Would Linux be more popular with LESS choice?

[MALsPa]

I'm not advocating forceful removal of too much choice in Linux. It will have to be voluntary.

Okay, then, lets wait around and see how many people voluntarily decide NOT to put out new distros, DEs, and apps so that perhaps Linux can become more popular. Is it realistic to think that this will ever happen? I just can't see it. People will come up with new ideas and will want to implement them, and they'll continue to do so because they can.

What would you rather have: quantity, or quality?

Of course I would rather have quality. But I have no problem with the quality of the distros, DEs, and apps in Linux right now. And I have no problem with the quantity, either.

If the Linux desktop market share is only 1%, then it really doesn't matter.

I counter that it really doesn't matter if the Linux desktop market share is only 1%. It certainly doesn't matter to me.

You're not as "selfish" as you make yourself out on this one. I don't see you at this forum just to get what you need and then take off. I see you helping people, so obviously you're not just here for you.

I don't know about that. Helping out at the Linux forums is one thing, but I don't care if more people use Linux or not. As long as Linux is out there for ME to use, I'm happy.

[mzsade]

they would rather stick with their boot-legged copies of XP.

Why not? Both are free.

Well, apart from being not the 'right thing' (don't you laugh), there's the benefit of being free from virus/anti-virus and fragmentation headaches. But no-----o! You see, it has nothing to do with the mind-boggling number of of Linux distros or Linux itself, it's the mindset that a Windows user develops that makes him reluctant to try anything different, having gotten comfortably ensconced in a particular way of working right from the start, whereas in Linux, one has to make an initial effort--not anymore, especially with Mint, but the Linux reputation of being difficult precedes it.

[markfiend]

Why are you guys laughing at me?

Because you keep on missing the point. What you are asking goes directly against the core ethos of the whole GNU/Linux project.

No-one cares about market share. The choice inherent is developer choice. If you actually look at some of the stuff RMS has written: the GNU project is for developers by developers. The rest of us just get freebies as an offshoot of that; the developers would (for the most part) keep doing what they do for the love of it, even if no-one else used it.

What you are asking, that developers stop working on some distros and apps, and concentrate on a smaller number of distros and apps, does sound very much indeed to me like you are proposing "Get all the Linux developers to stop developing what they want to develop and start developing what I want them to develop."

And finally:

they would rather stick with their boot-legged copies of XP.

Why not? Both are free.

Free as in free beer maybe (technically free as in stolen beer) but not free as in free speech. And that is the sort of freedom I cherish about GNU/Linux. And mintnoob's ideas would take away from this freedom.

[randomizer]

Free as in free beer maybe (technically free as in stolen beer) but not free as in free speech.

I know that, but to the typical Windows user there is only what you pay for and what you don't. There's limited concept of true libre in their world view. There is a warped perception of it, however. Many believe they have a right to use software that they didn't pay for and which is not legally available for free (gratis) in any way they want, even though it's not actually possible since the source is closed. There is also a second group which believes that when they pay for something that they own it and can't stand the concept of restrictive licencing (such as with Windows, since you never actually own a copy of it). Sadly this second group typically don't go the extra step and dump restrictively licenced products. But I digress. My point is simple: too few care about freedom, only free beer.

We must be running out of this free beer by now with how many times that analogy has been used.

[markfiend]

Oh agreed, I didn't mean to imply that you didn't know, sorry.

We must be running out of this free beer by now with how many times that analogy has been used.

*hic*

[mintnoob]

For all of you who think choice will suffer greatly if the Linux community decided to slim down on the amount of distros.

Let's say opponents of slimming Linux worst nightmare happened and Linux's number of distros went from an insane 500+ to only 10.

Here's the amount of choice you would still get from only the current top 10 distros: 67 choices!

