A few weeks ago, dot.kde.org featured a great why-should-this-be-amazing story about Linux being used as the day-to-day desktop operating system for city employees in Largo, Florida. Roblimo got a chance to see the system in action to find out how ordinary office workers are proving that the old "Linux is tough to use" shibboleth is nothing but FUD, and how a medium-sized city is saving buckets of money by minimizing the tax dollars spent on licenses and hardware. Oh, and they've also pre-empted the kind of costs (in hassle and money) that can face any organization that Microsoft suspects may have some licenses out of order. This is the kind of thing every elected official should have politely waved in his or her face by concerned taxpayers. The Largo system uses KDE on Red Hat, but since both KDE and Gnome are paying much attention to user interface, similar systems could easily be running on various combinations of hardware / distribution / desktop system.
I'm glad they're using Linux. Unfortunately, there's no way the medium-sized business I work for could do that. For us, the problem isn't usability, it's software. And I don't mean a lack of MS-Office. Let me explain.
Like most businesses our size, we use a variety of custom, semi-custom, and prepackaged applications. While, yes, we could use free alternatives to our operating systems, office software, and email/scheduling software, there's no way in heck we're going to find a free replacement to our inventory, financial reporting, or human resource management software, for example, any time soon.
Even if someone came up with one, the cost of switching to the new packages would be enormous, given the complexity of such software and its impact on day-to-day business.
What's more, the specialized software we use requires that MS-Office be installed or it can't do simple things like generate reports. So, if we have to stay with these inventory, HR, and financial reporting packages, that means staying with MS-Office and MS-Windows.
Oh, how I'd love to move the whole corporation to free software. But there's just no way that's going to happen until all the mid-size apps are moved to Linux.
Until then, I'll continue to use FreeBSD, OpenBSD, and Linux on the server side as much as possible. But that's where they'll stay for the forseeable future.
One thing I noticed when I installed Redhat on a machine (I'm a Linux newbie) was that it was very usable by anyone, provided someone else configured it.
Easily, my non-tech-savvy friends could get used to KDE and become comfortable in it, but I don't think they could set it up to be usable (nor do they care, and rightly so, they shouldn't have to). They could actually install easily--the Redhat install was exquisitley easy--but as far as installing programs, setting things up the way they like, etc., I don't think they could.
Then again, many people can't do the same with Windows (installing programs and configuring it to their liking).
One thing I noticed when I installed Redhat on a machine (I'm a Linux newbie) was that it was very usable by anyone, provided someone else configured it.
And that's a good reason to prefer Mandrake over its Red Hat cousin (or Caldera or Corel over either) in that situation. Half the bitching here and elsewhere about Linux could be eliminated if people would pick appropriate distros for the job at hand, rather than becoming too attached to a particular one, often for ideological reasons, which are often the wrong reasons. Try several - cheapbytes is your friend. When you find one you like, buy a boxed copy.
The "right" distro depends on what I'm trying to do: I prefer Caldera or Mandrake on the desktop, Red Hat on servers (it reduces support problems), e-smith for garden-variety multi-purpose office servers, Turbo or Red Hat on mainframes, and Lineo or BSD for embedded devices. Choosing wisely (and widely) keeps you from the IT equivalent of hammering in screws with a pair of vise grips...
Oh, and be flexible - the costs for switching between distros are low and becoming less as we move to LSB-land, so don't be afraid to switch when it makes sense.
KDE was likely the most crusial choice. Even moreso than Linux. For the stuf office workers do all day KDE is realy the best Unix based solution (Not counting MAc OS X which I havn't seen myself but have herd good things about).
I am all for being nice to "the other side" on these things but what I see is people strugling to use Gnome for ideological reasons and other people getting work done with KDE for financial reasons.
You know Finacial reasons like "Less money spent on Asperin", "fewer monitors shot at" and best of all you can fix the problems that do come up for less than it costs to fix the stuff you pay a grand more per seat for.
I am all for being nice to "the other side" on these things but what I see is people strugling to use Gnome for ideological
reasons and other people getting work done with KDE for financial reasons.
What are you basing this on? As a consultant in St. Louis, MO, I have had the pleasure of working with a few firms who provide Red Hat/GNOME solutions for corporate desktops and workstations. And GNOME use is rapidly expanding from Red Hat (and other distributions such as Debian and Turbo Linux) to other Unix variants. With Solaris switching to GNOME 2.0 in place of CDE as the default desktop environment, and HP-UX likewise embracing GNOME, you're going to see even more validation for GNOME on the desktop in the next 6 months to a year.
