The Penguins Are Angry
Ok, I am writing this blog to help explain my previous blog on Linux. If you missed that piece you are welcome to catch up here or here.
Now, the nice folks over at Linux Today were nice enough to post my Linux Blog. And of course as soon as they did that I received some “enthusiastic feedback”.
Some of you might not believe this but I really appreciated the comments. It is nice to know what other people are thinking and how they interpret my writings. So, to that end I am going to try and clarify my argument and respond to some of the complaints.
But before we get to all of that let me make a few things clear:
1. I don’t hate Linux
2. I am not a troll. In fact, I had no idea Linux Today would even pick up my article and since I posted that blog on a Mac site I really wasn’t looking to get in a fight with anybody.
3. The “Euro FUD” thing was taken a little out of hand I think. It was a joke ok? A joke. Ha ha, joke. With me? I don’t hate all of Europe just France. See, that was another joke. Ha ha. Anyway.
4. I would love to have an intelligent conversation with anyone who considers themselves a Linux power user. I have a few questions and would much rather have them answered in an thoughtful way than just have some one degenerate into name calling.
So, to that end lets get on with the response.
The point of my article was two fold: First, to explain where I thought Linux need to go and second, how I thought they should get there.
This is a key point that most of you missed. I am merely saying that I think if Linux wants to reach X then they will have to do Y.
With me? Now, you may disagree with X but at least consider if Y would get you to X, ok?
Think about it this way. I am working from the following assumptions:
1. Linux, as a whole, wants to become a force on the average consumer (not corporate and not server) desktop.
2. If Linux wants to do this in a reasonable time then they will have to change the way they operate.
You can disagree all you like with my position but at least argue the point I am trying to make. Every single post save one on Linux Today missed the boat entirely. I got some of the dumbest responses I have ever heard. Sadly only a few people actually tried to respond in some sort of coherent manner. Everyone else just called me names and said my article was stupid.
Ok, you think its stupid. Fine. Great. Wonderful. Now, tell me how you would reach the goal I mentioned without taking the path I charted. Don’t tell me you don’t want to reach that particular goal (because thats not what we are talking about) but instead try to evaluate my plan in that sense.
I feel like I am trying to talk a friend into buying a car. I tell him that he needs a driver’s license, driving lessons, insurance and all the rest. His response is to call me stupid and say he would rather ride a bike. Ok, you want to ride a bike. That is great. Does that make my statement about getting a license any less true? Just because you disagree try at least to articulate why it is you disagree. Do you find the path wrong or the destination?
To recap, I am trying to say that the way Linux, as a whole, operates needs to change if they ever hope to compete with Apple and Microsoft as a viable consumer desktop OS. That is all. I am saying they need to change if that goal is the one they want to achieve.
So, lets go with that. That is what they are aiming for. So, if that is the case then I think they need to do the following:
1. Converge distros
2. Charge for product
3. Standardize
That, in a nutshell, is my argument.
Now, here is what I want. I want to debate those 3 points with an intelligent Linux user, not one who is going to call me names, compare my article to Stalin and then run away and pout.
Conversely, if you feel that my entire premise is wrong please feel free to explain why you think this is so. By that I mean if you think Linux shouldn’t strive to be a desktop OS argue that point. If you feel that they will one day become a presence on the desktop by doing what they are currently doing, fine, argue that. If you agree with my goal but disagree with my methods then argue that. I don’t care what point you take all I ask is that you make a logical argument.
Almost every comment I received could be boiled down to this: “I am right because I said so”
Please people, lets try and put a little more thought into it than that.
So, who is up for a debate?
del.icio.us
Digg
Facebook
Slashdot
StumbleUpon
Comments
A company I work for has a PC box and they would like it to serve some rudimentary server functions, mainly NTFS file-sharing and ftp for some clients to access certain files. They didn’t really want to pay for Windows 2003 Server or anything. Being well-versed in Mac OS X, and having had experience with Solaris, I figured hell, why not download a Linux distro and throw it on there.