1) Ubuntu (8): GNOME, KDE, LXDE, Openbox, Xfce, + Studio, Christian, Edubuntu
2) Fedora (5): GNOME, KDE, LXDE, Openbox, Xfce5 -
3) Linux Mint (5): Fluxbox, GNOME, KDE, LXDE, Xfce
4) openSUSE (6): Blackbox, GNOME, IceWM, KDE, WMaker, Xfce
5) PCLinuxOS (8): Enlightenment, Fluxbox, GNOME, IceWM, KDE, LXDE, Openbox, Xfce
6) Debian (10): AfterStep, Blackbox, Fluxbox, GNOME, IceWM, KDE, LXDE, Openbox, WMaker, Xfce
7) Mandriva (10): AfterStep, Blackbox, Fluxbox, GNOME, IceWM, KDE, LXDE, Openbox, WMaker, Xfce
8 ) Sabayon (4): Fluxbox, GNOME, KDE, Xfce
9) Arch (8): Blackbox, Fluxbox, GNOME, IceWM, KDE, Openbox, WMaker, Xfce
10) Puppy (1): JWM


Now I'm not advocating the number of Linux distros be reduced to 10, but just showing that even with just the top 10 distros, there's still a ridiculous amount of choice! Now show a newbie this list and tell them "Here ya go, let me know which of these distros works for you!" and see how many of them say "Screw that, I'm sticking with Windows/Mac." (And this is just with the top 10 distros!)

[Snydar]

I don't think having a large selection of distros is really a negative thing... a good majority of those hundreds of distros are run by only a couple people, and many do not contribute upstream, they just repackage things, include different apps, and label it as their own.

To me the frustrating part about Linux fragmentation happens by the crazy amount of applications that do the same thing.

How many music players exist that run on GTK or Qt? And how many of them are actually considered "complete"? (Past a beta version, suitable for use among thousands of people with few bugs...)

I have noticed that a lot of projects are started by people who want to learn how to program. They kind of copy the main ideas and layouts of other existing applications, then make their own implementation of it. I'm happy they get some practice, people can report bugs and they can learn more... but couldn't their time be focused on another existing project? I feel like half the time the project gets fairly popular, and then development just stops... which helps nobody. It seems to me that with music players, the main difference between them is just the layout and the features included. Most "developers" aren't changing packages upstream, they are just creating their own app, with certain features, plug-ins and certain layouts. Why can't they just contribute to an already existing project and offer their layout or new features as plug-ins for the same supported version of the application? Instead of having 20 different music players... why not have at the most 3 or 4? One that includes just the basics which the average user would need, one that includes advanced features, and one that is a beta test ground for new and experimental features. Wouldn't centralizing projects like this propel development and encourage progress?

This story holds true for many types of applications. Just look at which distros include which default applications, and look how similar they all are in features. Am I wrong and is the fragmentation actually helping things out? One person gets an idea, then another person can copy that new feature, make it better and then implement it into their own program? Why not just have fewer applications with fewer problems? Does the competition to have the coolest music player encourage better development? Or would bringing people together and brainstorming encourage better development? I know both ways can be effective, but with Linux being free (as in beer) wouldn't the latter (centralization) encourage more progress because (for the most part) people aren't making money of their programming in Linux? I get frustrated if I need more than one application for different features when there could be one project, with all of the features available by having plug-ins. Isn't this an everyone wins scenario? It still offers a depth of freedom in features for the user to choose. Contributions of new plug-ins would be available to all, they could be evaluated, improved, and then included by default. Sounds good to me.

People who use Linux usually like helping others, contributing to projects and are happy to answer questions. Why not focus our efforts a little more and contribute to centralizing some projects? Two-hundred heads on 3 or 4 different projects are better than one head on one project. We should swallow our pride on having our own cool application or distro, and focus our efforts on fewer projects. Maybe that is what is happening already though? There are stragglers making their own efforts, but the big contenders in applications and distros do have big teams. Linux is doing good for its current target audience, but maybe focusing development will bring in more users and more contributors and blossom a new future for Linux. (A little over the top, but you should get my point.)

[MALsPa]

Well, let's say I saw things about Amarok that I didn't like, and that I had the brains to make it a better app, to be something closer to what I thought it should be. I could contribute my services to developing Amarok.

But what if I wanted to make some serious changes to it? How much freedom would I be allowed to do what I wanted to do with Amarok? The Amarok devs are not going to just let me waltz in and make my own changes to it, right? I might be able to contribute some things, but to a large degree, my hands would be tied. I couldn't go in and change Amarok into something that I felt would be a better app.

So the only solution would be to create my own music player. Perhaps it would be similar to Amarok, but with my own touches. Something that does what I want it to do.

And now we have another music player out there. If enough people like it, it might influence Amarok devs and any other people working on other music players. Or they may decide that what I've done doesn't fit their vision of what a music player should be.

