OSNews

I don't think the FOSS community should start to use lock-in tactics just because everyone else does. We should do what we believe is right, instead of trying to achieve Pyrrhic victories.

I understand your point and thats mostly how FOSS project development operates today. It works well but not for the success of alternative platforms.

Perhaps Linux could only move forward in the respect to adoption if a new software development ideology were formed; an idea that focuses exclusively on innovation for FOSS platforms like Linux.

Edited 2008-01-23 23:25 UTC

The idea has struck me a few times and is the primary reason I use "Linux based OS" or a specific distribution name. Here's why:

Linux is only a single commodity part of a bigger thing. Linux is the kernel, the core OS itself and nothing more. We all agree that one requires the rest of the distribution on top of the kernel or it's of no use. The smallest distributions specifically designed for a phone or R/C radio even include the kernel plus something. That something may be only the slim program between the user's four buttons and the kernel wrapped hardware but it's more than Linux. It's like referring to all firearms as fireingpins; but different distributions of fireingpins. Or, referring too all cars as Hemmy's but different distributions of the core engine.

It may be something as simple as referring to specific distributions; the kernel plus user space programs. Linux seems to cause an overwhelming fear of choice. If it's "Linux" it must be that complicated thing of which there are hundreds of programs for each function. But, they are all different. They are all seporate distributions onto themselves even though they are assembled from the same parts. Why not refer to Ubuntu or Ubuntu Linux rather than Linux. I don't run Linux, I run Mandriva which happens to be a Linux based OS.

The answer is surely not to reduce the number of choices but maybe start to recognize that these distribution things are not all the same even though they happen to have a Linux kernel at there core.

"linux based", i can live with that.
better then gnu/linux at least...
(sorry, i guess this can become inflamatory)

In that case, you can count me out. One of the big selling points of FOSS for me is its ubiquity. The fact that it doesn't matter if the underlying OS is Linux, Irix, Solaris, Windows, FreeBSD or OS X. The tools I need to use will be there waiting for me, because those tools are Open Source. Lose that and it's back to vendor lock in, of a slightly different type.

Exactly. Plus, the "halo effect" of OSS should not be underestimated--the fact that users come to be familiar with and dependent on FOSS software means that they will be that more likely to be willing to switch to Linux in the future, because all of that same software they're used to runs there.

With Linux on the desktop going from a slow crawl to verging on an explosion ...

While it would be nice if this were true, and perhaps it is, what is the evidence, such as a few facts?

As for the rest, I found it very hard to work out what this guy's big idea is. Let's stop arguing and all pull together? Most folks think that. Boil everything down to a few uber-distros? Less of an idea, more of a fantasty. Fitting the right distro to the right customer? Perhaps, but it's all hidden in a thicket of verbiage and tales about what sounds like a rather complicated past. I'm sure it's not the guy's fault, just not very cogent journalism perhaps.

In the end, this interview left me with the feeling that there's one in every town, but if you're lucky you'll never have to meet them. I think I'd rather put my bets on Mark Shuttleworth.

I vehemently disagree with that statement. Allowing users to get comfortable with open source applications like FireFox, for example would make switching to an another OS seamless.

I think we need to agree on why people would move from Windows in the first place. My opinion is that it will be because they are fed up with Windows. There are some like me who just feel more comfortable having all of my tools at my fingertips (like bash, grep, find, ls, sed, awk and more). But, there aren't really that many people like me in that respect. Windows users don't know about these tools and consequently don't know what they are missing. And, they aren't going to move to switch for end user applications they can't get on Windows either.

The reason they will move is because the want the freedom that comes with Open Source OS's. They will get tired of the never ending BS from Microsoft. They will move *AWAY* from Windows in search of something else. They won't move *TOWARD* Linux for the applications.

The best thing that can happen for open source operating systems is to be "open". And, that means encouraging cross platform applications that remove some of the barriers that prevent people from moving for the above stated reasons.

Your POV is myopic.

See my previous OSNews post regarding this very topic.
http://www.osnews.com/permalink?295348

In theory, yes. But what actually happened is Firefox became popular as a Windows app. I would say between 80 to 95% of Firefox's browser marketshare is shared between Windows and Mac OS X users. In fact, there has been some speculation that Mozilla might be treating the *nix version as a second-class citizen.



If that were true then Linux might have a credible market score on the desktop by now, right? Instead, most people just complain and continue using Windows for the sake of their applications, while a minuscule few might leap over to Macintosh.

I therefore came to believe great exclusive applications are the only way the little guys can prove themselves worthy on well established/owned territory. After all, Windows is so successful for that reason.

