Slashdot Mirror
Linux On Netbooks — a Complicated Story
An anonymous reader writes "Keir Thomas has responded to the recent raft of news stories pointing out that Linux's share of the netbook market isn't as rosy as it used to be. Thomas thinks the problem boils down to a combination of unfamiliar software and unfamiliar hardware, which can 'push users over the edge.' This accounts for the allegedly high return rates of Linux netbooks. In contrast, although far from superior, Windows provides a more familiar environment, making the hardware issues (irritatingly small keyboard, screen etc.) seem less insurmountable; users are less likely to walk away. 'Once again Microsoft's monopoly means Windows is swallowing up another market.'"
My gf knows that Linux is on her computer, but even so, she can't understand why she can't go to BestBuy and get software. Or why she can't download Silverlight. If you put Linux on a machine and don't explain the difference between it and Windows, then you're just asking for trouble.
I am no fan of Microsoft, but it's not like they are doing anything illegal or unethical here. Even Redhat's CEO commented he didn't believe in Linux's desktop future.
Frankly, netbook looked like worth a shot for Linux. If it fails, then maybe desktop market is just too hard for Linux to win.
Of course the actual reason Linux's share of netbooks has dropped is simply because netbooks have changed from a nerds' thing into a mainstream thing.
UNIX's marketshare of all computers did the exact same percentage decline over time as netbooks are having now. It's the early adopters, stupid!
The United States of America: We do what we must because we can.
That's "effect", not "cause." 9 months ago, there were tons of Linux netbooks in stores. They've gone away because they're unpopular, and get returned a lot. (Well, I can't speak for Germany, but that's the case in the US.)
Comment of the year
A large percentage of Windows users do not understand what an operating system is and assume if they can buy it in a store, it'll work. Manufacturers need to put giant stickers saying:
Not a Windows system, does not run Microsoft anything, none of your programs will work on this, Apple* made it.
*that is a lie, but Mac users won't be on the cheap end of the aisle.
Not that I think it will help much. I've had too many acquaintances think "ooh, cheap computer", buy one, and then ask me if Microsoft Ubuntu is newer or older than Office 07, and if it will run Vista Excel.
They usually end up returning it and I buy another bottle of aspirin.
Windows has something Microsoft once identified as critical mass in the market. It was no accident that they arrived at that point. The choked, cheated and killed IBM's OS/2 making it the only desktop operating system for PCs. Had Linux begun to mature during that era, we would be telling a very different story as Microsoft would never have achieved critical mass.
What is critical mass? I am probably wrong or incomplete in my understanding of what that means, but to me it means they control enough market share that every software and hardware vendor must heed what Microsoft says and does or face the consequences. It also means that all users have come to expect only one user experience and is cursed to be unaware of other options and what they mean. When they don't get what they expect, they believe something is wrong.
People are okay when that "something else" is Mac OS X. They know it is different and usually comes on an Apple branded PC. It is a conscious decision that users make and are aware that "It's not Windows."
Just keep chipping away... keep chipping away. Eventually Linux will begin to mean something to users. It may mean the equivalent to the pictures that come in wallets, purses and picture frames. It may mean something that works, serves its purposes and doesn't get viruses. It may mean something that kinda works, but everything they want isn't quite available yet.
One thing that changes user perception is "standards compliance." Users don't have a clue what that means, but if it works fine in Windows and not in Mac OS X or Linux, the PERCEPTION is that there is something wrong with Linux and Mac OS X. The more pressure put on Microsoft to comply with standards on the web, the greater the possibility that alternatives could be perceived as viable.
"Critical Mass" means that people think it's the standard. "Critical Mass" means it is the defacto standard. Toppling a standard is no easy task.
One thing you should do is thank Linux for forcing MS to keep XP available for you at $15 instead of the normal OEM of $70. It is better for customers and hardware manufacturers that Linux is available as a viable alternative.
One of the problems that I see in the Linux world is that many of us are quick to cry "monopoly" and blame it on unfair practices.
So if it's because of Microsoft's dominant market share, why does Apple do so well in the markets that it is in (at least in terms of return rates)?
Blaming it on Microsoft is a cop-out because it lets people avoid the harsh reality that the fault really lies with Linux. Linux is far, far from passing the Aunt Tillie test. Ubuntu is nice in that it's trying to be more consumer-oriented, but so far, most of its changes are superficial.
