I recall the marketing attempt to call a bunch of stuff "Polyserve". I worked in Digital UNIX Software Engineering from 1995 until 1998 and the entire time I was there, the two projects were called AdvFS and Trucluster in the project builds and were commonly known by those names - but NOT Polyserve - in the engineering circles.
You are right on with this. HP wanted to close Spit Brook Road entirely. Several of the original AdvFS and TruCluster developers were long gone - I knew quite a few of them, and I used to regularly eat lunch with one of the early AdvFS developers. This was about HP getting rid of their former Digital (called Compaq) engineers and closing facilities. They may not have saved money by dumping AdvFS and Trucluster projects by their cost, but by closing Spit Brook Road, I am sure they saved plenty.
I worked in the Digital UNIX engineering group from 1995 until 1998. The name Digital UNIX was already being used at the time that I joined the group, though the UNIX product in the Version 3 era was called Digital OSF/1. Digital UNIX V4.0 definitely used the UNIX name because it passed all of the branding requirements. I cannot recall if we also named V3.2 Digital UNIX or not - that was around the turning point from OSF/1 to UNIX in naming.
You are wrong about Compaq making the name Tru64 UNIX. We decided that before Compaq grabbed us. Whether we had a release out the door, perhaps not. Internally we calleed V4.0 the Platinum release. We called V5.0 the Steel release. I left after Build level 16. There were at least 20 builds in the release. Sometime near that time Compaq acquired Digital, so that is why it appears that the Tru64 name is a Compaq name, but that is not true, the Digital engineering and marketing people came up with the name.
Reminds me of a fun story. We had some cool sweatshirts made. We wanted to call it butt kicking (and a bunch of other stuff) Linux, but marketing named it instead and called it Digital UNIX. The engineers had a good sense of humor and a great deal of pride in the OS. Too bad the rest of the corporate culture stifled it. I believe in the 1995-2000 era it was the best Linux available.
It's great that Shuttleworth is trying to improve the availability of Free (as in speech) software. But I hope he doesn't move too much efforts over to this.
The reason I use vanilla Ubuntu is because I don't have to put a lot of effort into setting up my OS to agree with all my hardware. Instead, from the start I can work more on customizing how I interact with my OS. I remember the hell I had with a Radeon x800 and Fedora Core 4 a few years ago. If "closed" (as if it's always a bad thing) software provides a better solution, I'm more likely to use that. So, I hope Mark still is going to put his best foot forward for plain Ubuntu. I bet the corporate interest is more aimed to that Ubuntu.
Like many things, this is a balancing act. Pragmatism suggests that we create something that gets the job done. We've done that for years, and we've had a lot of success with that. It is more than high time to do more than that. Yes, I want to be able to do things, access content, etc., but for a great deal of what I do, simply accessing the Internet through Web pages, Email, and news content is quite enough.
Why not work with those who can live within those constraints and try to expand what is truly free until every single component can be freely modified and improved using free software?
If that ever becomes the norm from the bottom to the top of the infrastructure, that just makes it possible to create so many more free and innovative components on top of that platform infrastructure.
We keep getting better at this, and the result is that we are starting to see more and more products based on a free infrastructure. Network centric computing, especially so-called "Web 2.0" based applications greatly accelerate the importance of further building upon what has already been established.
There is a certain amount of "nobody got fired for buying IBM" in the corporate world which results in inertia. That is most certainly true. However, when the combinations of reliability, stability, security, and cost effectiveness can be not only demonstrated, but well packaged - maybe even by IBM - that kind of perception can change in a hurry.
I do not see such a change being initiated in the US corporate world. However, if a foreign competitor develops a great advantage, particularly if it undercuts and outsells an US opponent, or if a third world country starts grabbing business from major international enterprises, that is the kind of momentum shifter that could change these perceptions - cost savings and market retention can - and just may - be the forces that cause industry to move on to something different.
No guarantee that it will happen, but at the very least, I expect Linux inertia to cause existing products to compete more effectively. If they don't, Linux really will get the inertia that it has lacked.
I think that DSL has a great niche working with really old hardware. The only distro I know of that is still actively being developed that is smaller than DSL is SliTaZ - very interesting, but very new.
DSL has an old 2.4 kernel, an old Firefox browser, but you can count on it to work with old stuff.
Puppy works with pretty old stuff, but really shines when you load it into RAM on equipment made within the past three years. Wireless support is something that Puppy handles better than DSL.
Zenwalk has a relatively unknown, but fast package manager called Netpkg and a snappy implementation of the XFCE desktop. Derived from an earlier implementation of Minislack, Zenwalk comes out of a stable Slackware heritage. With a fast package manager and a fast desktop implementation, Zenwalk carves a nice niche out of the Slackware landscape.
Arch Linux really is another distribution that once grew out of the Slackware space and has now come into its own with the pacman and AUR package management tools and the idea of giving you total and complete flexibility to build exactly and only what you want. It aims for simplicity rather than coddling the user with its own notion of ease of use. People really either love Arch Linux or avoid it for these very reasons.
Xubuntu is an easy to use system with very current software from the Hardy Heron Ubuntu project, replacing GNOME with XFCE on the desktop. Good solid stable software with excellent wireless network configuration.
TinyME is brand new, as far as a Version 1.0 implementation, but the project has been going on for a couple of years now as a community supported effort to provide lighter versions of the well regarded PCLinuxOS software. This one uses OpenBox instead of KDE. Like other PCLinuxOS systems, it really benefits from the good hardware detection algorithms from Mandriva and the solid packaging from "TexStar", expert RPM packager and founder of PCLinuxOS.
As you can see, each of the distributions mentions has a nice niche. They won't all be appealing to everyone, but each of them is solid in several respects - certainly a credit to the modularity of both Linux and GNU software.
I like this idea best of all - just make sure that you, as vendors, make sure that there are openly available drivers for all of the systems that you sell, then make an entry level option to buy them completely bare. I tend to install different stuff anyway.
I actually did buy a laptop recently with Vista on it - wanted to see how good or bad it really is. I found it to be horrible to configure initially because the OS and the vendor provided AntiVirus software from Norton were both attempting to get updates as soon as my network came up, and they totally consumed the entire laptop for at least an hour. Once they settled down, Vista was OK, but nothing to get excited about. It took about three reboots - including one in safe mode - to finally get the initial configuration properly set up. I've run Vista only three or four times since. I've put Xubuntu 8.04, Fedora 9, AntiX M7.2, and sidux 2008-01 on it. Run AntiX when I want to get on and off quickly just to read Email. Run sidux most of the remainder of the time.