*nix is popular today with businesses because of it's lower TCO and great scalability, not so much because of the "freedom" it provides--though I'm sure they take advantage of the freedom.

Edited 2008-01-24 00:25 UTC

In theory, yes. But what actually happened is Firefox became popular as a Windows app. I would say between least 80 to 95% of Firefox's browser marketshare is shared between Windows and Mac OS X users. In fact, there has been some speculation that Mozilla might be treating the *nix version as a second-class citizen. "
I see Firefox's popularity on Windows as a good thing. "There has been some speculation" .... what does that mean? Are you saying that you bet someone out there feels that way?



If that were true then Linux might have a credible market score on the desktop by now, right? Instead, most people just complain and continue using Windows for the sake of their applications, while a minuscule few might leap over to Macintosh. "
I think you just validated my point. Making the applications cross platform prevents that.


We'll just have to disagree on that one.

Yes, it is a good thing but it proves my point that people just embrace the software on their existing platform rather than try it on the other platforms it runs on.

The speculation/rumor occurred on Mozilla mailing-lists and Slashdot, Digg, and other news sites. Mozilla denied the accusations, of course. If this were true, it certainly isn't a trend I wish to see happen as a *nix user.



Cross-platform and FOSS have been largely ubiquitous for a long time now; at least in my opinion.

If anything, FOSS is becoming more Windows-specific because how some new FOSS projects target proprietary development tools and libraries on Windows, rather than the cross-platform options.



Regardless who agrees upon what... its evident that whatever is happening now is clearly not helping to bring Linux and other FOSS platforms to the masses.

Edited 2008-01-24 00:49 UTC

I therefore came to believe great exclusive applications are the only way the little guys can prove themselves worthy on well established/owned territory.

I thought the basic idea behind FOSS software was to give people more freedom to choose and do things their way, not to force them to a single platform?

Sure...

But this discussion is about why Linux is a failure (in terms of adoption).

That is why I was pondering a new development ideology; one which focuses more on *nix and making it shine in the face of competition than freedom for all users.

WELL DUH!

I mean jeez - are you saying Firefox was originally developed for Linux and ported to Windows? Why would you simply assume that Open Source software is primarily developed for an Open Source OS and then ported to Windows?

The fact is - Windows has the market share. If you're going to write an application, whether it be open or closed source, commercial or free - you're probably going to build it on the platform that is going to have the most market-share for your target audience. Firefox was developed primarily for Windows, secondarily for Linux and other OSes, period - stop pretending that it wasn't.

edit: fixed quote tags...

Edited 2008-01-24 03:29 UTC

You're getting it wrong. Making applications cross-platform isn't a means to bring more people to Linux, but rather to lessen the impact of applications on OS choice, i.e. it makes it less of a hassle for people to use the same application across platforms. In other words, it's not going to convince people to switch, but it won't impede their decision either.

You're basing your argument on the premise that making exclusive applications would bring people over to another OS, but there's no indication that this would happen. In fact, most Linux applications are *not* available on Windows, and yet this doesn't seem to have convinced people to switch - despite the fact that there are *excellent* Linux apps.

Your suggestion would only make life harder for people who use the same apps over many platforms (a trend which ultimately makes the platform irrelevant) without giving *any* guarantees that it would make more people switch. Personally, I'll side with convenience (and more open-source software on *all* OSes) rather than some dubious strategy based on exclusivity.

indeed, linux, bsd, osx, windows.

all this is really secondary to being able to move data across os borders.

Hi,

I vehemently disagree with that statement. Allowing users to get
comfortable with open source applications like FireFox, for example
would make switching to an another OS seamless. "

Seamless or pointless? If you're a Windows user running KDE, openOffice, GCC, etc then what reason do you have to become a Linux user running running KDE, openOffice, GCC, etc?




Despite wishful thinking, Windows is "adequate"...



Can't you just port all of these tools to Windows instead of using Linux (or download the Windows binaries and/or Cygwin)?



No. Most people (except for everyone who is already using open source OSs) really don't care at all, and it doesn't matter how much wishful thinking open source advocates do, people still won't care.

When you buy a car do you ask if the engine management computer is open source? I doubt it - most people don't even think about it.

The same applies (for most people) when they're buying mobile phones, games machines and other devices. Normal people think the same way with computers - they buy the computer, it comes with Windows, they plug it in, it's "adequate", then they use it and continue to use it (without ever thinking about it).



