The Linux world’s very own version of Paul Thurrot, Steven J. Vaughan-Nichols, lists 5 things Linux can learn from… Microsoft. “Linux does a lot of things right – open-source, security, reliability – but it’s far from perfect. In fact, Linux and its vendors could stand to learn a few things from Microsoft. Yes, Microsoft. Like what? Here’s my list of the top five things that Linux could learn from Microsoft.” Next thing you’ll know we’ll have Apple switching to Intel and… Oh, wait.
I like some of the stuff… I do. What I don’t agree with is:
“The Linux vendors can no longer count that their message about a new distribution or service being read by anyone except people who have already committed to Linux.”
Red Hat Continues Booming Growth in Fiscal Q1:
Red Hat reports software subscription sales of $71.5 million, up 45 percent compared to the prior year’s first quarter.
Total sales during the quarter were $84 million, an increase of 38 percent compared to the year ago period.
So.. Saying Linux is not going any where… Is wrong. And thats just people that went with Red Hat.
He also says Linux should place more Ads:
“For every ten, heck probably ever one hundred Microsoft ads, I see one Linux ad. ”
I would aruge that MS has about 10 times the marketing budget of most companies… So.. thats really a money thing, not a becuz they are being “Idiots” thing.
I’ve never seen a linux ad! Does anyone have a link to one?
This is probably the most famous Linux TV ad:
http://digg.com/linux_unix/IBM_linux_commercial.._its_cool
It came out in September 2003.
To what extent is the question.
His point about MSDN and developers is really good. Where to start? Microsoft provides a flashlight.
His point about advertising is also a good one.
OEM support? OEMs are going to support linux when people demand it, that’s up to us, not the vendors.(more specifically, the vendors voices will go unheard unless there’s a larger linux community)
Common format and common interface I’m not quite so sure I agree with. But as is with everything, your greatest strenghts can and often times do turn out to be your greatest weaknesses.
OEM support? OEMs are going to support linux when people demand it
That’s not true. There was no ground swell of voices crying out to the OEM’s, “Please give us our Windows!”. It is just what came with it. The OEM deals built the demand, not vice-versa.
I believe there is something else that could tip the scales. Schools! If kids used Linux throughout school; if Linux was the standard at Universities, everyone would expect to run it at home or elsewhere. This to me is the great opportunity. Why do schools run Windows, including servers? There is no real reason other than inertia and ignorance – and this OUR tax dollars going to support a big business. Schools would greatly benefit from free AND Free software. Our kids should be learning to truly innovate – becomes leaders in the tech world. America is growing old and stale in the tech world. Let’s change it NOW!
————There was no ground swell of voices crying out to the OEM’s, “Please give us our Windows!”.———–
If I remember correctly, there wasn’t much of a personal computer market before windows.
And arguably a mac/apple(which is what was around during those days) aren’t really PCs because of how closed they are.
———–The OEM deals built the demand, not vice-versa.————
It’s a classic chicken/egg scenario.
I agree with you about schools. Instead of spending thousands on new macs and new computers as they’ve done for generations now, just reinstall everything with linux. It will run just as good on existing hardware and the budget for upkeep of the inhouse IT will plummet.
Though one thing is wide open. Red hat is not exactly small company anymore. Let’s say that all US schools switched to RH. They’d still be supporting big business. This aspect does not bother me because they need to have tools to get the job done.
Edited 2006-07-21 22:24
Schools have large IT budgets in the UK, the theory is that the children will learn better with them.
GCSE/VCE/GNVQ/A’Level is all based around the Microsoft office suite. Go check out the Edecxel lesson site or pop to a bookstore and look at the students books.
Also easy training packages like CLAIT and EDCL also. have a look.
If that was the reason for Linux small marketshare, things would have to be different in germany. During my school years i encountered many Unix Variants, usually used with terminals (mostly ancient stuff), or networked PCs running novel netware. I haven’t even seen one single Windows PC in a german school before sometime around 2000. I don’t know if thats the case in every part of germany but in the city i live windows PCs where very rare in schools.
That obviously doesn’t have an big inflluence.
Apple tried to market to schools in order to build its longterm market potential. While it certainly garnered support in the education market, it never really translated into greater market share in the home market; primarily because people used PCs at work, and they wanted to have compatible machines at home. The only way to get Linux into people’s homes is to make sure that they use it at work.
I like you man, I read your articles all the time, but I’m afraid this time you’re full of cow dung.
Not a single one of your points was even remotely valid. You need to get completely free of your past Microsoft influence and then see what Linux truly is.
Could you provide some details on why “not a single one of [his] points was even remotely valid”? It would make your post harder to ignore.
read my post below
You need to get completely free of your past Microsoft influence and then see what Linux truly is.
A community of zealot, unable to get the facts ?
I’m not a huge company, however I do offer OEM support for linux. http://linux.v2w.org
There are some very good points in the article. Such as DRIVER SUPPORT, Linux definately could use driver support. And the concept of an MSDN-like database for Linux is a damn great idea. This article makes some very valid points. As a long time linux user, I have ran into many problems. I’m not one of those linux zealots, I can see the many flaws in linux, but it’s advantages far outweigh the cons.
/2 cents
There are some very good points in the article. Such as DRIVER SUPPORT, Linux definately could use driver support.
Wow, gee, you think?
This is such a non-statement: everybody knows this, noone gets any better by repeating it. Of course an OS can use more drivers. Even Windows could use more drivers (although the need is a lot less urgent). I would think that, anno 2006, people don’t need to be reminded that niche OSes (yes, 4+-% is a niche) have less driver-support, and that not having your devices working sucks, and that it would be best if that were remedied.
Mentioning it again isn’t helping.
/blow-off steam
Whatever man. Personal attacks won’t get you anywhere.
Driver support? I’ve seen hardware (one piece, but proves the point), that was easier to get running (some 1.5 years ago) in Linux (Debian/Sid) than Windows (XP SP2).
Namely a Bluetooth USB dongle, whose drivers got borked in SP2.
Except of 3D acceleration on graphic cards (NDA problem, mainly) and specialist sound cards, I personally see little problem with drivers on Linux.
