 |
 |
 |
|
 |
Emmett Dulaney
|
|
|
 |
 |
What the Linux World REALLY Needs... |
Emmett looks at the excuses for Linux's lack of widespread adoption, shares his take and invites you to post your own. |
by Emmett Dulaney |
7/17/2006 -- It has been quite a while now since the first release of Linux. In fact, it has been 15 years since Linus Torvalds, a computer science student at the University of Helsinki at the time, made freely available a kernel that mirrored many of the features of Unix and Minix. In this time, there have been small pockets where Linux has grabbed a foothold and "stolen" some market share, but for the most part it is still not setting any adoption records. By the way, I use the term stolen loosely since many of these areas were new to begin with and not that many were migrations.
Why is this?
I can think of seven excuses often tossed about, and one real reason of my own creation. Let's look at the excuses first:
Excuse #1: Some will say this is because there aren't enough applications for Linux. That's a line of crap that anyone with access to a search engine and gumption enough to check can easily dismiss. There are applications -- and I'm talking decent to great ones -- that can be used for productivity suites and application servers as well as everything in between.
Excuse #2: Others will say the market is too diverse and confusing. I'll agree this is a problem and attempts have been made to unify what is "Linux" (you can create just about anything and call it Linux form the operating system for your phone, to the OS on the mega-million dollar server). I still don't think this excuse is a very valid one, but more one of convenience: Dell has said that they don't ship Linux on desktops anymore because they don't know which one the market wants. In reality, if you want to buy Linux, you basically have three choices now: Red Hat, SuSE (Novell) and Ubuntu. Each has its own niche of the market and each serves that market very well. Given the strength and commitment of these three, it would not surprise me at all to see five years down the road that they are the only real entities left.
Excuse #3: Yet another excuse slowing adoption is that the mainstream does not know about it or understand it. After 15 years, I would suspect that most administrators have heard of Linux, and most have probably even loaded it on a machine in their basement and played with it at one time or another. Surely, they've watched television at some point in time and seen the commercials from IBM and read some trade magazine where they've seen ads from vendors. No, this excuse doesn't hold much water either.
Excuse #4: The brainwashing of Microsoft: It has convinced everyone that their offerings are the only ones to have, or so this excuse goes. This is true, supposedly, for everything from server and desktop operating systems to applications and game consoles; the world mindlessly adopts everything leaving their shipping department. Really? Maybe I'm getting old, but I remember a fair number of products they came out with that fell flat on their face (Bob, anyone?). The reason their products are adopted – those that do become successful – is because they work hard to understand the market and give it what it wants. Yes, I know all about vaporware and empty promises, but they are not isolated in that practice (it is employed by car manufacturers, politicians and others on a daily basis). Love or hate Microsoft, you have to admire the way they stay with a product, make it better, and make it (usually) what the market wants.
Excuse #5: There are issues with Linux and certain hardware. Here is a surprise for you: There are issues with some piece of hardware and every operating system. Aside from the old kernel/hard drive issue (long since resolved), the only hardware issue I can think of immediately is with Winmodems – legacy (and very cheap) modems controlled mostly by software. In the first place, they are notoriously buggy to begin with, even when you have the right OS and drivers. In the second place, modems aren't commonly used today as opposed to several years ago. Lastly, if you do need a modem for business, I would certainly hope you would be smart enough to buy a decent one and not try to get by with one of these in the first place. All that said, I know of no disadvantage Linux has when it comes to hardware compared to any other operating system.
Excuse #6: Lack of technical support. If I have a problem, no one here knows how to take care of it. Maybe you need to hire or train someone. The odds are good that no one at your site is proficient with Windows Vista at the moment either, but when it comes out, someone will probably be responsible for learning more about it. Linux -- and I am being as honest as I can be -- isn't that difficult to learn. Gone are the days when you needed to memorize hundreds of command-line tools and their options because the main files took too much space to load them on the hard drive. You can pretty much administer all you need to through a graphical interface now and yank out a reference guide when you get stuck.
Excuse #7: The costs of adopting/migrating are too high. Compared to what? The costs of adopting a new implementation should be based upon total savings of one operating system to another and nothing else, and Linux is often the best solution. The cost of migrating is high because it often involves more than just the operating system –- it includes applications and services as well. If the applications you are using are proprietary, then the migration may not make sense -- pure and simple. More and more, though, solutions/applications are becoming less proprietary and more open and thus the operating system migration can be beneficial.