Unfortunately not, people are silly and don't know the difference between price and worth. They'll think "Linux is free so it must be worthless" and "Apple is expensive so it must be better" then they'll buy an Apple/Mac... :-)

Imagine if I had 2 identical fire extinguishers. One cost me $2000, came with a nice pretty box and has a 20 year guarantee. The other one cost me $80, came in a clear plastic bag and has no guarantee. Even though these fire extinguishers are identical, which would you choose in a life or death situation?

The funny part is that (for these people) if you tell them there's an excellent piece of software they can get for free called Linux they won't be interested; but if you tell them you bought a copy of Linux for $800 they'll probably beg you to "pirate" a copy for them before they know what it is... :-)

More secure. No need to run Windows ... so no Windows Update backdoor. No spyware. Phishing sites and malicious websites and spam in general no longer hold any danger to your system. No forced upgrades. 23000 searchable packages available to you to install from one easy-to-use point, guaranteed malware free and all zero cost. There are other reasons, but those will do for starters.



Matter of opinion. It is not good enough in my view, it is fundamentally flawed, and one of its main design objectives seems to be to try to lock you in and limit your choice. Ugh.



There are bits that are in the core of the OS that you can't really port, such as (as examples) SVG support, multiple filesystems, execute permissions, true multi-user (as in, more than one user on the one machine at the same time), lack of the single-point-of-failure registry, and so on.



When you buy a mobile phone, would you buy one that only works with only one network (lock-in), or would you want to have one that can work with any network you choose (open)? A television that works with only one broadcaster? A CD player that plays only Sony CDs?

Most people will avoid lock-in where they can see they have a chance to do so.



You have something of a point here.

http://en.wikipedia.org/wiki/Friedrich_von_Schiller#Quotations
"Against stupidity the gods themselves contend in vain."

Edited 2008-01-24 13:30 UTC

Actually, the exact opposite of what you said is true. As a friend of mine says:

I don't like Windows - I like what I can run on Windows.

You concentrate on building better apps and you will get a HELL of a lot more converts than spewing a bunch of political crap to people who could care less about politics. Take it from me, a Windows user. Drop the preaching and the 'Bill Gates is a seal-clubbing bastard' drivel. I've heard it before and I'm not impressed. Is MS the spawn of Satan? Maybe. Do I care? Not really. Perhaps I would've cared about 10-15 years ago when I was still in my teens, but I'm too old for that sh*t. Understand? You give me some real-world examples of how I'm going to be more productive and get real work done faster on Linux and I will listen. And remember, I'm a power user. I don't get viruses, my computer does not have spyware, and I don't have any major problems to speak of on Windows.

So why should you care about users like me? If you manage to convert power users like me, you'll also convert a lot of other people by default whom I provide tech support for. They will switch if I tell them to. And I will tell them to switch if I switch myself, because I'd want them all to be on the same platform as what I'm using.

If what you say is true then you should already have been converted by now. Linux already has tons of great apps. The fact that you don't know this just reinforces the point that making more OSS apps Linux-only probably won't raise the visibility of Linux, or the apps, whereas making them cross-platform will at least raise the visibility of the apps, and will make a switch to Linux at least able to be considered by the users of said apps.

It really depends on what you're used to and what you are using your computer for.

I too regularly use Windows (XP) as a platform to produce and DJ electonica. I've tried using Linux as a studio PC and it didn't even come close to the power and ease I already had with Windows, a select few sequencers and VSTi's.

However for surfing the net, reading e-mails, watching videos, etc I find Linux a much quicker and easier platform.

And for my home server I run FreeBSD. It's quicker and easier to set up that windows and runs smoother on a box.

So basically there is no "one size fits all" OS. It's just picking the best platform for the best job.

I also have to jump in here in opposition. My logic is simple: more users == good.

We want to engage lead users and developers on the Windows platform in our quest to produce better software through transparency and inclusiveness. We want more testers, more bug reports, more patch submissions, and more feedback. There's a massive untapped population of skilled users on the Windows and Mac platforms that can be a huge asset if we can get them involved in the free software community.

KDE4, for example, could be a "gateway drug" that leads people to explore the various free software platform options. But even if this is not the case, it's important to consider the strengths and weaknesses of free software systems in terms of quality. As Dave Jones prominently argued, it's mostly userspace that sucks, particularly at the application level. The Linux kernel, glibc, and other systems-level components receive a lot of testing and tuning from deep-pocketed corporate contributors. It's the top of the stack, projects like KDE, that could use some more exposure.