I really like Dell and IBM computers (which are now the Lenovo line). I have two Dell computers at home, a Dell Dimension 4100 - 2000 vintage technology which has greatly impressed me with years of reliability and surprisingly decent performance. While it is getting a bit ancient now by computer standards and slow compared to newer equipment, it still works. Huge positive points to Dell for such a good machine. Another positive is that virtually any operating system I have tried on this home PC just works - ranging from Windows 2000 on the original box, to QNX, a POSIX oriented real time system, to SliTaZ, a 25 MB micro Live CD, to Arch Linux, an ultra flexible, but hobbyist oriented system. My usual desktops on this box are sidux and SimplyMEPIS, with PCLinuxOS getting air time when my kids use this old box.
Dell Latitude D600 - a great moderately priced used laptop system. I now use this for my every day home use as my primary system.
Compaq/HP D530 - a refurbished box, runs well for distro testing, but the case doesn't close well and the fan is noisy, so I do not use this as much as the others. Still, with a 2.9 GHz processor and plenty of space, it is a great distro test box.
IBM Thinkpad - used to have the T42 at work, was highly impressed with it. I now have the T60, which actually has the "IBM Thinkpad" label on it, but it is really a Lenovo T60 - on the inside, it does in fact say, "Lenovo T60". One of the best current generation corporate laptops. Thee laptop keyboard is better than the docking station keyboard for key feel, if you can believe it - one of the main reasons I love Thinkpads.
Given this, I decided to buy a Lenovo 3000 series laptop. The reason I did so was to get consumer features - a built in camera in particular. I gave up just a bit of speed - not much, to get a Lenovo 3000 Y410 - a Duo Core (1.5 GHz per core) with 2 GB memory and a 160 GB hard drive. Great keyboard, nice glossy laptop display, good Intel Pro Wireless 3945 network card, decent compatibility with current Linux distros.
I'd buy Dell or Lenovo models with little hesitation and probably HP models too. All three have been really reliable for me, and as end of model closeout or used models, they are an even better value.
There are all kinds of options. FreeBSD is but one of many.
The Debian project is all about freedom. Many probably do not realize it, but the Linux kernel is not the only kernel to be embraced or to be used with the Debian environment. I know the GNU Hurd project had worked with Debian at one point, so did FreeBSD, NetBSD, and OpenBSD. Those are only the ones I know about.
There are many free and open OS efforts out there and there are many free and open applications out there. The only way to slow down the movement would be to close off the networks. Even then, this movement had started before the prevalence of Broadband networks. Today's networks only accelerate collaboration and innovation. Those who are not innovating (you can figure out who they are) are gasping and grasping, hoping not to drown under a torrent of rapidly moving current.
I would say that Red Hat and other companies are probably in talks with Microsoft all the time. If they are not, they probably should be. Both companies have customers that use both sets of products, so it is in everyone's interest to communicate and to improve interoperability between distinct, independent products.
I do not see Red Hat caving in and signing any kind of patent agreement, but I could easily see Red Hat working with Microsoft if the work was in their mutual interest and in the interest of customers.
Red Hat is probably very reluctant to make those kind of deals because we have seen the kind of press Microsoft has given so far - stuff like "We will not sue this partner", yet in another breath that same day, "We think that Linux infringes on..." ad nauseum.
I do not think anyone's software should rightly infringe on anything else, because ninety percent of functions are common functions that should not be possible to patent, protect, or license, but there could be open collaborations to share the cost and benefits of implementing new features and improving the interoperability of existing features.
What we have today is a bunch of baloney. I hope all of it comes to an end soon and we usher in a new generation of collaborative software. Before that happens, there is likely to be quite a bit more nonsense, but perhaps we will get there sooner rather than later after all.
At one time, the idea of patents was created to encourage investment in innovation and protect new ideas, rewarding the founder and investor in new ideas for a set period of time. This is fine for completely new technologies, but I seriously question the usefulness of patents for well established ideas, especially when someone attempts to place a patent on an idea that really is well established and wants to own every nuance. That kind of patent ought to be abolished completely.
I can understand patents for brand new ideas, and I can tolerate them for that, though my first preference would be to get rid of patents altogether. Perhaps their time of usefulness has passed. Instead, let's collaborate more on innovations, change laws to encourage open collaboration, and only protect against closed collaborative efforts of giants who seek to squash any potential newcomers into their markets.
What I advocate is protecting freedom and protecting choice. Politically I see freedom and choice getting reduced. I wonder how much longer we will be able to speak freely like this?
I worked at Digital Equipment Corp. myself, and again as a consultant for Compaq prior to the HP merger-acquisition. I also attended a Debian conference where SEVERAL HP employees were openly favorable to Debian Linux and who were actively promoting the use of Debian GNU/Linux software within HP. From my Digital days, I know that Jon Hall, when he was still at DEC, actually brought in Linus Torvalds (I met Linus personally during one of those meetings, and Jon also saw to it that Linux had his own AlphaStation to port Linux to Alpha. So there is a long history with BOTH HP and DEC in using and promoting Linux internally.
Externally, it has always been a bit below the radar, even though HP does in excess of $2B a year in hardware, software, services, and other line items. Maybe for a $100B+ company that is not very big, but with a little more effort, they could scale that five or tenfold. Maybe it will happen if they have as much success on the desk as they have had with servers. The sheer numbers of systems sold could far exceed the server numbers, though admittedly the profit per unit would be smaller. But if you can get into the millions of deployed units, critical mass can be reached, then it is all profit after that.
I hope HP goes after it big time, Dell and IBM see their success, and follow suit. The industry will then follow and at last we will have a solid Linux desktop presence. Hope it actually happens that way!
Apple Computer and Apple Records HAVE been in the news once again - this very week! Perhaps what we need is a "Love In" and a final settlement between the long bickering between the record company and the computer company over logo issues, media rights, and even love.
Who knows, they may find more to argue about if they don't come to a fresh new aggreement. The latest they are arguing about is the considerable amount of revenue that Apple Computer has been getting from the sale of iPod systems, which happen to include music. Apple Records is all up in arms about that. They want some of that money.
Concerning the source, I do not recall the reference, but it might have come from I Cringely or whatever that PBS commentary is. Whether it was his column or elsewhere, a number of years ago I recall that we were somewhere nearing fifty percent of American households owning at least one computer connected to a network. IF that number was accurate at the time, then the seventy percent number, nearly ten years later, is believable. Whatever source I was reading noted that it will be extremely difficult to get much higher than that, but it could approach or slightly exceed that if television usage, especially if computer systems, televisions, video production, and spoken communication are tightly integrated and transmitted over common media. One source, (again, I *think* is was that PBS source) suggested that the Internet would make a very viable medium for all four of the media mentioned above, but there are both political and economic implications that are very large that would need to be overcome in order for that to actually happen.
