Slashdot Mirror
Linux Needs Critics
An anonymous reader writes "Keir Thomas berates the fact that the world of Linux almost entirely lacks critics. In fact, he says, Linux people tend to see genuine critical evaluation as a bad thing. FTA: 'The problem with this anti-criticism approach is that it's damning Linux to an eternity of navel gazing. Nothing can ever get any better. The best hope we have are the instances where a few bright sparks, with their heads screwed on the right way, get together and make something cool (as happened with, say, Firefox back in the day). But that's rare and can't be relied upon.'"
Linux has plenty of critics. Developers are critical of their own code. Just look through the lkml or read the code, there are plenty of places where there is constructive criticism about how something is done.
-- Erich
Slashdot reader since 1997
The critics in open source world are the young ones that get a big head one day and call your project stupid because it uses language X instead of their favorite language. Then they fork the code, write their own crappy software, get some distribution to decide to use it and then the original project gets dumped one day.
I have to admit, the fanbois are making me homicidal.
I LOVE Linux. I love plain old Unix. I love the command line, and the cryptic commands, and man pages, and lynx and apt/yum. I like X windows and MC. I love building from source. The whole environment is clean, somehow. It's got a sort of serenity for me that I don't see very often in my job.
And yet...It's just a tool. It's a good tool. It's my favorite tool. But it's just a tool. There is room for improvement, and, like any tool, there are places where it's not useful.
The thing that drives me nuts is the pure unthinking zealotry. I got started on old proprietary unix, and while linux has more zest and more wild features, there are things that were worthwhile in the old systems. But if you say that, then you get slapped down as a heretic.
Everything benefits from criticism, so in that sense, he's right, but really Linux has plenty of critics. Install linux for someone who is used to something else, and you'll get plenty of criticisms. What I think Linux needs is the same thing I think Mac needs and Windows needs: the people on the inside need to start listening to people who aren't already sold on their product. We have just as many fanbois as the Mac and Windows people, and we've got some of that persecution complex that makes the fanbois extra loathesome.
Just calm down, take a breath, go use something different for a while. Get some perspective. The real zealots make it harder for me to sell *nix solutions to the phbs because they're coming to expect a bias.
ad logicam Claiming a proposition is false because it was presented as the conclusion of a fallacious argument.
Linux Needs Critics
Oh how true that is. I myself love to criticize things (I'm reading Slashdot, after all). But why don't I constructively criticize Linux?
I think a lot of has to do with what every argument or analysis starts with: base assumptions. So let's start with comparing Linux to the leading commercial operating systems and the most important thing to consumers--price. And the guy mentions this in his blog. But we can't get to questions like "Is feature X really worth Y dollars to me?" Because Linux does not cost money to install. It's like dividing by zero. It makes criticism of a missing component difficult because it doesn't cost me anything! How can I criticize it?! You will see people like Steve Ballmer have to dig and dig into imaginary costs of retraining, supporting and maintaining Linux to give it a "hidden cost" so that Windows can even begin to contend with Linux in price (you'll notice these concerns were suspiciously left out of advertisements when discussing the switch from XP to Vista).
Another important aspect of operating systems (at least to me) is security. And, being a pedantic ass, I cannot even comment on the security of the Microsoft operating system because I have no idea what they are doing. I can get the Linux source code pretty quickly if I felt the need to understand why it is that the userspace/kernelspace concept has failed (although, I have never done this, the option is there). So, again, we enter this point where I can't even get to criticizing Linux for susceptibility to a botnet or trojan because it doesn't practice security through obfuscation like leading operating systems.
On top of this, as a Linux user (and as evidenced above) my priorities and performance parameters are all out of whack and completely divorced from the mainstream (or so my perception goes). If they weren't, I would be using Windows primarily at home.
So I think that unless more free open source operating systems arise to compete with Linux, criticism will remain low. And you've got the cult barrier to break down where people have lived with the burden of paying out their ass for software so how can you criticize something after suffering for so long under the blah blah blah religious spiel blah blah blah.
My work here is dung.
There are critics out there for Linux. But how many of them offer quality criticism, instead of complaining? And there are developers out there who are willing to listen to quality criticism, but how many of the few critics out there comment on any specific piece of software that goes into a complete Linux system? Both sides could do more- critics could write white papers with suggested corrections. Developers could take the "Linux sux" as an indication that they need a top down audit of their project. But both of those solutions are asking too much of either side. There should be better practices on both sides. And of course, this all ignores the good work on both sides that are being done, where there is constructive criticism and receptive developers. You can always use more of both, so there is never enough of either.
http://bgcommonsense.blogspot.com
There is plenty of disagreement about every aspect of Linux. The kernel, the GUI, the apps, everything. And if you examined Linux, or a distro you aren't familiar with, you would probably find something you didn't like about it and you are quite free to criticise it.