The promotion of cross-platform development environments should be amongst the top priorities of the free software community. We're not going to close the "application deficit" with Windows by pushing *nix-only development frameworks. Software vendors want to "go cross-platform" just as automakers want to "go green". They really don't care about the relatively small increase in target audience. It's a branding and corporate image thing. Whatever, let's indulge them anyway.

Protectionism is never the answer. A free exchange of goods, services, and ideas between communities is essential for progress. Communities ought to have a framework in place for ensuring that this exchange is conducted such that all members have an equal opportunity to benefit. In the Linux community, this social contract, is the GPL, which ensures that we can export our work freely to other communities without compromising our values, our identity, or our destiny. It's our "fair trade" agreement.

OSS is OSS. It will be fragmented every time somebody makes a distro in their garage, and ported every time somebody takes the time to port it. This can't be stopped. Everybody can use it, including people you don't like, and there's no way to change this. This is the whole point of open source.

Which is at least part of the reason it pisses me off every time somebody suggests that everybody should use one graagh-borg-ASSIMILATE distro, window manager or toolkit and take over the world. Forks and competing standards happen because existing tools don't do what developers want the way they want, and the ability to make them (and use whichever one perfers) is what makes open source fun.

Not porting software to windows is a new one to me; I think maybe it hasn't been suggested before because it's impossible. There are actually windows users who can port software to windows, and good luck trying to stop them.

So everybody who does the open source versus windows thing, take a deep breath and repeat after me: Open source is not designed to take over the world. Open source is designed to be free. Open source does not hate windows. Windows is not the enemy.

See? Our blood pressure is lower already.

Edited 2008-01-24 01:49 UTC

The problem is that Open Source is just that open source, porting to Windows is not only accepted but permitted. The problem with no porting to Windows is that there are no killer apps that are Linux only and not only that, if you dont want your software on the Windows platform dont open source it.

This might be helpful (arguable), but it's definitely impossible. Free people do what they like. As many people converge there will be many more people springing up and doing their own thing.

Getting major distributions to converge on best practices is about all you can do. Freedesktop.org is good in that area, but it's slow work.



While this would indeed help Linux and other Free platforms, it's impossible. Just accept it! If you give me the source code, as you must to be a Free software advocate, then I can port it to Windows myself. Once I do I will release the code and binaries. Now what have you done? You've pissed off your fickle freedom-loving users and the software STILL runs on Windows.

Wow stop porting to Windows...

I dare you go ahead... Oh wait it will not work...

Let me give an example. Apache. In the early days the Apache team for years ignored Windows. They said that they were a Unix server only. That raised quite a bit of flack and the result is that Windows is now a completely supported platform.

The reality and I think people are missing it. If people don't want Linux on the desktop it is because Linux on the desktop is not working. Or the people doing Linux on the desktop are not doing something that people want.

Think hard about this. Windows = cost, Linux = free. Yet here we are Windows = market leader, OSX = catching up quickly... Linux? toil toil toil...

On the server different situation completely...

To suggest that BSD code can be 'closed off' so that people can no longer access the source code is disingenious in the extreme.

BSD code can never be closed. You can fork it and close that, but what's to stop you using the existing code base?

As for freedom, I fail to see how a license which *restricts* usage can somehow be considered 'free'.

To suggest that BSD code can be 'closed off' so that people can no longer access the source code is disingenious in the extreme. "

BSD code can be used as the basis of closed software, hiding changes, improvements, regressions and problems of all kinds. The closed software can then be pushed by its backers and succeed over the open software, for a variety of reasons including (possibly) that it might be better. This cannot (legally) happen with GPL'd software.

GPL'd software is good for mandating a future in which code-visible-and-editable is the norm. BSD code is nice, but it does not mandate this. Because it does not *force* code visibility it implicitly supports code hiding, even though it is itself open.

Suppose a company takes some BSD'd code and incorporates it into their product, makes sales, makes a profit, but does not contribute money or patches back to the original application. Society is harmed by a closed application existing, society is harmed by the company's customers choosing the closed application instead of the open application from which it was in whole or in part derived, and society is harmed by improvements not flowing to all users.

It is not sufficient that existing and future users of the BSD'd version continue to use it and be free. The code has been closed off to the detriment of society.

If said application had been GPL'd instead then the company would be faced with reimplementing the functionality in a probable inferior way, or they would be required to publish their changes. Even if they chose to roll their own society would still be served by its probable inferiority driving customers to options which *were* based on the GPL'd code.



This argument is always raised whenever the BSD license's drawbacks are discussed. Nothing can stop you from using the last free version the day after the fork, and nothing is likely to stop you at any time. That's not the point I was making and I think it is as disingenuous for you to suggest that it was as it is for me to suggest that BSD code can be closed off.