And finally, one person's "superior" is another person's design flaw. Apple is "superior" and "innovative" (that's debatable) mostly because Apple doesn't give a damn about its ecosystem. Microsoft does. It bends over backwards and even consciously duplicates buggy behavior, all in the name of backwards compatibility (given the HUGE diversity of software and hardware in the Windows ecosystem, the (relatively small) amount of breakage between each version of Windows is actually a testament to Microsoft's ecosystem cultivation). Is this technically superior? Probably not from an orthodox perspective. Does it make sense? I think so. THIS is why Microsoft has its monopoly. Until Linux can start cultivating such an ecosystem (no, telling someone that they can just download the source and compile it for their system does not cut it), it will always remain on the sidelines. Period.
I think people would like Linux more if they were familiar with program names. Notepad, Paint, Wordpad, Calc... whatever. When I boot Linux on occasion, I'm more confused with what program does what than how to use them.
While I applaud the work of thousands to build such robust amazing programs and give them each their own special name, I'm of the opinion that if you give someone KDE with a few programs labeled generically "email" "internet browser" "calculator" "text editor" "Office Text/Spreadsheet/Presentation" "Network - Wireless" "Printers" and so on and so forth instead of each programs' real name, you'd be a lot closer to the #1 goal of usability: making an intuitive interface.
My favourite computer still is my 1024x768 screen 12inch iBook.
It is ONLY acceptable because of the UI feature that quickly shows miniaturized versions of the windows of all my running applications, and lets me pick one and get back into it in one click. That gets rid of most of the need for a large screen.
And the iphone ui is optimized for its screen size, etc.
Linux might do better on netbooks if a similar gui optimized for the screen size was available and worked well. I understand a few of these may be available but haven't tried any.
Have to say I'm holding out for an Apple netbook. UI of MacOSX is too much better.
I am an extreme comp-sci geek, but I have way better things to do than configure the low-level settings of my laptop.
Where are we going and why are we in a handbasket?
Bad thing about eees with Xandros is that it sucks. And people who'll see it would think that Linux generally sucks. MSI had a more acceptable pre-installed Linux offering; that's why return rates are about the same with Windows and Linux MSI Winds.
Is it any wonder why so many people are used to the rules and non-regulations of a Windows machine?
That being said, every flavor of Linux I've tried has some different scheme to it, making basic operations unnecessarily complex.
Standardizing basic ops like install/uninstall, media player/ect. would be a good start, but probably terribly unrealistic among mainstream distros.
Maybe there is a misunderstanding by you of what your companies' needs are. Most of the time more is involved in running a corporate network than hooking computer to router with NAT like at home. Many companies utilize Active Directory to administer their network and switching workstations to Linux makes that somewhere between extremely difficult and impossible.
Second, companies are averse to changing anything that currently works adequately because there are (usually significant) costs involved in moving to a new platform in the form of testing, rollout, training, and support time.
Finally, even outside of corporate environments you need to provide a compelling reason to switch to something else from what's been in use for a long period of time. Debatable feature parity simply does not qualify.
Linux is for Geeks and it always will be. When I built my own mini computer with a micro atx mb some time ago I was curious about how linux would perform on it vs Windows 2000 and Windows XP. XP was a total dog. The processor just couldn't handle it. Win2k ran nicely. But when I tried to install Fedora I quickly realized it was impossible. The Fedora install didn't recognize the hardware and wouldn't even boot. Of course any geek will tell you just reconfigure the discombobulator and invert the thingamajig and then recompile the root and then burn new install discs and then boot from Alt-Shift-Tilde. Which to a non-Geek is the equivalent of performing brain surgery on yourself. What the hell are you talking about? I have to do what?!? I had to hunt around on the web for several hours just to find out what the problem was, then after reading all the various explanations of how to do it and telling myself I am a software engineer and I have no idea what I just read, I finally found someone who wrote a hack to boot the system, then allow the install disk to run. And after all that I found that it ran worse than Windows becuase it wasn't optimized for the cpu. So I went back to win2k and was happy ever since. Which is why Linux will NEVER be ready for laptops or desktops or anything else besides servers. Because only Geeks have the patience to hack around and kludge it up so that it works. I just want to turn it on and have it work so I can get on to more important things like doing my job or searching for pr0n ;) Oh, and I just bought a Dell mini 9 and I didn't hesitate for 1 second between choosing Windows over Linux. It was worth the difference in price. And if want to install another OS it will be OS X and NOT Linux.