Linux has no deficiency of people who criticise and no deficiency of people who listen and act on it.
Windows has no deficiency of people who criticise and a seemingly complete deficiency of people who listen and act on it.
MacOSX has worse than deficiency of people who criticise as they have people who actively criticise the critics and even attempt to silence them. MacOSX has a deficiency of people who listen and act on it. ...just to put it out the way I see it.
More developers that can handle being criticized.
If you can't see the value in jet powered ants you should turn in your nerd card. - Dunbal (464142)
it's free, why are you complaining?
This should be the OSS equivalent of Godwins Law. As soon as you trot out the "it's free, why are you complaining?" argument - you lose!
Has this dude visited any community involving Linux users... ever?
The standard general Linux criticisms:
1. Driver support. Usually from a lack of manufacturer support.
2. No central focus on meeting business needs (tech support). This complaint is changing with such a large amount of development occurring with programmers employed by business communities for open source development.
3. Have to give up favorite Windows programs (apps & games). This improves over time, but yes, it is a different environment, again with a different historical focus.
Plus lots more, like programmer IDEs, look & feel issues, etc., etc. Criticisms, constructive or otherwise are everywhere Linux is discussed, including countless published sources.
I've certainly encountered folks with an unconstructive beef against Linux who make complaints that it gets unfair praise for being mediocre, merely catching up to Microsoft. With those folks, yes, complaints are sometimes muted because the target of their ire is usually changing so often that their rants are stale before they speak them - so they can become embarrassed by being contradicted in the heat of a discussion too often. But even then, such complaints are still extremely commonplace in both print and online.
I really don't understand where this dude is coming from.
Ryan Fenton
The article is quite right; there is too much groupthink and myopia. The Linux Hater's blog is a must-read as an antidote to all that, and he or she has some useful points to make. The articles on Linux Weekly News still have a Linux-centric viewpoint, naturally, but usually aren't afraid to point out shortcomings (especially when quoting the latest Linus flaming on the kernel list).
-- Ed Avis ed@membled.com
I critisize Linux all the time, and I try and criticize it for very good reasons.
Let me give you an example.
The most wide spread Groupware Suite that is freely availible in the Linux world to challenge Exchange that I can see is eGroupware. eGroupware is an excellent suite in my opinion. Now. Linux has three dedicated Groupware Clients. Kontact, which is part of KDE-PIM. Evolution, which is part of Gnome, and Thunderbird.
Now. To do anything with eGroupware other than E-mail you need XML-RPC.
Kontact has XML-RPC Support, but it has a nasty bug where if it becomes De-Synchronized, it will respawn the same events on the Calendar over and over.
Evolution has no XML-RPC support. You can rig up GroupDAV
Mozilla Sunbird has no XML-RPC Support.
What does it say about Linux's productivity-ware when two of the three Groupware clients produced by Linux developers cannot communicate with its intended native Groupware servers?
Linux just doesn't love critics who won't roll up their sleeves and fix things.
Ideas are cheap.
Game... blouses.
There are some things that simply won't be corrected out of love of the subject. Heck, some of those things won't even be found (and reported).
What a successful software company does that can ONLY be done by PAYING people is persuade people to analyze & create requirements, code, test, and fix ALL of the system. Yeah, the OSS community does most of it pretty well, but they simply won't do it as pervasively or as rigorously as needed unless motivated to, which usually comes in the form of being PAID (to wit: be able to eat).
This is why corporations pay managers: people who are responsible for figuring out what all actually needs to be done, paying other people to get it done, and confirming that it actually has been done. Managers are paid critics who are on the hook for following thru on their criticism. Much of the success of Linux comes precisely from companies like Ubuntu, Red Hat, IBM, Mozilla, and others who actually do pay people to get those annoying unpopular little things right.
In contrast, we end up with the situation that keeps driving me away from Linux: stuff that I need to work just doesn't, and nobody has sufficient motivation to announce the problem, and nobody has sufficient motivation to fix the problem even if known. So instead, I go to someplace like Apple & friends, who - being PAID - are fanatical about making every little thing right (ok, they make mistakes too, but are more motivated to find & fix the little things).
Hence the ultimate failing of "free software": like it or not, money motivates people to do necessary but unpopular jobs, including finding & fixing software flaws.
Can we get a "-1 Wrong" moderation option?
The problem is that most critics are dismissed summarily as people who don't understand *nix properly. This is where the "snob" syndrome comes into play.