I think that's the kind of thing that happens with Linux. With distros, some people didn't like what Debian was all about, the way Debian was organized and run, and the way the distro was run. So they created spin-offs: Mepis, Ubuntu, etc. It goes on and on. Same thing with desktop environments, same thing with applications. You can't just go in and work on an existing project and be able to turn it into whatever vision you have for it. If you really want something, if you have a vision for something, then you have to create it yourself, like Clem did, like Warren Woodford did, like Tex of PCLOS did. Like the creators of Xfce and LXDE did. Like the creators of some of the various music players did.

This is the reality. This is why there's "fragmentation" in Linux, right? And this is why there will always be fragmentation in Linux -- although I still maintain that the fragmentation, or "choice" as it's been called here, is not necessarily a bad thing, and that furthermore it can't be demonstrated that it hurts Linux popularity, anecdotes aside; and as I've said before, Linux' lack of popularity isn't necessarily a bad thing, either, IMHO.

[randomizer]

Now show a newbie this list and tell them "Here ya go, let me know which of these distros works for you!" and see how many of them say "Screw that, I'm sticking with Windows/Mac." (And this is just with the top 10 distros!)

Well, perhaps you could do them a favour and not show them 67 flavours of Linux? You're pulling an Einstein here, looking at worst case scenarios and making an argument on them. What if I showed people a list of available web browsers and told them to pick?

The other option is to simply say "ok, use Windows/Mac." Who cares if they don't like the look of all the choice? It's their choice not to like it... right?

[Zwopper]

Now show a newbie this list and tell them "Here ya go, let me know which of these distros works for you!" and see how many of them say "Screw that, I'm sticking with Windows/Mac." (And this is just with the top 10 distros!)

Well, perhaps you could do them a favour and not show them 67 flavours of Linux? You're pulling an Einstein here, looking at worst case scenarios and making an argument on them. What if I showed people a list of available web browsers and told them to pick?

The other option is to simply say "ok, use Windows/Mac." Who cares if they don't like the look of all the choice? It's their choice not to like it... right?

+1
Although the web browser thing is happening here in Europe due to some EC ruling - "Browser Choice" is now mandatory in windows after the upgrade to IE8 - you're presented with 20 or so browser choices - although most people stick with IE cause it's what they know.
But that's beside the point, choices are always good.
If someone asks me about Linux, I analyze their needs and install something according to that, usually, Mint, #! and as of late PCLOS.
An "ordinary" user has no clue which OS is good for them, so they need help in choosing, be it Windows, any Linux distro or MacOS - the needs are what decide what OS you should use, and in order to find that out you need help, that's what every company does, so why shouldn't it apply to everyday users as well...

[monkeyboy]

Its like beer and women. I would absolutely hate to be limited to the lowest common denominator of either because somebody can't figure out what they are doing.

[lexon]

Until there is a driving force in Linux, Linux will never be more than a footnote.
Look at Apple. Apple PC's are at least twice the price of a Windows PC and are still selling very well.

Quote:

"It's not a traditional computer company. It's a mobile devices company."

http://www.sfgate.com/cgi-bin/article.c ... 1E4VDA.DTL

Lex

[lexon]

Its like beer and women. I would absolutely hate to be limited to the lowest common denominator of either because somebody can't figure out what they are doing.

I can agree with that. Until micro breweries started to make beer in the USA, we had very little choice. Actually for some years we had no choice for “good” beer.

We will never run out of women.

Lex

[mintnoob]

Usage share of web client operating systems. (Source: Median values from Usage share of operating systems for May 2010.)
Windows XP (50.50%)
Windows Vista (21.42%)
Windows 7 (14.22%)
Mac OS X (5.80%)
Linux (1.24%)

Now when you say some of the problems with Linux is lack of hardware support (for printers, etc) or software compatible versions from commercial companies (iTunes, Yahoo IM, etc) and lack of games for Linux to bring in larger number of users, I believe whole-heatedly that the above reason is mostly why; with such minuscule market share, why would commercial companies bother?

[mintnoob]

I have noticed that a lot of projects are started by people who want to learn how to program. They kind of copy the main ideas and layouts of other existing applications, then make their own implementation of it. I'm happy they get some practice, people can report bugs and they can learn more... but couldn't their time be focused on another existing project? I feel like half the time the project gets fairly popular, and then development just stops... which helps nobody.