Bringing Linux back into this picture, my previous argument suggests that there is little difference in the degree of difficulty of using graphical interfaces to common applications. What Linux lacks are several of the most popular applications, but it does have many of the common network centric applications (which is one of its greatest strengths). Another big strength of Linux as it pertains to the original question is low cost coupled with familiar, easy to use interfaces that anyone who has used a recent vintage computer could easily use.
So if a system were set up with a kind of kiosk usage, where only certain functions could be performed (something you can achieve quite easily with Linux software) you could put together a system tailor made for children, for airport terminals, for teaching, for Web browsing, etc. There are numerous special purpose Linux distros that are already honed to do specific things like that. Either using a general purpose system, locating a special purpose system that already meets the needs, or creating one from one of the building block distributions would be a relatively straightforward task for someone capable of installing and building systems.
Talk about jumping to conclusions! You conclude that because I use Linux that I have taught my own children how to use Linux. I have done no such thing. I also have not taught my wife how to use Linux. In spite of that, both my wife and my children, without instruction of any kind from me, have figured out (rather easily I might add) how to use a desktop Linux environment. They are not really using core Linux features, they are merely using the graphical user interface and the applications that can be easily started from that interface.
It would be much more of a challenge to get any of them to master the internals of how to manage a system. However, with a graphical user interface and tools, they can do the basics, just as people have done with Windows for years. People don't manage Windows either, they use it until it breaks, then call for help. Linux can be used in precisely the same way, and my family is proof of that. They don't even know, most of the time, what they are using. It is either fun (kids) or functional (wife browsing the Web for shopping and NFL trades and scores).
Tom, I read your article from LXer earlier today. Like you, I have a bias toward free and open software, but I have not been one to blast other alternatives. I believe in free enterprise and in choices. I am actually glad, to a certain extent, that Microsoft has had success, (though at other times, I wonder about that).
I believe that Apple has taken the original user interface inventions and crafted very usable features around them.
I believe that Microsoft has taken usability and put a relatively consisent, and certainly familiar, interface around it. They have also done one thing better than anyone else, they have successfully commoditized their software to the point that nearly everyone who uses computers recognizes Windows based software, and as a result, it enjoys the lion's share of the market.
I believe that Linux based software has brought flexibility, freedom of choice, and an inexpensive alternative that works to both general purpose and special purpose software.
When comparing desktop software, Microsoft is the volume leader, and that is because they market what they sell better than anyone in the history of computing. That counts.
Apple has created systems that are decidedly different, enough so that many people are willing to pay a premium to use both the hardware and the software that runs with it. They are the king of high end desktop software.
Linux is the nimble alternative that can be adapted to most anything. It lacks name brand commercial software, yet it is capable of handling many routine desktop chores inexpensively and effortlessly.
Where I thought Tom was particularly astute was in his final conclusions. Apple is likely to remain a high end player, Microsoft is likely to remain a high volume player, and Linux is likely to remain a low cost alternative in the middle.
Critical mass may sway the market percentages a few points, but overcoming inertia is going to be extremely difficult, (though not necessarily impossible) for either Apple's OS X or Linux based operating system platforms.
I think that all three can, and do, play a role today. Tom is probably right that it won't be easy to change any of the roles we see in the marketplace today.
Yes, there are such kids, but they are not in the majority. About seventy percent of households have at least one computer system, and most of those are homes in which there are school aged children. You underestimate children. Most children merely lack the opportunity, not the ability. Children learn different things at different rates, so yes, not all of those children know or will know computer skills. Hopefully, the majority of them will. The ones that do not will have a difficult time succeeding unless they have other skills that are in such high demand that they overcome their difficulty at using the common communication tools of their generation.
I have three children, ages 11, 7, and 5. All three of them are able to tinker with games, and they could care less which operating system is on the computer, as long as the games are cool. My eleven year old son is much more savvy than most people (including me) gave him credit for. This past week, he was grounded from using his GameBoy and PSP. So what did he do? He tried all four computer systems in our house, looking for ways to connect to the Internet, find games,and look for resources.
I had set up my system to allow me to automatically boot and login to a KDE session on one of the systems and password protect it once up. He couldn't get in while it was up, so he rebooted the system and FOUND the WAY IN!
Needless to say, kids are resourceful. Not finding Internet Explorer, he rather quickly found Firefox, then used it to access Game Boy and Pokemon sites. Too bad he wasn't smart enough (yet) to cover up his trail! Busted!
Anyway, though, kids from elemenary school to high school age are more able to figure out differences in applications and even operating systems than most adults are, whether they already know them or not. My son's experience with Linux has been playing a very few games in my presence.
Therefore, since these systems are refurbished, possibly aging systems, and the desire is to keep them presumably inexpensive and easy to update, I suggest that there are a number of general purpose Linux distributions that can meet general requirements. In addition, there are also quite a few special purpose Linux distributions targeted at specific tasks, for example, Kiosks, Firewall servers, simple desktop system, evaluation Live CD, etc. The flexibility and choices are there; many of these choices work very well.
I think that IBM is smart to try this in developing countries first. For one thing, when you do not have a well established standard, it is much easier to try something else - and maybe even make that "something else" a standard in those countries. There have been well documented cases concerning the resistance to change in the U.S., even though there are now quite a few viable alternative desktop environments.
I have been using alternative desktop environments for the past twenty years. For many years, I used a development environment containing the proprietary VAX/VMS and OpenVMS environments, Then I used several different UNIX environments. These days, I use Windows XP, connected to UNIX and Linux environments in the workplace using Tarantella's terminal server software, which provides a desktop graphical user environment into remote systems (of various UNIX and Linux varieties). At home, I use predominantly Linux desktop software, but I also test BSD and Windows software. To me, there is no functional difference in which of them I use because my tools are predominantly Web and Internet based.
I think that prepackaged Microsoft solutions offer convenience and appear to reduce complexity and cost. As long as the environment is rigorously set up and you never have to do anything out of THEIR ordinary, that approach works for many people.
I have a really difficult time believing that kind of an approach works optimally for 40,000 people. If you force them to all work the same way, you save on your visible support costs, but you cost your business nimbleness and flexibility when you want to do something different.
While Microsoft has certainly improved in this regard, we have all heard horror stories about worms, viruses, denial of service attacks, and all kinds of other issues. Microsoft constantly fights back with reports that their systems have fewer issues, but twenty five years of actual experience running UNIX desktop and server systems and over ten years of running Linux desktop and server systems tell a different story.