I took a RHEL class a while back, and the teacher (who was FROM Red Hat) spent ample amounts of time talking about BSOD and other stupid things, most of which I looked at him and just wondered, why didn't he check XYZ, because that would have solved his issue. No, instead, he went off on a rant about how useless Windows was.
It's no wonder that Linux fans can't take criticism, when a lot of their nature is built up to bash other OSes. Mac gets a pass because it has a bash terminal, but in reality no real security (see: Pwn2Own), and Windows is the giant on the hill that gets bashed regularly. Occasionally I'll see a Slashdotter here mention that IIS is a pretty decent product or something slightly complimentary, only to be modded down as flamebait.
And Linux has critics? Please. The critics TFA is talking about is more about code, it's about 'the whole package'. And while *nix on the server side is something I'd readily deploy for database backends, for web farms, for much more -- there's no way in hell I'd deploy it on an enterprise desktop because it's simply not ready to be used or administered.
But I'll expect my bad mod rating shortly, arguing that *nix fans have lots of critics, and that I'm simply one more of the "uneducated" ones.
The price is always right if someone else is paying.
Yes, I totally agree. Linux has many, many critics, from users, "prousers / powerusers", developers, and other corporations e.g. M$ (although agreed, these are more often just attacks.)
Where I thought his article was very insightful, however, was the typical response "Linux is free..." where I agree with his analysis. This, however, reflects on us, the Linux community. I cringe when I see somebody say, "It is free, what do you expect?" or "you have no right to complain." Users have every right to (nonabusively and in a civil fashion) criticize software.
If Debian (stable) suddenly stopped working, my organization would lose thousands of person-hours of lost productivity. In many ways, doing somebody a half-favor is often worse than doing them nothing at all:
Imagine if I volunteered to repair your garage, but then did a half-assed job and quit halfway through. It would cost you MORE in the end to clean up and switch to another provider. Would it be then ok to say "I did that for FREE, how can you complain?"
Obviously this is a continuum, and many of the criticisms are unfounded or just whining. But, as a whole, if we want Linux to continue to succeed we, as developers and users alike, should listen and respond constructively ourselves to any (also constructive) criticism that is provided by the community.
Slashdotter, ID #101. UIDs are in binary, right?
You are ignoring the fact that 100% of the users benefit from improvements made by the few that have the time and resources to get involved with the code.
Being serious for a moment, a large part of the problem with Linux - at least in the "getting more people to adopt it" sense (chicken and egg) - is the fact that Linux developers/proponents tend to be unable to understand that while something is "not their fault", it is still their problem.
For example: I have a DVR box that I put together a few years back. It uses an ATi All-In-Wonder capture board. A "clean rebuild" of the box, including all the recording software and OS (it uses WinXP), takes ~3 hours.
I am consistently told by Linux-using friends that I should "absolutely be using Linux instead", that all Linux software is inherently superior, etc. Yet when trying to install any of the various Linux/MythTV flavors, I've consistently found all sorts of problems. The ATi Remote Wonder doesn't work well for most of them. The recording software either doesn't work at all, or is "spotty at best." Video playback quality is lower.
When I've asked about this on Linux help boards, the response is always the same (and I'm sure I'll get a bunch of raving loons attacking me here for saying so as well): "well it's your fault for having an ATi board you should go spend $$$$$$ on a hauppauge and a nvidia board and buy this and buy that because that's what my box uses and anyways the ati drivers suck because ati sucks."
Now, I recognize that ATi hasn't been as "forthcoming" with driver source / documentation as some other companies. This is where the Linux folks can say it's "not my fault." The reality, though, is that it is a barrier to entry, and therefore it is their problem.
The other problem is that the Linux world lacks consistency. The same command structure, driver package, installation routine often has to be "tweaked" to work - if it works at all - on any given random distribution or even between versions of the same distribution.
Now of course, merely by saying something like this in the open, it's a good chance I'll be branded a Linux heretic. Maybe even a slew of nasty downmods will come my way. After all, criticisms like these are part of the whole "not in front of the goyim" mentality of Linux users whenever there are non-Linux users about.
Dude. Just get the nVidia board.
Kidding. Your point is valid -- use the best tool for the job. If you've got an ATI card that works well with Windows but not with Linux, and both OSes provide most of the features you want, then use Windows (assuming the cost of Windows is below the cost of that nVidia card).
Linux remains a server OS. It's coming around very slowly to the desktop and I've no doubt it will get there -- it's doing quite well on the netbooks where people don't want to install everything in their closet: they just want a functional, email-checking, web-surfing laptop and for that almost any OS will do.