Not to mention, that on Linux, almost all drivers come with the OS, get written/approved/checked by the kernel team. On Windows, they come from the HW vendor, and often it shows on the drivers’ quality.
I agree with the MSDN point.
You want new customers? You want Microsoft’s customers? Stop getting excited about placing — oh wow — a single-page, one-shot ad in the New York Times. Start spending the big bucks you need to get the Linux message, your message, in front of customers every day.
Can you really support a large scale marketing program selling a FREE operating system? No way, at least not on the scale that Microsoft does it.
SJVN (for short, if you will) really is the Linux version of Paul Thurrot. Both perceive themselves as a representative of their respective communities, but both write this kind of predominantly naysaying articles. Not that this is wrong, but it makes them similar.
His argument is really easy to pick apart, though, so for once, I’ll be that guy that picks it apart:
1. Developer support – First, in OSS, the code is open. If you’re not sure how to program, let’s say, a Kicker applet for KDE, you can just take a look at how other experienced KDE developers wrote them. I respect the argument that code != documentation, but for a prospective contributor to OSS projects, having unfettered access to all of the preexisting code is in some ways much more valuable than something like MSDN. Plus, many larger projects do have best practices documents, HIGs, complete public and private API documentation, submission procedures, and more. They vary from project to project, but then again, the needs and goals of no two OSS projects are identical.
2. UI consistency – This is doubletalk. He compares Microsoft’s track record of keeping its UIs consistent across releases to OSS’s track record of keeping them consistent across desktop environments. Then he goes on to say that Microsoft isn’t even very good at preserving their familiar UIs anymore. Well, I’ll add one more point: Microsoft’s ability to keep UIs consistent across Microsoft-brand software sucks, and it isn’t getting any better. The consistency amongst GNOME or KDE apps is far superior, and it’s clearly getting better with each release.
3. File formats – I don’t get it, what can OSS learn from Microsoft on the issue of file formats? Many of Microsoft’s supported file formats are only reliably readable or writable using Microsoft software. I can’t really think of any file formats native to OSS applications that would be difficult to support by proprietary software vendors, nor can I think of any popular OSS office applications that don’t support ODF yet.
4. Marketing – Commercial Linux vendors do market their products, and they do so fairly effectively given the admittedly small quantity of funds devoted to marketing. Show me a CIO that doesn’t know about Red Hat, and I’ll show you a CIO that isn’t worth his $1000 suit. As for marketing to consumers, it just isn’t worth it. Canonical, for example, would be wasting its money if it started buying banner ads. Maybe someday direct marketing to consumers will be worth it, but today is not the right time, which brings me to…
5. OEM support – This situation is undeniably the number one issue holding Linux back from penetrating the consumer desktop market and arguably the corporate desktop as well. However, the title of the article concerned things that Linux can learn from Microsoft, not things that OEMs can learn from Linux. If amatuer programmers, enthusiasts, and hobbyists can figure out how to get the vast majority of OEM hardware running smoothly on Linux, then I should hope that the likes of Dell could do this, too. If community projects can provide useful support services (forums, tutorials/HOWTOs, documentation, etc.) with little or no funding, then I should hope that commercial service providers could do much better.
In general, I submit that Microsoft has more to learn from Linux than the other way around. Maybe this wasn’t the case 3 or 5 years ago, but it’s true today.
His argument is really easy to pick apart, though, so for once, I’ll be that guy that picks it apart:
1. Developer support – First, in OSS, the code is open. If you’re not sure how to program, let’s say, a Kicker applet for KDE, you can just take a look at how other experienced KDE developers wrote them. I respect the argument that code != documentation, but for a prospective contributor to OSS projects, having unfettered access to all of the preexisting code is in some ways much more valuable than something like MSDN.
Read the article again. He specifically mentions that being open-source is one thing Linux has done right, over Microsoft.
That still leaves Linux without a one-stop unified resource site (or at least a big professional one) for programmers who want to learn how to program for the Linux platform, regardless of their skill. And MSDN is just that, for the Windows platform.
And that was his point.
Right… and you read the rest of that very paragraph you quoted. I noted that OSS projects vary greatly with regard to their needs and goals, and therefore a “one-stop unified resource site” just wouldn’t work. Reusing my example, would such a resource on programming for the Linux platform be of any use to someone who wanted to program a Kicker applet?
We already have a site that documents what could be considered the lowest-common-denominator of how to program for the Linux platform:
http://freestandards.org/docs/lsbbook/lsbbook.html
Does that help me if I want to write an address book using FUSE? No, for that I would go here:
http://fuse.sourceforge.net/
Users need a one-stop resource for help. That’s why we have commercial Linux distributors and comprehensive support options. However, anybody that’s willing to devote nights and weekends to developing free software will be able to find the information they need in less than 30 seconds using their search engine of choice.
The crucial difference between the Microsoft development model and that of the OSS model is that, in the case of Microsoft, the vast majority of the code is derived from a common base, a common set of APIs, and a common set of language bindings. In the OSS model, projects are free to derive from wherever they want on the giant tree of the free software “stack.” The interfaces one would use to code a Kicker applet are completely different from those one would use to make the address book on FUSE.
This principle of “arbitrary inheritance” is, in part, responsible for the rapid development of free software. When you understand this, the idea of creating an MSDN for Linux becomes simultaneously a daunting and useless endeavor.
4. Marketing – Commercial Linux vendors do market their products, and they do so fairly effectively given the admittedly small quantity of funds devoted to marketing. Show me a CIO that doesn’t know about Red Hat, and I’ll show you a CIO that isn’t worth his $1000 suit. As for marketing to consumers, it just isn’t worth it. Canonical, for example, would be wasting its money if it started buying banner ads. Maybe someday direct marketing to consumers will be worth it, but today is not the right time, which brings me to…
But with Ubuntu, the main carrier for pushing it isn’t the marketing department, but word of mouth, “oh, you’d like to try Linux; Ubuntu is really good for that” and another convert joins the ranks.
