Quotes of the week

[Posted December 16, 2015 by corbet]

We would like more people to use Emacs, but we should never think that we _need_ more users. When developers of a free software package think they _need_ more users, it is a lever that can be used to push them into bad decisions.

— Richard Stallman

Prepare for Git on-disk repository representation to undergo backward incompatible changes by introducing a new repository format version "1", with an extension mechanism.

— Junio Hamano preparing for interesting Git changes

I particularly like that RMS quote

Posted Dec 17, 2015 14:46 UTC (Thu) by pizza (subscriber, #46) [Link] (33 responses)

Especially within its wider context. I've been (quite poorly) attempting to articulate this for some time.

I particularly like that RMS quote

Posted Dec 19, 2015 16:55 UTC (Sat) by jospoortvliet (guest, #33164) [Link] (32 responses)

I both agree and disagree with the quote. Sure you don't need others to use your software. But if your goal is to bring software freedom to users, a meaningful result requires a meaningful number of users... So, the Linux desktop, for example, is IMHO a failure. It has no relevance, with the few million users it has...

I particularly like that RMS quote

Posted Dec 19, 2015 18:57 UTC (Sat) by lsl (subscriber, #86508) [Link] (31 responses)

So bringing software freedom to a mere few million users is not a worthwile endeavor? You need to have at least hundreds of millions of users or you're a failure? Nope, I don't agree with that at all.

I particularly like that RMS quote

Posted Dec 20, 2015 15:58 UTC (Sun) by jospoortvliet (guest, #33164) [Link] (30 responses)

Oh, sure it is worthwhile. Just not relevant.

Just like it'd be if, due to a war or drought or something, millions would be dying of hunger. It is awesome to save a few dozen, but unless you have the ambition (and will) to do more, the impact of your good deed is tiny. Don't count on a Nobel peace prize, a spot on the front page of the Time, TED talks or anything like that, they tends to require a tad more.

I particularly like that RMS quote

Posted Dec 20, 2015 19:10 UTC (Sun) by liw (subscriber, #6379) [Link] (28 responses)

Saving one person is an important and valuable thing to do. It doesn't matter you get a Nobel peace prize, a spot on the front page of the Time, get to give a TED talk, or any other reward.

Belittling people's achievements when they have saved someone, or made the world a better place, is counter-productive and useless, however.

I particularly like that RMS quote

Posted Dec 21, 2015 1:55 UTC (Mon) by bronson (subscriber, #4806) [Link] (27 responses)

It's all about expectations. If your goal is to feed a million people and you only end up feeding a hundred then people will be disappointed. (yes, ten years ago Linux on the Desktop was a goal. Alas, it seems like everyone's given up on that.)

I don't understand your statement of belittling people's achievements... You're saying that jospoortvliet is doing that by noting that the goal wasn't reached?

I particularly like that RMS quote

Posted Dec 24, 2015 8:21 UTC (Thu) by micka (subscriber, #38720) [Link] (26 responses)

> ten years ago Linux on the Desktop was a goal

Well, it certainly is on my desktop. And it already was ten years ago.
And you should say for whom it was a goal. Because not everybody has the same goals.

I particularly like that RMS quote

Posted Dec 25, 2015 21:38 UTC (Fri) by bronson (subscriber, #4806) [Link]

Ubuntu, Gnome, and KDE aimed for it. It's easy enough to find the quotes.

Remember: "But if your goal is to bring software freedom to users, a meaningful result requires a meaningful number of users."

I particularly like that RMS quote

Posted Dec 30, 2015 8:39 UTC (Wed) by jospoortvliet (guest, #33164) [Link] (24 responses)

Hey, I am a happy Linux user, too. But, when I started to contribute to KDE over a decade ago I wanted it to become a replacement for Windows in the market. Not just for me (it already was), but everybody. I wanted my parents to, some day, come home with their new computer from Dell, running Linux.

