Let's kick off this forum: open source in the workplace

Originally posted by LineS chort, were did you have those problems with? What distro? What versions of KDE / Openoffice.org / Gnome?

At various times I've used several versions of Red Hat, Debian, and Mandrake. Plus I read reviews from time to time just to see where things are at (most of the reviews are grossly biased because they're done by Linux zealots, but some times they're useful). Also one of my coworkers is a Linux zealot and he contributes to various projects, so I always hear his stories, too.

Many windows programs have different layouts. Almost all the windows versions have different layouts. Especially Windows XP is a hassle. What about the ever changing 'Personal menus', what about 'hide inactive icons', what about the choice between the classic menu and the new XP start menu. What about the classic view of the folders, control panel, My Computer and other icons missing from the desktop??? What about Office XP hiding items that aren't used often. Seems to me that's not very standard either. If I have customer on the phone, most of them are not to good with computers, most of the time i don't even know where the button is.

Well this is linear advancement, i.e. upgrades from one version to the next, not concurrent development. Also, someone please correct me if I'm wrong, but they still use the same API(s).

Windows has package formats also. What about MSI, MSI 2.0, Nullsoft Installer, InstallShield, WISE installer etc etc. At least if you have RedHat or Debian you can install a package on all machines in your company just with one command!!! ie. apt-get install openoffice.org It will take care of the packages for you.

Ahhh, but all those Windows formats are self-extracting, yes? Also, again they are not all (although some) deliberately concurrent, there's a progression (with MSI, for instance). I don't recall ever having to go out and download a new package manager for Windows just so I could install software. Even when an application installed it's own installer, it was always an automatically launch and intuitive dialog box driven thingy. Next, Next, Next, Done. Not so with the jumble of Linux package managers. This is because in Linux the package mangers try to do it all, i.e. be the installer, but also the registry too. In Windows there is only one registry (to the user, any way).

By the way, apt-get only works if there are no conflicts. More on this later.

I don't think windows has less suites available for productivity. Is this really an issue? I guess most people didn't bother looking for other office suites. But i can tell you some of our customers are seriously thinking about using openoffice! Office 2003 is just so expensive, and who needs all those features?

Well there's a clear standard format. It's very simple, if you are in the business world you must be able to read and create Word documents, there's no avoiding it. It's bad enough that OOo can't save in office formats, but why have a whole bunch of different suites that can't do the same thing? If Linux wants to seriously compeat with Windows (is that really the goal, any way?) then it has to have it's own strong standard to rival Microsoft (while maintaining the highest compatibility possible). I simply don't believe this can happen with everyone doing their own special office suite.

I don't think depencies is a OS-related problem. For most games you need DirectX or a new video driver. For most Visual Basic applications you need runtime dll's. For .NET you need .NET framework. Most programs rely on other programs. That's not a problem. Just a fact. At least most packaging systems warn you for depencies before hand. And with tools like apt-get and yum you don't even have to worry about downloading it manually, like you would have to do with DirectX and stuff.

Well, I shouldn't have said "dependencies", I should have said "the dependency hell". It seems like many Linux apps are linked against every library in existance. One of my friends told me he was trying to remove some of the Gnome bloat from his system when he got the dire warning that /sbin/shutdown was linked against that Gnome library... is this really necessary???

How many times has the situation happened to you where you have foolib-1.5 install, then you try to install bar-1.1, but whoops, that depends on foolib-2.0, but you can't just remove foolib-1.5 because 30 of your existing packages depend on it, but then again you can't force foolib-2.0 because it conflicts with the existing foolib-1.5, ARRRRRGHHGHG! This is the biggest thing that made me hate Linux. It's not that things *couldn't* be done right, it's the fact that they *aren't*. This is a sign of very unprofessional and unstructured development. The fact is that Linux is *too* decentralized and *too* open to contributions. The quality really suffers.