But then again, the story is different if one were to compare Red Hat to another UNIX distributor, such as Sun (Linux, for all intensive purposes, is a UNIX; whether the nuts at opengroup like it or not); the Red Hats and Canonical don’t need the marketing, because they’re going after different cliental; those who are using Linux right now, are companies who LISTEN to their IT staff rather than having some clueless ranting CIO buzzing on about the latest thing he has read in his monthly gold magazine he has picked up whilst charging another over priced lunch to the company account.
2. UI consistency – This is doubletalk. He compares Microsoft’s track record of keeping its UIs consistent across releases to OSS’s track record of keeping them consistent across desktop environments. Then he goes on to say that Microsoft isn’t even very good at preserving their familiar UIs anymore. Well, I’ll add one more point: Microsoft’s ability to keep UIs consistent across Microsoft-brand software sucks, and it isn’t getting any better. The consistency amongst GNOME or KDE apps is far superior, and it’s clearly getting better with each release.
True, run all GNOME software on a GNOME desktop, and it is more consistant than Windows; run all KDE applications on KDE, and you’ll find that all the applications integrate alot better together than Windows. In KDE, for example, spellchecking is everywhere, in the chat application, when typing in forms; thats true integration of tools – equal to that of Apple.
1. Developer support — You really missed SJVN’s point. He’s saying that a single source of information is what’s missing. Linux doesn’t have an equivalent of MSDN.
2. UI consistency – Linux apps generally don’t have the same kind of CUI consistency that Windows and Mac apps have. Their UI is all over the map, some good, some atrocious. Sure, you see similar patterns in Windows and Mac apps, but to a lesser degree. It isn’t simply about Gnome.
3. File formats – Agree.
4. Marketing – Disagree. Microsoft builds brand recognition across the board in all markets by doing broad advertising. The problem for most Linux companies is that they simply don’t have the kind of advertising budget that MS has; hence, this is tough for them to address.
5. OEM support – No, wrong. This is about Red Hat and other Linux companies making inroads with OEMs, not amateur devs.
MS has adapted itself to compete with Linux and has done reasonably well. But it still needs to improve time-to-market in order to keep competing in the future.
The thing is Microsoft can be a single source of documentation because they make everything. When it comes to Linux there isn’t a single-source of documentation because there isn’t a single source of products.
The GNOME website has documentation for GNOME, the KDE site has documentation for KDE, etc.
I would love to see centralised documentation, but I think that expecting people to keep up-to-date copies of documentation from N projects would be too much. Similarly I can’t expect projects to move their docs into a common format and central site.
Of course distribution makers are the centralised source for people wishing to use their packages, its just the developers that are left hanging here.
Although some of the points have arguments to the contrary, I appreciated this article. It attempts to get realistic about how Linux can benefit the mainstream user. I found it thought-provoking.
Two things come to my mind. Firstly, the problem with advertising: This requires having huge amounts of cash on hand (like MS) and even Red Hat’s strategy won’t work by spending money in that way. Also, something I noticed from the Firefox campaigns is that Open Source projects have a hard time presenting a unified front to wedge into consumer consciousness, a side effect of the flexibility of Open Source projects; They are entities that shape to fit the user and don’t have a singular interest (eg. a marketing dept or CEO) behind them. A marketing push requires more than just a product and a logo.
Secondly, the OEM problem. To be brutally honest, I can picture a lot of Average Joes ignoring the “giggaherts” and “linix” stickers, buying a cheap box, and then angrily returning the PC to the retailer complaining that his computer doesn’t have “Microsoft” on it like he’s used to. These people could be in the minority, but then again, maybe not. That said, I also agree with the “chicken/egg” diagnosis. There is no demand for Linux from OEM’s because the OEM’s ship PC’s almost exclusively with MS software.
I have seen observed before, that PC OEM’s are sort of like independent extensions of Microsoft, compared to Apple which is OEM and software vendor both.
It will be hard to change the system. Most successful IT businesses, MS especially, are more “integrators” than “innovators”, who try to provide a lowest-common-denominator, out-of-the-box solution to a single market demographic (MS, behemoth that it is, has multiple vectors, but still makes most of its profit from the Windows/Office hegemony).
Gotta say. I think Microsoft gets the cheap things right.
The Microsoft logo and the windows brand are just the greatest.
Tux and a meaningless name nobody is is sure how to say it.
Personally tux should be moved to linux gaming.
Is that for every 10 people you have 10 opinions. There’s no consensus and no single organization to lead the way.
I wanted to disagree with you but then I thought, if I did you would be right. hehe
1) No Gaming
2) Poor Hardware Support
3) Search for patches
4) You have to edit Config Files
5) Dependency Problems
6) Compile stuff
The reality these are all true, or were true, and true to someone who tried linux once and it was then.
1) Linux gaming is good enough for occasional gamers and its cheaper and suitable to get a wii
2) Its true that if it works on windows it won’t always work on linux, and why should it, Live-CD’s have been a blessing and a curse for exposing this. Its better to buy your components to work with the OS. I would even for Windows.
3) ATI/Nvidia being the biggest problem in this, I can’t think of many reasons why a sane person would want to, see point 2
4) You may have to but they are easier to find than a dialog box, often contain inline documention, but more importantly you almost never need to edit one.
5) I don’t hit these. I suspect these are often obscure packages, I suspect they have alternatives, I suspect they will be fixed, I don’t know I just haven’t had a problem sine I updated Xfree from source on slackware 3.2.
6) When do you need to compile anything, unless your developing. The program I have on my system that I compile by hand is scummvm becuase that program is just too long between releases.
1) “No gaming” is false, yes. “Poor gaming” is on the contrary no myth. Well you know, most PC games doesn’t work on Linux.
2) “Poor” is an overstatement. I would call it just “worse”. As you say, it doesn’t always work on Linux. Also the feature set can be smaller or performance lower compared to the official drivers. Like my own printer. In Linux it prints at maybe half speed and the drivers have less features.
