“The plethora of Free Software applications available today, none working perfectly, is a problem which stands in the way of major adoption of Linux on the desktop. In order to conquer the desktop, we have to stand united.” Read the controversial editorial at Freshmeat, by Marius Andreiana.
Too Much Free Software
2003-04-04 Open Source 94 Comments
What I think OSS needs is more Commercial software… If companies agreed to open up the source after a certain amount of sales (or time ) went by you could have good quality software and developpers could get paid too.
Either that or someone with the right ideology and deep pockets that could fund some projects.
I use linux ( Gentoo ) and love it. But this article is exactly right. We dont need 30 different sound servers and the like. I know that part of this is what makes linux great, being fully customizable and always being able to meet your needs, but vi will do what vim will do which nano will do. I for one would like to see more work done on a few specific ones then to have a hundred thousand different ones.
Because no matter if what he says is the truth (I am with him personally), there will always be people on this site who will paint him with trolling colors.
The author makes a pretty nasty jab at Qt.
Not sure how the author could claim that he wants KDE to be one of the standards and at the same time endorse Gtk as the only gui library free software developers should use.
I don’t really get editorials like this. Sure, there are a lot of browsers, mail clients, music players. There are alot of these on Windows, too. This always happens until one application rises to the top and people lose interest in the other projects. For example, OSS is disapearing and being replaced by alsa.
As far as graphical toolkits go, there are not that many popular graphical toolkits. You have Gtk, Qt, Tcl, and WxWindows. Most apps are using either Gtk or Qt.
Plus, writing window managers, mail servers, and the like are fun.
Er, no, there are not too much free software. Just because YOU dont want all of it, doesn’t mean that it is bad. Just because the average user don’t want all of it doesn’t mean that it is bad.
Sure many projects might seem like a waste of time, but if thats what the developer wants to do with his time then why should he have to work on something else that he might not care about?
GNOME and KDE for the majority, Blackbox and Window Maker for modest hardware. The rest are simply wasted time, both for the developers and for users who try them and then delete them in disgust.
Maybe the author should learn the difference between window managers and desktop environments before making such a rant? And yes, if the user is a “normal desktop” user, then the rest probably is a waste of time, but not all free software is aimed at said user. While i use KDE myself, i find things like fluxbox very interesting, and can even run KDE on top of it if i want to.
Then the author lists a bunch of things that does sound reasonable for KDE and Gnome to share, only problem is whatever the 2 sides will ever agree which kind of system to use.
GTK+. That’s it. Qt still has licensing problems, being non-Free for commercial applications. Motif, Tcl/Tk, wxWindows? Die!
Well, we are some who couldn’t care less about qt being non-free for commcercial applications, especially since QT is extremely cheap. I seriously doubt you will get KDE programmers to switch to GTK anyway, unless you make a GTK that works like QT, in which case the current GTK users probably dont want to use it. (sure GTK– and a GTK-qtlike could both exist, but good luck finding people willing to create the latter when they can just use QT today)
For me, and many others, free software is all about choice. If what this author suggests was to happen, then that choice would be heavily limited. This might benefit computer illiterate users, but why on earth would the developers who make these things limit themselves to benefit others, who don’t even help with the development?
I sure would like to see linux more used on the desktop too, but i am not willing to pay the price of hurting the software that makes linux such a nice platform for technical users too.
I will have to agree with your Qt remark Chris. That was quite uncalled for, while the rest of the editorial was ok.
The problem somewhat exists on Windows as well though. If you try to find some specific app for windows on a site like download.com you will have to go through plenty of badly written/designed apps to find one a usable one. There is no way to prevent this afaik. The only reasonable thing to do is to make an app database (for end users) where you review apps and remove them if they don’t follow certain standards. Also, settle for one app from each category in a distribution instead of several apps doing the same thing. It’s not that hard to download an alternative if you badly want one.
If you make people focus on certain products, make them more widely used, there will be more bug reports and the developers will feel more responsible for their application, there will also be more devs interested in joining the team.
The main problem with free software in this case is that most of them are one or two peoples pet project. Their only interest is to develop an app, and let people use it if they want to. They don’t really feel responsible for what they do, cause they don’t have to. That’s why rewriting an app from scratch is no problem to them, it enables them to try out new ideas and have some fun.
It seems like nowdays, it’s the makers of the desktop environments that are responsible for providing the user with apps of their choice and make them work, and that’s in my opinion a good idea. Crappy apps will always exist, making the good apps harder to find. We will need someone to find the good apps for us and bring some pressure on the devs. But make sure that the devs actually want to make an app for people to use and not just because he thinks it’s fun to develop software. There’s a big difference there.
I guess I’m some kind of odd-ball who doesn’t really have much of a problem using a plethora of applications for the same purpose.
I use Bluefish and Quanta Plus equally. I use GNOME and KDE equally. I use Nautilus and Konqueror equally. I switch my browser daily. I don’t have much issue with my sound server on any distribution (Any distribution worth it’s beans configures sound during setup, with or without user interaction, and any user who knows what he’s doing is more than welcome to change that configuration later on if so desired).
It’s not like GEdit and KEdit write files that are incompatible. Regardless of what application you use for most anything, another free application is going to be just as able to read and begin where you left off. We’re not talking about incompatabilities and horrible issues like Microsoft Office and the other office programs.
Quite frankly, I wouldn’t call this a troll, just a bandwagon from people used to having one or two programs only, and just being intimidated by having so much choice. Nothing’s forcing you to use this and this, and if you feel one choice is better than another, fine, use the choice you want. Complaining about a plethora of free software? Jesus, I think there are more important issues to address, and aren’t as superficial.
Since I have to make my post complete to keep it on topic… sigh…
The possibility to open the source of great software like Gobe Productive would be a step in the right direction. Too bad that effort failed.
After the failed release of their code under the GPL and their supposed adoption by Apple they just sort of disappeared, their website seems to be down for a while now (anyone know anything about the future of the company?)
Freeware isn’t necessarily software for linux. That said, the author does have a point, because linux does depend on freeware in a very special way. We don’t really need Abiword, Kword, OpenOffice, and the other 1000 half-assed software to every solution. We don’t need 20 thousand terminals or 100 text editors. And yes, its precisely about choice and freedom, because too many half-assed word processors will leave consumers with only once choice -Microsoft Word.
But what all don’t understand is that we need 385+ text editors. It provides choice. What people here don’t understand is that its not about providing choice to the user, its about providing choice the programmer. For example, lets say we have a hypothetical programmer named Bob:
Bob has written a multimedia application over the past few months and has now released it. Unfortunately though its found to have a few bugs. So when Bob goes to check his email, he finds a whole lot of bug reports. This is where “choice” comes into play. He can either mundane go thru and fix his code so it actually works well. Which being the super duper programmer Bob is, he would most likely find boring and beneath him. Or he can tell everyone he’s to busy with his new project – the text editor to end all text editors – BobWrite. Effectively telling the community go stuff it and fix the program yourself, but in a very linux time honored fashion.
Had Bob been limited to only correcting current text editors, not only would the text editor not have a cool a name as “BobWrite” but it would just as boring as fixing bugs in his own multimedia application. Obviously, if Bob or any other programmer were faced with the kind of a situation, they probably decide then that they would want to get paid for their boredom. So, Bob would have to close the source, and move to windows, or apple – where people are more willing to pay for software itself. Instead of Bob having to provide some kind of service along with his “free” software like: Limited Edition Sale – BobWrite + 1 month of BobNET online help from Bobsemail@bobnet.com for only 39.99.
To me it seems extremely arrogant to tell anyone what they can or cannot do in their own free time.