This is the biggest problem I see with Linux right now: The fact that anyone who thinks they're a programmer can write code and post it on Sourceforge, or Freshmeat, or something like that and have people download it. If it does something slightly different than existing software, it might interest people, eventually it finds it's way into a distro, now it's gone completely from just a weekend hack to some software that's being bundled with a system, but it was never tested against that whole system (nor likely, any other system). It was only tested on the author's box and perhaps a few friends.

Take hdparm, for example. That's a really widely used utility in Linux now, lots of people rely on it. A few of my friends who are professional, paid software developers looked at the code and just about gagged. The sytle was horrible, there were gigantic routines instead of tight functions, etc... This is not especially shocking, even Linus admits that his early code was really, really bad (he said something about it looked like a newbie trying to learn C, which he was at that point).

The quality must come up, which means that people with real skills need to be dedicating time to this software. In most cases, that will mean people getting paid to write code. This is why enterprise Linux needs to take off, to fund the development, but there's a very real chicken & egg problem because how do you afford developers if enterprises aren't buying? The best hopes at this point are Red Hat, Novell, and actually the eeeevil corporations, IBM, HP, Sun, etc. Those are the companies with resources.

I partly agree with you on this one. Although almost all hardware on my pc works on my laptop. You should check if your hardware is supported with linux. I have the following things working (right out of the box): Wireless networking : Rock Solid! On windows I had problems sometimes with it forgetting my key every once in a while My mousepad. USB 2.0. Sound. DVD-player. CD-burner. APM suspend. CPU frequency adjustable. NVIDIA 3d opengl stuff. What would you need more? Just make sure you get the right software. It feels to me that is a while ago since you tried LotD because all the problems you mention don't seem to affect me.

Well it all worked for you, that's great. Your one laptop does not represent all the laptops in existence being used by enterprises. I've tried a couple of live CDs on my work laptop, some worked, some didn't. All required me to know that I had to disable SCSI prior to attempting to probe hardware (found that out the hard way). Some live CDs that were actually based on a common source behaved quite differently. One had severe problems with my LCD display, while another one (based on same distro and same original live CD layout) had no problems with the display.

I personally haven't tried LotD (well, aside from live CDs) in well over a year. I have no desire to because of the very poor documentation (the man pages in most distros are pathetic) and all the dependency hell with linking everything to world + dog. I use BSD now and the most clear differences are in the control of outside packages (pretty much avoiding those dependency nightmares) and excellent documentation.

This is not a "let's bash Linux" post, though. On the countrary, I'm showing what has to change in order for Linux to be widely accepted in businesses. These are things that cause countless hours of frustration for helpdesk people and staff, who want things to "just work". For all the complaints with MS software, (cost, insecurity, monopoly anti-competitive tactics, etc) you have to admit that for the most part, their stuff "just works".

The two main weaknesses of MS are cost (they have the most eeeeevil licensing schemes), and the horror of their patch management. Price has flexed a lot lately, especially in government and educational contracts (they're trying very hard to not loose to Linux on price), and they've done a number of things around patch management, although the underlying problem is all the vulnerabilities which generate the huge number of required patches.

If Linux can stay competitive on cost, while solving the problems above, and clearly having a better story than Microsoft on patch management, then it'll catch on. The danger is that in the rush to get out new features, Linux will be "release early and often", which is very much like Microsoft's "rush to get it to the market now, then patch it later... lot's of times..." approach.

Basically it all boils down to quality, since that's what's biting MS in the butt now, but it's also a major problem for Linux as well. It gets really simple: Whomever can deliver better quality software, cheaper (TCO-wise), wins.

Oh, while I'm at it, here are a few more pointers in moving Linux into the enterprise:
1.)Enterprises are not religious. They don't hate Microsoft in a philosophical way, they hate them in a "wow they really cost us too much money and their vulnerabilities cause us a lot of problems" kind of way. Many, many Linux proponents seem to assume that the world is just seething with anger against Microsoft--this is not true. Enterprises don't care for RMS and his crusade for destroying Microsoft's whole model of software (create, own, license, relicense). For the Enterprise "it's not personal, it's business". Besides, a lot of these enterprises are actually software companies and they don't exactly want to fight against their own profit model.