I have never once, either at home or in the workplace, ever encountered any issue at all that resulted in even a minute of down time because of integrity or security issues. Many years ago, I was working on some wide open VMS desktop and server systems and we had a big scare - a prankster got into our network and put a silly display on every workstation in the company. We closed that hole the same day and never had a problem since.
I have other tales to tell about Microsoft based systems. At one QA employer, one person opened a worm Email message and affected the entire company's Email address books. We spent about a day of full time engineering resources to solve the problem itself, but we also wasted countless hours of individual employee time, getting rid of junk.
At two other employers, I was hired, along with around ten other people, to eradicate a virus and upgrade all software to the very latest service pack and anti virus data. The cost to each company was in the tens and hundreds of thousands of dollars - for ONE incident!
No system is impervious to intrusion, and best practices can mitigate many of these problems on any system, including Windows. However, the incidence of these kinds of issues and the general ignorance toward them is significantly greater on Windows based environments. That ought to say a lot, even though Windows is also extremely more widely used.
What are the real costs of moving to another environment? Training, staff, and support changes. Are there real costs and real changes? No doubt. Are they worth doing? That is a call that each project must study and review for themselves. There are real migration costs, and the biggest cost is severely messing up a change. Don't undercount that aspect. However, any well managed effort can be done successfully when you first lay out the business reason for considering it, then lay out the requirements, set up a prototype, then a limited deployment, and identify issues. Move forward if and only if all issues can be documented, resolved, and implemented. Stretch out the deployment to a wider group and see if it can also be effectively deployed. Then come up with a strategy to deploy en masse. It CAN be done, and it can be done very effectively, saving considerable long term repeating costs, but the cost of change is not zero. It takes time, research, effective management, and deployment.
Regardless of how well written a piece of code is, there always need to be at least a few comments. First of all, I believe that every program should have the name of the program at the top. I believe in code reuse. Therefore, when you grab some code, identify the current name of the file. Identify who wrote the original code. Identify who has changed the code. Identify the purpose of the code - what it does, what, if anything, it expects as input, the environment needed to run it, the tasks it performs, and the outputs it produces. These minimal things ought to be in every piece of code, no matter how small, even if it takes up more room than the code itself!
Any code that either you or someone else might have to think about ought to be documented as well. Document each block of code. IF there are any lines of code that use some kind of optimization or shorthand, explain what you are doing and why.
Any good code really ought to be used, even if it was developed originally as a one time hack. Most of my good stuff can be copied and reused many times over. I like to remember what *I* did, but I also want to make my code usable for others. I sure appreciate it when someone else does the same for me.
Some code is more easily read than other code, but I maintain that no code stands completely on its own. Documentation provides a useful and important context and should never be minimized or ignored. However, that also puts the onus on the documentation to be useful, accurate, and relevant, otherwise it wastes more time and causes more problems instead of being a useful resource.
I may be late on this, but from what I can now discern, neither Mandriva nor Turbo Linux has agreed to be part of this "joining of forces". In fact, a few weeks ago, I got a rather sharp reply by Email from a Mandriva employee when I suggested that Mandriva ought to consider moving from the use of urpmi and RPM packages to Debian.deb packages and synaptic or one of the other Debian package interfaces.
Sure, you can use Alien, to a certain degree, to install packages with only moderate levels of packaging dependencies. But unless I read the follow-up reports that the Mandriva employee's comments incorrectly, suggesting that Mandriva will become part of some cooperative Debian project is unlikely, unless it helps them effectively gain market share from Red Hat and SUSE. My read is that Mandriva isn't interested in Debian. Not sure about Turbo Linux, but I doubt that they will go for this, either.
I find the most recent releases of both Windows XP and the vast majority of Linux desktop software distributions to be easy to install, easy to use, ready to go, and usable. For the kinds of things I do, Linux desktop software is actually a bit quicker to install in most cases (I can get a typical desktop system installed and working in about twenty minutes; it takes two to three times as long to get XP installed, then I need to install XP Office, then my own utilities and tools that I prefer, and my the time I am done, three hours have passed.
On the system I am using right now, XP runs fine, that's not the issue at all. Most people probably already HAVE XP installed on their systems.
The real question and issue here is whether or not Linux can provide an acceptable alternative to Windows and OS X on the desktop. When I am home, I rarely run anything else, and that has been the case since the winter of 2002. Before that, I used both Windows 98 SE and various Linux distros at home, but I had a lot of mail saved in Outlook Express, so I kept 98 SE running for a long time until one day, my disk wore out. At that time, I made the plunge.
When I am work, obviously I have to use whatever my employer provides. I have been working in a school district, and they not only use Windows, they also control what is installed and don't provide privileged access to the systems, so you really cannot easily hack at them.
In the past, I have worked at technology companies. There, you can find more than one desktop alternative, at least in labs, but even so, on the office desk, you still find mostly Windows software.
I worked in an operating systems group many years ago, and we used our own products on the desktop. Even there, Windows had a presence. Product Managers and group managers often had PCs as their primary workstation instead of UNIX workstations. If a UNIX shop even runs Windows on the desk, that ought to give you a good idea of just how entrenched Windows software has become in business.
I think that those are but a few of the reasons why Linux software is slow to be adopted on the desktop. It simply has many things to overcome. Market share, in my opinion, along with a general resistance to change, are the strongest factors preventing widespread desktop Linux adoption. Standardization and perception are probably a few of the other largest factors standing in the way, but none of these factors can be ignored. They are reality.
Desktop Linux is an alternative right now, but for the foreseeable future, it will remain primarily an alternative, not a primary option. Costs and improving features will whittle away other markets, but very slowly, and not with as much success as we've seen in server markets. That's just what has been happening... wish it were not so, but it is, at least for a while longer.
I am very pleased to see people giving MEPIS some attention, and some positive attention at that. While MEPIS certainly owes a debt of gratitude to Knoppix for the outstanding work that has been accomplished on that project, I've been suggesting and recommending SimplyMEPIS to anyone who wants to try out a Linux system for the first time with little risk. More than that, though, I've been recommending SimplyMEPIS to moderately experienced Linux users who want to learn how to use a Debian-based system effectively. SimplyMEPIS is simply one very fine desktop system. It installs effortlessly, has a well chosen set of default applications, and it doesn't get in your way, should you happen to be an experienced user that just wants to use a Live CD as a quick way to install a flexible system.
I recommend SimplyMEPIS, not just to beginners, not just to Linux enthusiasts, but for anyone who wants an immediately useful and usable desktop system, but for people who ALSO want to expand, extend, customize, and build a flexible desktop computer system. This is as close to an ideal way to start as you can get right now.
