a least they code sever tests on the real world and work some time on the problem.
Software has also some part of exploration on it. To have small focused teams exploring those ideas makes total sense.
For example the original software for the Iphone was done with a small team of 10-20 people. And it took them some iterations(years of work) to really know what people wanted or needed.
To same something is the standard because we say so is highly presumptuous.
When it comes to Open Source Software, hate is not the option, you use it or take the Source Code to change according to your needs, but if you are a common user just change the distro or change your opinion.
This is why you should Love Linux or any Open Source/Free Software : https://goo.gl/eP6sy
After 6-9 they revealed what they worked upon and made loads of incorrect statements about Wayland. Obviously this resulted in pretty harsh response.
Wayland is quite usable in Fedora as well as various other distributions, so your "never go beyond talk": eh? This while Canonical delayed Wayland by not announcing they wouldn't help out with Wayland anymore.
Anyone who knows open-source hackers should know that their bullshit detectors are highly attuned to typical political malfeasance, and the transparent overlays of conventional corporate politics are not going to work on them (other approaches would work supremely well; hackers are not invincible).
I'm actually worried about Canonical after their repeated political failures. Despite their best efforts, Red Hat continues to stomp all over them, with systemd winning over upstart and now Wayland over Mir.
They need to re-evaluate their strategy for land grabs. Do they not have the technical skill on staff to conceptualize technologies that get people on board? Is Red Hat intentionally undermining them (hard sell since most of these projects existed before Ubuntu decided to try to copy and win), and if so, how can they compete?
EDIT: Noticed how I'm wasting karma point to be proven right by the ones that this article is all about: people that 'love to hate'.
He needs to chill out and listen more to the people who do the real work.
For example I (German) have to deal with east European mathematicians in my job (they are actually really nice people). The kind of strictness that they treat people who present mathematical bullshit (they are typically right in their mathematical reasoning) is sometimes on the borderline of what is considered as impolite in Germany (and in the US probably even more).
I also went climbing with former Russians. I just say: What in their culture is considered as normal language is in Germany already considered as sexist (and US citizens would be deeply horrified :-) ).
TLDR: What is perfectly acceptable behavior can be deeply different in cultures.
Turns out that he had the impression that whenever we were in a meeting, and he presented an idea, I'd pounce on it and tear it apart. It took him a long time to recognise that I was generally presenting valid criticism, and that it was not a personal attack.
But my reason for being surprised was that my impression of it was that he was someone I respected for often coming up with good ideas.
Good ideas are worth taking the time to figure out how to improve. Bad ideas you shut down in the least amount of time possible - they're not worth spending time on.
To me, the fact that I spent time going through his ideas and finding lots of smaller flaws in them was evidence they were good ideas. All ideas have holes at first, but good ideas can be turned into great plans.
To me, investing time in his ideas was praise. But he at first saw only the criticism.
It taught me to take care to praise explicitly as well, because I could not expect people to understand my motive for spending time criticising their ideas. But I might never have realised if he wasn't the type to be willing to be abrasive enough to start that conversation, as this was how I was used to people communicating in my circles.
This culture disagreements combines with only text communications in which people can't see the body language and tone of voice of others.
You add to it that people don't care about what they say on Internet like they care on reality. Specially after "text only" communication would require orders of magnitude more effort to express yourself without misunderstandings and most people only have limited time for it.
As a result you get an ugly thing.
My impression about this is different.
My personal hypothesis (which might also be wrong) rather is that the internet is a mouthpiece for people who have strong opinions about some topic (which does not imply that they are wrong here) that few people in the "real world" care about. Since few people in the real world will be interested in this topic, these people often appear as nice people in the "real world", since hardly any discussions about the "heated" topic will accidentally happen there. The same happens if these people discuss about different topics in the internet. The main difference is in the internet there are enough people that care about this exotic topic deeply enough to give contra (whether they have good points or not - at least they have a similar level of opionatedness). This leads to the impression that these people are toxic, etc.
TLDR: I believe the difference is not about how you say things in internet vs. real life, but about how you talk about your "opinionated topic" vs. something different both in internet and real life.
I also firmly believe in order to accomplish something, you sometimes have to be a pioneer and charge ahead in a direction the majority doesn't deem worthwhile, even if that means making mistakes and incurring aggressive dissent. Not everything can advance by committee or mass consensus.