These are the top seven excuses I have heard for the slow rate of Linux adoption. There are a handful of other excuses that are occasionally bandied about, but they tend to contradict themselves even in their phrasing and disappear like smoke as soon at they are voiced around anyone who has an inkling of what is being discussed.
Now that I've listed the excuses, let me propose what I believe to be a real factor: the lack of an adoption spokesmodel.
When you think of Microsoft, you think of Bill Gates. While he has been called everything in the book (and a lot of things unprintable), a few key words pop out: nerdy, highly intelligent, productive, driven, business savvy. While everyone knows he doesn't write each line of code for Microsoft, his image is known by those who evaluate the products, those who use the products, and even those who have never even touched a computer. In other words, through his image, he plays to the stereotype and is able to attract the customer base: When it comes to work, who doesn't want solutions that are productive, business savvy, etc.?
When you think of the Mac, you think of Steve Jobs. Whether he is in the company or out of the company at the moment, it is still his image that fits the stereotype and makes the product trendy.
I once knew a girl who ran a Mac and she wasn't "cool." What? Isn't everyone who runs a Mac cool and artistic? No more so than everyone who uses Microsoft's OS has bad hair and glasses.
Linux does have a spokesmodel: Linus Torvalds. Far be it from me to speak ill of someone so meaningful, but unfortunately, he remains quiet about most things and this does not help speed the adoption process. You can't hate the guy, no matter how hard you try, and that is something that has to be possible (look at how many don't like Microsoft's leaders and how that has only made them sell more).
If I ask my almost-blind grandmother who Bill Gates is, she'll have an answer. If I ask her who Steve Jobs is, she'll come close. If I ask her who Linus Torvalds is, she'll turn to her applesauce and pretend she can't hear me rather than admit she has no clue. And while you have to love the penguin mascot, it doesn't say much to those not already familiar with Linux and using it.
If anyone could benefit from a spokesmodel, it is the commercial versions of Linux that would be at the top of the list. How many, though, could name the CEO of Novell (SuSE) or Red Hat? When Jack Messman was recently released from Novell, the shareholders responded by driving the price up substantially – the news of someone few could name being fired made the company more valuable than it had been for a while. Who replaced him?
In the absence of a single visible spokesmodel pushing for Linux in the workplace, what has sprung up is an image that the media has artificially created: That you need a ponytail and earring to run Linux. Now that we are out of the 1990s, that is not an image that finds an audience readily these days within the Fortune 500. Or a lot of small companies trying to become big. Or a lot of universities. Or...
If Linux is ever to go mainstream, I assert, there must be a face put with its movement. That face must be someone you can like and hate -- someone you can associate with, cheer for, heckle and throw a pie at. It has to be someone you can get so mad at when something doesn't work like it should, and someone you can wish you were like when you read about their exploits outside of work in the trade rags.
The operating system is sound. The excuses are just that. The missing piece is the face.
Do you agree, or am I as full of bunk as everyone else? Let me know by posting below.
|
Emmett Dulaney is the author of several books on Linux, Unix and certification. He can be reached at .