This means that, while Debian might be very popular with Linux enthusiests and lot's of people might agree with their stance on "not including any non-free software, firmware, or documentation" with their base distribution, an Enterprise would just see that as irrational and would be troubled by the fact that they had to add in a lot of basic pieces manually because the distributors were too purist to allow it. That's why you'll almost never find Debian being used in any kind of business situation, while Red Hat and SuSE are.

2.)Enterprises hate risks. This means that, as a general rule they don't want to do business with a small, unheard of, and inexperienced company. That sentiment is increased 10-fold when talking about something as critical as the Enterprise-wide operating system of choice. They want a company they know will be around for the next several years, they want support, they want integration assistence, etc. All that talk of "free" Linux replacing ____ will never happen. I've seen a lot of companies that use Linux, but only one of them that used completely "free" (as in beer) Linux (and that's because it was a box the sysadmin basically kept secret). Linux will start invading the Enterprise, but not Debian, Slackware, Gentoo, etc... It will be Red Hat (which dominates currently), SuSE, and SJD (Sun Java Desktop).

So to wrap up, again I didn't mean this as a "bash Linux" post, but as constructive criticism. More than anything I want to get across the thought that what motivates the average LQ reader, and the majority of Slashdot readers, is not the same thing that motivates enterprises. It's also irrational to think that a hobbiest OS can be mainstream while continuing to behave as a hobbiest OS. The quality needs to improve dramatically, the focus has to be might tighter, and developing towards clear common goals (of the target audience, in this case, enterprises) has to replace the current trend of "developing for me" or "developing what my buddies and I think is kool". Commercial Linux vendors "get" this, but most Linux users on the street do not.

I work as a sys admin at a small boarding school. I have been steadily converting desktops and work stations to Linux. I didn't really give them a choice. The Windows boxes that I converted did have the liscenses and the school could have been in trouble. Linux was the only way to go or by liscenses. Being a nonprofit they didn't want to spend the money. Solutions solved. Sure I had and still have to teach classes but they are really getting a better deal from that as I teach not only Linux applications and their usage but also general computer tech. In my opinion everyone wins.

First off, thanks jeremy for adding this forum. It will be interesting to see how businesses are implementing linux and the problems or success they face.
Chort, I see your points and they are good ones, there are a couple that are questionable.

quote:
This means that, while Debian might be very popular with Linux enthusiests and lot's of people might agree with their stance on "not including any non-free software, firmware, or documentation" with their base distribution, an Enterprise would just see that as irrational and would be troubled by the fact that they had to add in a lot of basic pieces manually because the distributors were too purist to allow it.

Does not windows come as a base os, but to have any fuctionality for an enterprise or just plain business, applications have to be bought and installed? I understand your choice of debian, but a business would make their choice of a distro, based on their own needs. Adding applications would take place no matter what os is used.



For all the complaints with MS software, (cost, insecurity, monopoly anti-competitive tactics, etc) you have to admit that for the most part, their stuff "just works".

Why then do bigger businesses(enterprises), universities, etc. that use microsoft products have help desks? Any os that is being used will have problems, but saying that ms software just works is questionable, (compared to linux?)

Just my thoughts and not picking on you chort as I do value your contributions to this board.

Originally posted by twilli227 quote: This means that, while Debian might be very popular with Linux enthusiests and lot's of people might agree with their stance on "not including any non-free software, firmware, or documentation" with their base distribution, an Enterprise would just see that as irrational and would be troubled by the fact that they had to add in a lot of basic pieces manually because the distributors were too purist to allow it. Does not windows come as a base os, but to have any fuctionality for an enterprise or just plain business, applications have to be bought and installed? I understand your choice of debian, but a business would make their choice of a distro, based on their own needs. Adding applications would take place no matter what os is used.