And I am 100% sure this goal was shared widely. Today I feel this ambition is gone - the main reason I put both my paid and free time in ownCloud is that the community there certainly still has such ambition. See for example Frank's keynote from the conf this year: http://youtube.com/watch?v=AAIXIKQUvRw

I don't want to diss other people and their motivation, just saying what I think: the Linux desktop teams seem to have lost ambition and to have settled for doing it for themselves and a small circle of users, rather than aiming for world domination.

Tell me I am wrong.

I particularly like that RMS quote

Posted Dec 30, 2015 9:03 UTC (Wed) by anselm (subscriber, #2796) [Link]

I don't want to diss other people and their motivation, just saying what I think: the Linux desktop teams seem to have lost ambition and to have settled for doing it for themselves and a small circle of users, rather than aiming for world domination.

I personally think the main goal of the Linux desktop projects is to provide entertainment for Linux desktop project developers. This is not a bad goal in itself, because bored Linux desktop project developers tend to find more interesting things to do with their lives than Linux desktop project development, so it is very important to keep them entertained as much as possible or they will leave. However, many of the things you need to do for “world domination”, especially in the long term, are not entertaining at all, and commercial companies who are out for world domination attempt to offset this by paying people to do the non-entertaining things.

For a Linux desktop project, actually trying to achieve world domination seems to compromise the entertainment aspect to a point where the world-domination goal no longer remains viable. (Remember that for the GNOME folks, even considerable corporate funding wasn't sufficient to get anywhere remotely near that goal.)

I particularly like that RMS quote

Posted Dec 30, 2015 13:23 UTC (Wed) by seyman (subscriber, #1172) [Link] (21 responses)

> Not just for me (it already was), but everybody.

Why? When I think of Microsoft's majority market share, I can't help remembering the pain they've inflicted on users and the crimes they've committed to keep that market share. And I can't imagine why any FLOSS developer would want to do the same thing (which, I suspect, is RMS' point in the original quote).

I particularly like that RMS quote

Posted Dec 30, 2015 18:58 UTC (Wed) by bronson (subscriber, #4806) [Link] (20 responses)

What on earth? Strong-arming customers into the Linux Desktop was never anyone's goal. (unless you can find evidence to the contrary?)

I particularly like that RMS quote

Posted Dec 30, 2015 22:17 UTC (Wed) by seyman (subscriber, #1172) [Link] (19 responses)

> Strong-arming customers into the Linux Desktop was never anyone's goal.

Then how was Linux supposed to dominate the world?

I particularly like that RMS quote

Posted Dec 31, 2015 5:36 UTC (Thu) by bronson (subscriber, #4806) [Link] (18 responses)

By being better than the alternatives. Much better.

I particularly like that RMS quote

Posted Dec 31, 2015 9:00 UTC (Thu) by seyman (subscriber, #1172) [Link] (17 responses)

> By being better than the alternatives. Much better.

This sounds like wishful thinking (actually, it sounds like the "doing it for themselves and a small circle of users" that jospoortvliet is complaining about).

Replacing Windows in the desktop market requires either a playing field more level than it's been seen the mid 1980s or far more agressive tactics from the Linux desktop guys. Simply making a better desktop isn't enough, imho.

I particularly like that RMS quote

Posted Dec 31, 2015 18:43 UTC (Thu) by bronson (subscriber, #4806) [Link] (16 responses)

Sadly, I agree: it probably was impossible. But it sure seemed like it could happen at the time. The desire and volunteers are there, and there was visible progress every month. In a decade or two the result seemed all but inevitable.

But the nay sayers were right: when everyone scratches their own itch, you end up with a bunch of buggy, poorly thought out barely-working environments. For me, Gnome 3 and KDE 4 were the final nail in the Linux Desktop coffin. The dream is dead. Open source developers (desktop anyway) will only work on shiny fadware (hey, tablet!) and love to break things. Even when paid by Red Hat.

What kind of aggressive tactics are you thinking of? Putting out a better system in every way, and charging nothing for it, is pretty darned aggressive right there.