3) Yes, you said it yourself, ATI & nVidia. Wireless too in some cases. ATI & nVidia will never open source their drivers. I don’t see why they should either. It is the Linux kernel that should have had a real, working, stable driver system. But the kernel developers don’t want that…
4) A Linux system is based on RTFM and editing config files. All users will have to do this sooner or later. This philosophy will probably never change. For ordinary users this is worse because typing stuff on the command line or in files isn’t as self explaining as clicking buttons and checkboxes. They don’t know what to type and they don’t want to read the manual. Even if they would find the man page (say xorg.conf) they wouldn’t understand it.
5 & 6) If your distribution doesn’t provide you with a particular application or you need a newer version you will often need to download the source package and compile because every author cannot provide binaries for every version of every Linux distribution. To compile you have to have a LOT of additional packages but an ordinary user cannot easily know what packages…
“But, what users want is one common interface.”
False; depends on a user. Furthermore, what some users do not want is to be forced to use some “common” interface. Common != Good.
In my opinion, the biggest strength of Linux software is that users can mix and match interface components. For example, I don’t like the default GNOME or KDE environment ( namely, the panel with the equivalent of MS’s “Start” menu, and the icons ), and so I instead use Fluxbox as my window manager. Since GNOME is not my cup of tea, I use KDE programs, simply because I like them more than GNOME ones.
Basically, what I’m saying is that I’m not sure that freedom of choice should be sacrificed in the name of user-friendliness.
I bet you aren’t used to people that write down on pieces of paper simple things like steps to change the screen resolution on Windows for example. But, believe me, they exist. And I would go further and say that they are a very significant amount of computer users, at home or at office.
So a “common” interface, even if it’s not a champion in design concepts, has a lot of inherent goodness. You train your people only once.
OSNews.com is still up and hasn’t been bombarded with hundreds of hate messages because of this article.
He has one valid point in the five, which is MSDN. MSDN is very good. Apple’s documentation is also pretty good. GNOME’s could use work. An integrated “GNOME platform doc” describing everything from GTK+ to GStreamer would be great for developers.
He’s off on the other points, though.
1) Marketing. The Linux companies large enough to do marketting do it. You’re not going to see Linux ads in “PC Gamer”, but Linux is at this point for suits, not home users, so that’s really irrelevent.
2) OEM support. How do you “learn” OEM support? I agree its important, but its not something you can “learn”, and its not something that’s really under your control anyway.
3) The format bit is weird. He doesn’t even make a suggestion here, because he can’t. You can’t make your format common, it’s not under your control.
4) Common interface. Microsoft interfaces are pretty common, except when they’re not. XP is just different enough than 2k to be frustrating, and new versions of Office and Visual Studio take advantage of the user’s knowledge of previous versions to actively mess with their heads. As for KDE to GNOME to Enlightenment, that’s quickly becoming a non-issue. Nobody uses Enlightenment, and corporate distros are quickly standardizing on GNOME. With SuSE’s move to GNOME, the top desktop distros are now all GNOME. KDE will likely continue to be very popular with individual Linux users, but in the context of corporate support, targetting GNOME only seems like an increasingly sure bet.
1. MSDN
that’s probably true…
2. Common Interface
oh we don’t want a choice do we? There should be a standard config for the firt time user that is standard maybe… but I think the philosophy of supplying the user with modules and not with just one default solution is a great thing!
3. Common Format
I pretty much agree with what other people sayd… it’s not a standard, it’s a monopoly, that’s different.
4. Marketing
they will… as soon as they get the money to.
5. OEM Support
do I have to mention MS blackmailing OEM vendors? do I have to mention MS trying a bit too hard to become the nr.1… they have not been fined several times by our courts just because we don’t like US companies
I hope Linux learns how to be better by being different from windows and not by copying it.
1- MSDN
I dont know about others, but I personally find lost constantly when trying to figure out how to do stuff. the OS-Level APIs are redundant, unicode/non unicode interfaces are mixed up, and the overall low level functions are not docummented or clear enough.
2- The look is “somewhat” the same (although not always), but widgets are different depending on the app and not always behave similar. I think even if there is KDE/Gnome, at least all KDE apps are used the same way, and most gnome ones too. Every microsoft app is different.
3- Common format, What is to learn about this?
4,5 – Dont think I can comment on this, just note that even if hardware support is not perfect, nothing avoids you to sell OEM computers that come with fully supported linux-compatible hardware
…the free software philosophy.
1) MSDN – It is evident that the author doesn’t mind paying for documentation. So how come he doesn’t mention the endless number of books on the subject of FOSS programming, from publishers such as O’Reilly, McGraw-Hill, Manning, etc. There are tons and tons of commercial documentation and tons of free documentation. Seems that the author just doesn’t know, which means the author doesn’t research. Another false argument.
2+3) Common Interface + Formats – While it may be in the interest of the “general public”, the author should know that free software is about choice. And please, oh please, don’t talk about standards.
4) Linux does not have the financial back for grandiose advertising. Since MS pays almost every computer distributer known to mankind to write “X Inc. recommends Microsoft Windows XP for Home/Business/Sewerage”, we’re pretty much left with nothing. The rest, Linux simply cannot afford. That said, I’m sure the major commercial Linux distributers, such as Red Hat and Novell, do a pretty good job when it comes to Marketing. Just look at the web server world.
5) Once again, turn to 4. Microsoft pays companies to distribute their computer with Windows pre-installed. Linux simply doesn’t have the money to do it. And when complaints are made (See BeOS), MS simply buys us out.
This is not a fair world, and MS is to blame.
Some of the points are fairly good.
However, I will take some exception on the common interface issue.
I have watch the so called common interface change over time to the point that early interfaces appear quite unfamiliar to some one using the present day versions. New versions will change that some more. What I find far more annoying is when the way some program changes the way it works – usually to accommodate some new “feature” most of which I probably will never use. An excrescent example is MS Office multiple cut and past feature (This just about destroyed the usefulness of cut and paste in MS Office). Yet KDEs Klipper is a perfect example of how to do the same thing right (works just as before yet has more functionability without confusion).