You missed what I was talking about. I wasn't talking about applications, I was talking about fundamental documentation, and firmware for things like network cards, etc. Those things you just can't live without, and in every other operating system you get them (actually the firmware thing is very weird). This doesn't have to do with just keeping the OS "lean and mean" (like OpenBSD and NetBSD do), it's about being FSF supporters to the point of zealotism. Being purist might win points with people that agree with you, but it doesn't win customers. Now of course, I have never seen anything to indicate that Debian in particularl wants corporate customers. I'm just pointing out that this behavior is considered exemplary to many Linux users, but is in fact very detrimental to the pursuit of trying to get Linux in the Enterprise.

For all the complaints with MS software, (cost, insecurity, monopoly anti-competitive tactics, etc) you have to admit that for the most part, their stuff "just works". Why then do bigger businesses(enterprises), universities, etc. that use microsoft products have help desks? Any os that is being used will have problems, but saying that ms software just works is questionable, (compared to linux?) Just my thoughts and not picking on you chort as I do value your contributions to this board.

All companies have help desk personnel (or at least some unlucky person in IT that gets to answer the phone). Some users would get things wrong even if they were given a tool with a single button that says "make it work".

What I mean by "just works" is that you don't have to do tweaking and configuring to get things to work for you. There's very little knowledge assumed. You don't have to compile things. You never have to edit a config file. When you go to install something, you just click next, next, finish and it works correctly.

Put it this way, Linux presents users with lots and lots of options, and very little steering as to what to choose. The responsibility is on the user to know which options they want. Which terminal emulator do you want? Which text editor do you want? Which browser do you want? etc... Choice is great for people that have religious reasons for using every app they ever execute, but if you're just trying to do you job that has nothing to do with computers (answer the phone, sell things, pay people, fill perscriptions, etc) you want one button that says "the Internet" and another button that says "e-mail", etc. Commercial Linux' "gets" this. They're trying to come up with unified desktop suites with once choice in everything that is pre-configured and just ready to use.

The ESR example with CUPS is sterotypical of Linux in general as a desktop OS. Linux applications, as a general rule, are not configured to be friendly and forgiving towards unskilled users, they're designed to be open-ended and flexible. Flexibility is something that geeks and tinkerers cherish, but not something you want to confront a secretary with. Give a user enough rope, and they'll figure out a way to hang themselves. The trick is to play out just enough rope that they can do their job, and make sure they only have one rope to choose from to minimize the chance of tangled knots.

Hi all, first of all I'm sorry for my poor english.....I like this forum, but I don' t understand if this is a technical forum or only a philosophy one ;-)

I'm new in the Linux world, in about one month I begun to use it in my home pc and it rule! I am a sys admin in a big fashion company and I'm tring to install and use a LotD (I like this :-) in my office .....We mainly use MS Office, Lotus Notes and AS/400 emulation client in our desktops plus some other small utilities. Well, sure it isn't very simple but it works....
I've found a lot of programs and now my LotD is up and running....what's the matter if I have 2 (or more) desktop environment, 4 (or more) web browser, bilions of word processor, editors, spreadsheet and so on? This is usefull I think....We can choose the one we prefer and use it.
MS windows is a point and run environment, maybe Linux is a bit (more then a bit I guess) more difficult to setup for the IT but at last it has the same opportunities.

Now, a technical question, does anyone now if exist a good 5250 (AS/400) emulator for Linux? Thx !!!!

Great post!

All I really have to had at this time of the night is:

1) I have a licenced copy of Windows XP and Office 2003 Professional and I choose not to use them anymore. So this post really has nothing to do with cost of ownership, becuase I own MS product but choose not to use to because there is a more stable, rewarding and friendly alternative with a fantastic community willing to help you should you encouter a problem.

2) Linux for laptops needs more work... I'm running Linux on a Sony Vaio Z1 (real big in the business world)... an it has real issues suspending (hibernating) at the moment and consequently I have to actaully reboot more often with Linux that I ever did with Windows. I don't mind as I'm a big Linux fan and know with a bit of time and/or hacking around this will eventually be sorted.

3) KDE is a better desktop environment than Windows ever was. It seems they must be getting more investment now as they are releasing updates more often and if you look at the features that come as part of the basic KDE package Windows doesn't even come close.