Slashdot Mirror
I recall the marketing attempt to call a bunch of stuff "Polyserve". I worked in Digital UNIX Software Engineering from 1995 until 1998 and the entire time I was there, the two projects were called AdvFS and Trucluster in the project builds and were commonly known by those names - but NOT Polyserve - in the engineering circles.
You are right on with this. HP wanted to close Spit Brook Road entirely. Several of the original AdvFS and TruCluster developers were long gone - I knew quite a few of them, and I used to regularly eat lunch with one of the early AdvFS developers. This was about HP getting rid of their former Digital (called Compaq) engineers and closing facilities. They may not have saved money by dumping AdvFS and Trucluster projects by their cost, but by closing Spit Brook Road, I am sure they saved plenty.
I worked in the Digital UNIX engineering group from 1995 until 1998. The name Digital UNIX was already being used at the time that I joined the group, though the UNIX product in the Version 3 era was called Digital OSF/1. Digital UNIX V4.0 definitely used the UNIX name because it passed all of the branding requirements. I cannot recall if we also named V3.2 Digital UNIX or not - that was around the turning point from OSF/1 to UNIX in naming.
You are wrong about Compaq making the name Tru64 UNIX. We decided that before Compaq grabbed us. Whether we had a release out the door, perhaps not. Internally we calleed V4.0 the Platinum release. We called V5.0 the Steel release. I left after Build level 16. There were at least 20 builds in the release. Sometime near that time Compaq acquired Digital, so that is why it appears that the Tru64 name is a Compaq name, but that is not true, the Digital engineering and marketing people came up with the name.
Reminds me of a fun story. We had some cool sweatshirts made. We wanted to call it butt kicking (and a bunch of other stuff) Linux, but marketing named it instead and called it Digital UNIX. The engineers had a good sense of humor and a great deal of pride in the OS. Too bad the rest of the corporate culture stifled it. I believe in the 1995-2000 era it was the best Linux available.
It's great that Shuttleworth is trying to improve the availability of Free (as in speech) software. But I hope he doesn't move too much efforts over to this.
The reason I use vanilla Ubuntu is because I don't have to put a lot of effort into setting up my OS to agree with all my hardware. Instead, from the start I can work more on customizing how I interact with my OS. I remember the hell I had with a Radeon x800 and Fedora Core 4 a few years ago. If "closed" (as if it's always a bad thing) software provides a better solution, I'm more likely to use that. So, I hope Mark still is going to put his best foot forward for plain Ubuntu. I bet the corporate interest is more aimed to that Ubuntu.
Like many things, this is a balancing act. Pragmatism suggests that we create something that gets the job done. We've done that for years, and we've had a lot of success with that. It is more than high time to do more than that. Yes, I want to be able to do things, access content, etc., but for a great deal of what I do, simply accessing the Internet through Web pages, Email, and news content is quite enough.Why not work with those who can live within those constraints and try to expand what is truly free until every single component can be freely modified and improved using free software?
If that ever becomes the norm from the bottom to the top of the infrastructure, that just makes it possible to create so many more free and innovative components on top of that platform infrastructure.
We keep getting better at this, and the result is that we are starting to see more and more products based on a free infrastructure. Network centric computing, especially so-called "Web 2.0" based applications greatly accelerate the importance of further building upon what has already been established.
I do not see such a change being initiated in the US corporate world. However, if a foreign competitor develops a great advantage, particularly if it undercuts and outsells an US opponent, or if a third world country starts grabbing business from major international enterprises, that is the kind of momentum shifter that could change these perceptions - cost savings and market retention can - and just may - be the forces that cause industry to move on to something different.
No guarantee that it will happen, but at the very least, I expect Linux inertia to cause existing products to compete more effectively. If they don't, Linux really will get the inertia that it has lacked.
I think that DSL has a great niche working with really old hardware. The only distro I know of that is still actively being developed that is smaller than DSL is SliTaZ - very interesting, but very new.
DSL has an old 2.4 kernel, an old Firefox browser, but you can count on it to work with old stuff.
Puppy works with pretty old stuff, but really shines when you load it into RAM on equipment made within the past three years. Wireless support is something that Puppy handles better than DSL.
Zenwalk has a relatively unknown, but fast package manager called Netpkg and a snappy implementation of the XFCE desktop. Derived from an earlier implementation of Minislack, Zenwalk comes out of a stable Slackware heritage. With a fast package manager and a fast desktop implementation, Zenwalk carves a nice niche out of the Slackware landscape.
Arch Linux really is another distribution that once grew out of the Slackware space and has now come into its own with the pacman and AUR package management tools and the idea of giving you total and complete flexibility to build exactly and only what you want. It aims for simplicity rather than coddling the user with its own notion of ease of use. People really either love Arch Linux or avoid it for these very reasons.
Xubuntu is an easy to use system with very current software from the Hardy Heron Ubuntu project, replacing GNOME with XFCE on the desktop. Good solid stable software with excellent wireless network configuration.
TinyME is brand new, as far as a Version 1.0 implementation, but the project has been going on for a couple of years now as a community supported effort to provide lighter versions of the well regarded PCLinuxOS software. This one uses OpenBox instead of KDE. Like other PCLinuxOS systems, it really benefits from the good hardware detection algorithms from Mandriva and the solid packaging from "TexStar", expert RPM packager and founder of PCLinuxOS.
As you can see, each of the distributions mentions has a nice niche. They won't all be appealing to everyone, but each of them is solid in several respects - certainly a credit to the modularity of both Linux and GNU software.
I like this idea best of all - just make sure that you, as vendors, make sure that there are openly available drivers for all of the systems that you sell, then make an entry level option to buy them completely bare. I tend to install different stuff anyway.
I actually did buy a laptop recently with Vista on it - wanted to see how good or bad it really is. I found it to be horrible to configure initially because the OS and the vendor provided AntiVirus software from Norton were both attempting to get updates as soon as my network came up, and they totally consumed the entire laptop for at least an hour. Once they settled down, Vista was OK, but nothing to get excited about. It took about three reboots - including one in safe mode - to finally get the initial configuration properly set up. I've run Vista only three or four times since. I've put Xubuntu 8.04, Fedora 9, AntiX M7.2, and sidux 2008-01 on it. Run AntiX when I want to get on and off quickly just to read Email. Run sidux most of the remainder of the time.