As for common file formats – Whats this guy smoking! Have you tried to open an old Doc File in a modern version of Word. More than a couple of versions of MS Office back and you are more likely to get the file opened correctly in Open Office than MS Office. Add to this the fact that MS is now starting the shift to their own proprietary XML file format. If MS follows their previous form lots of luck opening any DOC or XLS file in MS Office15. There even problems opening MS Office XP files in Office 2000 or 97. I know of offices that purchased one or two new computers and had to upgrade all their versions of MS Office because of file compatibility problems. (Actually worked for one firm this happened to).
On OEM Support – on the buying computer front – 100% amen brother. People will use what they have in front of them. I run a small non-profit that gives computers to families that cannot afford one. We use Linux on these computers – mostly for cost reasons but also because it is generally easier to install Linux and set it up on a wide range of hardware than for Windows. This activity has made two point clear.
1. People will continue to use what they have. We have just started a survey of computers we have handed out – but I can tell you that nearly all them have remained Linux over the last year and a half. Nearly everyone has found no deal breaker problems and are still using the systems as delivered.
2.Linux install and setup is easier than Windows at least for us. When you are mixing a wide variety of old and new hardware you find that Window will require hours of searching for device drivers and often you wind up discarding parts for lack of a good device driver – because you have old hardware and a new version of Windows or vise versa. I cannot count the times I have had problems installing some hardware on a Windows system for any of one a dozen reasons. The biggest problems we are now facing is a stack of Compaq computers we are now handing out (We have to manually set up the sound chip driver for these machines). Sometimes being able to go in and manually do something is an advantage. And yes the end user will not do something like this – they go find help, often a professional.
So I grant this guy a 2-1/2 out of 5.
Is that for every 10 people you have 10 opinions. There’s no consensus and no single organization to lead the way.
You say that like that’s a bad thing… 🙂
Look at it this way, at least Linux (as an OS) is immune to the dangers of being tied to a single company (which could then be bought or destroyed by MS).
Linux is certainly more chaotic, but personally I like it, it makes it very dynamic.
This might fall as a subset of #5, but the primary driving force of any OS is usually the applications. It will hard for “free” Linux developers to compete with software houses that pay people 40+ hours a week to develop applications for Windows. Because of that, Windows apps are usually better…. far better (putting on flame suit). And as a side note, it doesn’t help that there are so many flavors of Linux and package managers, but I digress.
Let’s face it. DOS was a crappy OS, but people wrote for it. Same with Windows. For Linux to succeed, it not only needs software as good as the choices for Windows, but superior. There has to be a reason to switch. There has to be somebody who says “Oh? You want to run something like that? Well, it’s only available on Linux.” Right now, I can run everything I need under Windows, and for things like DVD & video manipulating, capturing, authoring, Hhome Theater PC apps, etc. – there is nothing near as good in the Linux world as some of the apps I use frequently. I simply cannot switch, even if I wanted to.
http://www.aslab.com/
http://www.emperorlinux.com/
http://www.eracks.com/
http://www.fifthedimension.net/Merchant5/merchant. mvc?Screen=PROD&Store_Code=fifthedimension&Product _Code=BBasic&Category_Code=S5
http://groovix.com/
http://www.ibexpc.com/
http://www.koobox.com/
http://laclinux.com/
http://www.linspire.com/
http://www.linuxcertified.com/
http://www.linuxsyscorp.com/
http://www.microcenter.com/
http://www.microtelpc.com/
http://www.outpost.com/
http://www.seascape.us/
http://shoprcubed.com/
http://www.sub300.com/
http://www.swt.com/
http://www.system76.com/
http://www.systemax.com/divisions.htm
http://www.walmart.com/
http://xtops.de/
http://ztechshop.net/
http://tuxmobil.org/reseller.html
http://www.us.debian.org/distrib/pre-installed
http://www.linux.org/vendor/system/index.html
http://tuxmobil.org/ (general information)
No OS
(Sabio made by Quanta, like Dell-latitudes)
http://www.avadirect.com/
http://www.asimobile.com/
http://www.powernotebooks.com/
Nice FUD from the Bendster…
1) Gaming: the point is moot, since console Gaming blows both Windows and Linux gaming out of the water. In fact, it blows them right off the surface of the Earth. And sends them into the sun.
Even MS knows that: why do you think MS put so much money into the Xbox/Xbox 360? Hardcore “PC” gamers are a dying breed, and there are lot more 3D games coming out for the consoles, in a greater variety of genres, than for PCs.
Casual gamers do continue to maily use the PC (don’t forget that Solitaire, Minesweeper and Bejeweled are the most played games out there), but guess what: Linux is great for casual gaming. The average distribution comes with a ton of casual games, including clones of the aforementioned classic.
So for this point, really, Windows’ advantage on Linux is mostly irrelevant, especially since the biggest games (such as HL2 and WoW) can be easily played under Linux with Wine.
2 and 3) The fact that hardware support for Linux lags behind that of Windows is not in dispute. However, the truth is that such support is progressing faster than new hardware is coming out, therefore the gap is closing. Witness the big progress Linux has made with wireless.
ATI and Nvidia remain small issues, fortunately installing their drivers is most often than not trivial.
In some areas, Linux hardware support is more convenient that for Windows, as you don’t have to fish out drivers from vendor web sites (which is often more complicated than installing the NVIDIA Linux driver…).
4) Do I really need to point out that the following is complete and utter BS?
A Linux system is based on RTFM and editing config files. All users will have to do this sooner or later…
Come on, I know from your previous posts that you have a strong anti-Linux bias, but this is just false. There are no reasons that a well-configured Linux box would require anyone to edit text files: there are graphical tools for virtually all administrative tasks. The fact that people often use the command line is that, despite being intimidating at first, it is often more efficient than a GUI.
The false assumption in this particular bit of FUD from you is that, just because it has a GUI, Windows is somehow simple to administer by newbies. If that was the case, why do I keep getting calls from family and friends who need help solving Windows problems? And let me tell you, it’s a lot easier to handle Tech Support by phone when you’re dealing with a command line instead of a GUI!