...and make it ubuntu.
I've stopped recommending people try linux. Rather, I recommend ubuntu, 'a type of linux'. Trying to explain how each distro is built on the linux kernel and uses a specific desktop environment tends to send people running so instead, I rave about ubuntu - which they can then go and test right away from a live CD.
On netbooks, the problem of different distributions is amplified by all the custom distros. As much as this is open source in action, it splintered the 'linux' option, especially when XP always comes as the same recognisable package. Hardware manufacturers could have put the effort into ensuring upstream hardware support and supporting key software development (ooo.org, ffox, rhythmbox? mplayer?) rather than developing their own "OS".
I think we would be seeing a different story if customers were offered ubuntu as the option to XP across the board.
... if it weren't riddled with fanboyism and aggressive language.
In contrast, although far from superior, Windows provides [...] Once again Microsoft's monopoly means Windows is swallowing up another market.
Wrong. Fail. Abort. Windows is swallowing up another market because Linux doesn't belong on the average user's netbook, for the same reasons it doesn't belong on the average user's desktop. It is a usability nightmare, you need to be a network engineer AND programmer to fix it when it breaks, and perhaps most importantly the Linux community is hostile and unhelpful toward non-techies.
I am a network geek and programmer, and I still get pissed off at Linux on a daily basis because things that should just work, do not. Usability issues never get addressed, no one wants to touch them. "My app is fine, go fuck yourself" is the general attitude I see among app developers/maintainers. Maybe they're sick of replying "RTFM" to every single question, but to me that is a symptom of bad code. Joe Random doesn't read the README, nor should he need to. If you can spend the time to write a long, complicated README, you could spend that same time writing a small script that does all those contrived pre-installation steps for the user.
The problem is that we programmers are terrible users, because we don't use computers the way non-programmers do. The goofy little apps and utilities I make for myself, they have the most spartan, militaristic interfaces because I write the code first, then wrap buttons and knobs around it. I know how to use my stuff, because I'm the guy who built it. I know which bits of code fire when I click this or type that. Joe Random does not. We need to fix our apps to be so intuitive, even Joe Random's retarded stepchild can use them.
The netbook does not matter. Other than the size factor, it is hardly different from 3-4 year old laptops, and like any laptop, usability is top priority. If we want Linux to rock netbooks, we need to make it usable.
-Billco, Fnarg.com
no.
Wealth is the gift that keeps on giving.
I think people would like Linux more if they were familiar with program names. Notepad, Paint, Wordpad, Calc... whatever. When I boot Linux on occasion, I'm more confused with what program does what than how to use them.
Excel, Visio, Quicken, Outlook and Visual Studio aren't exactly self-explanatory.
While I applaud the work of thousands to build such robust amazing programs and give them each their own special name, I'm of the opinion that if you give someone KDE with a few programs labeled generically "email" "internet browser" "calculator" "text editor" "Office Text/Spreadsheet/Presentation" "Network - Wireless" "Printers" and so on and so forth instead of each programs' real name, you'd be a lot closer to the #1 goal of usability: making an intuitive interface.
In the Applications -> Internet menu from Ubuntu on my EeePC, I have "Firefox Web Browser", "Mozilla Thunderbird Mail/News", "Pidgin Instant Messenger", "Transmission BitTorrent Client", and several others. Compare with the Windows debacle of Start -> Publisher -> Weird Program Name.
I agree with your point, and apparently so did the distro maintainers a few years ago that made Linux much better on this count than Windows.
Dewey, what part of this looks like authorities should be involved?
There are many reasons to care, but I can think of two that are general enough that they would apply to most people.
1. More software, particularly polished software. It doesn't have to be commercial, but the bigger the audience, the more people who will be interested in helping bring that software to Linux.
2. Better and supported hardware for Linux. This has gotten a little better over the years, but the more people who have Linux in the general population, the more time that vendors will spend on both creating compatible hardware and writing good drivers for it.
Both of these affect you, the the current linux user.
Now, it's true, you may be all set with what you are running now and happy with it. The thing is, people in the past had computers that did exactly what they wanted them to do too. The first ones broke codes and created ballistics tables. Eventually, there were enough of them, and enough understanding of them that people had time and experience to do other things with them. Printers, networking, better storage devices, business software, games, the Web.