I particularly like that RMS quote

Posted Dec 31, 2015 19:51 UTC (Thu) by madscientist (subscriber, #16861) [Link] (1 responses)

"A better system in every way" was always completely impossible. No amount of work, even fully coordinated work where everyone agreed on a common outcome and sacrificed their own interests for the goal, would accomplish that. That would mean that every piece of hardware that works with Windows also worked as well or better with Linux. Remember what the hardware landscape looked like a decade ago? Additionally that would mean that every application that people used on Windows had a 100% equivalent or better implementation on Linux (and by "equivalent" I mean completely interoperable; i.e., 100% Word etc. compatibility): another impossibility.

"Charge nothing for it" is irrelevant because for almost everyone who buys a computer Windows is effectively free: it comes with the system and there's no obvious choice. And once you're running Windows you're locked in.

As seyman says, by the time Linux got into the traditional desktop game it was already over: network effects had ensured that "different" was equivalent to "marginal". The only way to win is to change the game completely: that's why the iPhone was successful, and why things like ChromeOS etc. on tablets have a chance.

The traditional computer desktop is a stagnant market and one not really worth worrying about. Within a short amount of time phones will have enough power to do everything a "normal" user needs to do. Once they solve the interface problem there will be no looking back. At that point the only desktop users will be, by definition, "power users" and perhaps then Linux will take over.

In the meantime I've been using UNIX since before Linus ever heard of it: I started with a REAL vt100(-ish) terminal. Right now I'm using Ubuntu GNOME 15.10 on a beefy desktop system with dual monitors, and it works just fine for me. IMO the oft-heard "barely working" claims are simply supportable.

I particularly like that RMS quote

Posted Dec 31, 2015 19:52 UTC (Thu) by madscientist (subscriber, #16861) [Link]

> are simply supportable

Er... *not* supportable
*sigh*

I particularly like that RMS quote

Posted Dec 31, 2015 20:16 UTC (Thu) by halla (subscriber, #14185) [Link] (9 responses)

Bronson, _you_, yes, *you* personally are part of the problem. You are part of the reason why fewer people work on Linux desktops than before. You have had a consistently and unreasonably negative attitude to the work people who are actually doing work, are doing. As far as anyone is aware, you haven't done anything yourself. What's your achievement? Until I know, to me, you're just a commenter with a perpetually negative attitude. You keep whining about how the really courageous and innovative work that went into KDE 4 and Gnome 3 was bad for your poor self. What the heck have you done in the past _decade_ that advanced Linux? Anything at all?

Well, right now, the thing that keeps Linux and free software back is these two things:

* lack of capital to fund development and, more importantly, market the work that is being done
* the kind of negative shit you, and anselm and a lot of other people on LWN and elsewhere keep spouting.

I guess that being a negative kind of guy, being able to dismiss the hard work of other people off-hand makes you feel good about yourself, makes you feel you're the smart person in the room, and I guess that that's good for you, but it's really, really bad for free software. Everyone lives their life in an echo chamber, and since LWN is part of the echo chamber for free software, the sound you make sounds disproportionately loudly.

It's not the Gnome developers or the KDE developers who've "lost" us the desktop wars. It's you and people like you. Just like people like you made going with free software a losing proposition for Nokia and Intel. The idiotic infighting over RPM vs DEB in the Maemo/MeeGo days alone was a killer.

KDE 2 was a better desktop than KDE 1. KDE 1 was pretty good! KDE 3 was better than KDE 2, and KDE 4 ended up better than KDE 3. And Plasma 5 is a _really_ good desktop, produced by a small band of really motivated, ambitious and dedicated people. Step down from your hobby horse, admit you've been a small-minded person and admit that the Linux desktops are tremendous achievements! And then, go on and contribute in a meaningful way!

And if you can't do that, it's time for you to take the needle off the scratched phonograph record. Your sound has been heard, and it's found wanting.

I particularly like that RMS quote

Posted Dec 31, 2015 21:39 UTC (Thu) by bronson (subscriber, #4806) [Link] (6 responses)

You're shooting the messenger. I am nothing more than a symptom of a problem, and probably not even that.

Let's be clear: I have nothing but excellent things to say about the Linux ecosystem as a whole, and I believe fundamentally in open source. I love working with it.