After a day or so of research, I settled on Debian, as pretty much everything I read swore up and down it was by far the best, easiest, and most stable distro around. Ok, great. So I go through all this rigamarole to use some stupid program to download the ISOs, when in fact just downloading them over http (which they pretty much say is the worst possible way to do it) it actually the easiest and definitely the fastest way to get them.
Then, after two weeks of installing, reading man pages, reading websites and forums, reinstalling, repartitioning, reinstalling again, apt-getting, apropos, etc ad infinitum, I cannot get either gnome or KDE to work.
Now, I know I’ve got all the right services running on the machine, and I know I can configure it from the command line, but wtf? I could do that 12 yrs ago. Maybe I am spoiled but I don’t want to spend all day in vi editing config files. And you pretty much absolutely need a book handy to be able to understand all the subtle differences in where everything is located. Man pages are not always well written, and I could find tons of what appears to be documentation in XML format, but as yet cannot seem to find a way to read any of it.
For the record I am a 29 yr old computer engineering major. As far as I’m concerned anyone who says Linux is ready for primetime is either full of shit or has been drinking too heavily lately. I know, you get what you pay for. I wish I could pay $129 and put Tiger on this thing!
I understand your points but I think that the whole problem stands in what people see in a desktop environment.
Mac OS X is more like an ecological environment and provides everything to sustain it. The naming scheme used to identify parts of the system, using either terms related to nature (Quartz, Core, Safari, Carbon, Cocoa, Darwin) or productivity (Spotlight, Dashboard, Exposé, Automator, QuickTime), is done on purpose to convey the idea of a place where applications collaborate.
Now try to find such scheme for Windows or Linux… Gnome, KDE, Kernel, Distro, X11, Aero, OLE, COM, VBA… Very enticing, doesn’t it? The desktop is just a place to run applications and then it is up to the users to glue things. If they can’t, they call an integrator or use an integrated package.
Both environments are different beyond any understanding; one side encouraging user experience, the other side targeting experiencing users.
It happens that there are more people enjoying the experience of gaining control on their environment than people enjoying the environment as a playground for their experiments. Tons of money was already poured into the first environment while the second one is just a footnote in the long history of paradigms with very low market/mind share.
The lack of convergence and standards amongst the distros is just a side effect of this will to control the environment. If Linux was a real alternative, we would have seen the next Smalltalk, not a xeroxed operating system.
MY point is I couldn’t parlay 22 years of computer experience (a lot of which centers around using some similar OSs) into a FUNCTIONING desktop. I’m not talking about good, better, best, or any of that. I’m talking FUNCTIONING. I am a total computer geek, in almost every possible sense of the phrase. I am the person everyone I know calls when they have some kind of computer problem, no matter how obscure. But I couldn’t get a running desktop. I also think it’s fair to say a “desktop” in this sense is beyond a CLI and even beyond pure X11, we’re talking OS X, Windows, and what Gnome and KDE purport to offer.
If this is the case, then you are essentially proving what the Author and I are trying to say. Personally I feel Linux is targeting experience LINUX users. As long as that is the case, Linux is NOT moving beyond a niche hacker playtoy outside of the server space. Sure, I can’t change the color of every OS X UI element at will (though I think there are certainly ways to do so if one so desired), I can use it day in and day out to get real work done. And there are tons of UI hooks into the underlying BSD layer that I rarely need to rely on a Terminal window just to get things working the way I want them too.
Actually, after reading your post again, I can’t really decide if you’re responding to the author or me, so maybe the above post is moot. I guess that’s what I get for reading without my glasses. <shrug>
Yes, I was responding to the author.
Nevertheless, I agree with you that the amount of knowledge required to use a single piece of technology like Linux or Java is becoming ridicilous.
You are wrong because LINUX IS NOT A COMPANY. Linux in itself is not a PRODUCT. Who do you buy Linux from…? When your Linux breaks who do you complain to…? Companies can take Linux and turn it into a PRODUCT, much the same way APPLE has taken Unix and FreeBSD into a PRDUCT.