Ubuntu de facto cancelling their desktop program to focus on the cloud sounds to me like the premature death of something that worked pretty well, at least for my use cases. I really hope that decision was not entirely based on petty open source community squabbling. I love Brian Lunduke (he has a pretty good open source podcast by the way), but I really find it hard to get behind his long-standing criticisms of Mir and Unity.
Among the arguments against Mir and Unity, I find the one based on duplication of effort the hardest to swallow. Innovation is an exploratory process. Sometimes that means many people coming up with many solutions. In other areas of open source development this doesn't seem to be perceived as a problem. In fact, it's celebrated as a marker of freedom for giving the user a lot of choices.
But somehow, Ubuntu lost all of its public support, for no reason I can (as an outsider) readily identify. If this is symptomatic of the future of open source, it's a bad precedent.
That seems hyperbolic. Ubuntu is popular and pretty mainstream for a Linux distribution. That will always attract hate and virulent nay-saying, as well as valid and constructive criticism.
I disagree. While it's probably impossible to come to an objective consensus about any matter of public discussion, at least the vocal people all fall over themselves celebrating the death of Mir and Unity. It also seems to me the hatred toward these software packages prompted Ubuntu to abandon them in the first place, and mind you these were integral pieces that set the Ubuntu desktop apart from every other distro.
So, no, I don't think it's hyperbolic at all to come to this conclusion since we're talking about losing public support to the point of Canonical outright canceling two of their most distinguished programs.
I'm close to celebrating the death of Unity but won't because:
1. the damage is done. Ubuntu and Linux already lost a whole lot of momentum. We had a really nice desktop. We were selling it everywhere, looking forward to be able to justify paid support agreements.
2. Some people actually seems to like Unity and while I really cannot see myself using it I think it would be offensive toward those if I celebrated loudly
Having a display stack specific to one distribution makes things quite a lot more difficult than just another desktop environment. That was heavily criticised.
Mir supporters often mention that the toolkit will easily handle this; but that's not the case. The Wayland version of Firefox still has quite a few issues. It took a huge amount of effort to have X11 applications work nicely under Wayland using Xwayland (copy/pasting, dragging, etc). There's still various things left.
Various things in Wayland are still a bit iffy. E.g. some Wayland solutions are specific to one desktop. Fixing all of this and getting an agreement across desktop environments is a huge amount of effort.
Then this has to be doubled because hey 'Mir'. Obviously that's heavily criticized. Though various developers just said "No" (e.g. Kwin developer).
With Wayland, we had the opportunity to present a unified, modern desktop environment to them and say "Here, support this and we'll be happy". Then Mir showed up with absolutely no substantiation or value proposition and retriggered the prospect of nightmarish half-support across a fragmented ecosystem, which tires everyone out, not just vendors.
It was a great illustration that open-source and Linux haven't changed: they're still difficult to work with, and you still don't know if the engineering effort you're dedicating is going to be flushed away due to distro infighting. NVIDIA had already said they were playing a wait and see game before committing more resources to the next-gen Linux desktop, and Canonical only proved their decision right.
That's because in other areas more choice is always a good thing as it doesn't equal fragmentation. Yet another choice for parsing JSON? Yet another email program? Great, why not? But operating systems are platforms, not just libraries or tools that you can combine ad libitum.
For an OS to be an attractive target for developers it needs users. If you split the user base into several smaller user bases, the OS becomes that much less attractive for developers and as result it becomes less attractive for users as well.
A big player like Canonical should understand that and consider consensus building as much part of its role as creating new software.
https://en.wikipedia.org/wiki/Mir_(software)#Controversy
This isn't a hate parade, it's the community speaking out against something they didn't want. Mostly related to licensing.
Actually, that Wikipedia link is not meant as evidence against Shutterworth's comments or in defense of my own. It's provided merely as context to those who are just now hearing about this situation.
No, my comment regarding the lack of a hate parade is supported only by common sense. A position easily taken once one digs a bit deeper into this subject. You should click through some of the actual references mentioned in (and outside of) that Wikipedia article:
Matthew Garrett's comments, http://mjg59.dreamwidth.org/25376.html
Bradley Kuhn's comments, http://ebb.org/bkuhn/blog/2011/07/07/harmony-harmful.html
Jonathan Riddell's comments, https://blogs.kde.org/2013/06/26/kubuntu-wont-be-switching-m...