The same thing goes for Linux in a more specific way. The more people who use similar things to you, the more potential value your own Linux box has. Someday, you may surprise yourself and find that your Linux box that you were happy with, can now do something else that you never thought possible and it will improve your life. You had nothing to do with it, so someone else had to create it. What allowed for that was the increase in "market share", "penetration", "mind share", or whatever you want to characterize it as.
Linux simply isn't ready for the mainstream consumer market. Few hardware devices and peripherals come with Linux software and drivers. People shouldn't have to search the Internet for drivers. All Linux consumer applications should come with easy to use installers which will work with all of the major Linux distros. People should never have to install drivers and software from a command line unless they want to. I sometimes use Linux, but it's not my primary or secondary operating system.
Blaming the person using the computer is the reason we still have 12 step processes for codec installs on many distros. Often, these things don't become more intuitive, we just blame a gov't conspiracy or call the end user a lazy moron and tell him to RTFM.
Blame it on the end user or blame it on MS, it doesn't change the fact that the people that are receiving these netbooks with linux don't want them. If people want linux to be accepted, they need to make it more appealing to the target market.
Personally, I could care less if it gains in popularity. I use it for what I need it to do. I'm not concerned with what Joe Enduser has on his netbook.
The likes of the big box stores - Best Buy in particular - are not selling the Eee with Linux for a couple of major reasons:
1) They can't sell support for it because ... and ...
a) they don't have anyone who can offer support for Linux
b) there's precious little to support which can be charged
2) They can't sell software for them, because there isn't any.
3) Being a lower-priced item, I'd guess there's a lower profit margin.
So, basically, there's business case impetus to "stick with Windows". I mean, seriously: for the kind of person who shops at Best Buy, which would sell better: that it has XP, so it's familiar, or it has Linux, which is free and secure?
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
Why should you care if your neighbor is using Windows, or something else?
There are thousands of good reasons, literally, but here are the ones which might stick in your maw more readily:
* Conficker
* Sasser
* Bugbear
* Blaster
* Melissa
* Love Bug
* Code Red
* (insert next bug here)
It's not so much a matter of why they should use Linux (or OS X) but why they shouldn't be using Windows. It's bad for them, their data, their friends and family, and their data. And it makes many a professional IT person suffer through cleaning up their messes. (How do you think water sanitation workers would feel if you shoved caustic substances down -your- toilet?)
Yes, I realize that as Windows/IE loses market share, other software will start being targeted. But as that happens, software variety will increase. A larger software ecosystem not only improves the quality of various competing projects but it results in a much higher cost of doing business/lower return for spam, spyware, and worm writers.
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
how does it help if there are more ex-windows newbs on ubuntu?
Those guys are going to get angry if a main service does not work on linux. It creates pression groups that tends to get more open interfaces. Internet Explorer specific website tends to die due to other web browsers. Perhaps one day we will no longer need MS office compatibility.
Is usability/training/help for non-technical users. Technical issues aside, one of the ways Photoshop kills GIMP is in usability. This doesn't just mean the UI itself, it also means the materials available to help you learn about it. Adobe has some first rate stuff. Their help files are quite helpful and include things like pictures of what to do, they have online video training docs, and their books are awesome. That goes a long way to making it usable for the non-tech type. My mom is an art teacher by profession and while she's not scared of computers, she's not good with them. She needs things spelled out step by step. Well she tells me that Adobe does a good job of doing just that. She's found it easy to use. She knows the art aspect of what she wants to do, and their docs help her figure out how to make the software do it.
That is something major that much OSS seems to lack. The software itself might be badass, but if it is hard to use, it'll be a geek tool only. Normal users aren't going to read text man pages, Google through newsgroup posts, and try stuff on their own to make things work. They need an easy experience. To them the computer is a tool, not a toy, thus it'd better be easy to use because learning how to use it isn't fun for them.
So if Linux ever wants a big share on the desktop, that is something that is going to have to happen. All the common tools that people use will have to be nice and easy to use, and nice and easy to learn about.
However, none of that is going to happen until, as you say, they start taking a more realistic look at their products.
Exactly! The industry will need to settle on maybe 3 desktop distros (light, medium, pro) before there will be enough de-facto standardization for driver writers etc. to bother with.
Table-ized A.I.