What would happen if I would tell Eugenia to shut down OSNews, and go work on Slashdot? “There are hundreds of geek sites out there, and they all suck, you should all work on 1 site”. I would be called an arrogant troll. When somebody says the same thing to volunteering software developers it is “insightful” and “truthfull”. And everybody is *shocked* if someone dares to paint him in “trollish colors”.
Why do you guys care about the other office suites? So what if you think they’re crap? Other people don’t or the project wouldn’t be there.
Use what you want, for God’s sake. Nobody’s forcing you to have a bajillion applications for a single purpose on your system. It’s quite easy to remove what you don’t want.
We don’t need 20 thousand terminals or 100 text editors.
that’s why I would argue that maybe OSS and “the Desktop” aren’t likely to happen without more traditional software companies. open source has tended to be popular because people need an app to serve their needs, not your needs or my needs.
traditional software companies worry about customer needs, so try to attract them and let them do what they do best. OSS software developers do what they can, but in the end most of them are scratching their own itches. not that I’m not thankful for all of their hard work.
as far as QT’s licensing, this whole anti-GPL garbage seems a bit odd to me. I mean developers on every commercial platform pay for their development tools, why is it such a bad thing for them to pay for their development environments for QT? I would think that most major software companies out there would actually prefer commercial development tools to whatever Jimmy@gnome decides to slap together this week. if they start helping Trolltech pay the bills, I’m sure it’ll mean more developers and more development.
can the licensing costs on QT really be as bad as the licensing costs on windows? too much free-as-in-beer mentality and not enough real world common sense?
No no, it’s not about that at all. It’s about wanting to have a few apps that are focused on providing the users with a working solution. It’s a good thing that experimental projects exists, cause that’s often where innovative ideas are born. But someone has to provide the users with some usable software if they want linux to be more widely used. And that can be done if it was organized a bit better. It’s not about telling people what to do, it’s about asking them if they would like to.
Who are you to tell someone else that they should work on another project instead of one that they honestly feel is or will be superior. It’s none of your business, take what you get or become a developer yourself and help make the program you like even better.
This is like saying why don’t we have just one computer manufacturer company? Who needs all the complicated choice. Eliminate the small vendors and merge Dell, Compaq, Hewlett-Packard, and Gateway. All the better for the consumer when we consolidate all the efforts, right? Wrong.
Nice very cool. What about Bob?
OSS/GPL really is a belief system. RMS is clear about this but most of the middle pretend it is not. I am cool with that idea. It won’t ever be mainstream. Maybe that is why I am cool with it.
Some of the compromise licenses are resonable ways to exchange ideas and standards.
Being a programmer myself i totally agree with the above post by Vincent. It may be a shame to admit it but fixing bugs is certainly not interesting nor enjoyable most of the times. I do that all days long at my job so i don’t want to do it when i spend my little free time on OSS projects. What i want to do is to create a program *i* will enjoy programming and using. I’ll fix the bugs that bothers me and that’s it. I could have kept the source for me but i decided to share it so that people could use it. If they use it great, if not, doesn’t matter.
That’s the main reason there are so many free software right now. I don’t think the problem is that they are too many free softwares out there, it’s that the quality of the major OSS projects may not be always on par with commercial ones. I don’t see myself spending myself studying and learning the internals of a big OSS project (Mozilla for example) and only doing debugging. They are many people that are doing it and i’m glad for their work but as i said, i’m doing that already 40 hours+ a week…
It’s not about telling people what to do, it’s about asking them if they would like to.
GNOME and KDE for the majority, Blackbox and Window Maker for modest hardware. The rest are simply wasted time, both for the developers and for users who try them and then delete them in disgust.
Motif, Tcl/Tk, wxWindows? Die!
Please stop developing and using some obscure application when there are better alternatives.
Drop the “not made here” syndrome and your 15 minutes of fame on freshmeat when making the announcement, and unite with the rest of the community.
Is that asking?? Over here this is just plain arrogance.
There’s an old saying, and it’s taken me some time to realize that it applies to computers and software as much as it applies to anything else in life: You get what you pay for.
There’s nothing inherently wrong with having lots of free software, but the “price” you pay for free software is the time and effort it takes to do the research, sort through the bad software and find the software that works best for you.
So what if there’s a hundred text editors? If it’s not very critical to the user, then the first or second one they come across will work fine–end of search. But if your text editor is critical to you (as a programmer, writer, publisher, etc.) then it’s worth it to spend more time finding the “right” one.
Choice is a good thing, even if you don’t really want to make a choice.
Bravo!!! Well put. Yes, there are too many computer-news sites out there linking to each other. We should shut them down, and just work on the uber-news site: News.com. Of course, that means we’d need to kick Cnet out on their collective arse. But, if they ask nice enough, maybe we’ll let them help.
Do you realize how assanine this guy’s argument is? There’s too much free software?!?! WAHHHH! Cry me a friggin’ river! What we need is less ‘I know what’s best for the GPL community’ articles. But, what’s that you say? If I don’t like what he writes, I should stop reading his articles? That’s exaclty what I plan to do. If you don’t like 100 text editors, don’t install them! Don’t use them! If this guy wants to do something useful, he should stop trolling and write some documentation for his favorite projects. At least that would be a benifit to the community.
Now, about free-as-in-beer mentality. Too many people have an absolute refusal to open up there wallets to pay for anything. Opensource is not (should not) be about free-as-in-beer, but free-as-in-speach. Just my 2 cents…
Louis, THANK you for saying what I also believe.
In the OSS world, NO ONE wants to fix bugs! Only the very bad bugs they fix, or the ones they can easily reproduce. The rest of the bugs that bug otgher people are staying unfixed. This is why I find _all_ OSS apps crashy and buggy. Because they are (I am talking about GUI apps, NOT stable apps like postgres or apache).
And then, we get Red Hat, SuSE and Mandrake trying to fix as much as they can, but they can do only that much when they got to debug 3 CDs full of software….
Sure, OSS development is for fun, but _please_ do not hold any hopes high to “conquer” the desktop when the stability and buginess of most of the GUI OSS apps are so pathetic. It is funny how people used to say how buggy Win9x is, but XP and SErver 2003 and OSX are _extremely_ stable for me, while I can crash the apps that come with all distros in less than a half an hour.
People need to realise what they want and act towards this goal only: world domination on OS market, or just have fun coding away.
There is no middle ground, it is black and white.
>Now, about free-as-in-beer mentality. Too many people have an
>absolute refusal to open up there wallets to pay for
>anything. Opensource is not (should not) be about
>free-as-in-beer, but free-as-in-speach. Just my 2 cents..
If Open-Source software becomes free-as-in-speech then its Free-Software and no longer Open-Source. Do not try and mix everything up here…
OSS/GPL really is a belief system. RMS is clear about this but most of the middle pretend it is not. I am cool with that idea. It won’t ever be mainstream. Maybe that is why I am cool with it.
Well, I don’t really see OSS as a philosophy, more of an extension of College-and-University-Student developing in the real world. Having fun, learning stuff and scratching an itch. It’s not about having as many users as possible, making it nicely shirnk wrapped with a nice commercial finish.
It’s bragging to another student: “hey, i made this program that downloads my shedule from the university network, and sends a message to my cellphone when there is a class coming up.” The other student is impressed, asks “can i have a copy of the sourcecode”. It’s not like it will end up in a blue box at the local computer store..
I for one would like to see more work done on a few specific ones then to have a hundred thousand different ones.
This sounds like Linux itself.