I was mostly quiet during the Gnome 2 days... It wasn't great but it wasn't bad. Regular people could learn it and use it and rely on it. I participated on some mailing lists and helped out in a few places where I could. The future wasn't here, but it seemed like it was on the horizon. I had a number of people in my office using and enjoying Linux.

Then Gnome 3 and KDE 4 arrived. They wasted huge amounts of my time and forced me to switch my team to Macs just to get back to work. Talk about an insult. It still makes me unreasonably angry. Any time I go to a conference and see a sea of 95% glowing Apple symbols, I think back to eight years ago when it was 80% Apple, 15% Linux, and 5% MS. I guess that was the peak.

So, you're wrong about the first thing that's holding the Linux desktop back. No amount of capital can make up for poor and mercurial leadership.

These days, other than Cinnamon (which is too small to make a difference), there's no hope on the horizon. Gnome, KDE, and Ubuntu are all chasing desktops and users that don't currently exist. Apple is cleaning up. And MS appears to be dumping a lot more capital into attracting developers and backpedalling from their everything-like-a-tablet mistake. They're getting interesting again.

Windows 8 was the opportunity. We had the infrastructure and manpower to take advantage of it. And, by shipping terrible design and premature technologies, we blew it.

No, I'm not the problem. But I do care, and I'm trying to describe the problem as best as I can. One day I'd like to be able to bring developers back onto the Linux desktop. One day.

I particularly like that RMS quote

Posted Dec 31, 2015 22:33 UTC (Thu) by halla (subscriber, #14185) [Link] (2 responses)

"You're shooting the messenger. I am nothing more than a symptom of a problem, and probably not even that."

No, you're not. You're not the messenger, you're not the symptom, you're the problem. Admit that, and you can help with things to progress, but as long as you're in denial, you're the problem. Just check how many whiny posts you've made over the last couple of years -- it's appalling. This, really, is a time for _you_, and I mean you, personally, to take stock of what you are about and what you are doing, and, maybe, figure out a way to be constructive in the future.

I particularly like that RMS quote

Posted Jan 1, 2016 0:17 UTC (Fri) by viro (subscriber, #7872) [Link] (1 responses)

I probably will regret stepping into that, but... What exactly is the problem? Would somebody considering GNOME and KDE as excellent software emetics (just RTFS and go) qualify? Or considering both projects as proof that letting feng shui grifters^W^W^WUX theorists run wild is a Bloody Bad Idea(tm)?

PS: since you appear to be hell-bent on playing this ridiculous DSW... to forestall the questions as to what have I been doing the last decade - this and that, git log on the kernel repository and see for yourself.

I particularly like that RMS quote

Posted Jan 1, 2016 0:41 UTC (Fri) by nix (subscriber, #2304) [Link]

It's more than two decades now, I believe. (Where the hell does the time go...)

I particularly like that RMS quote

Posted Jan 1, 2016 20:21 UTC (Fri) by madscientist (subscriber, #16861) [Link] (2 responses)

I really don't agree with boudewijn's accusatory tone, however:

> Then Gnome 3 and KDE 4 arrived. They wasted huge amounts of my time and forced me to switch my team to Macs just to get back to work.

I guess I just don't understand this. As I mentioned above-thread I've been using UNIX (lots of different variants) since the 1980's and I started with Linux in 1993 (lots of different distributions). I don't see anything about Gnome 3 which would justify the anger you have.

Certainly the difference between Gnome 2 and OSX is *far* greater than the difference between Gnome 2 and Gnome 3. My current employer only provides a macbook and I used OSX for a while: it drives me absolutely nuts: I really hate it. I now dual-boot Linux and try to avoid rebooting into OSX as much as possible.

I did skip the first few Gnome 3 versions (I think I jumped from Gnome2 to something like Gnome 3.6 or so): maybe the early versions were problematic and things have gotten better? Maybe the type of work you do causes the DE to impact you more than me?

I make the following customizations: change capslock to CTRL, enable focus-follows-mouse mode, install two extensions: Frippery panel favorites and system-monitor (to put CPU etc. graphs on the panel), and make a change to force the title bars to be smaller. Of these only the last one requires any "magic" to implement, and it's an issue with the default theme, not the Gnome3 implementation.