|
|
|
 |
More articles by Emmett Dulaney:
|
There are 40 user Comments for “What the Linux World REALLY Needs... ”
|
Page 2 of 4
|
7/20/06: nuna says: |
Poblem?: A vendor Locked-In software distribution model from the perspective of non-tehnical end users. i.e. The lack of a Linux-wide, standardized, installer, for all third party software, that would work the same regardless of the distribution of choice. With Microsoft, users are locked into the basic OS. But, all third party software used on top of it is highly competetive. Clik-N-Run type installation models are vendor specific, and highly un-competetive. Hence, non technical end users are totaly dependent on a single entity for all their third party applications, availability, and fair pricing. Bad business, and a Lousy choice for an end user to make. |
7/20/06: werner from cayenne says: |
Im using Linux since long time, but its problem is the lack of professionality of the progs. OpenOffice, when the keyboard stops to work, you cannot storage your file what you typed. KOffice, you cannot shift the page margins to bigger or smaller size, and so on. When a CD cannot be read because its scratched, it dont stop more and you have to reboot, etc. Lili and Grub both are the worst - examples how most easy tasks, by a stupid direction of thinking and aproaching to the problem, can be programmed much to complicated and even so dont work correctly (both have often problems with extended partitions on big HDs, and/or with buggy BIOS), and both should be substituted by something what ALWAYS ANYHOW finds correctly the partitions. Similar, the kernel should manage it always to find the partitions and boot anyhow instead to crash with kernel panic when the init ramdisk or vmlinuz have a wrong information about the root filesystem on the HD. When after checking/repairing you should type in CTRL-D for reboot or the root-password for reparations, then when you have no root password and type in ENTER, the sistem reboots and repead that eternally. The autoplay of CDs should work under KDE without udev, hal, dbus - but dont do it, on Slackware them dont run correctly at all. And so on. It is necessary to found a commission which verify and repair the lacks of 'being practical' of Linux. Also, the distros should refuse to include half-ready progs, in order to force the programmers to bring forwards their programs. |
7/20/06: nuna says: |
Problem?2: Can't realisticaly leverage "support externality". Externality, being an industy term for off-loading to others, some of their external costs. i.e. Windows is Windows. Anybody, including some pretty non-technical users can support it. Hence, Dell's actual support costs/resposibilities are greatly (externalized) reduced. Linux's plethora of incompatible core distributions makes general purpose support by normal users, their friends, and aquaintances, anything but simple. Linus Singing-In-The-Rain with a Top Hat, Tails, and a Cane, won't change, or fix it either.. |
7/20/06: Stuart DeGraaf says: |
I think the problem limiting desktop Linux uptake is much simpler: M$ preloads and its ownership of the hardware vendors, combined with human sloth. Until this injustice is resolved - maybe courtesy of the EU, certainly not the DoJ - M$ dominance will continue unmitigated. Undoubtedly Office 2007 and Vista will break CrossOver Office and OpenOffice compatibility; this is clearly M$ primary goal. Linux's golden window of opportunity will slam shut when the next generation of M$ crapware appears on corporate executives' new computers and all the serfs are compelled to be compatible. Most "executives" are too "busy" and "valuable" to learn anything new, and the "grunts" must be kept in lockstep. This is how M$ will cram Vista down our throats... |
7/20/06: nuna says: |
Problem?3: Wrist Wrestling In The Locker Room. No-one has a problem with Linus's community owned, Linux, being the only kernel used by all of the various Linux Distributions. We learned from UNIX's mistakes the hard way. But did we really? No! We just moved our Unixization to the full Distribution level. And, for the same reason. Market-Share Wars. Linux's slow uptake is due, at least in part, to the fact that the various core distributions are stuck in the locker room, fighting over who will be Captain of the "Linux Team". Leaving a now even, screwed up (can't deliver on time) Microsoft on the field to score further goals unopposed. Answer?: A community owned, standardized CORE Distribution that the rest of the various (now niche) Distro's would build upon. Everyone, including the extreme Zealots could live with this solution. And, the other problems (nuna 1-2) afore mentioned would disappear as a result also. We don't need a note from the teacher to do it either. If Debian simply went core distro, and only the core, with a general purpose installer, instead of their current full package repository model. A Ross Perot style "Giant Sucking Sound" to this much simplified, lowered barriers, supremely supportable (by even normal users), core, would likely soon follow. Or, we can just build bleachers in the Locker-Room, and continue to watch the game from there. |
7/20/06: Anonymous says: |
I nominate Bob Arctor to be the face of Linux. Just the kind of guy anyone can relate to and identify. |
7/20/06: Rufus Polson from Vancouver says: |