Every software package mentioned in the article supplanted something else. Linux largely stole the thunder from Minix and to a lesser extent *BSD. Using Mr. Andreiana’s logic
Linus should have just worked on Minux or BSD rather than writing his upstart OS that was originally far inferior to its predecessors. Instead of Gnome, KDE, etc. we could just soldier on using twm from X11. Who need office suites, just write everything in TeX/LaTeX, calc, etc.
Free software works a lot like evolution. Things change over time and occasionally die out and their nitch taken over by something else.
Lots of free software is to computing what biodiversity is to ecology.
I was under the impression that Minix was for educational purposes.
we need more spell checkers.
Sure, OSS development is for fun, but _please_ do not hold any hopes high to “conquer” the desktop
I think it’s pretty rare for an OSS developer to say: “my program will conquer the world!”. (or they are kidding like linus). They just do it for fun. It’s those damn OSS advocates and lunatics that are screaming “we’ll conquer the woooooorld”..
“We must beat microsoft, because they are evil” isn’t much of a motivation for making a program in my opinion. “I’m bored, let’s make a program that plays mp3s” is a much better motivation.
Working to put food in the mouths of your children, or for that new 17″ apple powerbook, is an even better motivation, but that’s maybe just me.
I was under the impression that Minix was for educational purposes.
I was under the impression that Linux was too when Linus first started it.
Ok, I agree, there are too many applications to choose from. Also many of them are half baked or suffer from poor support. The chance of OSS to dominate the desktop would be better if only a handful of well written, very tested and world-wide supported were written. But let’s be realist, that w’ont happen.
I think an option could be a site (or even a full project) dedicated to qualify/catalog applications according to differente categories (multimedia, office suites, bla bla bla) and its status. This is a method to help “natural selection” as someone said. A group would have to support its project and fix bugs to have a better grade, if not, application could be qulaified as “Good, but unsupported”.
And yes, I understand that one problem to start this “solution” is the problem itself, there are TOO MANY applications to try to check/test/qualify them. I think may be to start at least with popular ones. I don’t know, may be someone has a better approach.
No, it was a hobby, and people liked it, so it blossomed.
As far stability goes, I can only speak from the point of view of a RH user (one that uses KDE and 95% KDE apps). My stability is fairly good. When I first used Linux in 1998, I never seemed to encounter any crashes, but then again, those programs I did use were tremendously basic in comparison. The only regular crashes I encounter are with MythTV and that is not even at 1.0 yet. I do think XP is tremendously stable too. There are a lot of Free Software fanboys that claim XP is unstable but in my experience it is as stable as my RH desktop. So, while I’ll argue against the instability of the Linux desktop, I’ll also argue for the stability of the XP platform. I have to admit though that I’ve gone out of my way in the Linux world to only get well-supported hardware, and use reputable, proven software. The moment I feel like taking a risk, I do encounter instability. But for a tit for tat comparison, Evolution vs Outlook, Mozilla vs IE, OpenOffice vs MS Office, I’d be hard-pressed to find any discernible difference in stability.
The apps you are reffering ARE stable because are the main apps for the platform. But Konqueror is not stable. Gnome-panel, kopete, everybuddy, opera and many other wildly used apps are not either. And these apps are coming with the main distros, so it is important for the commercial distros to only include apps that are truly stable.
Too much free software is bad, as it leads to substandard programs. While having a choice between 5 good programs is a good thing, having a choice between 100 bad programs might as well be having no choice at all.
Example: KDE comes with how many text editors preinstalled – three? four? I keep forgetting the exact number because while each of the editors is a decent effort, none of them is in the same class as vim or emacs. Now, if the authors had concentrated on making a kvim or kemacs, KDE would have one excellent standard editor instead of offering a choice between several nice-tries.
Some are now probably going to say ‘You can always install $EDITOR afterwards’, which is true – but what is then the point of KDE offering editors at all?
(And for the record: I did not start my own OSS project, instead I inherited one and spent most of my time fixing bugs and documenting it.)
” so it is important for the commercial distros to only include apps that are truly stable.”
Thats putting the onus where it belongs.
I can’t believe what I’m reading! Do you guys bitch that there are too many brands of jeans? Too many types of cars? Too many types of restaurants? Too many types of *tools*? There was a time when the whole software industry was like the current OSS one. It was a time (early 1990s) when competition was active, and the software industry as a whole as vibrant and lively. When PC Magazine actually bothered to do a comparison of the major Office suites. When it’s “best software of the year” awards weren’t largely ceremonial. Now, it’s largely a stagnated mess with only one or at most two competitors in each field. That’s a bad thing. In evolutionary terms, it’s a very clear sign that the whole schebang can be wiped out in an instant. And now, people are asking for this dysfunction to infect the OSS world? Those who know nothing about history are doomed to repeat it.
>Do you guys bitch that there are too many brands of jeans?
Personally, I don’t mind choice. I don’t mind having different apps that do similar things. But what I DO mind, is for ALL of them to be STABLE apps. That’s what matters to me. Feature-complete, stable, bug-free. And if they are 100 of them, then great! But if we have 100 of them that none is stable, or good, then the whole thing is useless and pointless for a user.
The point of KDE offering their bundled editors is because some people like them better than the other options. That seems like a no-brainer, dude.
What’s wrong with having Kedit, KWrite, and Kate bundled? Three programs for DIFFERENT PURPOSES. Kedit is the notepad, KWrite is the word processor, Kate is the developer tool.
This article is totally off base. The author doesn’t account for us that don’t WANT to run the programs that all the developers should commit to rather than making their own. It took me several YEARS before I found a window manager I was truly happy with. I went through Window Maker, Blackbox, Fluxbox, Sawfish, Metacity, and have only within the past few months found what I settled on, Waimea. The features which waimea bring to the user is a very lightweight window manager with such a high level of configurability that you can make it act and look almost any way you want it to. If I remember correctly the code is smaller than that of Blackbox and Fluxbox, so much so that subtracting the size of blackbox from fluxbox yields a number only about 31KB larger. It does everything I want it to with some great eyecandy possibilities aswell. I’m certain there is no way to turn any one of the more resource intensive window managers into something as streamlined as this, and even so I’m aware of few that allow for configurable “actions” of how the window manager actually works. Do I think for a minute that the developer of Waimea should have worked together with blackbox or fluxbox? No… cause I honestly don’t believe they would have incorporated it all the same or with the same efficiency. This is the heart of what makes linux so great. Because for every “bad” option out there, there is an equal number of good options, all which meet very different needs. If the community worked on the ideals of the author Waimea would have never come into existence, and I’d still be unhappy using Blackbox and Fluxbox or WindowMaker. Right now linux usability offers similar benefits of microkernels… we have a number of programs which may do small things but they can be combined together to get the whole thing done. And if you don’t want one of those things, you won’t need that part of the program. Unless we recode every single program that is going to be the “best” and the one everyone uses into a highly modular design we won’t have a very good amount of choice or options to get done specifically what we want. Workbone is a great console CD player… I used to use it all the time, but recently I found abcd… a player which better fits my needs and wants. I’m certain there are many more players out there that fit other peoples needs and wants, if we simply combine all the best functions from all of them we may just have another bloated program that the majority of users don’t care to use because it has a bunch of stuff they don’t need or want. As we all know, competition is a great thing, it only feeds the innovation that development teams need to keep their software being used. I may only be .0027% of the linux community, but I for one am happy that there are a number of small applications that most people have never even heard about that specifically meet my needs and nothing more. And at the point there isn’t and application like that, I’ll program one. It’s a lot easier for someone to obtain a small, efficient, but fully functional for the purpose they want it for application by coding it from scratch than it is to attempt to streamline some bloated monster that has every single function possible done in the best way, even if those functions are not something you want or need. I do feel it’s pointless to code another project from scratch that will simply add a single function to a program that’s already out there that does exactly what the user expects except for that. I’m also, however, unaware of anyone who does that. If such projects existed KDE would not be just KDE, but we’d have KDE-light, GTK-KDE, and a bunch of other things that do the same exact thing as KDE, in about the same size, however do it in a different way.