There is only one single thing about Gnome3 that really bugs me, and that's the decision to force password entry as a modal dialog: https://bugzilla.gnome.org/show_bug.cgi?id=688434 It's fine if it's your login password being requested, and possibly even SSH passphrases, but email account passwords, network passwords, etc. MUST allow you to go look them up somewhere then cut and paste them. This is being "fixed" by apps changing to not use the Gnome3 password dialog, which is not great either.

> But I do care, and I'm trying to describe the problem as best as I can.

Well, honestly, "a bunch of buggy, poorly thought out barely-working environments" cannot be considered a useful (or, from my experience, accurate) description by any stretch :).

I particularly like that RMS quote

Posted Jan 4, 2016 20:58 UTC (Mon) by bronson (subscriber, #4806) [Link] (1 responses)

> Certainly the difference between Gnome 2 and OSX is *far* greater than the difference between Gnome 2 and Gnome 3

That makes sense but it wasn't my experience. Once we got used to Homebrew (bleah), it was a disturbingly easy switch. The Mac and Gnome 2 feel pretty similar, especially to a worker that doesn't care much about the platform they're using (IME).

I started with 3.0 (forced upgrade because we needed the latest Ubuntu), commented on a bunch of bug reports, and finally gave up around 3.2 or 3.3. If Mate had been available when Gnome 3 shipped, things might have gone a lot smoother.

> Well, honestly, "a bunch of buggy, poorly thought out barely-working environments" cannot be considered a useful (or, from my experience, accurate) description by any stretch :).

That's true, oversimplified to uselessness. I'm just trying to find a quick way to describe my issues with KDE 4 (especially Nepomuk) and Gnome 3.low. I'll find a less accusatory and grim way of conveying that thought, thanks!

Hm, your desktop prefs sound very similar to mine.

I particularly like that RMS quote

Posted Jan 7, 2016 5:42 UTC (Thu) by flussence (guest, #85566) [Link]

> Once we got used to Homebrew (bleah), it was a disturbingly easy switch. The Mac and Gnome 2 feel pretty similar, especially to a worker that doesn't care much about the platform they're using [...]

It now occurs to me that copying a competitor's UI isn't a blunder in itself. It's getting everyone used to that design, then following up by burning as many bridges as humanly possible while your competitor *is still offering* the same user experience you've got your users hooked on...

Maybe this partly explains why the “glowing Apple sea” phenomenon in FOSS conferences.

I particularly like that RMS quote

Posted Jan 1, 2016 18:27 UTC (Fri) by anselm (subscriber, #2796) [Link]

For the record, I've used Linux on my desktop since 1993, and KDE pretty much ever since it first came out. I like both and I greatly appreciate the effort that people continue to put into them. I do, however, reserve the right not to have to be an uncritical fanboy. There are obvious problems of various sizes in various places, and pretending, for whatever reason, that these don't actually exist or, if they do, are entirely somebody else's fault, doesn't really help anyone in the long run.

Also for the record, I support a number of non-technical people who are running Linux desktops. They're reasonably happy (I'd guess at least as happy as they would be if they were running, say, Windows), but, for example, gratuitous regressions of the type we had between KDE 3 and the earlier versions of KDE 4 don't tend to make things easier, and a more disciplined approach towards backwards compatibility on the part of the developers and/or distributors would, in my personal opinion, certainly do a lot towards making Linux on the desktop a more viable option for more people.

Finally, I do my bit for advancing Linux and free software, and have done so, in various ways, for more than two decades now. It's just that my personal interests do not happen to lie in the development of desktop environment infrastructure. I do take exception to your insinuation that I'm a free-rider who does nothing except complain, and I'd like to ask you to stop with that.

I particularly like that RMS quote

Posted Jan 2, 2016 2:14 UTC (Sat) by flussence (guest, #85566) [Link]

> KDE 2 was a better desktop than KDE 1. KDE 1 was pretty good! KDE 3 was better than KDE 2, and KDE 4 ended up better than KDE 3. And Plasma 5 is a _really_ good desktop

KDE3 was pretty good, but it's strange how the entire thing is slandered as a steaming pile of insecure bitrot any time people call it TDE.