I really like Dell and IBM computers (which are now the Lenovo line). I have two Dell computers at home, a Dell Dimension 4100 - 2000 vintage technology which has greatly impressed me with years of reliability and surprisingly decent performance. While it is getting a bit ancient now by computer standards and slow compared to newer equipment, it still works. Huge positive points to Dell for such a good machine. Another positive is that virtually any operating system I have tried on this home PC just works - ranging from Windows 2000 on the original box, to QNX, a POSIX oriented real time system, to SliTaZ, a 25 MB micro Live CD, to Arch Linux, an ultra flexible, but hobbyist oriented system. My usual desktops on this box are sidux and SimplyMEPIS, with PCLinuxOS getting air time when my kids use this old box.
Dell Latitude D600 - a great moderately priced used laptop system. I now use this for my every day home use as my primary system.
Compaq/HP D530 - a refurbished box, runs well for distro testing, but the case doesn't close well and the fan is noisy, so I do not use this as much as the others. Still, with a 2.9 GHz processor and plenty of space, it is a great distro test box.
IBM Thinkpad - used to have the T42 at work, was highly impressed with it. I now have the T60, which actually has the "IBM Thinkpad" label on it, but it is really a Lenovo T60 - on the inside, it does in fact say, "Lenovo T60". One of the best current generation corporate laptops. Thee laptop keyboard is better than the docking station keyboard for key feel, if you can believe it - one of the main reasons I love Thinkpads.
Given this, I decided to buy a Lenovo 3000 series laptop. The reason I did so was to get consumer features - a built in camera in particular. I gave up just a bit of speed - not much, to get a Lenovo 3000 Y410 - a Duo Core (1.5 GHz per core) with 2 GB memory and a 160 GB hard drive. Great keyboard, nice glossy laptop display, good Intel Pro Wireless 3945 network card, decent compatibility with current Linux distros.
I'd buy Dell or Lenovo models with little hesitation and probably HP models too. All three have been really reliable for me, and as end of model closeout or used models, they are an even better value.
There are all kinds of options. FreeBSD is but one of many.
The Debian project is all about freedom. Many probably do not realize it, but the Linux kernel is not the only kernel to be embraced or to be used with the Debian environment. I know the GNU Hurd project had worked with Debian at one point, so did FreeBSD, NetBSD, and OpenBSD. Those are only the ones I know about.
There are many free and open OS efforts out there and there are many free and open applications out there. The only way to slow down the movement would be to close off the networks. Even then, this movement had started before the prevalence of Broadband networks. Today's networks only accelerate collaboration and innovation. Those who are not innovating (you can figure out who they are) are gasping and grasping, hoping not to drown under a torrent of rapidly moving current.
I would say that Red Hat and other companies are probably in talks with Microsoft all the time. If they are not, they probably should be. Both companies have customers that use both sets of products, so it is in everyone's interest to communicate and to improve interoperability between distinct, independent products.
I do not see Red Hat caving in and signing any kind of patent agreement, but I could easily see Red Hat working with Microsoft if the work was in their mutual interest and in the interest of customers.
Red Hat is probably very reluctant to make those kind of deals because we have seen the kind of press Microsoft has given so far - stuff like "We will not sue this partner", yet in another breath that same day, "We think that Linux infringes on..." ad nauseum.
I do not think anyone's software should rightly infringe on anything else, because ninety percent of functions are common functions that should not be possible to patent, protect, or license, but there could be open collaborations to share the cost and benefits of implementing new features and improving the interoperability of existing features.
What we have today is a bunch of baloney. I hope all of it comes to an end soon and we usher in a new generation of collaborative software. Before that happens, there is likely to be quite a bit more nonsense, but perhaps we will get there sooner rather than later after all.
At one time, the idea of patents was created to encourage investment in innovation and protect new ideas, rewarding the founder and investor in new ideas for a set period of time. This is fine for completely new technologies, but I seriously question the usefulness of patents for well established ideas, especially when someone attempts to place a patent on an idea that really is well established and wants to own every nuance. That kind of patent ought to be abolished completely.
I can understand patents for brand new ideas, and I can tolerate them for that, though my first preference would be to get rid of patents altogether. Perhaps their time of usefulness has passed. Instead, let's collaborate more on innovations, change laws to encourage open collaboration, and only protect against closed collaborative efforts of giants who seek to squash any potential newcomers into their markets.
What I advocate is protecting freedom and protecting choice. Politically I see freedom and choice getting reduced. I wonder how much longer we will be able to speak freely like this?
I worked at Digital Equipment Corp. myself, and again as a consultant for Compaq prior to the HP merger-acquisition. I also attended a Debian conference where SEVERAL HP employees were openly favorable to Debian Linux and who were actively promoting the use of Debian GNU/Linux software within HP. From my Digital days, I know that Jon Hall, when he was still at DEC, actually brought in Linus Torvalds (I met Linus personally during one of those meetings, and Jon also saw to it that Linux had his own AlphaStation to port Linux to Alpha. So there is a long history with BOTH HP and DEC in using and promoting Linux internally.
Externally, it has always been a bit below the radar, even though HP does in excess of $2B a year in hardware, software, services, and other line items. Maybe for a $100B+ company that is not very big, but with a little more effort, they could scale that five or tenfold. Maybe it will happen if they have as much success on the desk as they have had with servers. The sheer numbers of systems sold could far exceed the server numbers, though admittedly the profit per unit would be smaller. But if you can get into the millions of deployed units, critical mass can be reached, then it is all profit after that.
I hope HP goes after it big time, Dell and IBM see their success, and follow suit. The industry will then follow and at last we will have a solid Linux desktop presence. Hope it actually happens that way!
Apple Computer and Apple Records HAVE been in the news once again - this very week! Perhaps what we need is a "Love In" and a final settlement between the long bickering between the record company and the computer company over logo issues, media rights, and even love.
Who knows, they may find more to argue about if they don't come to a fresh new aggreement. The latest they are arguing about is the considerable amount of revenue that Apple Computer has been getting from the sale of iPod systems, which happen to include music. Apple Records is all up in arms about that. They want some of that money.
Send love!
Concerning the source, I do not recall the reference, but it might have come from I Cringely or whatever that PBS commentary is. Whether it was his column or elsewhere, a number of years ago I recall that we were somewhere nearing fifty percent of American households owning at least one computer connected to a network. IF that number was accurate at the time, then the seventy percent number, nearly ten years later, is believable. Whatever source I was reading noted that it will be extremely difficult to get much higher than that, but it could approach or slightly exceed that if television usage, especially if computer systems, televisions, video production, and spoken communication are tightly integrated and transmitted over common media. One source, (again, I *think* is was that PBS source) suggested that the Internet would make a very viable medium for all four of the media mentioned above, but there are both political and economic implications that are very large that would need to be overcome in order for that to actually happen.