Why didn’t Linus just go work for Microsoft and convince them to create a more Unix-like OS?
Why didn’t Linus just add to Minix?
If the said approach was the RIGHT approach we wouldn’t be where we are today. I wouldn’t have Waimea… I wouldn’t have FancyLauncher, I wouldn’t have epic, I wouldn’t have half of the niche applications I use. And although I’m only a fraction of a percent of the Linux community, I’m glad some developers took the time and made brand NEW programs that did exactly what I want and little to nothing more. There’s already enough WRONG with some of the mainstream Linux apps that we don’t have to go about combining it with other WRONG stuff from other mainstream Linux apps. What this guy is talking about is what Microsoft refers to as “default choices” being made for the user, something I, and i’m sure many other Linux users, do not want to have made for us.
The reason there is lots of free software is b/c there isn’t a program that does what someone wants. Sure we have 5 million text editors but that’s because each on does it a little bit different. I like having the choice b/c most likely one of those 5 million text editors will work the way i want it to. If we only had one, it may be difficult to have every developer agreeing on how it should work. And with only one, its not going to suite everyone’s needs. I’m using linux b/c i want choice. Choice is good, it may be scary for newbies, but soon they will realize what power choice has.
It was a time (early 1990s) when competition was active, and the software industry as a whole as vibrant and lively. When PC Magazine actually bothered to do a comparison of the major Office suites. When it’s “best software of the year” awards weren’t largely ceremonial.
I would argue that it was worse then. not because there was more choice, but because whichever choice tied you to a proprietary file format. try opening a Quattro file in Lotus or Excel, or a Wordperfect document in WordPro or Word. even now, the law firms(and other, I suppose) that are still bound to Wordperfect need to have MS Office to communicate with the outside world.
But Konqueror is not stable. Gnome-panel, kopete, everybuddy, opera and many other wildly used apps are not either. And these apps are coming with the main distros, so it is important for the commercial distros to only include apps that are truly stable.
I can’t even remember the last time konqueror crashed on me. In fact i cant remember the last time any KDE application crashed on me. On the other hand i just had explorer.exe crash on me a few minutes ago, but it did restart itself, so no harm done.
No, i will not claim that windows xp is unstable, because it isn’t. It does crash once in a while for me, and it does pause for 10 seconds once in a while, but so did my redhat system until i replaced the kernel, and i have had linux freeze on me a few times as well, so not much difference there.
I dont know what you do to or with your system to cause programs to crash that much, it is entirely possible that the crashes are due to real bugs, i just don’t see them or encounter them.
And speaking of which, where are your bug reports on bugs.kde.org? I don’t see any bugs reported with your name on it. (at least none containing the word eugenia)
Sure you might not have enough time to help debug it, but if you dont report the bugs, then how do you expect them to be fixed? I know that i can’t do anything to help fix them, because i have no info on how to reproduce them…
Do we really have to get into this whole “what is free software” again? The article was very sound. The fragmentation has to go and it wouldn’t be such a bad idea if major players would dip their beaks into “Free software”.
Free (Liberty) != Free (Gratis)
Since the Open source community has shown time and time again that it is totally inept at coming together and making standards and getting some sort of unification going, then the (last?) hope of Linux as a major player is that Companies with big wallets bet involved. Go ahead and let major players establish the standards then, since 9.999 teams of “hobby developers” lead to 9.999 mostly incompatible solutions!
I don’t want to troll, but it just seems that every time someone points out that there are 9 billion competing standards out there (and that it’s HURTING linux) that someone gets flamed for “trolling”.
Unification and standardization, please!!!
I agree with him. I don’t use Linux but I closely follow the development because I’d like to use it when it’ll be at my taste.
My impression of the Linux development is that the energy is spread out too much. About a years ago, I was reading a paper where they compare all major linux distributions. Well, there was like eight of them! So eight of them (plus the others) share a minimal % of the total users. There should be like just a few that covers all the needs (novice, intermediate, pro, server, maybe one or two more).
You might say that, because I don’t use Linux, my opinion is worth nothing. But I’m simply saying why I’m not using Linux, and that is most probably the opinion of many in my situation. Having a lot of software is not a problem. But I just feel it could be a very good thing if Linux can just get out of his marginal position. It deserve better.
I can’t even remember the last time konqueror crashed on me. In fact i cant remember the last time any KDE application crashed on me.
konqueror is great, but the embedded media stuff tends to crash every time I try to open a media file in konq… of course, it was just a matter of not using that feature, but they should keep that stuff in the unstable branch if it isn’t 100%. if KDE was commercial, I’d be the first to bitch, but I’m pretty amazed at KDE in general, so I work around it.
konqueror is great, but the embedded media stuff tends to crash every time I try to open a media file in konq…
Oh, yeah, i had happily forgotten about that. Used to happen to me as well, until i removed xine-libs, and hence got rid of most of konquerors multimedia capabilities.
I will agree that multimedia players on linux is currently not in a great state, though considering what the developers have to work with (like almost no documentation and even patent problems) i think they are doing a great job.
Lucky for me that i dont use anything but sound playback often, and this at least works just fine. And for what little multimedia files i do playback over the internet i use codeweavers crossover plugin, mostly because i was too lazy to figure out how to use mplayer as a plugin.
I agree completely with your comment about either coding for personal pleasure or for the purpose of OS domination. The biggest problem with the ‘bazaar’ (as opposed to the ‘cathedral’) approach to OSS is lack of accountability and direction. Like it or not, software projects need specific direction and defined goals. Having someone or something spearhead an effort is NOT inherently evil. OSS needs to corporately decide what their objectives (if any) should be both on a global and per project level. This simply cannot be accomplished in a purely democratic/anarchistic setting. Yet because of the idealistic politics of OSS and many developers, this never really happens.
FWIW, this happens to be one of the reasons I find FreeBSD so much more satisfying than Linux. Perhaps it IS just because of personal and/or political reasons, but the bottom line is that I see FreeBSD as a much better and more polished product.
Linux is within shooting distance of becoming a real competitor to commercial desktop OSes. The OSS community has the code-fu to compete with commercial devlopers (and many OSS developers wear both hats), and Linux has made incredible progress in a short time.
However, the devil is in the details. Linux/OSS will never become a solution that _just_works_ without FOCUS AND DISCIPLINE that is is currently lacky in many (most?) OSS projects. Passing the finish line will require tedious hours of bughunting, testing, code auditing, etc. Does the community have the resolve to _really_finish_ projects, or is self interest the primary concern (I wanna code what I want, when I want, and only work on the fun stuff)?
There are people out there who enjoy finding and fixing bugs and auditing code. They are (in my experience) an odd breed of people who can be difficult, at best. But IMO we NEED these people. Take a look at what Theo DeRaadt and his band of code cleaners have accomplished over at OpenBSD. This is exactly the sort of focus and discipline I think OSS projects need.
That’s my $0.02. Flame away.
I would love to give away stuff on the net for free. I would get a kick out of helping out the communty and being an active member. But instead I make a living managing this software, learning about it, working with it and helping others. One of these days I will be a great programmer. But until then I can only offer my thanks.. perhaps one day my code will follow.
If the world were a simpler place. If I didn’t have to pay for food and housing and network bandwidth and provide for my future I would already be releasing free code. But the world is what we make it.