+1

[mintnoob]

Well, let's say I saw things about Amarok that I didn't like, and that I had the brains to make it a better app, to be something closer to what I thought it should be. I could contribute my services to developing Amarok.

But what if I wanted to make some serious changes to it? How much freedom would I be allowed to do what I wanted to do with Amarok? The Amarok devs are not going to just let me waltz in and make my own changes to it, right? I might be able to contribute some things, but to a large degree, my hands would be tied. I couldn't go in and change Amarok into something that I felt would be a better app.

There's almost 20 music players I counted:

aTunes, Amarok, Audacious, Banshee, Bmpx, Clementine, Decibel, Gnome Music Player, Gudyadequ, Exaile, Juk, Listen, Muine, Music Player Daemon, Quod Libet, Rhythmbox, Sonata, SongBird (now Lyrebird)

What are the odds that NONE of these projects would love to implement your idea if it was worthy?

Your fears are unfounded in the case of music players which there are a flood of apps in that category. With app categories that have little choice, there would be greater odds one would need to start there own project if the few existing projects were not open to your ideas.

[mintnoob]

Well, perhaps you could do them a favour and not show them 67 flavours of Linux? You're pulling an Einstein here, looking at worst case scenarios and making an argument on them. What if I showed people a list of available web browsers and told them to pick?

The other option is to simply say "ok, use Windows/Mac." Who cares if they don't like the look of all the choice? It's their choice not to like it... right?

I would never do that to a noob. I was just making a point.

However as it stands now, almost every noob needs guidance from a somewhat experienced Linuxer to guide them through the maze of Linux choices.

Even with Mint (which I personally think is the best distro for noobs), the choice Mint offers (GNOME, KDE, Xfce, LXDE, Fluxbox) can be daunting enough for a noob and that's only 5 choices!

A noob attempting to convert to Linux alone hardly stands a chance IMO.

[MALsPa]

Well, mintnoob...

To sum up, I don't agree that the level of Linux' popularity is a problem at all. I don't agree that less choice would necessarily make Linux more popular. I don't believe that very many people will voluntarily hold back from going forward with any Linux project simply because other people think there is too much choice in Linux. And nobody can force anyone to not go forward with any open source project.

Good luck with your crusade.

[randomizer]

However as it stands now, almost every noob needs guidance from a somewhat experienced Linuxer to guide them through the maze of Linux choices.

And almost every noob needs help from an experienced Windows user to, well, use Windows. That's the nature of complex software, you don't just know it, you have to learn it. Once you've got into your comfort zone you're fine as long as nothing changes. For most users of any OS this comfort zone usually excludes even the most basic of troubleshooting knowledge so as long as everything works they are fine. The moment something doesn't work they need help because they don't know what they're doing (or what their computer is doing).

The trouble with potential Windows to Linux converts (and I dare say vice versa, even though it would happen less often anyway) is that any problems they've previously solved or solve in a trivial amount of time regularly do not seem like problems any more, just part of everyday work. When a different problem occurs on a different OS, their small repository of troubleshooting information gets them nowhere and suddenly what could possibly be a trivial problem seems like a behemoth to overcome. That's where Google and the geek kid down the street come in, or where they give up and say "X operating system sucks because it doesn't do Y!"

Technological illiterates will always have problems. Choice is not a problem unless you perceive it as one. Choice does not break your system, but choice can make it hard for you to decide what your system will be. If you are not used to choice, it will seem worse than it is. You can throw market share graphs around or hypothesise all day about whether less choice will make Linux more popular or not, but this fundamental question will not be answered by doing that: is the problem with the choice or with the lack of choice that you had to begin with?

[markfiend]

mintnoob: anything to say to this?

What you are asking goes directly against the core ethos of the whole GNU/Linux project.

No-one cares about market share. The choice inherent is developer choice. If you actually look at some of the stuff RMS has written: the GNU project is for developers by developers. The rest of us just get freebies as an offshoot of that; the developers would (for the most part) keep doing what they do for the love of it, even if no-one else used it.

What you are asking, that developers stop working on some distros and apps, and concentrate on a smaller number of distros and apps, does sound very much indeed to me like you are proposing "Get all the Linux developers to stop developing what they want to develop and start developing what I want them to develop."