Bringing Linux back into this picture, my previous argument suggests that there is little difference in the degree of difficulty of using graphical interfaces to common applications. What Linux lacks are several of the most popular applications, but it does have many of the common network centric applications (which is one of its greatest strengths). Another big strength of Linux as it pertains to the original question is low cost coupled with familiar, easy to use interfaces that anyone who has used a recent vintage computer could easily use.
So if a system were set up with a kind of kiosk usage, where only certain functions could be performed (something you can achieve quite easily with Linux software) you could put together a system tailor made for children, for airport terminals, for teaching, for Web browsing, etc. There are numerous special purpose Linux distros that are already honed to do specific things like that. Either using a general purpose system, locating a special purpose system that already meets the needs, or creating one from one of the building block distributions would be a relatively straightforward task for someone capable of installing and building systems.
Talk about jumping to conclusions! You conclude that because I use Linux that I have taught my own children how to use Linux. I have done no such thing. I also have not taught my wife how to use Linux. In spite of that, both my wife and my children, without instruction of any kind from me, have figured out (rather easily I might add) how to use a desktop Linux environment. They are not really using core Linux features, they are merely using the graphical user interface and the applications that can be easily started from that interface.
It would be much more of a challenge to get any of them to master the internals of how to manage a system. However, with a graphical user interface and tools, they can do the basics, just as people have done with Windows for years. People don't manage Windows either, they use it until it breaks, then call for help. Linux can be used in precisely the same way, and my family is proof of that. They don't even know, most of the time, what they are using. It is either fun (kids) or functional (wife browsing the Web for shopping and NFL trades and scores).
Well balanced and reasonable comments.
Tom, I read your article from LXer earlier today. Like you, I have a bias toward free and open software, but I have not been one to blast other alternatives. I believe in free enterprise and in choices. I am actually glad, to a certain extent, that Microsoft has had success, (though at other times, I wonder about that).
I believe that Apple has taken the original user interface inventions and crafted very usable features around them.
I believe that Microsoft has taken usability and put a relatively consisent, and certainly familiar, interface around it. They have also done one thing better than anyone else, they have successfully commoditized their software to the point that nearly everyone who uses computers recognizes Windows based software, and as a result, it enjoys the lion's share of the market.
I believe that Linux based software has brought flexibility, freedom of choice, and an inexpensive alternative that works to both general purpose and special purpose software.
When comparing desktop software, Microsoft is the volume leader, and that is because they market what they sell better than anyone in the history of computing. That counts.
Apple has created systems that are decidedly different, enough so that many people are willing to pay a premium to use both the hardware and the software that runs with it. They are the king of high end desktop software.
Linux is the nimble alternative that can be adapted to most anything. It lacks name brand commercial software, yet it is capable of handling many routine desktop chores inexpensively and effortlessly.
Where I thought Tom was particularly astute was in his final conclusions. Apple is likely to remain a high end player, Microsoft is likely to remain a high volume player, and Linux is likely to remain a low cost alternative in the middle.
Critical mass may sway the market percentages a few points, but overcoming inertia is going to be extremely difficult, (though not necessarily impossible) for either Apple's OS X or Linux based operating system platforms.
I think that all three can, and do, play a role today. Tom is probably right that it won't be easy to change any of the roles we see in the marketplace today.
Yes, there are such kids, but they are not in the majority. About seventy percent of households have at least one computer system, and most of those are homes in which there are school aged children. You underestimate children. Most children merely lack the opportunity, not the ability. Children learn different things at different rates, so yes, not all of those children know or will know computer skills. Hopefully, the majority of them will. The ones that do not will have a difficult time succeeding unless they have other skills that are in such high demand that they overcome their difficulty at using the common communication tools of their generation.
I have three children, ages 11, 7, and 5. All three of them are able to tinker with games, and they could care less which operating system is on the computer, as long as the games are cool. My eleven year old son is much more savvy than most people (including me) gave him credit for. This past week, he was grounded from using his GameBoy and PSP. So what did he do? He tried all four computer systems in our house, looking for ways to connect to the Internet, find games,and look for resources.
I had set up my system to allow me to automatically boot and login to a KDE session on one of the systems and password protect it once up. He couldn't get in while it was up, so he rebooted the system and FOUND the WAY IN!
Needless to say, kids are resourceful. Not finding Internet Explorer, he rather quickly found Firefox, then used it to access Game Boy and Pokemon sites. Too bad he wasn't smart enough (yet) to cover up his trail! Busted!
Anyway, though, kids from elemenary school to high school age are more able to figure out differences in applications and even operating systems than most adults are, whether they already know them or not. My son's experience with Linux has been playing a very few games in my presence.
Therefore, since these systems are refurbished, possibly aging systems, and the desire is to keep them presumably inexpensive and easy to update, I suggest that there are a number of general purpose Linux distributions that can meet general requirements. In addition, there are also quite a few special purpose Linux distributions targeted at specific tasks, for example, Kiosks, Firewall servers, simple desktop system, evaluation Live CD, etc. The flexibility and choices are there; many of these choices work very well.
I think that IBM is smart to try this in developing countries first. For one thing, when you do not have a well established standard, it is much easier to try something else - and maybe even make that "something else" a standard in those countries. There have been well documented cases concerning the resistance to change in the U.S., even though there are now quite a few viable alternative desktop environments.
I have been using alternative desktop environments for the past twenty years. For many years, I used a development environment containing the proprietary VAX/VMS and OpenVMS environments, Then I used several different UNIX environments. These days, I use Windows XP, connected to UNIX and Linux environments in the workplace using Tarantella's terminal server software, which provides a desktop graphical user environment into remote systems (of various UNIX and Linux varieties). At home, I use predominantly Linux desktop software, but I also test BSD and Windows software. To me, there is no functional difference in which of them I use because my tools are predominantly Web and Internet based.
I think that prepackaged Microsoft solutions offer convenience and appear to reduce complexity and cost. As long as the environment is rigorously set up and you never have to do anything out of THEIR ordinary, that approach works for many people.
I have a really difficult time believing that kind of an approach works optimally for 40,000 people. If you force them to all work the same way, you save on your visible support costs, but you cost your business nimbleness and flexibility when you want to do something different.
While Microsoft has certainly improved in this regard, we have all heard horror stories about worms, viruses, denial of service attacks, and all kinds of other issues. Microsoft constantly fights back with reports that their systems have fewer issues, but twenty five years of actual experience running UNIX desktop and server systems and over ten years of running Linux desktop and server systems tell a different story.