Well, if Linux doesn't have a spokesmodel, maybe it's time to get radical. Call it GNU/Linux and your spokesmodel becomes Richard Stallman. Now there's a personality who has impact. Some love him, some hate him, some are frustratedly affectionate, but nobody who's heard of him is indifferent to him. That was a joke--mostly. Actually, in many countries around the world Stallman IS the spokesmodel for GNU/Linux and FOSS. But it would never fly in the US. I don't think spokesmodels have much to do with it. The real reasons depend on the market segment. On the desktop it's a fairly simple structural thing--Microsoft is a monopoly, and in the absence of deliberate effort desktops come with Windows on them. Where there are two or more choices, but if no choice is actively made you always get one of them, it's pretty clear what choice most will end up with. Elsewhere that isn't applicable--and Linux actually does fairly well, certainly much better. To the extent that adoption has been slowed, it's probably been mostly a combination of simple inertia (there are still plenty of setups out there using programs written in COBOL, for Pete's sake) and the market confusion thing (excuse 2). The market confusion thing is partly an indication that there still isn't really an understanding of Linux. It's always been fairly true, and is getting more and more so, that the variety of flavours of Linux don't matter. They're not all going to go away, but the same programs run under all of them and someone trained to support one can pretty much support another. People who use Linux as their desktop will often switch between different ones, or even use different ones on different computers; they experiment like people trying different kinds of soft drink. I think that's pretty much unprecedented in the operating system market. Once people get used enough to Linux to understand that, the "market confusion" factor will diminish--which is good, because holding one's breath until all the distributions go away is not recommended. |
7/20/06: Hendra from Singapore says: |
I'd like to comment about the missing "face" part. I agree with you for the most part that Linux need a spokeperson but I don't think it's that crucial, not to that extend, so IMHO the other excuses still more relevant. But if you really think we need someone to be associated with the Linux image, I think Mark Shuttleworht of the ubuntu fame may be a good suite. He's a more outgoing person that Linus and not media-shy. He didn't invent Linux, but ubuntu may be just the one distribution that we've been looking for all this while to replace all the Windows boxes out there. Bill Gates didn't invent DOS either, but he made it mainstream. I'm not in for some corporate honchos (CEO this, CTO that) who may be here today gone tomorrow, and their only contribution to the Linux community is to sign the paycheck for their employees, so any top dog from Novell or Red Hat is out of question, at least for me. |
7/20/06: Ram from Miami Florida says: |
I disagree with you about having a central representative for Linux. The world has changed dramatically since the early nineties when the Cold War ended; it took about another decade for those backward countries to wake up. Now you have brilliant people hungry for progress in every corner of the world and not just he top countries. You are going to see a lot more changes that will shake economies and move people around like never seen before. The old business models are obsolete and people acostumed to it will soon realize that geology and beurocracy no longer play a big role in economies. Linux is about colaboration, community, sharing, putting the mind to work, creating newer more sophisticated solutions; each compact and modified to do one thing but complimenting all the other things possible. Linux does not need hot shots because hot shots are overrated. However, I do admire Linus but he's just a kernel programmer. What about all the other guys. |
7/21/06: Victor from Bloomington, IN says: |
Ok. I agree with most of what you have said. The one excuse that I think does still hold some water is the hardware one. You are right, winmodems are the biggest problem with linux, and they don't really matter, however, I have read of many wireless card problems people have with linux, myself included. While they aren't as bad as winmodems were by a long shot, they have basically taken their place in the pantheon of pain in our butt hardware on linux. The other thing is I would like to propose a refinement of you posit that we need a face. I don't think we need a face, I think we need a commercial vendor to market linux in the commercial space. Many of the Tech people who write up propositions in major companies know what they are talking about and don't care what new commercial with a "Mac" and a "PC" guy is playing on TV. However, many of the higher up guys who aren't techies who have to sign off on migrations and new deployments sadly know computers by commercial space. What we need is a company like Xandros or Ubuntu to parter up with a large hardware company, preferably a new name as many people associate all the major names on the market these days with Windows or Mac and create a platform. Non tech people don't care about the operating system. The software should be transparent. They buy a computer take it out of the box and it works. Linux works better on base installation than Windows or Mac. In a full linux install there are more applications, with some interface spiff up of most graphical respository tools there are far more immediate upgrades available, and so Linux should be an obvious choice for the commercial space. It's that whole opt in to getting linux and which distro do I want, and I'm probably going to do the software install myself thing. Linux needs a company to stop pushing it as an OS and start pushing it as full hardware included product. It isn't perfect, it's not even close to perfect, and it's no where near as ready for non techie prime time as windows, and if you look back at the history of Mac computers and their software issues you will know that doesn't matter. When you can buy a "Linux Computer" instead of a computer with a Linux OS then we will be looking at the building blocks of the psychological shift that needs to happen for linux adoption and not a moment sooner. |
First Page Previous Page Next Page Last Page
|
|
|
|