There’s never enough free software!!! I live for trying out new software! I go on Bebits every day to see what’s new and usually download svery new app that comes out for BeOS(mostly all freeware there)It’s like watching TV if you don’t like it change the channel(or delete it) but hasving many options gives the user choices,what one person likes another may hate!and BTW what’s up with the statement:
‘and the rest of the world (e.g., Gobe) ignored, which is great’?
I have never used the Linux oe Windows versions of this app(I already had StarOffice on Windows and no longer run Linux at the moment) but I have the BeOS version and it is a fine program,it does just about all the office type stuff a home user would ever care to do,in fact I use this instead of StarOffice for all my needs for an app of this type,it’s all I need! so I don’t think it’s great ,I think it’s a damn shame it was ignored!
For God’s sake, people! If Linux, or BSD, or BeOS, or Zeta or $(put the most disgusting OS in your opinion here) don’t have the same features of Windows, or Mac OS X or $(put your favorite OS here), why don’t you just use whatever you like and leave the OSS developers and users alone?
I happen to like the option to choose between 385+ text editors, 45+ web browsers, 150+ mail clients, etc and I think I’m the rule, not the exception. What might be feature-less to you could be exactly what I need. But no… You guys must be the kind that wonders about the following issues:
1) Why the hell AOL produces Netscape, Opera Software produces Opera if we already have Internet Explorer?
2) Damn it, Apple. Stop this Quicktime thing! We’ve already got Windows Media Player. And that is valid for you too, Real.
3) WordPerfect, Easy Office, 602 Pro… Who needs this shit? We’ve already have Microsoft Office. I have an idea: I’ll ask their developers to join Microsoft to make Microsoft Office even better.
4) Hmmm… Why those trees all looks different? We’ll need to fix that too.
5) Oh, my God! 16 million of colors?!?!? WTF? Who needs so much colors? We must create a standard to use only red, green, yellow and blue (I’m wondering if black or white will be allowed too).
If $(the worst OS in the world in your opinion) don’t please you in any way, look elsewhere.
Oh! Sorry about the rant and please forgive any mispellings, because English is not my native language (although some of you might think that English is good enough to rest of the world, too).
As far as I know wxwindows is not a gui lib but a framework. Meaning it is based (uses) on a gui lib (gtk on linux, win32 on Windows) to give users a crossplatform framework (framework = abstraction layer).
So putting wxwindows in the list with gui libs is just wrong.
[quote]I use linux ( Gentoo ) and love it. But this article is exactly right. We dont need 30 different sound servers and the like. I know that part of this is what makes linux great, being fully customizable and always being able to meet your needs, but vi will do what vim will do which nano will do. I for one would like to see more work done on a few specific ones then to have a hundred thousand different ones.[/quote]
Good point, we should not have to look at tons of movie players to find one we want just a few would be good, we don’t need 100+ text editers.
Well said, I agree.
most users cant install that “plethora of free software” anyway, as
it’s in targz or cvs
Why re-inventing the wheel ? Better focus on the winning solutions and make them better. If this is not possible in the OSS world because of fragmentation, it means it will never seriously compete with commercial applications.
Well, they’re right. On Linux, there’s no decent movie player and no working sound recorder (like the one in Windows 95) shipped as the default by GNOME, but hey, there are more than 385 text editors!
Xine is a “decent” movie player. It looks and operates about exactly the same as the divx player I use with windows. mplayer is also a “decent” movie player. I’ve used both with no problems.
Then there is a tricky play on workds of “no working sound recorder… shipped as default by GNOME”. So he is admitting there are working sound recorders for linux, just GNOME doesn’t have a default one! Well that is terrible! Every DE and WM needs to ship with a sound recorder I guess. Forget the fact that there are plenty of free (superior) ones available.
I don’t know exactly what to write here, but it’s bad. Hopefully, ALSA will be the only API to access sound devices, OSS will disappear, and maybe there will be only one sound server used by default on all distributions.
Never had a problem with it myself. ALSA is on its way to becoming the default sound system. OSS is just here for the temporary.
There’s Gaim, GnomeICU, Licq, and many more. Gaim has a plugin system which allows the use of all major protocols and, for example, sending messages encrypted with GnuPG (a feature planned to be offered by Yahoo! in an “enterprise” version of their client).
In Windows, there is AIM, ICQ, Powow, Trillian, Yahoo Instant Messenger, MSN Instant Messenger, god only knows how many others. How is it more confusing to users who nothing about messaging than windows?
Ok, now that I have been harsh, I would like to say that there ARE good points in his article. However, those points are hard to apprecciate when he makes them by insulting other applications that people have worked hard on. I mean, I don’t use IceWM, but there are people that do. IceWM is something good for the linux community to have. To say, they should just use blackbox is laughable.
The overall point of the article I agree with. It would be best if people would put work BACK into the original project instead of forking at the drop of a hat. But that doesn’t mean that all alternative projects are bad. Some grow into things that are honestly better. Projects like enlightenment shouldn’t be dissed for trying something radical.
The author has a point. Too many software of the same kind makes the world of users migrating to Linux from other OSes very complex. But, suggesting the upcoming development projects to quit is arrogant, childish, and impractical. To be part of a OSS project, a developer must believe in it. New projects also bring a air of freshness and look at the problems in a drastically different way then already popular projects. Of course, lot of projects are simply reinvent-the-wheel or yet-another projects. I think the distros should figure out what is a good combination and install that as default. Like they can choose – XFree86,GDM, Gnome, Metacity, OpenOffice, Mozilla, Evolution, and Gaim as the default and not confuse the users by asking to choose between applications that do the same work. If they do, they should provide a item by item comparison of the different applications. Otherwise, it serves no purpose. Let the user get familiar with something that is stable and widely used. When the user gets more adventurous or hears about a particular application doing a certain job better, he/she can just find it in the distro CD and install it. If it turns out to be good, the user can uninstall the old app. Simple..
How can a single app be extended to cover all requirements without being massively bloated? If the app is modularised to let everyone add their fave features, yet still avoid bloat, you end up with the mozilla situation, where they have just split the thing apart again.
One of the reasons they gave for this split is because: “Unintegrated applications tend to be faster to load, smaller on average in dynamic memory consumption, and more robust and crashproof.”
That is why you need many seperate apps!
How can you compare OpenOffice to an editor like Jed?
Somewhere, out there, amoung the hundreds of text editors is the perfect one for you. And if it’s not, well, I guess you will have to write it and add to the list.
as others have pointed out here,Windows also has a ton of
shareware, freeware. There is a lot less monoculture in
the windows world than people are prepared to admitt.
So people can deal with a least the same amount of diversity in linux as they get in Windows.
( various versions of word, differences between w95,w95b,
w98, w98se, w2k,wxp,and their respective apps.)
… his suggestions that follow are mostly personal just his personal preferences.
So the question is what gets installed
Less is More, should be the guiding principle. That doesn’t have to mean just one DE.
Desktop environments just need more interopability so using them is similar enough that swithcing between them isn’t that difficult.
Its up to the Distros to make their choices, but in each
Distro, have the software available,but pick one app as
the main default , and have two back ups in the wings.
That gives you 3 apps per category. Not the one , that some would like, but a big improvement over any Distro
that is stupid enought to give you more than that in
a default install.
It is good to have two major DE’s . Just like it is good
to have two engines on a plane. If one fails..
And don’t forget XFCE. 4.0 should be really something.
I take great umbrage at his dismisall of all window manggers other than Widnow Maker, and Blackbox.
Icewm is a great light weight wm, especially for
those coming from a Windows environment.