Except MacOS X. That only runs on the shiniest of hardware.
I think what he was saying is that he uses Windows XP cos it works, and people are constantly telling him to switch to linux even though it's obviously not viable with his setup. Open-source users are going to have to learn to adopt a live and let live mentality, linux is not the holy grail, it works brilliant in some cases, and not at all in others, same with any other bit of software.
You missed the point. You make a statement "its not their problem" but then the argument you use is about assigning blame. Like the guy said, its *not* the developer's fault but it *IS* their problem. The problem is that a lot of people have all this hardware that won't work with Linux and won't just spend money that they probably don't have just so they can use Linux. Why switch to something that only works with x% of the hardware out there when you can use something that works with (x+y)% of the hardware, where "y" is usually a significantly sized number. So its their problem insofar as they have to figure out a way to pass this hurdle, otherwise you'll never reach critical mass in terms of people adopting Linux.
Well don't get mad at individuals when they say, I'm not going to use Linux BECAUSE IT DOESN'T WORK. People aren't going to switch to a free OS when it requires them to replace their hardware that costs a significant amount of money. Linux may be free, but chances are, any new users already made the investment in hardware beforehand. Don't get upset when people complain that Linux has issues when, in fact, it DOES. If hardware doesn't work with Linux, it's not Linux's fault, but it IS a problem for them. Just because something isn't someone's fault doesn't mean they shouldn't try to solve it, especially when its a huge obstacle to their success. Thats elitism and lazy all bunched together. You're expecting people who are used to OSes where all this hardware works and then you're trying to explain to them that this superior OS is so great but oh yea, you're gonna have so many headaches if you use it with that machine.
Don't you realize that the problems introduced by incompatible hardware like this makes Linux have MORE PROBLEMS for that particular user THAN WINDOWS? Why would someone switch to an OS that has more problems than their current OS?
Your use of "fault" suggests you don't understand what Moryath said. And your example seems to support Moryath's conclusion very well.
If a manufacturer makes a MOBO that doesn't support a type of HDD that PC World is selling, it isn't the manufacturer's fault, but it is their problem. At least, it is their problem if they want more people to buy their MOBOs. If you support a fraction of the hardware out there, you get a fraction of the customers. That fraction consists of people that bought pieces specifically to work with your product and people that just happen to have pieces that work with your product.
Its not about fault. Its about figuring out a goal. If your goal is to get more desktop users to run Linux on their DVR, then available hardware thats incompatible with your software becomes your problem. It becomes a challenge you have to overcome to reach your goal.
As far as linux spreading on the desktop, I couldn't guess at how many developers actually have a goal of making an OS for the average desktop user. I've spoken with a few that set goals of making the platform stable, or fast, or capable of doing a specific task they are interested in. All of that is fine and, I believe, necessary. But if there are people that actually want linux to gain market share on the desktop, they will have to face the problems of achieving that goal, even if its not their fault the problems exist.
1 (short ton / firkin) = 89.1432354 slugs / keg
But what you are saying is exactly what his point is, if I have it right.
You don't care if Linux works right for someone else. However, if Linux proponents want people to adopt their OS of choice, they have to deal with the fact that they have to do more than provide their part of the bargain. They have to go the extra mile to make it work for *the other person*.
Is it fair? No. But fairness doesn't matter in the slightest. It's not really fair that Microsoft has to support legacy junk too, but they do in their own crappy way.
People want their stupid video card or whatever to simply work. It doesn't have to work perfectly, but it does need to actually be installable and functional with no more than a reasonable effort. We need to face the fact that Microsoft, for all their piles of fail, has set the bar for compatibility that Linux needs to follow if the goal is to increase share of general purpose PCs out there, especially desktop workstations.
Well being a programmer myself and using Linux as my main desktop, this is "also" my problem. I'm an end-user as well.
The last irritating things I have found: I just installed opensuse 11, running KDE 4. Well I love it and I don't understand all the fuss around it. Anyway I had to zip several files. I had to send them by email to a client....So simple is it? I openned this new dolhpin. I selected them, right click,actions and then the list appeared:
What kind of paranoid menu is this ? I've got 4 encryption features out of six. I simply need a very simple "compress..." or "Archive" like KDE 3.X to make a simple "zip file" or a tar.gz, I will attach it to my email and that's it. I couldn't find it, so I openned terminal and I typed the proper command line...
See how stupid it is?
Sometimes you feel like what Linux lacks the most is simply "common sense". Sure I will customise that annoying/stupid action submenu when I will have the required time to document myself...But It is truly annoying, even if you are a developer. Some guy out there was so proud of his encryption scheme that he puts 4 commands.