4) Call me stupid... but isn't typing emerge xxxx a lot easier than having to click next next next.. when you install? Incase you thought InstallShield was windows specific, here is an extract courtesy of their web site:

"InstallShield X has built in support for virtually every platform, including Mac OS X, Windows, OS/400, Solaris/SPARC, Solaris/Intel, Linux/Intel, Linux/PPC, HP-UX, AIX, FreeBSD, Tru64, and every other UNIX platform."

I believe strongly that with the industry is opening their eyes to Linux with the support and collaboration between many world players (i.e. IBM, Sun, HP, Novell) occuring in the last couple of years based around Linux OS (way of thinking), which also has the support of a dedicated and skillful community. I for one am staggered at the rate at which Linux application are being developmed at the moment and have never had any real problems with software hot off the press.

No doubt Microsoft will have to pull more dirty tricks if they want to maintain their pole position, but it's about time the school bully met his match.

Fastly

"InstallShield X has built in support for virtually every platform, including Mac OS X, Windows, OS/400, Solaris/SPARC, Solaris/Intel, Linux/Intel, Linux/PPC, HP-UX, AIX, FreeBSD, Tru64, and every other UNIX platform."

Very interesting quote, and while that may be so, I haven't seen a single *n*x package available with an InstallShield installer--ever. That goes not only for Linux and BSD, but for commercial UNIX like Solaris, etc (granted I haven't used commercial UNIX nearly as much, but I did run a lab with Solaris boxen for two years).

Now, I suspect that the reason is because InstallShield is a commercial product and they charge for licensing (for the full software, any way), hence none of the Open Source projects want to use it, and of course Sun, etc have their own packing systems that were developed (so far as I know) well before InstallShield was around.

The point really isn't about InstallShield, though, it's about the lack of a standard for Linux packages. In fact, there doesn't seem to be much desire for a standard package system amongst Linux distributions.

I think what it boils down to is choice, knowledge and flexilility.

There are advantages and disadvantages of each distributions package management system (and not using one at all).

I've been happy with package management systems, but it seems the problem lies in interoperability between distributions and their homegrown package management methods... i.e. using say a redhat RPM with a Gentoo install... however most RPM's for standard packages are ebuilds anyway... and if I am running a commerical application such as Zend Studio (for my PHP development) I belive it comes with a neat InstallShild installed that has yet to fail and I've used it on Redhat, SuSE, Slackware, Debian and Gentoo...

So... you want free software download the tarball and deal with the dependencies or use one of a number of mature and reliable packagement management systems...

Or... if you buy commercial Linux software use their installer - it will be just like installing a windows app.

I do belive there is room for an opensource IntallShield type project?

Fastly

Open source is about choice. With all this talk about a single environment and a single productivity suite we will become nothing more than a single choice. Let's face it, KDE and Gnome will be the big two desktops for corporate users and the big three RPM based Linux's (RH, MD and SuSE) will rule the enterprise. Gentoo threatens to go enterprise soon as well.

RPM has it's strengths, ports based distros are great as well. Competition is what drives ingenuity. Staroffice, Openoffice, MS Office, Corel, Lotus and so on...good for them. Whoever makes the product that someone buys and likes then everyone is happy.

CDE was supposed to be the desktop uniformity push and look what happened to that.

Originally posted by ghight While I see some hurdles that don't yet make Linux desktop a viable option, I disagree on many of your points. I think what you have mentioned is the very thing that makes Linux as good as it is for the masses. Many of your points should and could be settled on a business by business basis. They can standardize just like anyone else. If "somebody" decided to standard Linux on something I didn't like, they've done the very thing that made me switch to begin with.

One of the things you have in major distros (like Mandrake) is the possibility of installing the software in the machine and make a installer diskette that saves all configuration packages so it is really easy to reinstall it all over the place.

Another solution that is gaining more and more potentiality is the "Make your own distro" approach where you can build an instalable LiveCD.