All Linux needs is
-preinstalls from oems with some spine and vision,
-better hardware driver support,
-a little polish and integration of what we have now,
-a few of the major commercial apps like Photoshop.
Gentleman, we have the techonology.
…and why much of is not that good is because there are a lot of people learning to program. The same thing happened in windows where everyone learned how to program an editor and then went on to greater things (or not).
I think it’s the sign of a healthy developer and user base.
Everyone compares Linux to what Windows is right now, but if you look back to Windows 3.1 and Windows 95/98 many people were complaining about the same things: lack of drivers, lots of useless software, a lack of useful software etc.
Windows even had a huge monolithic and ( somewhat ) competent competitor, although it being the start of the PC busines, IBM did not have nearly as large an installed base to use to its advantage.
True, Microsoft may loom larger now but MS revisions are coming out only every few years, and they have a talent for making anyone they deal suspicious.
Meanwhile, linux is coming out with revisions every sixmonths or less, and since they cannot be a monopoly, and the source code is available people are not suspicious of its source, merely its capability. Linux has a very good chance of catching up.
The one place where Linux could improve its capability is to improve its installs. Everyone who writes a program for windows does not consider it complete until there is a setup program for it. Everone who writes a program for linux should be thinking the same way. If your install merely uses rpm, why not include a setup script for it. The same for apt.
A recent example would be the new Nvidia driver installation. It’s ok as far as it goes, but why couldn’t it, with some decisions made by the user, go to the X config file and edit it for itself, without relying on the user to do it ( like the Texstar Nvidia rpm for mandrake )?
Software, free or otherwise, is like food and money. There may not be a need for more, but there is always a use for more.
Too much free software? Not possible. Choice is always a good thing, without exception.
You might say that, because I don’t use Linux, my opinion is worth nothing. But I’m simply saying why I’m not using Linux, and that is most probably the opinion of many in my situation.
Yeah I might say that, and if I did I would be right. So I’ll say it, but more nicely. Because you do not use linux you have no frame of reference. Those of us who use linux deal with these supposed “problems” every day, and let me tell you what, they don’t exist.
If I decide I want to watch a dvd in linux or in windows I go through the same exact process. I open up a player, I chose (ogle in linux, powerdvd in windows) and I select my source (my dvd drive) and I watch it. I can’t tell any quality difference between either. Power dvd has a prettier interface, but ogle was free. In either windows or linux I would have TONS of options of how to watch this dvd. In fact, in Windows I would probably have MORE. In linux all I know of is ogle, xine, and mplayer. In windows there’s powerdvd, windvd, windows media player, and a whole bunch of others on download.com.
This choice doesn’t bother me in windows and it doesn’t bother me in linux. I don’t see why it would bother anyone. I feel the more I have to choose from, the better chance that one (or more) of them work, and the better chance that of those that work, I can find one I REALLY like.
I am currently working on a free software projects. An ambitious one, that is under 5% completed as of now. Not only that, but a stable solution exists. That solution is widely used, and is one of the best free software actually (namely, the gimp).
But it doesnt act exactly like I want, and Im not going to develop for it. Why? I dont know, I just dont feel like to. Its already written, (in C, with GTK, which I dont like), so all I could do is fix bugs.
So I started my own. What are my chances of becoming as good and stable as the Gimp? close to none I guess… Why do I bother? Because I actually enjoy doing this. When I feel it will have reached a certain level, I will probably post this project to freshmeat. That will be Yet Another Incomplete Graphics Apps. I will post it there because someone might find it useful, or interesting enough to help with the development. If no one use it, I dont care, because I will still have leanred a lot in the process, and actually had fun.
If the project succeeds, I cerainly hope I’ll have a chance to have my 15 minutes of fame on freshmeat, because that will be the only pay I will ever receive (excluding knowledge and fun). If it does become the best software in its category well it won’t seem like I’ve wasted so much time, will it? If not, that time was mine, not yours, so it’s my problem. If I feel like I was wasting time I would do something else, but tv is so boring
But as I said, I would not work on similar free projects, because I’m the kind of guy who needs to start my own, from scratch. I hope I can do it better, to learn and to show off, to kill time. That may or may not be successful, but PLEASE don’t tell me what I should code, or not. It is MY free time. I do as I please with it. Hire me and I’ll do as YOU please, but for now I’m trying to have fun. Thank you.
Wow lots of rambling and typos…
Dont drink and post, kids.
Anyway my point was that you can not say that “resources are wasted on smaller projects” because that is just NOT how it works. Many people wouldn’t work on other projects anyway.
So the resources are BETTER spent doing useless projects (that might become the best one day, linux anyone?) than nothing at all.
as the default and not confuse the users by asking to choose between applications that do the same work.
I couldn’t agree more, and the fact is that is what is happening. Redhat installs Gnome and a set of applications by default depending on whether you select server, workstation, desktop, etc. You only have to go through and handpick if you hit customize. In this manner a user can easily set a set of package and get what he reasonably expects to be there.
The open source world is about choice. When you balance that choice with sane defaults I really see no problems for the end user.
I agree with dwilson’s post — we need to have sane defaults and choice. I’ve installed Red Hat 8 and 9 over the past few weeks, and installation seems to be moving in that direction. The problem is, it isn’t good enough yet.
While I’ve been able to fix all problems I’ve run into (such as getting my Nvidia card to work as it should automatically after installation, and dealing with problems installing software that I like), other people I know would not have had the time or patience or ability to deal with these rather simple problems. Some people may not see that as a problem, but anyone who wants to see Linux compete with Windows as a viable OS for the masses has to see it as one. I would not recommend it to my grandfather, and until I do, I won’t think of it as a truly competitive OS for the home market, even though I love it, and it has become the only OS I use personally.
Those who think that choice is good are right. Those who think that much more work has to be put into standardizing are also right. The answer to the problem (which is an expression of different values and motives) is to have both, and things are moving in this direction. We need distributions aimed at different things; one doesn’t rule out the other.
The author of the article went gone too far in suggesting that people should stop developing what they want to develop; what he should have said is that people who want to help Linux compete for the home market need to get together and decide some things. His underlying point is important, though. The distros that focus on ease-of-use have to have a very simple, stable set of defaults that a person with no technical knowledge can install and use, never having to touch a command line or dealing with driver problems or whathaveyou. This isn’t the current state of affairs. I think a fair modification of the article’s point might be: if you hate Microsoft’s domination of the home market, and you want Linux to compete against it, then these are the things you ought to be working on.
All in all I think this is a bad idea. Developers write code because they want to. They may contribute important features to big projects but occasionally write a some clever little program they want and decide to release it on fm.net. Flaming them that they shouldn’t write their own * and they should just contribute to G* or K* is probably not the smartest thing and will lose developers (and that is the only thing the opensource movement really has going for it).
My problem is that reading over this his views on Free software look like mine, except toned down (a little bit less than I would) for publication.
I favor Gnome and gtk+, AbiWord and Gnumeric, and Mozilla (even over Chimera or Safari which run on my most used platform).
The only point I really disagree with him on is I have long been a user of everybuddy ( http://www.everybuddy.com/ ), but that may well just be because I am tired of G* programs and didn’t want something with aim in the name–plus, based on Gaim’s name I didn’t assume it could do multiple protocols. I might have to try Gaim (hope the choice won’t offend me as a user).
I also agree with him that KDE and OpenOffice are around to stay.
Mostly I just think it is funny that he didn’t tackle the issue of RPM’s unjust atempts to unseat APT: the People’s Choice&tm;.
No, i don’t agree totally.
There’s not TOO mutch software, but i think there are too mutch similar apps.