And by the very people responsible for that code in the first place, no less.

I'm not convinced the users are KDE's worst enemy here. It would do well to stop treating them as such.

I particularly like that RMS quote

Posted Jan 1, 2016 17:51 UTC (Fri) by seyman (subscriber, #1172) [Link] (3 responses)

> For me, Gnome 3 and KDE 4 were the final nail in the Linux Desktop coffin.

I have pretty much the opposite reaction. Gnome 3 was the first step towards making a Linux Desktop that actually works, instead of trying to being a bad copy of Windows. The Gnome devs decided to work on an achievable goal instead of one that was never going to happen no matter why.

> What kind of aggressive tactics are you thinking of?

I can't imagine a desktop replacing Windows without its main sponsors committing blackmail and/or fraud. Why anyone would want to do this, I have no idea.

> Putting out a better system in every way, and charging nothing for it, is pretty darned aggressive right there.

As I've already said, this requires a level playing field. The desktop market hasn't been like this since the mid-1980s.

I particularly like that RMS quote

Posted Jan 2, 2016 2:30 UTC (Sat) by flussence (guest, #85566) [Link] (2 responses)

Gnome 3 is a bad copy of Android 3. And although they both ignore Linux conventions and standards in many of the same ways superficially, the Gnome designers don't seem to want to understand *why* Android's did so.

I particularly like that RMS quote

Posted Jan 2, 2016 21:48 UTC (Sat) by seyman (subscriber, #1172) [Link] (1 responses)

> Gnome 3 is a bad copy of Android 3.

JFTR, Gnome 3 and Android 3.0 came out less than 2 months of each other (Android in Febuary, Gnome in April) so this is obviously false.

I particularly like that RMS quote

Posted Jan 2, 2016 23:02 UTC (Sat) by raven667 (subscriber, #5198) [Link]

I think it is more accurate to say that GNOME and others are all responding to the same trends and forces, that we are in the first generation whose primary computing experience will be with mobile touch OS and not the traditional desktop, whose UI expectations are definitely influencing design decisions today that are made with an eye on the inevitable future.

I particularly like that RMS quote

Posted Dec 30, 2015 14:18 UTC (Wed) by andresfreund (subscriber, #69562) [Link]

> the Linux desktop teams seem to have lost ambition and to have settled for doing it for themselves and a small circle of users, rather than aiming for world domination.

From my POV all the plays for world domination in the linux DE environments led to massive delayed releases, significant frustration of existing users *and* existing developers, and to, at least, a decrease in adoption & development rate.

It's often a valid approach to forgo power users and focus "normal users" instead. But that doesn't work if most of your development comes from those power users, if they're your only multiplicators, and your only QA and support.

I also have the impression is that that world domination is not as much the goal anymore. But I think it's because making ambitious and widely unrealistic plays burned out a significant fraction of contributors and users. With naught to show for it.

I particularly like that RMS quote

Posted Dec 20, 2015 22:33 UTC (Sun) by pizza (subscriber, #46) [Link]

> Oh, sure it is worthwhile. Just not relevant.

If it's "worthwhile", then it's relevant to the person doing it.

Quotes of the week

Posted Jan 5, 2016 9:11 UTC (Tue) by mgedmin (subscriber, #34497) [Link] (1 responses)

What's the context for the git repository format change plans?

I remember when Bazaar used to change the repository format once a month. It was painful.

Quotes of the week

Posted Jan 5, 2016 10:58 UTC (Tue) by MrWim (subscriber, #47432) [Link]

> What's the context for the git repository format change plans?

It seems to have come from this patch: http://thread.gmane.org/gmane.comp.version-control.git/28...

There's been various discussions on the git mailing list about adding more efficient ways of storing and manipulating refs:

http://thread.gmane.org/gmane.comp.version-control.git/28...
http://thread.gmane.org/gmane.comp.version-control.git/28...