Not all problems are driver related, user-friendliness is also a "big" problem.
it IS ready for the desktop. Either in a managed environment with a guru at the top, for those who know what they're doing, and for locked down spoon fed distros.
I realise this is beside the point, but that's the exact same situation with Windows. In any given corporate environment, for example, there are a handful of people who know what they're doing, and then the slavering masses who know nothing and are constantly screwing things up until someone decides to lock down the desktops.
Considering how quickly the average user can completely obliterate a Windows box without even trying, it amazes me that anyone can say Windows is "desktop-ready" while something like Ubuntu isn't.
Half the problem in this particular area isn't the choice of OS -- it's that users aren't ready for the desktop.
mirrorshades radio -- darkwave, industrial, futurepop, ebm.
Wow. Thank you for proving my point so well with your vitriolic, hate-spewing post.
The stated goal - at least from a large portion of the linux community - is to see as many people using Linux as possible. Even as "volunteers" and "freeloaders", it is to Linux's advantage to try to reach this goal. The more market share that Linux has, the more support that they should see from software and hardware developers in the form of things like more open sourcing of drivers, better relationships with vendors (perhaps convincing companies to ship computers with Linux rather than Windows for example), and in terms of companies being willing to produce Linux-centric versions of their game (or in the case of MMO's, their game client). In other words, if Linux gets to a critical mass, then problems like "hey, Video Board X doesn't work in Distro Y Funkey Monkey" will be much easier to address.
Functionally, however, this is a "chicken and the egg" issue. At its current market share, and given the way many so-called "linux advocates" behave towards companies that don't do exactly what they want at the drop of a hat, I can understand why there are companies out there not handing out their design documents and driver source code. Linux zealots have burned a lot of bridges, made a lot of enemies, and generally made asses of themselves at just the wrong time during many points in Linux's history.
Again: whether this is what you, personally, have done may have bearing on whether it is "your fault." The fact remains that whose "fault" it is, in terms of support and in terms of making Linux friendly enough to get that coveted "critical mass" of market share where the various companies all start having to play along or else risk losing sales, is completely irrelevant.
Linux, IF you want it to reach that "critical mass" market share point, needs to reach a certain bar of compatibility. This doesn't mean that it needs to be compatible with everything known to mankind, but it DOES mean that you need to support, say, the major product lines of the "big three" video board market share holders (NVidia, ATi, Intel), the "big three" styles of audio card (built-in AC'97, Realtek, Creative), and so on. And these need to work without users having to go hunt down some obscure repository, post to 5 messageboard forums, and then follow instructions written like "well obviously you have to bleep fraggle this and sudo command toggle bashznz that and then it'll work, what kind of a lame n00b are you if you don't understand that."
I'd happily volunteer my services.
I'm blunt, hypercritical and am allergic to excuses, stupidity, bullshit, and responsibility cop-outs.
I'll even give a couple for free:
If I have to tinker with it to make it work, it's crap and needs improvement.
If I have to edit a text file instead of using a configuration GUI, it's crap and needs improvement.
And yes, I use Linux at work, but Windows XP at home because game support is crap (see #1 above).
Question everything
Like the guy said, its *not* the developer's fault but it *IS* their problem.
No it's not. That kind of absurd expectation (that every piece of hardware in the world will work flawlessly, despite the apathy and even animosity of manufacturers) is put on no one else, because it's ridiculous.
Look, it's like this. You buy a doodad, let's say a TV card, from LittleGuyPCI Inc. Try to stick it in your Windows box, it doesn't work. "Goddamn LittleGuy!" Try to stick it in your Mac, it doesn't work. "Goddamn LittleGuy!" Try to stick it in your Linux box, it doesn't work. "Goddamn Linux!"
Do you see a disconnect there anywhere?
Hey, I finally got my first freak! Took you long enough!
Look, it's like this. You buy a doodad, let's say a TV card, from LittleGuyPCI Inc. Try to stick it in your Windows box, it doesn't work. "Goddamn LittleGuy!" Try to stick it in your Mac, it doesn't work. "Goddamn LittleGuy!" Try to stick it in your Linux box, it doesn't work. "Goddamn Linux!"
Do you see a disconnect there anywhere?
Yes, absolutely. When I stick it into the Windows box, it actually works, so the rest is all meaningless conjecture. ~
Seriously though, I wouldn't say that the problem is "developer's problem", really - this implies that developers care about adoption, which they often do not. But it is a problem of those who wish Linux to be adopted. It doesn't matter whose fault it really is - but if it hinders adoption rate, and you want that to change, then it is your problem; it's really that simple.