Aaron Seigo's comments, https://plus.google.com/+AaronSeigo/posts/76Nd9RSTZWp
Then, tell me if these opinions or the community comments they generated actually seem hateful.
In fact, the only substantial hate generated by this debacle was really in response to Canonical's unfounded swipe at Wayland, which was later retracted, and Shutterworth branding of opponents as "The Open Source Tea Party". And even that level of grandstanding only generated a modicum of hate from the community in comparison to things like systemd or SELinux.
They didn't really engage with the community and kept all the discussions and development behind closed doors, releasing dumps of code once in a while once they reached a milestone that they were satisfied with.
That's what Google does with Android, for example, and that would be okay... but only if they were able to take on such ambitious plans.
You can't expect the community to welcome you when you blindly follow your ideas and you can't even produce anything in quality. And when I mean in quality, I mean something that justifies going your own way, not just another "GNOME clone" without any added features.
I'm pretty sure they had amazing ideias in mind and that they had reasons on why they avoided going to Wayland but... is anyone able to point out any public mailing list, or public blog discussions, where they discuss all of this? I really tried to follow all Ubuntu/Mir/Unity/Phone/Convergence project but all the information I found was poor and outdated. Even working Ubuntu Phone images for the Nexus 5 was hard to find.
Another problem with Unity was that it was totally not like the thing that made many of us like Ubuntu in the first place.
I'm not saying it was bad but I didn't like it. I guess this holds true for a number of earlier Ubuntu users.
Well not so much a "problem" per se, but this was always something that amazed me: Unity was the last thing I was/am looking for.
I get that choices for the DE are great. Unity isn't my cup of tea, but having the option is nice. However, my main quibbles with Ubuntu lied - and still lie - totally elsewhere.
For example, I have never done a dist-upgrade with Ubuntu that didn't break everything and force me to do a new installation. This confuses me a bit, since Ubuntu is a great go-to distro for newcomers to Linux. And from that perspective I get that some people might be upset by the "NIH" of some parts of Ubuntu while others still could use more improvements overall.
That said, I like what Canonical does very much. Ubuntu is - amongst other things - a great live-system. I always make sure to have an Ubuntu Live flash drive around in case I have to save a friend's files because their Windows system kicked the bucket. And I suspect Canonical wouldn't have ended up where they are now if they didn't have the guts to go their own way on some things.
So you'd rather destroy a promising program just because you weren't asked about your opinion? Doesn't this effectively mean that we stopped caring about open source, the only thing that actually matters now is the perception of broad community support? It doesn't seem to be about users anymore either, it's all about the approval of gatekeepers now.
It used to be anyone, a guy in a garage, or even a huge company, could make something, share the source code, and we'd be happy for their contribution. Now that's apparently shifted to "don't start anything new, just fall in line with the existing stuff". What company or individual would want to publish anything new in this kind of environment?
There is a difference between ignoring a piece of software you don't like and actively campaigning against it.
No project exist in a vacuum. I like it that way.
But it's a bit odd that once you realize that you can't build your own vision without asking for my opinion, you start throwing a tantrum saying I'm a negativist and that I don't embrace the community spirit.
No one forced Mark to shut down the project, he made the decision on his own.
Also, community and open source is all about contributions and public discussions, not _just_ code dumps.
Personally, I think software should be the central point, not populism. Ideally, an open source project would surround itself with a support community of people who actually use the software instead of wasting so much time fighting.
If I build something that 10 people like, I might not care that 3 billion people choose to ignore it. But I would probably care a great deal if 5 people spend a lot of effort and air time campaigning against it.
> No one forced Mark to shut down the project, he made the decision on his own.
I'm disappointed in that as well, in case that wasn't clear. But as a vocal critic you also don't get to wash your hands completely if something you hate goes down the drain.
> The same way you can build anything you want on your own, I also can say it's crap and I don't want it.
Sure, I just don't like how much enthusiastic assent you can get out of the implication that the main reason for your dislike is simply that I built something on my own.
As long as you do freedom the right way.
At the time Mir was announced what I heard and what a lot of others heard was "Ubuntu's paid developer resources with relevant expertise are being diverted from Wayland".
Replacing X is sorely needed in the Linux desktop space, and it's a huge project. So big that unless it's frightfully mismanaged then it's extremely dubious what a split in development effort is going to accomplish. And what did it accomplish? I have absolutely no idea what Mir actually shipped that was usable. I have no idea what they were doing substantially differently and why that was good (something about convergence by using bionic so it ran on Android too I think?) And I do remember Canonical at the time arguing they would beat Wayland to something usable... But here we are, I'm still using X, my video payback still tears, and I can't for the life of me think how the Linux desktop has benefited.