5 & 6) These myths have already been disproven many, many times. If you need those kind of obscure packages, chances are you know how to compile from source. “Ordinary users” rarely need these obscure, hard-to-find apps, and most distros offer up-to-date “bleeding edge” versions of packages for those users who “just can’t wait.” Again, these tend to be power users (most users don’t care to have the latest version of a program). So in fact your argument concerns a very, very small percentage of actual users. It makes for good FUD, though.
BTW, it is not the developers who make packages, but the distro maintainers.
Oh, and in words of the immortal Bender: BMSMA!
1. Currently there is no MSDN because there is no single Linux vendor. Linux itself has nothing to do with GNOME, just as GNOME has nothing to do with Solaris.
This isn’t an excuse, this is just a reason. Currently if you have system with a lot of dev packages installed (or Gentoo) then you can probably sniff about /usr/share/doc, on man and with info and get nearly everything you need.
It’s a nice idea for a project though… a publically accessible open documentation project, which could have from Linux kernel and GNOME docs right through to the stuff written by people when they are reverse engineering hardware so new drivers can be written. Currently if you know where you’re looking you can usually find docs at a standard equal to or higher than MSDN.
2. GNOME and KDE have consistent interfaces. I’d say that after you load Windows up with all the tools that you need (not just Microsoft Office) then you have quite disparate interfaces.
Take Windows Aqua, Internet Explorer, Microsoft Office with it’s shiny toolbars, Picasa’s custom interface, Lavasoft Adaware, Live Messenger, Photoshop, WMPlayer, Google’s jabber client etc. and everything seems to have it’s own quirks, a slightly different manifestation of the open dialog, custom buttons, etc.
I’ll talk about GNOME since that’s where my experience lies… GIMP, GAIM, GNOME, Firefox, Openoffice/GNOME Office, FSpot Totem, Rhythmbox… everything fits in. There are some little bugs and quirks but they are thought of as bugs and quirks, not shiny interface features.
About jumping between KDE, GNOME, E17, Xfce… you do that because you want a different interface. If they were all the same, why would someone change between them?
3. Does he have a point here? Microsoft’s formats usually work in one or two programs, ODF currently works many programs. And like when a new Microsoft format is released, you need to wait while some people upgrade their software that isn’t compatible with the latest version or export it as a previous version. The difference being that all formats here are open and ODF is supposed to become the standard.
I agree that ODF support is a good thing but it sounds like he’s just bitching that it’s not being supported fast enough for him. It’s not exactly been around long. (Since last year — Microsoft IE still doesn’t *properly* support CSS1)
4. That’s nice, he wants to see Linux adverts. I assume he’s also got a few hundred million dollars to blow on it too. Novell/Redhat etc. are trying to compete with one of the richest companies in the world (is it a coincidence that they probably also have one of the largest marketing budgets too?).
I think they’re putting as many resources as possible into keeping their product competitive against the massive gorilla that is Microsoft. Marketing is important but what good is it when you’re marketing a product that nobody would want anyway.
5. OEM support is increasing, but it’s up to OEM’s to support it. Driver support isn’t an issue if there’s OEM support, though. No OEM is going to ship a computer that doesn’t have the correct drivers for the hardware. Driver support again isn’t something that will just magically get better. Reverse engineering drivers is extremely difficult and time consuming, and full of legal problems.
Better hardware drivers will only come when OEMs start pushing for them, likely. Infact, if OEMs support it it might fix a lot of the driver issues immediately, considering the most trouble I ever have is with little proprietary bits on a laptop like LEDs and fan control that follows no standards for some reason.
It’s “dint”, not “dent” in this context.
All to often development can’t keep the pace of marketing.Scrapping features isn’t exactly good practice now is it?
MSDN is something to be further reviewed though to be honest.
Edited 2006-07-22 06:41
>>MSDN is something to be further reviewed though to be honest.
On the basis that women seeking equality with me lack ambition;
I think open source developers seeking parity with MSDN
lack ambition.
Resourcing the legwork is the challenge. Making the
knowledgebase “accessible for improvements” is perhaps
the key.
Edited 2006-07-22 07:47
Advertising is annoying. There is a certain limit to that. Consumers need to know about Ubuntu. Then make a good OS and you get great reviews from extremely popular websites. There’s your advertising. Advertising isn’t needed in the way it once was because of the Internet. If I need to find saomething I just look up ConsumerReports.org etc. The only reason Linux didn’t make it before is becase the OS wasn’t that great for the desktop not because of Freaking crappy ads.
Did they need ads for the server. NOOOOOOOOOOOOOOO
…at least for me.
Whenever I had a specific question about what APIs are available on Windows, and how to use them, I tried to search the MSDN Website. And it always provided me with utterly useless stuff that I didn’t care about. The answer was always close, but just exactly besides the point.
I never had that problem on GNU/Linux, I’d rather use google and freshmeat to find out what’s available, and then refer to the documentation that is specific to the project or library that I want to use. Having a unified documentation for *all* libraries that are available for GNU/Linux, would propable be a total mess, and always outdated. What’s the problem with using google anyways?
The biggest problem facing linux adoption has nothing to do with Microsoft, it’s the current core users, the early-adopters, that are likely holding it back and preventing it from “jumping the chasm” into the mainstream.
There seems to be an attitude of entitlement from many in the community that does not translate well for profit-driven commercial vendors in a competitive market. People genuinely act surprised that the vendors haven’t tripped over themselves to invest millions in designing, testing, releasing and supporting commercial linux applications or hardware.
What is the “linux” market?
Is it the millions of people that download free distros and slag “Microshaft Winblows” (emphasis on that dollar sign) and extoll all the free applications they can run? Is this a viable market for investing in?
Can anybody provide metrics or market research indicating these legions of users would be more likely to actually pay for commercial applications and not just warez them as many already do for Windows?
I think one of the biggest problems is that the vast majority of the linux userbase hasn’t committed to using linux, they still keep Windows in active use or on standby. We keep saying “I’d ditch Windows if it wasn’t for this or that”, which automatically implies you’re still willing to use Windows when linux won’t cut it, so where’s the incentive to develop for a new platform if the vendors know you’ll ultimately still purchase or use their software on Windows? Why should Adobe or Autodesk or Intuit or similar companies start porting to linux? Ultimately you’ll choose your platform based on what they decide to support.