An example ?
DVDx and DVX could become a really cool crossplatform suite for DVD backup/conversion.
I think that software needs more abrstraction (i mean: routines porting and reusing should be more simply).
And, anyway, i think that similar softwares (eg. Operating System) needs a sort of “operative centrals” where developers can met themselves (there’s too mutch confusion in Sourceforge) to exchange ideas and methods.
I for one would like to see more work done on a few specific ones then to have a hundred thousand different ones
The problem is, most coders write for fun. Writing the basics can be fun, but once you get down to minute bug fixing and trying to perfect certain aspects of the program, the interest level decreases. So we have software with limited features.
But don’t complain, the only reason this software exists is ’cause someone wanted to learn how it worked (Or something similar) and so created their own program. Now this makes up the majority of Open Source Programs.
What you have to understand is this: Just because there are alot of developers that aren’t devoted to their software, doesn’t mean all open source developers aren’t devoted to their software. Their is a select few who stick in it for the long, hard run. As well as the fun coding, they fix bugs provide support and make their programs good. You just have to find a program made by someone like that. Doing so will limit your options severely. Rather than having over 50 web browsers, you’ll just have two or three good ones. Rather than having hundreds of text editors, you’ll have five or six that are worth your time.
The same can be said with windows software. Go to <download.com>. They have alot of poor programs their. Why don’t those programmers just go and work for bigger companies, like Macromedia and Adobe. There is plenty of low quality software for any platform. The trick is to find some (If there is any) that is worth your time. High quality, active programmers, and supported.
And linux has this
(Gimp, KDE, Open Office, Mozilla/Phoenix, Evolution, XMMS, the list goes on)
I find that much of the Microsoft bashing is a response to the mostly inaccurate reviews of non Microsoft software that takes place.
An example I do not recall which site it was but the reviewer concluded that linux was “not ready for the desktop” because there was no shockwave plugin. The author even gave a website that would not run do to a lack of the plugin. As a result of being a skeptic, I launch Mozilla on my Mandrake box and type in the address of the site. To no surprise, it works. Of course it does! I have a shockwave plugin installed in my Mozilla. And I do not have the crossover plugin.
This is just one of the examples of a Linux review (by a journalist, not a casual computer user) being so poorly researched and written. And just plain wrong!
But then, these reviewers often find something that they do not like about linux. A hardware item that was not recognized on installation or a program that does not operate they way they wish (typically not like a particular Microsoft program).
And then, that cliched non sequitor…
…not ready for the desktop.
Nonsense. Maybe there is a desktop operating system that they prefer, but this does not make the non preferred operating system “not ready for the desktop”.
So what many linux enthusiasts end up doing in response is pointing out the many flaws of Microsoft products. This is viewed as bashing by the Microsoft supporters. Often it really is bashing. Most of the time it is disspelling the myth that Microsoft products “just work” and all other products are inferior.
Trolltech has had no licencing issues since 1999 when they descided to dual licence QT. When you install (./configure, make etc.) it asks if you want to enter a registration number for the QPL version or if you would like to continue without registration (GPL).
This is no issue what so ever.
Now, for my rant
The Windows and MacOS versions of QT are qpl and do require you to pay for a licence…which would make GTK a tempting alternitive IF gtk was anywhere near as mature as qt on these platforms…heh one could argue that GTK for Linux doesn’t rival the maturity of Qt ala Linux either
First, I would like to state for the record that I don’t care if Linux ever takes over the desktop, as long as it’s always available for MY desktop.
That said, he article makes an excellent point that too many people start their own projects instead of improving existing projects. If you want to create yet another text editor as a learning project that’s fine, but is it really helpful to the rest of us to post it on freshmeat?
One more thing: Eugenia, have you submitted your long list of bug reports yet? If you really want to help improve the Linux experience, that’s the best thing you can do.
OSS GUI apps unstable? What? I use Evolution all day every day, and I honestly do not *remember* the last time it crashed. Ditto X-Chat (stable releases, anyway). Almost ditto Galeon – it crashed about 20 times yesterday, but guess what I tracked that down to? The Flash plugin. Which is…closed software. Shock. Oh, and ditto gaim. Honestly…
“Gnome-panel, kopete, everybuddy, opera”
gnome-panel? Again, has never crashed for me in a stable release, though 2.1.x did quite often. kopete, I dunno. everybuddy isn’t developed any more, try ayttm…and opera is closed source.
Some time ago I have done some research which operating system is the best for the desktop. I can only conclude one thing: Windows is really the worst desktop operating system:
Notaun is easy to crash. Select some frontends or add an ALSA effect while you use OSS – however for the rest it works perfectly. Media integration in Konqueror works great, I click a media file and I get an embedded Kaboodle media player.
KDE is really the best tool for editing websites, as it is the only program which I know it has network transparency. I can just browse to the ftp of my website, open a document in Kwrite and save it directly to the server. Now that is handy.
I agree on Kopete, 0.5 worked quite good, but 0.61a crashes very often.
Personally I think a lot of users of open source/GPL software need to get their attitude sorted out. I started a window manager called Oroborus and I enjoyed programming it because it fitted my needs. I also decided to release my work to the world, and boy did the sh*t hit the fan. I had a lot of arrogant people DEMANDING features because another WM had the same feature. I was doing this for fun and for free and I was not having this – so I ditched the project.
Still, it was picked up by to OSX community (OroborOSX), XFCE 4 and it’s still being developed by another developer. I think that Open Source Software needs to be careful of people who want to join projects to change that projects direction. I firmly don’t agree with the author of the article because of my encounter with writing Open Source Software – I think people should write more stuff rather that hijacking someone elses project. It will be sometime before I consider writing any more open source software, and I will keep tight reigns on it’s direction. Next time, if my software doesn’t do what THEY want it to do I’ll tell them where to go.
I agree completely. It’s the users who need an attitude change. A lot of them forget that we’re doing this for free, and basically giving our software away as a gift. Yet they still try to *demand* things from us, sometimes even arguing with complete nonsense like “customer demand”. Fine, if you want those features so badly, then take out your wallet and give us some cash.
Heck, even the Windows community seems more friendly. Nobody tries to *demand* features from freeware authors, or even writing articles that there’s “too much freeware” and that all freeware developers must ditch their duplicate projects and work together.
Yes, it’s wrong of users to demand things from people they aren’t paying. Wrong and absurd. This doesn’t change the basic point that for Linux to succeed broadly as a Windows alternative on home computers, these users must be catered to. Again, while I think that the author of this article is wrong to say that people shouldn’t work on the things they want to work on, his points are still valuable because they suggest a strategy for people who do want to see Linux compete in a certain arena.
Daan, some of this ‘research’ is rather misleading. Granted, the installation procedure for WindowsXP gives you basically nothing in terms of choice, but for some users this is ideal. And to say that Linux installation can be as easy as clicking until the system is installed is great — until you try installing on a system that has unsupported or badly supported hardware, which can turn into a huge pain, and almost never happens in Windows.
That leads me to the ‘hardware support’ category — rating Windows and Linux the same in this department isn’t really fair. And granted, installing drivers in Windows is sometimes tricky — but not often. Usually it is quite easy.
In any case, people bring biases to their rating systems. You might rate Linux installation as better than Windows installation because it actually allows you to choose what you put on your machine — and this would be very good reasoning. Certain members of my family, on the other hand, might rate Linux installation very low, because if they ran into any problems whatsoever, they would be completely unable to deal, and the OS wouldn’t get installed *at all.* Likewise, you may rate Windows as the ‘worst’ desktop OS (and I’m tempted to agree); others might rate it as the *only* desktop OS because they don’t have the technical ability to use anything else or keep it running — for them, Windows is currently the only choice.