Many LiveCD distros supply scripts so you can make your own; Slax, MkLiveCD, Knoppix and Mepis are just some examples.

For instance: PCLinuxOS is based on MkLiveCD scripts and Mandrake Linux and they don't like the RPM manager that comes with Mandrake so they use Synaptic instead.

Another good example of a self maid stripped off Linux distro is RainbowLinux that you can see at http://www.MandrakeUsers.org/index.php?showtopic=10851

It is rather nice that you can have several X servers, Window Managers, Desktop Managers, Office Apps and so on. This gives you the Liberty of CHOICE.

Best,

Lopo

I think it's pretty clear that Linux has a wide degree of choice available, but you guys aren't listening to what I'm saying. Enterprises don't want 500 flavors of Linux to choose from, they want one (or a couple) of "corporate Linux'" and that's it. Really, at what point is 500 better than 5? Corporations (most of them) also do not want to "build their own" OS, they want it in a nice package and handed to them, or better yet already installed on all the PCs they get.

Now, the LiveCD route might be great for VARs (Value Added Resellers) and consultants who build firewalls, IDS devices, etc on top of Linux, but they aren't really great for wide desktop roll-out because now they're even less "standard" then the already fragmented an unstandardized Linux distros they started from.

Please try to understand here. I've been dealing with networking software, consumers, and enterprise customers for about 6 years (comparitively few I know, but I suspect that's a lot more experience than your typical Linux user). My last two jobs have been selling to and supporting some very large enterprises, and as part of that you really had to understand what the customer wants and how their environment works.

Trust me, they do not want 2342323 choices, they want to pick a category, have 2 or 3 of the "top contenders", and then evaluate them to decide which they like best. Customers get upset when there isn't a clear leader and they feel they have to examine 5-10 choices to narrow down the field. The vast majority of them also do no want something that is so open-ended that the customer ends up having to essentially build it; they want something that they can drop in with minimal modification and make it work.

LotD really isn't about all this free-ranging "choice" of hundreds of flavors of Linux, it's about the choice of Microsoft or ____, i.e. the fact that there is something other than Microsoft as a viable option (both as a real option, and also very much as a bargining chip to bring MS's price down).

This is what the typical Linux user does not get about selling software. While individuals (well, very geeky individuals) want lot's of choices and the ability to micro-manage their own preferences (favorite text editor, favorite browser, favorite mail client, favorite desktop environment, etc) corporations have completely different priorities. They want something that is easy to deploy, easy for their users to understand, easy for their helpdesk to support, easy to keep up to date with patches, and easy to install 3rd party software on. Having lots and lots of options doesn't really make any of those tasks easier, and it makes a lot of them harder (training, helpdesk support, patches, etc are all vastly complicated if you have a bunch of "options" i.e. 5 different pieces of software available for every task).

By the way jcookeman I'm sure you know that Mandrake natively uses the .mdk package format, and .rpm is only secondarily supported. Also, of all the office suites you mentioned only MS Office really has any kind of install base, so clearly it's not really the competition that drives office suites, it's the fact that MS Office always syncronized with MS Windows releases and took advantage of being the only suite to work with the new Windows whenever possible. Now they have software bundle deals with anti-competitive practices making it nearly impossible to break the strangle-hold, since .doc, .xls, etc are the "industry standard" file formats that you have to be able to send/receive in.

In many areas, it's not competition that drives the sale but bundleing, integration, package deals on software, etc. Now Sun is trying to chip away at that with their Java Desktop, and of course there are two primary reasons why they've had some initial success: price, and the fact that it's an integrated solution with no confusing options. You only get the Sun software, that's it. One e-mail client, one document editor, one spredsheet program, etc. Notice that Sun has been able to steal some deals away from Microsoft with this while there have been many other Linux trials, but few of them have resulted in Linux winning (even some sponsored by IBM).

If you want to sell something to the Enterprise, you have to understand the Enterprise. From everything I've seen on Slashdot and most other places, the people who are being most vocal about LotD and LitE (Linux In The Enterprise) have the least idea of what it will actually take to get there and what enterprise priorities really are.