We seem to forget how lost Apple was without its fearless leader Jobs. If a company can flounder for that long without releasing a competitive PRODUCT, then how do you expect an open source PROJECT to when there is no financial incentive, or a real leader…?
If people want to develop for KDE or develop for Gnome, that is their perogative. But a company that wants to make a PRODUCT will choose one and support it.
Sensationalist bloggers and journalists continue to create this fictitious battle between Apple, Microsoft, and Linux. They are comparing apples to apples to oranges (no pun intended
. The parallels they draw are fundamentally flawed. And to do this, it clearly proves that they don’t understand the open source business model. They don’t even truly understand Apple’s business model.
Linux is merely an open source PROJECT just like many others; firefox, gaim, wordpress. Linux is part of a PRODUCT on the server side. Companies like Red Hat sells and supports a Linux PRODUCT. Linux appears in set top boxes, mp3 players, robots, cell phones, and hundreds of other devices. In these cases Linux is part of a PRODUCT.
Until companies see the desktop as an opportunity to take Linux, improve it, become a competitor, and support that PRODUCT then Linux progress will remain in its current state, that is developers making improvements that are important to them. Companies are beginning to do create Linux desktop PRODUCTS. Redhat, Suse, Linspire, Xandros, Mandriva (or whatever they are calling themselves these days) are companies with Linux PRODUCT. They are building on top of the open source PROJECT. You should compare those companies to Apple and Microsoft. Don’t take the easy way out and generally compare “Linux.”
I still don’t think these companies’ PRODUCTS are on par with Apples OS or Windows yet. But I will say this… when a company does produce a competitive Linux PRODUCT, Linux will be an afterthought, much the same way Unix is with your beloved OSX.
Yes, and I use all of these projects (although I think it’s pretty easy to argue that Firefox at least wants to be a product) and you know what? They work. Even WordPress can be set up and running with minimal configuration, and I was able to do so without ever touching the command line. Just because something is a project and not a product (a fine line that doesn’t always exist if you ask me) doesn’t mean it can’t be good, easy to use, and successful. Your own examples prove that. When Linux as a desktop GUI OS is as easy to setup and use as Firefox, I’m sure the temperature in hell will drop a degree or two.
Unix is an afterthought in OS X? Could’ve fooled me, since it in the foundation the whole GUI sits on. It’s an afterthought only inasmuch as a user never has to touch it if she doesn’t want/need/know how to. One click on the Terminal in the Dock, however, and I’m shell scripting bash with the best of them.
I am glad you completely missed the point. My point was that Linux is not a company.
But to respond to the two lines you could dispute…
When was the last time you used a Linux Desktop env…which one was it?
The market that Apple is now going after is not you, the enthusiast… You only make up about 3-5% of the total computing market… The desktop is really about the consumer. The Mass market. This is what the Intel move was all about…do you think the avg user that buys a Mac mini because its “cute” cares that it OSX sits on Unix…? They care about how cool it looks and how it acts. That was my point.
Fair enough, your first sentence says:
However, you’re whole post really focuses on how Linux is not a product but instead a project. You use the words product and project with emphasis a total of 16 times. You say Linux is not a company once. And, no one is saying Linux is a company any more than anyone is saying Mac OS X or Windows is a company. However, Linux is an OS, so it’s fair to compare to Mac OS X and Windows. Especially since several companies and several projects aim to make Linux a viable desktop OS alternative.
As long as Linux comes in 15 different versions with two different desktop window managers/environments, it won’t succeed at that, just as the author and other posters have pointed out.
Uh, that was my point. “...a user never has to touch it if she doesn’t want/need/know how to.” That’s exactly what I said. However, Apple can and is going after the enthusiast as well, because we can and will get underneath the underlying GUI and directly access Unix. The fact is no average user could set up Linux and never touch the command line like one could with OS X.
This whole thing is still going in circles. The author posits, and I agree, that there is definitely a push within parts, if not all, the Linux community to make Linux a viable desktop GUI OS. For that to happen, all the various distros need to unify, standardize, and become easier to use and manage.