I suppose Cinnamon got created and I now run that on top of others vanilla Ubuntu?
Canonical made blunt claims about deficiencies in Wayland which did not exist, they even made claims that others like KDE would happily adopt Mir but those did not know about their luck and did not plan to do this. Oh and add some CLA requirement for contributions to the mix. This understandably provoked a very harsh response from the affected projects.
> What company or individual would want to publish anything new in this kind of environment?
Those who don't insult other people and their work I would say. Then everybody will either be happy about your contribution or at least don't care if your contribution is practically unusable because it's incompatible with everything.
Oh it's been that for... I dunno, 15 years now? Maybe more to be honest. It's just got worse recently.
Not that the other side doesn't exist too of course, but acceptance is hard - even for better ideas - if the community doesn't buy in.
Why would a volunteer coordinator (or paid contributor at a different company) even look at a contribution from another team that just called him anti-social if the code isn't perfectly in line with his goals.
Even if Mark is 100% right, his 'anti-social' shots are not a good way to build in-roads with the projects Canonical will be relying on.
If only Gnome was half as usable as Unity.
The Unity UI is more minimal and looks better, has great keyboard shortcuts to the point that you don't need to use the mouse at all, a great interface to search programs and files. Unity took the best approaches from other DEs and put them together.
Gnome, on the other hand, is a collection of things that don't make sense. From the people who
* initially removed the shutdown function, because it was too destructive [1]
* almost bankrupted the project by funding SJW bullshit [2]
You can't possibly expect these people to come up with a good UX.
[1] https://bugzilla.gnome.org/show_bug.cgi?id=755953
[2] https://www.phoronix.com/scan.php?page=news_item&px=MTY2Mjc
The funding didn't cause this. It's the foundation handling money for organizations and them not paying on time. This caused a cashflow problem. Cashflow isn't the same as being profitable, though it is a problem.
Phoronix is NOT a reliable news source!
[1] https://wiki.gnome.org/Foundation/FinancialSummary
I always preferred Gnome 3 to Unity.
Never hated them. I do however think they made themselves a huge disfavour by abandoning a proven recipe that people loved. I mean even I used Ubuntu because of its almost perfect Gnome, and I was traditionally a KDE person.
Summarized: they built a userbase of people who loved their flavour of Gnome, then suddenly swapped that for something completely unproven that might be better but was very different - and expected us to continue to be fans just because they were Canonical.
Suddenly it turned out we weren't so much Canonical fans as fans of a polished free desktop that worked well.
Unity seems to be one of the more polished Linux desktops and there are many options, even if one doesn't like Unity or Mir there is zero compulsion to use them. Surely people investing their own time, effort and money to build free software and attract users is good thing?
But what then explains the hoarde of angry 'self appointed community spokespeople' singling out Ubuntu in every online discussion for NIH and duplication of effort.
By the same logic these community spokespeople should be accusing Flatpak of duplication of effort and NIH and haranguing Flatpak developers to join hands with the Snap project which precedes Flatpak. But all quiet on that end. This is a bit too hypocritical and seems to be more about some Redhat supporters seeking control than anything sincere.
How many Redhat, Gnome or Freedesktop projects are being targeted for NIH and duplication of effort the way Ubuntu's projects are.
The fact that Canonical was switching to Mir was of great concern, particularly to gamers, who were worried that AMD or NVIDIA would start pushing Mir-only graphical/driver patches at the expense of Xorg/Wayland patches, resulting in gamers being "locked in" to Ubuntu.
Many things Ubuntu end up being Ubuntu-only, sometimes for historical reasons, sometimes because a solution solves the problems Canonical has better and sometimes because things are done and few people know about them. Sometimes it's because Canonical is seen as an aggressive competitor. Launchpad is wonderful, but Github won. Bzr is not bad, but Git won. Unity is great for small screens, but Gnome 3 is an incredibly polished desktop environment and is clearly the future. Ubuntu phone is interesting, but Android clearly is unstoppable (and that has been clear for some time now).
It's really hard to compete with a whole industry by yourself.
Their software is good. They do awesome things but far too often they do it by themselves, without any help.