Same thing with hardware. We complain up and down about hardware not being supported, yet we often choose for the sake of convenience or economy to use hardware that is not compatible, rather than taking the time to purchase hardware from and support those vendors that do make an effort to provide linux compatibility. So where’s the incentive?
If the linux community wants to induce change in the market, they need to be willing to make the sacrifices and suffer the inconveniences inherent with choosing linux as their prefered platform. It means not falling back and purchasing Windows applications, and it means going to the trouble of researching your hardware purchases rather than buying the cheapest brand-name hardware at Best Buy. Money is the universal language of business and it’s something all vendors understand, and it is the only leverage the community has available. Start supporting linux-friendly companies exclusively, and other will take notice, it’s that simple.
As for OEM system support, we keep looking at that as a magic bullet, but it ain’t gonna happen, for similar reasons. For a company like Dell or HP to suddenly pre-install linux, it’s not simply a case of pre-installing linux. If it’s pre-installed they’re expected to support it. What’s the point of paying a premium for a name brand system if they tell you you’re on your own when there’s a problem or something doesn’t work? This implies an investment in backend infrastructure, for both technical and customer support. Economies of scale come into play, and for them to be able to cost-effectively offer linux based systems at pricing parity with Windows they would need to confidently believe they could sell units in large volumes. Can anyone give them that assurance? At this point it has nothing to do with Microsoft controlling the OEMs, it’s simply that Microsoft provides a much more cost-effective solution for the OEMs.
I don’t think it’s all in vain though, there are signs of change in the air if they can be executed correctly. If Novell generates significant sales for SLED in enterprise use, that can provide application vendors with concrete metrics for a paid customer base, which is a start.
And as much as people like to extoll GNOME as the enterprise desktop for highly debateable reasons, KDE 4.0 has the potential to really bridge linux with Windows, since it will offer cross-platform compatibility. Being able to run fully integrated KDE applications on either Windows or linux (and likely OS X down the road) helps reduce the pain points of any transition and, at least for smaller software ISVs, simplifies development by providing a single framework for multiple platforms. KDE 4.0 will evolve from being a linux desktop, to being a cross-platform application environment, which interestingly provides an opportunity for a little bit of embrace, extend, extinguish of our own, albeit on a much smaller scale. Lowering the barriers reduces the level of investment required for the ISVs, which changes the dynamics of ROI equations and similar things. (Please save the Qt licensing cost arguments, they’ve been beaten to death, we’re talking about ISV’s that don’t hesitate to spend money on development tools and if KDE/Qt can provide a cost-effective development framework, the cost of one-time Qt licensing will be so incremental as to be insignificant) Of course this is also predicated on KDE’s ability to execute properly on 4.0, but given the amount of planning and development going into it, it looks very promising. Which is likely why Novell is still investing heavily in them as well, BTW.
Most of all, I think people just need to quit expecting linux to act as a replacement for Windows, it sets false expectations. Consider it an alternative, look at areas where it is already strong and focus there instead of dwelling on the weaknesses that will continue to hold it back. We need to quit whining, stay the course and continue to support our choice of applications however best we can.
I’m not a free software zealot, I’m quit comfortable with mixing closed/proprietary software with OSS where each makes particular sense and offers me the best functionality. But that said, I do believe that linux must always be accessible and community driven, zealotry needs to be minimzed because it carries a significant cost in terms of credibility, and we need to find a way to bridge community and commercial interests since there’s significant return for both if everyone can just act rationally and look at the big picture.
End of rant….
Sorry but your rant is full of nonsense and contradictions.
The biggest problem facing Linux desktop adoption has a lot to do with MS.
Also, this “article” smells like a troll, especially since it implies Linux is not mainstream and has poor adoption rate, and leave it to the reader to understand we talk about the desktop only.
Saying that the current users are a problem for the adoption rate is plain stupid too.
I can’t read that from someone and not think he’s a troll. After episodes like ODF, the european $1 billion fund, antitrust lawsuits, and other such things, you have to be stupid or purposefully ignore everything, and have some deviant mind to then say the current Linux users are the problem and not MS.
There seems to be an attitude of entitlement from many in the community that does not translate well for profit-driven commercial vendors in a competitive market
And yet it it no problem for many of them that sell Linux appliances, servers, embedded devices, … This is plain FUD.
People genuinely act surprised that the vendors haven’t tripped over themselves to invest millions in designing, testing, releasing and supporting commercial linux applications or hardware
The people that feel so surprised about these things are never part of the community, how strange !
Where did you get these millions people that extoll free explications and slag “Microshaft Winblows” ?
You don’t even have metrics and network and pull huge numbers of very negative people out of your a**.
You sure wouldn’t pull a high number of positive people, that would destroy your argument based on nothing.
You talk about the Linux user base that is not the Linux user base.
Do you even realise the one that say “I’d ditch Windows if it wasn’t for this or that” are not the Linux user base ?
You seem not able to.
You believe your own suppositions, which makes your rant look very stupid.
You ask what incentive have companies to develop software for a market they don’t even know about : but this has been debated and explained a lot of time already.
If the company don’t sell the product now, one day it will be too late, when the free software equivalent is good enough. There are plenty of examples.
You are saying hardware vendors have no incentive when we complain about their hardware that don’t work on Linux.
But we have at least to let them know every time they lost a sell because of no Linux support, you still can bring back what you bought then.
What tells you the Linux community doesn’t support vendors that sell Linux compatible devices ?
I always do this. For example, when I recently bought a CPL solution, I said to everyone Devolo was top notch for Linux, they even provide open source drivers.
When I created my Linux PVR, it was different, as no PVR card has real Linux support. You still have to buy one that works on Linux thanks to the talented community, and then complain to Hauppauge. I still complain to some web site that don’t work in Firefox, and say they lost a customer.
The OEM suppositions are BS too. In France, HP and another vendor already made tests of pre-installed systems with Linux (with Ubuntu and Mandriva).