EOL
I think the previous poster hit the nail on the head. Linux is a PROJECT not a PRODUCT. But you need more than a project to build a product, you need a PLATFORM instead.
A platform is something that’s designed specifically to be built on. There is a clear line where the platform ends and the building begins. And the platform is designed to support the product instead of providing some arbitrary functionality without regard to platform as a whole. Look at Netbeans and Eclipse. These are platforms with well designed APIs for building products.
With Linux, there is no clear line where the project ends and the finished product begins. And, to make things more confusing, the project it self is fragmented. There is no unified voice of what the project should do or how it should be done.
Personally, I don’t think that Linus wants to dictate which GUI Linux should use, nor should he. But I think someone needs to lead the standardization of Linux as a platform.
Linux should be a modular platform designed to be a foundation from which Desktop and Servers products can be built on. And these products should build on the standards exposed by the platform instead of rolling their own.
Look at Mac OS. Darwin is a PLATFORM, which OS X is built on. There are clear boundaries as to where Darwin stops and Mac OS X starts. Darwin also provides Foundation classes that are the basis for development in Cocoa and, in many cases, shares the same config file format as the Mac OS X GUI and applications.
As posted earlier, with Mac OS X, I don’t have to touch a command line unless I want to, but It’s there if I want to go deeper. This represents standardization between the platform and GUI to build a complete product.
To clarify my post, I think it will take more than selecting a single desktop, such as KDE or GNOME to make a good Desktop version of Linux. You need to start with a solid platform.
Yes, a selecting a single Desktop, such as KDE or gnome would help in the short term, but there only do so much a GUI can do to cover up inconsistencies in Linux. The issue is more than skin deep.
The Linux community have to realise that what they’re doing is not working. Linux is not making the inroads it should in the desktop market. And that’s despite the media loving Linux. I’m sick of it getting so much space in PC mags (whilst OS X gets almost none) when it just isn’t ready for the mainstream consumer market. If it was, and with this media support, it would go double digit marketshare overnight.
Even Apple know when it’s time to make an about face for the sake of achieving goals. Hence the Intel switch.
The Linux community as a whole, need to unite behind a single streamlined distro for the desktop. Sure they can keep their array of geek and server distros.
It drivers me nuts that every six months there’s a new “must have distro”. Red Hat, Fedora, Linspire, Xandros, Lycoris, Mandriva (formerly Mandrake), Unbuntu, Debian, Suse etc etc. Imagine how Joe Consumer feels?! No wonder he’s not buying.
At the moment the collective might and intelligence of the Linux community is being pulled in so many different directions, that they have become ineffective at conquering the desktop and especially the consumer desktop.
It’s time for the Linux community to do something drastic - because Apple are massing the troops for a massive assault on the desktop which could leave Linux in its dust.
Dear James Stout,
I like what you have said, but feel that it may be, somewhat, misinformed; not too far off. It is, more or less, what I would expect to hear from a person who runs, largely, commercial software.
GNU Linux is free software made by folks who do not necessarily get paid for what they do (monetarily, that is). Linux has several, many, different distributions, and each has strong points, as well as weak points. Many use Debian Linux as a basis for their particular distribution and each accomplishes different goals even though some of these goals are not very different from each other. Not every automobile user likes Ford, or Toyota automobiles, for example. I suppose it is only natural in the free software community to expect to find different distributions in use for different folks.
Therefore, converging such distributions would be tantamount to Microscam welding its “engine’s” bonnet shut; worse, there would be only one distribution to consider.
Charging for the product will violate the GNU/GPL free ware license; this will not happen in the Linux community. BSD Unix handles this free-ware venture for those who like to pay for their operating systems(OS), and run Unix/Linux like OS. BSD Unix can be copy written or patented by any manufacturer once a change has been made by such a manufacturer. Also such changes are not required to be returned to the BSD Unix community as free ware, as would happen with any Linux modifications; Linux source code must be available to all. BSD Unix runs almost if not the same as Linux, perhaps, not quiet as well; beauty is in the eyes of beholder. The pay for use of free soft ware is contributions by those who are qualified to contribute by making successful, reliable, and documented modifications.