I'm not trying to be critical to just throw a lot of cold water on your ideas. I'm trying to get everyone to see things from an enterprise customer point of view so you can actually be successful.

No joke. There are a couple of heavy hitter Linux distros for enterprise. I don't get what the long posts are all about.

And, yes, MS Office has a big stranglehold on that market, but there are a lot of people that use Staroffice, Corel and Lotus. This is mainly due to the reasons you spoke of.

RH, SuSE, Drake and SJD...

The beauty of open source is this, you will have 7000 distros out there because people have the power to create them.

The long posts are about your last sentence. Businesses don't see "beauty" in 7000 distros, they see lack of direction. That's the message I'm trying to convey.

Also the fact that Corel and Lotus exist is pretty much a non-factor, since they're the extreme minority of market share. Nothing is ever going to dominate 100%, but if you have 90-95%, that's essentially 100% of those who are willing to go with the "standard". The other 5% or so will never pick the "standard", for various reasons. So I have to say "who cares if they exist?" They aren't gaining market share, and they don't have a respectable market share right now, so I'd hardly point to them as a reason for the viability of Open Source office suites. Just because something exists doesn't mean it's successful.

Why do I post long posts? Very simple, because no one seems to "get it". Everyone talks of wanting LotD & LitE, but no one understands what it will take to get there. Case in point is Gentoo wanting to come up with an "Enterprise" release of their software and figuring it will "only take 6 months" to do it. That's rubbish. Nothing they come up with in 6 months will attract more than half a dozen customers of any size, let alone even medium enterprises.

I'll stop posting long-winded dissertations when people start indicating that they really do understand what it takes to market software to medium-large size businesses.

Ok....first off, you are preaching to the choir. Secondly, there is Linux that is Enterprise ready. They've already been pointed out in this thread. If the fact that other distributions exist is a distracting factor for anyone considering open source in their "enterprise" then they are very short minded.

I do not promote the destruction of Microsoft. I promote more competition. It is understood that they have a monopoly on the marketplace productivity software.

You need to stop the attitude that "no one gets it". There are people that come across this forum with vast experience and they understand.

As far as Gentoo being ready for enterprise release, no one knows except for the developers. I think we all know they won't be at the level SuSE and RH are at.

The thing that primarily stops Linux in most business applications is because of obscurity, period. People just don't know what it is, their people aren't trained on it well and they are afraid.

I do a lot of consulting for businesses in the area and they are all curious, but are afraid because they "grew up" with Windows and don't want to take the plung. In the DoD we are starting to use Linux. The reason we are using it more now than in the past is because of its previous immaturity for multi-threading and its SMP scalability issues. But, those problems have been solved now and we are much more ready to accept it.

Linux is being used in very large scale. Almost all F500 companies are using it, IBM is now 100% behind it and obviously Sun has started to get in the mix with SJD, a very nice product.

It's used in major corporations for clustering for crash tests, automation, modeling, etc. Almost all technical US government agencies use it as well. So Linux is getting ready, fast, but end users need to be introduced and get some exposure to the technology before they are comfortable spending the millions of dollars it will take.

Just think of the money involved when thousands of employees production slow down and you have to pay people to train and take time out for that and IT staff trained and hire new people.

The tools out there are becoming great. Yast was opened up by SuSE and Sun is using it for SJD. I got away from Mandy and Redhat some time ago, but the ones I've seen from others running it recently look like they have great package managment and good scalable solutions for desktop deployment.

So we're on the same page then, great

Just one minor nitpick, when you say Linux us "Enterprise ready" that has dual interpretations. The first is that it's ready as a server platform, which I agree with and I think pretty much anyone else would to. The second possible interpretation is that it's ready as a mainstream desktop OS, with which I disagree (except in the special case of SJD).

I'd be interested to know what distributions of Linux you've seen in use at large sites. So far I've seen about 95% RH and 5% SuSE (domestically here in the US, and mostly F500 businesses, not webhosting or other "providers").