I have never once, either at home or in the workplace, ever encountered any issue at all that resulted in even a minute of down time because of integrity or security issues. Many years ago, I was working on some wide open VMS desktop and server systems and we had a big scare - a prankster got into our network and put a silly display on every workstation in the company. We closed that hole the same day and never had a problem since.
I have other tales to tell about Microsoft based systems. At one QA employer, one person opened a worm Email message and affected the entire company's Email address books. We spent about a day of full time engineering resources to solve the problem itself, but we also wasted countless hours of individual employee time, getting rid of junk.
At two other employers, I was hired, along with around ten other people, to eradicate a virus and upgrade all software to the very latest service pack and anti virus data. The cost to each company was in the tens and hundreds of thousands of dollars - for ONE incident!
No system is impervious to intrusion, and best practices can mitigate many of these problems on any system, including Windows. However, the incidence of these kinds of issues and the general ignorance toward them is significantly greater on Windows based environments. That ought to say a lot, even though Windows is also extremely more widely used.
What are the real costs of moving to another environment? Training, staff, and support changes. Are there real costs and real changes? No doubt. Are they worth doing? That is a call that each project must study and review for themselves. There are real migration costs, and the biggest cost is severely messing up a change. Don't undercount that aspect. However, any well managed effort can be done successfully when you first lay out the business reason for considering it, then lay out the requirements, set up a prototype, then a limited deployment, and identify issues. Move forward if and only if all issues can be documented, resolved, and implemented. Stretch out the deployment to a wider group and see if it can also be effectively deployed. Then come up with a strategy to deploy en masse. It CAN be done, and it can be done very effectively, saving considerable long term repeating costs, but the cost of change is not zero. It takes time, research, effective management, and deployment.
Regardless of how well written a piece of code is, there always need to be at least a few comments. First of all, I believe that every program should have the name of the program at the top. I believe in code reuse. Therefore, when you grab some code, identify the current name of the file. Identify who wrote the original code. Identify who has changed the code. Identify the purpose of the code - what it does, what, if anything, it expects as input, the environment needed to run it, the tasks it performs, and the outputs it produces. These minimal things ought to be in every piece of code, no matter how small, even if it takes up more room than the code itself!
Any code that either you or someone else might have to think about ought to be documented as well. Document each block of code. IF there are any lines of code that use some kind of optimization or shorthand, explain what you are doing and why.
Any good code really ought to be used, even if it was developed originally as a one time hack. Most of my good stuff can be copied and reused many times over. I like to remember what *I* did, but I also want to make my code usable for others. I sure appreciate it when someone else does the same for me.
Some code is more easily read than other code, but I maintain that no code stands completely on its own. Documentation provides a useful and important context and should never be minimized or ignored. However, that also puts the onus on the documentation to be useful, accurate, and relevant, otherwise it wastes more time and causes more problems instead of being a useful resource.
I may be late on this, but from what I can now discern, neither Mandriva nor Turbo Linux has agreed to be part of this "joining of forces". In fact, a few weeks ago, I got a rather sharp reply by Email from a Mandriva employee when I suggested that Mandriva ought to consider moving from the use of urpmi and RPM packages to Debian .deb packages and synaptic or one of the other Debian package interfaces.
Sure, you can use Alien, to a certain degree, to install packages with only moderate levels of packaging dependencies. But unless I read the follow-up reports that the Mandriva employee's comments incorrectly, suggesting that Mandriva will become part of some cooperative Debian project is unlikely, unless it helps them effectively gain market share from Red Hat and SUSE. My read is that Mandriva isn't interested in Debian. Not sure about Turbo Linux, but I doubt that they will go for this, either.
I find the most recent releases of both Windows XP and the vast majority of Linux desktop software distributions to be easy to install, easy to use, ready to go, and usable. For the kinds of things I do, Linux desktop software is actually a bit quicker to install in most cases (I can get a typical desktop system installed and working in about twenty minutes; it takes two to three times as long to get XP installed, then I need to install XP Office, then my own utilities and tools that I prefer, and my the time I am done, three hours have passed.
On the system I am using right now, XP runs fine, that's not the issue at all. Most people probably already HAVE XP installed on their systems.
The real question and issue here is whether or not Linux can provide an acceptable alternative to Windows and OS X on the desktop. When I am home, I rarely run anything else, and that has been the case since the winter of 2002. Before that, I used both Windows 98 SE and various Linux distros at home, but I had a lot of mail saved in Outlook Express, so I kept 98 SE running for a long time until one day, my disk wore out. At that time, I made the plunge.
When I am work, obviously I have to use whatever my employer provides. I have been working in a school district, and they not only use Windows, they also control what is installed and don't provide privileged access to the systems, so you really cannot easily hack at them.
In the past, I have worked at technology companies. There, you can find more than one desktop alternative, at least in labs, but even so, on the office desk, you still find mostly Windows software.
I worked in an operating systems group many years ago, and we used our own products on the desktop. Even there, Windows had a presence. Product Managers and group managers often had PCs as their primary workstation instead of UNIX workstations. If a UNIX shop even runs Windows on the desk, that ought to give you a good idea of just how entrenched Windows software has become in business.
I think that those are but a few of the reasons why Linux software is slow to be adopted on the desktop. It simply has many things to overcome. Market share, in my opinion, along with a general resistance to change, are the strongest factors preventing widespread desktop Linux adoption. Standardization and perception are probably a few of the other largest factors standing in the way, but none of these factors can be ignored. They are reality.
Desktop Linux is an alternative right now, but for the foreseeable future, it will remain primarily an alternative, not a primary option. Costs and improving features will whittle away other markets, but very slowly, and not with as much success as we've seen in server markets. That's just what has been happening... wish it were not so, but it is, at least for a while longer.
I am very pleased to see people giving MEPIS some attention, and some positive attention at that. While MEPIS certainly owes a debt of gratitude to Knoppix for the outstanding work that has been accomplished on that project, I've been suggesting and recommending SimplyMEPIS to anyone who wants to try out a Linux system for the first time with little risk. More than that, though, I've been recommending SimplyMEPIS to moderately experienced Linux users who want to learn how to use a Debian-based system effectively. SimplyMEPIS is simply one very fine desktop system. It installs effortlessly, has a well chosen set of default applications, and it doesn't get in your way, should you happen to be an experienced user that just wants to use a Live CD as a quick way to install a flexible system.
I recommend SimplyMEPIS, not just to beginners, not just to Linux enthusiasts, but for anyone who wants an immediately useful and usable desktop system, but for people who ALSO want to expand, extend, customize, and build a flexible desktop computer system. This is as close to an ideal way to start as you can get right now.