I had a friend who won an EEE netbook, which came with Linux. He liked it until he tried to use it on a website that contained Flash content. This netbook did not have Flash pre-installed, and he was unable to install it himself by following the links on the website, so he asked me for help. I am not very familiar with Linux (I try a distro each year to see if it is up to my standards yet). I was unable to determine what type of package the Linux distro on this machine supported, and was unable (through trial and error) to get any of Adobe's Flash for Linux packages to install. I ended up having to do it manually from the command line with an APT GET command that I found after a Google search. That is not an acceptable user experience for customers who expect to use this computer the same as they would use any other computer. And that is why you can expect high return rates for Linux Netbooks.
- James
Excel, Visio, Quicken, Outlook and Visual Studio aren't exactly self-explanatory.
They don't need to be; everyone already knows what they are.
And here we are already, back to the problem pointed to in TFA.
The problem with your argument is that everyone already *knows* what Excel, Quicken, Outlook, etc are. Those brands are mainstream. As a result, they can get away with not saying Excel Spreadsheet and Outlook Mail/News and whatever in your start menu. The Linux programs can't, so they should be as self-explanatory as possible. Instead of Firefox Web Browser, it should just say Internet. Instead of Mozilla Thunderbird Mail/News, it should say E-mail. And so forth. The average user doesn't care what clever name you chose for your mail client, they just want to be able to get email working on their newly-bought PC with minimal fuss.
Because a bigger market share means that more companies provide Linux drivers for their hardware products.
I am a software developer, and the guy all my friends bring their PCs to. I LOVE the UNIX environment, and HATE the Windoze environment, and I have an 8 port KVM on my desk with Linux, Mac, Solaris, and Windows at my fingertips.
I always use Linux or Solaris for server stuff if possible. To me putting a database application or a WEB server on a Windows box is just silly.
But I am typing on a Windoze XP machine. (Vista does not exist in my world) I seldom use the Linux machine because I always need to switch to the Windows machine for something the Linux one cannot do, and there is nothing the Windows one can't do, so I just end up on the Windows machine. The MAC is nice, and I use it occasionally, but it too just cannot do everything I need, or is more frustrating to get it done. When I first got the MAC, I thought it was so cool and used the hell out of it for a while, but simple things can be quite hard to do sometimes because it tries to be so "easy" that it can become hard to do anything the MAC guys did not think of an easy way to handle. So back to the Windows machine... again because I need to do something the MAC has difficulty with. As a desktop, Solaris is useless for most stuff.
It is like a trap in a way. Once I go to the Windows machine for one application I cannot do easily or at all on whatever other machine I was on, I just start doing stuff on that machine, and soon forget about the others. Cygwin and PuTTY do not help either because with that working, grep, find, and ssh is there, and I'm just done. (The Cygterm hack is my console, cmd.exe is dreadful.)
I really wish this were not true, but ...
If MS's "monopoly" is really the cause of every "Linux on the desktop" setback then you might as well give up because MS's market share isn't going to go down if Linux can't grow.
Rather than use the monopoly excuse, Linux fans should figure out the specific reason for the setback and try to address it.
Or you could just sit on your hands for another 10 years and say it's all MS's fault.
Has anyone entertained the thought that people might actually choose windows because it's the best available option?
Part of the blame are the bug-ridden cheap distros that come with the computer.
I've seen a whole community trying to help a guy to get his notebook, mainly the wi-fi, to work.
Days later he gave up and installed Ubuntu. It just worked.
'Once again Microsoft's monopoly means Windows is swallowing up another market.'"
Which is why, if the rumours of Microsoft giving XP to netbook manufacturers is true, they are guilty of predatory pricing which is basically summarised as discounting heavily with the intention of forcing a competitor out of the market.
Open and shut case really although it'll probably take the EU stepping in to do something about it.
Hmmmmmm..... Deep fried and look like Squirrel.
Only if you've spend ages trying to learn how to use Photoshop first.
Coming to BOTH new, they use different paradigms and GIMP on Linux beats PS on Linux, whilst GIMP on windows loses to PS on Windows.
Photoshop SUCKS in usability.
Once you've learned how to force PS to do what you want, or, more commonly, learned not to do what PS won't let you do, you are not able to determine GIMP's usability. Each think you've learned how to make PS do, GIMP may not do the same way and that's considered bad. There are some things GIMP won't *do* that PS does. That's considered very bad. But you don't try to do things anew and so you don't realise that there are some things GIMP lets you do that PS doesn't. And because you don't TRY (you've learned from PS not to try) you don't see this as very bad for PS.