This Linux community has a system of checks and balances (For example,Woody, Sarge, and Sid for stable, testing and unstable Linux kernels.). It is important to note that changes must be approved by qualified select journeymen who accept or reject such changes. What could be better than this? Autocracy => Microscam? Not in our fore-see-able future!
Bug reports also provide a wonderful source for code corrective information. This source of data is perhaps larger than you might, causally, believe, and is invaluable to serious Linux distribution design-build entities.
Standardization may, perhaps, already be accomplished. Linus Torvalds wrote the original Linux kernel (~1991), and it is yet the standard today. I understand that kernel replacement by it will run any modern legitimate Linux distro? However, since Linus Torvalds wrote this kernel there have been many new features added to both the desktop and server operations which are now available and were not there at the time Linus wrote this Linux kernel. (For example, wireless Internet connections are now widely available. However, this is not primarily a kernel function, it is usually included inside each Linux distribution’s initialization script.)
One of the greatest attributes of Linux is: If you do not like it, then fix it! Try this with windoze or Apple. There is no warranty from Linux to worry about.
It is clear to me that you already know a great deal about Unix and Unix-like operating systems. Based on this assumption, you should become an active member of the Linux contributors community and enjoy the benefits of learning new and exciting computer soft ware useful tools. Join with us and have fun!
One of the the strongest points of Linux and GNU, in general, is that modern computer usage research continues on a 24-7 operational basis for this world-wide community, literally, by thousands, if not millions, of Linux users.
File transfers, and commercial operating system repair is also easily accomplished with, for example, a Linux run-from-a-CD distribution: Knoppix. Ask any system administrator who has had problems with repair of frozen OS or heavily virus infected windoze boxes who used Knoppix to fix the problem.
For one, I really appreciate and welcome your comments and hope that you will consider becoming an active member of this wonderful GNU community. Thank you for “stirring the pot”.
Respectfully yours,
Jim Mills
PS. I have apache server/s running, which come with most Linux distributions; free for the download (Because you mentioned servers in your piece.).
I just want to say that it is not against the GPL to sell software as the poster above me stated. It is against the GPL to not release the source of code that is GPL’ed. As well it’s not a “freeware license”.
Freeware is bad. Its free sure but when that person stops the project theres no source so no one can continue the project or improve it. So why make it in the first place. As well the GPL promotes community. A stark contrast to the BSD license. Not that the BSD license is bad, it just has a different purpose. And no a software company cannot copyright code that is not theirs. If they add to it yes they can copyright their *addition* but not what other people wrote. The original coders still hold the copyright to their code, they just give it to you to make changes without requiring compensation. Also, you only have to distribute your changes to the people you distribute the product to.
So technically a company could sell a Desktop OS based on a modified version of the Linux kernel. They could also charge for the cost of distribution of this source. They could charge $500,000 to send you the source on a CD. Crazy huh? It’s been done. Also if a company makes changes to a GPL’ed project but is not distributing it and is keeping it “in house” then they do not have to release their changes.
I suggest buying “Free Software, Free Society” from amazon.com. It explains why the GPL exists and the problems with patent law. Just to be clear, the GPL is not against copyright laws, it’s against patent laws.
It seems all the Mac Users here never experience any problem with the Mac OS X or with any Mac OS at all. Everytime I talked with a Mac user, he just tells me how great and perfect are the Mac OS. Please be down to earth once again. There is no perfect OS and there never would be even this Tiger that was full of bugs once month ago (I didn’t check the recent updates, no times and working on different machine for a while). And I experienced so many troubles with Mac, I can tell you… And haven’t you guys heard about about the linux Live CD like Ubuntu and Knoppix. There are really friendly users, typically for Mac and Windows users who don’t know how to use a computer if there are no nice icons with a nice GUI… But in the end, I do agree that there is no real leadership for the Linux community but there is no money either. My point is that there are good and bad points in all these OSes and maybe we better have to stop kicking eachothers asses and start working together (or kind of…)