And it was cheaper than the MS Windows equivalent. Not installing an OS on the laptop just can’t be more expensive than installing one, that’s just BS.
There was no more backend infrastructure to invest in than what they have already, this is BS too. Hiring one technical that can do Linux, and sure enough can do Windows support too, is not the deal breaker you talk about. Especially since Mandriva and Ubuntu did the support in these cases, so zero infrastructure cost to HP or the other vendor. Unless you think redirecting Linux users to Ubuntu or Mandriva is so costly.
The cross-platform compatibility is just nonsense too. Anyway, Gnome is cross-platform, perhaps more than the current KDE.
What you hear then ? Gimp has an unusable GUI. People forget to tell you they tried it on Windows of course. Same for GAIM.
The contradiction is that you talk about Linux tools on Windows and ease of transition, but you were talking before about the current Linux user base that has Windows as a fall back or that didn’t even use Windows (“I’d ditch Windows if …”). These very people you talked about are actually the Windows users that tried some Gnome or Qt tools on Windows.
You say we need to quit whining, but you just have no explanation why : I’ll never trust people who say anyone in the community “need” to stop this or that.
Most of the time, these people are FUDders or trolls. We don’t need any of these wrong advices.
Not long ago, FOSS world was in far worse shape than it is now. Thanks to all the “whiners” you would want to shut up, we advanced greatly.
Some years ago, people were laughing at the FOSS community, now, strangely, a lot of people tell us we “need to stop”, “need to quit” this or that.
No we don’t, this is not a religious community with one big brain, everyone is free to whine or not.
And we don’t need to lose time reigning in the “bad/angry people”, we just have to tag along like we did before, and beware of people that call us “zealot” or say we “need”.
You’ll notice some people can never talk about FOSS users as anything other than zealot, they will even find millions of us are zealots (that’s for not saying we are all zealots), they are easy to spot.
Of course, these people will remind you that we “need to” minimize them, lose our time on few people that do no real harm (but they’ll remind you that the “zealots” have a significant cost in terms of credibility), while MS puts huge ads and say things like Linux is for “terrorists”, or “anti-capitalist”.
But hey, believe us, your zealots are the problems, not MS !
These people will recognize themselves (I’m not even sure of that), these people want to reign us in actually (how dare you use Linux or FOSS, are you nuts ?).
End of rant
Very good article. Just one minor point, though: The writer says:
“Fortunately, some of the PC OEMs, like Dell, are now pushing their upstream suppliers to provide Linux drivers for their equipment.”
In my opinion, Dell is still doing very little to improve Linux hardware support. Dell is largely responsible for me switching from Linux to Windows, after using Linux almost exclusively for four years. I bought a new Dell B130 laptop, on sale from Dell, a really nice inexpensive laptop with decent quality and all the features I needed. It even worked well with Linux, until…. I tried to use the modem with my dialup internet. I absolutely depend on my dialup internet, as nothing else is available where I live. The Conexant HDA winmodem in my B130 has no free drivers available for it at all. LinuxAnt offers to sell me a Linux driver ($30) for the hardware I already own, but it is total junk. The connection quits regularly every 5 minutes or so. Unusable. So, I’m back to Windows XP Home, which I hate, but I really have no alternative. I wish Dell would really support Linux on every component of every device they sell.
I’m a developer for primarily non-Microsoft systems. I have had to write software in Visual Basic, however. I do not find MSDN anywhere near as developer friendly as just typing my questions into google and getting back a useful response. In fact, I have always struggled with MSDN. Just goes to show you that one man’s benifit is another man’s drawback.
The article editor has some good points.
1. MSDN. While it provide the best for subscribers, there is a lot available free like SDK’s, samples, documentation, and other. While Linux provide a lot of documentation on the net, sometimes is very difficult to find it. Distributions and OSS projects should join forces in creating a common site to handle that, and also allow users contributions.
2. Common Interface -> More difficult to do than #1. Each OSS project provides it’s own vision and it’s way to implement it. That’s the flexibility Linux provide. Providing a common interface force a special look to most people. IMHO the common interface is a distribution job instead of each project like Enlightenment. Mac OS X do not provide the same look & feel or common interface than Windows. Does it affect them somehow?
3. While MS Office is the mayor office suite, flawless compatibility with its file format is a must. When WordPerfect, and Lotus 123 where the champions, Microsoft provide import filters in it’s office suite. Simple, if you are not compatible, you will not be able to penetrate the market.
4. Marketing is also difficult, specially for the Free projects like Debian, Gentoo. The big corporate ones like Novell, Redhat can try, but it will take a lot of resources and probably the gains will not be that significant.
5. OEM support depend on us. If a hardware do not work on Linux, either reverse engineer the driver (takes time), or simple spread the word in the net to simple do not buy that, and go another route.
Marketing sucks, it influences people for the wrong reasons using intrusive and dubious methods. Why would gnu/linux want to be associated with that ?
Marketing sucks, it influences people for the wrong reasons using intrusive and dubious methods. Why would gnu/linux want to be associated with that ?
All the reason why Linux will not go mainstream. People are suckers for marketing. Most companies seem to need it and I’m sure its for a good reason.
It’s been said, but I’ll say it again.
I’ll take an officialy configurable GUI before common GUI any day.
By officialy, I mean things like changing the window manager, mouse focus behaviour or keyboard shortcuts without going to the registry, or using 3rd party software.
I more or less agree with the other points of the original article.
“Linux, on the other hand, follows the old Unix model of using many simple, small programs, libraries, and APIs (application programming interfaces) to build more elaborate programs. The majority of Linux and its applications’ source-code is relatively easy to read, understand, and debug.”
Maybe not 100% related but a lot of the UNIX apps seems more or less broken quite often, for example amarok a while ago and just recently the rhytmbox version in my solaris nevada build 42 + vermillion 45 installation, for whatever reason the whole app just loads and loads and loads and allocate more and more memory until absolutely nothing happens when I have closed it with all my MP3s added. There is quite a bit of beta experience for many apps even in released versions.