I know you do. I have this conversation often with my BSD-fan friends; we continue to fail to understand each other on this point. It's an ongoing problem. I run in to the same problem debating socialism vs. libertarianism for many of the same reasons, regardless of which side I choose.

We can argue all day about whether the greatest freedom for the greatest number will be achieved by limiting some freedoms today; I expect we wont agree. We can debate the fine semantic points concerning just who it is whose freedom is being abridged; I suspect we wont agree on that, either. I'd just as soon not start... people seem to like voting me down for saying things which offend their ideals.

I'll happily 'fill in the blank'...K3B. Simply the best burning program out there in my opinion.

I use Nero. Can't complain. It burns everything.

I have no idea why you were modded down for that...but it happens..*sigh*...I've modded you back up one point anyway.

No question that Nero is a fine piece of software, but with K3B as an option I just don't see the need for a commercial alternative. It would be like buying a copy of Wordpad to replace KWrite. A giant step sideways.

I will tell you why people mod him down. Every discussion about free software, someone comes up with a completely valid working free solution, and Tomcat has to reply with the version he has to pay money for.

As if paying for things automatically makes them better.

I use Nero and I also use K3b... K3b outshines Nero on every option. So much so, that Nero seems amateurish in comparison.

Another killer Linux app is Amarok.

Every Ubuntu install I am asked to do, (and they are increasing every day)....

It runs Gnome, but I type this...

sudo aptitude install k3b amarok gps gkrellm mc joe


then set /usr/bin/kdeinit and /usr/bin/gkrellm
as start up programs in session manager... lovely...

Anyway, Amarok, simply the best music application on any platform. Can't be beat.

That's only your opinion. I've tried it and personally found it overkill for when all I wanted was to listen to a few MP3s.

Personally I favour VLC for every day use (the bonus being it runs on all of my platforms) but then that's only my personal preference too.

Another killer Linux app is Amarok.
[..]
Anyway, Amarok, simply the best music application on any platform. Can't be beat.

I don't really think it's such a "killer app". There are a whole range of good audio players for Linux. I personally like Rhythmbox the best. On Mac I would use iTunes. On my XP installation I also use iTunes but only because Rhythmbox isn't available and I don't know of any good similar Windows apps (and whoever is reading this, please, don't suggest anything resembling Winamp cos I just don't like those at all). Anyways, my case is just a good example here for why I consider porting apps to Windows A Good Thing (TM): I can then use familiar and high-quality apps under Windows too whenever the need arises for me to boot into Windows. That's why I find the suggestion of stopping porting apps to Windows inherently annoying: I am forced to boot to Windows from time to time anyway but then I would also be forced to turn to Windows-only apps cos the apps I would like to use are no longer available.

Hey, tomcat, did you hear about the EeePC? Apparently it's doing quite well...

As for the usual rant against Gimp and OpenOffice, it conveniently ignores the fact that, for 95% of users, these software are perfectly adequate.

Anyway, if you're not going to participate in this debate in a meaningful way (i.e. not restate your well-known anti-Linux agenda), perhaps you should simply abstain. Just a thought...

How many of those users actually bought Office or Photoshop, though?

That's the point.

Why should one use pirated software or buy MsOffice, while for home use OpenOffice works just as fine.

Why use pirated Photoshop (Elements) while Gimp, Krita & digiKam (or Commercial Pixel32) does all a home user need.

Same applies to K3b, Amarok, mplayer, lazarus, and so on ...

I'm using kUbuntu around 90% for desktop use (BeOS: 5% for mail savings, Windows: 5% to see a game demo, and compile and test lazarus projects)

BTW: I'm in search for a mail program (or solution) in linux that treat mails as files and not as one big database file, and same for people (email addresses) files. (BeOS like) Any thoughts?

Don't all of them do exactly that?

Well, it does beg the question why they need to find ways to accelerate take-up if there is already an "explosion".

How is this circular logic?

Ken has come to a conclusion. One that the interviewer agrees with, along with most rational people in the world. Because the conclusion is largely subjective in nature, and it agrees with popular conception (and experience), it can legitimately be called correct.

Ken has come to the correct conclusion that the majority of people who use a computer haven't a clue that they have a choice in how it's operated.

Because the conclusion is largely subjective in nature [..] it can legitimately be called correct.

If a conclusion is largely subjective by nature then it actually means it's not possible to tell if it is right or wrong. OTOH if it was objective... Besides, there is nothing to back that claim up. As you said, it is very subjective and there are lots of opinions in both directions.