So you're not going to see the downsides of PS since you've learned to move around them.
PS sucks.
GIMP sucks.
It's a very sucky problem to solve. They solve it in different ways. But once you've learned PS, you want GIMP to do the same.
Don't.
Learn anew. Then you'll find yourself swearing at PS when you want to render that outline as a repeated pattern rather than as one of the selected options of line stroke. You'll wonder why there isn't a good macro language for scripting manipulation of your 10,000 stock images. PS will be shown to be sucky.
"is market share the goal? If not then don't begrudge windows for providing an end user experience that is preferred. "
meh. if windows did that i wouldn't begrudge them. In my experience, though, windows market share is not built on direct competition on experience but on:
-Your data will disappear into a black hole if you use components from any other platform
-The protocols (i.e. communication *rules) that windows uses are property, and if you try to figure them out you'll go to jail.
My turnips listen for the soft cry of your love
"Linux .. is a usability nightmare, you need to be a network engineer AND programmer to fix it when it breaks, and perhaps most importantly the Linux community is hostile and unhelpful toward non-techies"
You're talking total nonsense. Any modern Linux distro is perfectly usable. And to fix it when it breaks, which is a rarety - is just as easy as the Install-Program option in Windows.
As for the hostile Linux, you are equally inaccurate in that statement. Join a forum, politely ask a question and get a response else pay for a support contract.
"I am a network geek and programmer, and I still get pissed off at Linux on a daily basis because things that should just work, do not"
Maybe you should try an other occupation?
If it's more familiar, it's more usable, and therefore superior for users.
It's as if linux advocates measure "superiority" as an intrinsic quality, whereas users see superiority in terms of usefulness to them.
A few of the problems with Linux becoming more mainstream is:
a) Linux doesn't have any ads. People feel that a product that is advertised to them is superior then the one they have either never heard of or barely heard of. It's not unheard of for a company to make a no name product and turn around and make the same identical product for a brand name with only the tags being the difference. When people are looking at the 2 of them they will assume that the brand name that they have heard advertised to them is better then the no name just due to the familiarity even though it really is the same product.
b) Linux is free, Windows isn't. That effects the perceived value. This is one of the things that hurts Linux and helps other companys like Windows and Apple. People like to assume that the more they pay for a product the better it must be since you get what you pay for. If it's free then it can't be very good is the typical mindset.
c) Linux isn't Windows. This is the biggest issue people have when they get Linux the first time since a lot of people like to call Linux a Windows alternative. They hear Windows and assume it has to be an exact clone of Windows and freak out and declare its garbage because it isn't a clone of Windows. I find that the best way to help people use Linux is to tell them up front "This isn't Windows, it won't always do the same thing as Windows. It is also a brand new and difference OS then Windows so there is going to be a learning curve since it isn't Windows" I also point out that every problem I've had with Linux (Ubuntu 8.10 in my case) has been solvable by just googling "How to do X on Ubuntu 8.10" (that tends to be the biggest thing to help them relax). Giving these bits of truth from the start are good so they don't come into Linux with false idea's and feel like they were lied to and deceived.
You will fail to use Linux optimally on laptops for these reasons:
1. hardware is too new, and uses all proprietary hardware that is designed to be "made for vista". Vista was designed with DRM in mind, the hardware is going to be bent in favor of closed source.
2. hardware sold with linux is designed to be "cheap" or "low power", which is not really the best market for linux, although it can do that. Linux users are no dummies, and the dummies are the ones returning the laptops (or netbooks).
3. Linux is best at driver minimization and unification. If you can get linux to run on hardware, that means the hardware is popular, is going to be well supported, and is non-proprietary meaning you should be able to find people that are familiar with it enough to fix it. Linux should be the watermark by which to determine if hardware is reliable. If it is "wintel" designed, it is going to make use of memory-sharing hardware that sucks down the CPU to add more features.
4. Linux is best on used/older laptops, people who pay for new hardware are the fools who pay for the bugs, every linux nerd knows this.. The market analysis data is flawed considering how it determines market adoption, by new laptop sales.
Just say no to license servers!!
I've learned that many people will move to Linux if the have the right help. In the past year I have helped over a dozen people with their Windows addiction, and now they are happily running Ubuntu 8.10. In the end it really comes down to users having a friend they can goto with their questions.