Whenever an operating system is new and popular, there is always an overabundance of duplicated applications. It is part of the evolution of the platform.
So many of the people who complain about Linux claim to be longtime computer users yet seem to have no perspective or memory of computing history. Linux is a young platform. There is a gold rush of apps. Not all of them are great, polished apps. But not everything that looks like gold is pure gold either.
The author’s article and most of the anti open-source / anti-Linux comments are nothing new. Legions of stupid people, mostly Americans, uttering assertions that they have no knowledge to back up.
I like diversity in OS software, but I ascribe to the UNIX ideal (or
my reading of it) of small modular application’s that can plug into
others and have reasonably consistent methods of calling.
Anyway the thing that brought me to Linux in the first place was the
choice it offered in settings applications etc. I don’t like bloat
either, which means I’m not a fan of most app’s as certain people
equate us-ability with the number of features (that I personally will
never use, hence the UNIX ideal).
I use as my window manager “ratpoison” (I hardly have to touch the
mouse, I like this sort of thing), I do however recognise that this is
not a perfect solution for everyone. Thus why diversity is a very
good thing. As it means that we don’t all get hammered by the default
Anyway, I’ve found the less features the more stable a program usually
is (it’s the less to go wrong point of view).
However I don’t like people even suggesting that if they don’t think
you’re project is popular enough it should be scrapped is totally out
of order. It’s their time that their giving and offering the results
for free and to boot letting you look at the source code (if thats
Rainer: Do you guys bitch that there are … too many types of cars?
In a way, I happen to do that at times. I have a damn hard time finding a replacement for my current truck as none of the gazillion new models satisfies my needs.
Miles: The point of KDE offering their bundled editors is because some people like them better than the other options.
Yeah, but writing completely new apps instead of leveraging the years of work, experience and code in existing editors sounds like a major case of NIH syndrome to me.
Kedit is the notepad, KWrite is the word processor, Kate is the developer tool.
I give you KWrite, but notepad vs. developer tool? A good editor is both (and also I wasn’t too impressed by the ‘developer’ aspect of Kate).
Free software have an evolutionary development style, like animals. Only the strongest survive. It is the principal reason to the rapid development of linux.
See Internet Explorer for example. After M$ destroyed Netscape, IE development starved.
Freesoftware is like genethic algorithms…
I agree, however, with the need for more commercial software, like Photoshop, Coreldraw, Mechanical CADs, etc.
We have a saying here at work that all customers are right, provided they pay for what they buy.
Along that line, I guess the author can be treated like a customer who is asking for features that aren’t there yet. Normally, the easiest way of handling such customer is telling the truth. The truth that such a service or product doesn’t exist yet. Accompanied with a nicely phrased words saying they will be put on the waiting list. Just get the name and write it on the list.
There are probably thousands of users out there who belong to the waiting list, though. They want to use it but can’t. That’s just too sad for them. I share their sufferings in not being able to use such a fine tool.
But on the other hand, whiny customers like this one tend to create pressure on developers, which is good.
At the end of the day, when the customer has finally settled down and accepted the fact that they will never be able to switch, reality comes in and bites him to despair and envy.
Flooding the market with too many apps are both good and bad. Good in the sense that there will be choices around. Bad in the sense of having to know what “caveat emptor” means.
Trying to control what to do under this scenario is like preaching in the wilderness. They see you but can’t hear you.
I think that says it all. There is just no one-size-fits all in software. Even on Windows there is lots of choice of different types of application, but certain ones were adopted into certain niches by businesses (not by users) and became de facto standards. This should happen naturally with Linux, not be forced upon it by some ‘organisation’ – be it KDE, GNOME or whoever.
“I can’t believe what I’m reading! Do you guys bitch that there are…Too many types of cars?”
I agree that having a lot of choices is good, but when I choose a car I expect the brake pedal to be on the left and the accelerator on the right. I should also be able to change from driving on asphalt to concrete pavement without overhauling the car or having to choose a different one.
To survive, Linux needs the standardization that allows a developer to create a GUI app once, and having it run on almost any desktop. Too often when I search for a Linux application to download I have to choose between distributions and sometimes between different versions of a distribution or desktop.
I want a lot choices, just not the kind that put severe limits on other choices.
I agree with Ken:
“Personally I think a lot of users of open source/GPL software need to get their attitude sorted out.”
There seems to be a lot of moaning about what people are
getting for free when you offer something. If people want new
features they should either help to program them and improve the software themselves, *OR* put up some money to get it done by someone who can.
There are a lot of people out there who use free (as in freedom) software and never drop any coin on the people who make it. If OSS and free software people keep thinking ‘free’ as in ‘freeloading’ then the success that we’ve seen to date is going to dry up, and fast.
You want it fixed? Start paying for it.
the so-called “instability” of opensource software. I am running FreeBSD, rock solid, with GNOME 2.2.1, again, rock solid, with Mozilla 1.3, Balsa 2.0.10, Pan2 and Quanta 3.1 (with qt 3.1.1), and again, rock solid. Neither of these applications have crashed.
Every time I hear this moan about instability it reminds me of a young lad on COLA (Comp.Os.Linux.Advocacy) who complained that KDE 3 kept crashing and he was getting quite anti-Linux, I in return asked him to reboot and run memtest86 (which is a selection in the SuSE Bootmenu), after 2-3 days on silence, I contacted him via email wondering what happened, he admitted that one of his memory modules was stuffed. He replaced it with a nice brand new one and found that KDE IS actually stable and that the unexplained crashes had more to do with his hardware than software.
Sounds abit like Eugenia Loli-Queru and her so-called “Linux instability” which later on had NOTHING TO DO WITH LINUX but everything to do witht he crappy VIA chipset she was using which has a netorious reputation for instability and poor reliability.
So GTK+ should be our only toolkit? But the author wants to keep KDE; he wants to keep OpenOffice (which uses its own widget library).
Sadly, we cannot keep just the best tool in each category, and expect our choices to be consistent. The best we can do is to ensure that useful, mature, software like Gtk+ and Qt is always installed in a consistent way from one machine to the next.
What’s the problem with the Qt licence? The licence is GPL, which is “free enough”, or commercial: if you want to exploit Qt’s ease of use to write commercial applications, buy a commercial licence. No serious commercial developer has a problem with that. For a persuasive analyisis of this point, read:
The link is broken, but the article is in the Google cache at:
I love Linux, I’ve been using it since 1996. I love it for several reasons; the modular structure, the rock solid stability, the dedicated community behind it and the enormous evolution it is still going through.
A couple of years ago, when we didn’t have 1323 distros to choose from the future of Linux to me looked more “defined”. Today, I’m not sure where Linux is going.
Do I think Linux should try to replace Windows on the desktop ? Yes, because Micro$oft’s monopoly on the desktop or any other monopoly for that matter is unacceptable to me.
I love what Apple did, they have succesfully built an attractive, easy to use OS on top of Unix. That’s putting 1+1 together.
If the hardware they sell wouldn’t be so expensive (especially in Europe and Japan) and outdated they surely would have a broader user audience.
Can Linux be the OS X for the i386 platform ? I think it can. Do we have to have less OSS for that ? No, the huge collection of software available to the Linux platform is a great advantage Linux has over other platforms.
IMHO, the only way Linux is going to be more competitive on the desktop is if companies like RedHat, SuSE, Mandrake, Ximian etc. decide to do so. And I hope they will. Everybody , from system administrator to computer newbie, should be able to enjoy Linux as if it was another Windows. How strange and controversial this may sound.