All this is to say, where are the facts that support your statement?
All this is to say, where are the facts that support your statement?
I didn't make the initial statement, but I will respond. All my GNOME systems are up to date as of today.
The GNOME 1.4 release significantly diminshed speed and functionality to achieve only a baic file manager / web browser type app. Its very unusual that someone would create a new version that does LESS things than then old did, but I still can't create a new application launcher from the desktop or edit an existing one as with KDE
* The current GNOME control center (yes there's a new one on the way, no it isn't here yet) is confusign with its `test' `OK' implementation. This isn't consistent with many other GNOME apps
* 48 x 48 icons that are `supposed to look good' at 20 x 20 often don't. One size does not fit all.
* AFAICT there's no MacOS / Windows / KDE type style guide which can be used to define consistency between applications
* Defining a filetype -> program mapping is difficult in KDE but especially more so in GNOME.
* GNOME still has many programmerisms within it. Sawfish and GNOME might be seperate apps but from and end user viewpoint they should work seamlessly. Having a `meta' button under the GNOMECC which only defines options avaliable for the Sawfish branch is one such programmerism. And what does `meta' mean to a non tech?
The complaint that Linux is hard to use is not one necissaryly of GUI per se, it's set up. Basic quality assurance questions are hard to answer:
What hardware are you going to use? What software are you going to use? Where is that software located? How do you install that software? Where is the software located on the Hardrive after you instal it? How do you get OpenGL to install? Why did it take me 15 minutes to find the PPP dialer? How do you set up a network where hard drives are shared? How do you put things into the menu?
If a sys-admin takes the time to form a planned approach for Linux installs, makes a custom CD for installing the "supported apps", then Linux can be usable. Problem is the Linux approach of shipping not only with the kitchen sink but 10 different kitchen sinks to choose from, all of which are disassembled and in their boxes is not one for newbies.
1) Largo has been running a Unix shop and thin client network for years. Only the switch to Linux and KDE 2.1.1 is new. I doubt that one or two sysadmins leaving would change things.
2) This is a done deal, not a "someday" or "we plan to" thing. I wandered around Largo city hall and talked to actual, everyday users.
3) I'd like to go back and speak to Dave and Mike in a year, yes -- to see how their plans to use OpenOffice pan out. The biggest holdup (as I wrote in the NewsForge story [newsforge.com] linked to above) is the lack of a good OpenOffice filter for WordPerfect files.
<sarcasm> Oh, right. Let me guess: where you work, all the secretaries installed NT on their own, and as soon as those W2K boxes arrive, they're going to upgrade. </sarcasm>
elefantstn, it is quite easy to order in boxen with NT and MS Office already installed. (Probably easier than getting boxen with no OS.) It generally isn't an optimal installation, but it's often good enough for secretaries.
Where the network admins get involved is usually when it comes to interfacing with the main databases. With 400 seats, you _need_ a database system that makes MS Access look like a toy. (And I speak as one who has attempted to take Access right up to the limits MS admits to...) So unless you've somehow managed to put your whole database onto a IExplorer-compatible website, you are going to have to install a database client on each machine. The front-end one I'm familiar with (for a Progress database running in Unix) is a one-floppy disk 5-minute install, which is a heck of a lot easier than installing Win or Linux.
IMHO, the best way to provide corporate desktop computers is to buy hundreds of identical boxes, find the optimal installation for Linux and required applications, and then clone that configuration. But if they won't give you the budget to buy computers and put them in the closet until needed, then you wind up ordering one at a time for new hires or to replace broken down machines; almost every !@#$% machine is different (at least from lowest-bidder sources), so in Linux you'll spend hours on each one downloading drivers, setting up the configuration, etc. Windows pretty much requires the same time if you install it yourself and bother doing it right, but when you buy pre-configured boxen, it certainly looks like you've saved all that labor -- except for the six times that box is going to crash or catch a virus and have to be re-installed, but that cost comes later, and management isn't going to add it up and realize that buying the alleged industry standard was really boneheaded...
One thing that many of the people on these forums forget is how much of a special case we are. Your average office work is never permitted to install their own software. People have been arguing the case for thin clients for year because the average office user has no need to do their own installations and should be discouraged from doing so. I mean, we don't 'em wasting time with net Quake or anything. Not to mention ungodly mess a user administered machine can become. And, if you're running the type of place where people share work-stations like in a call center or clerk desk then you need to preserve uniformity from one session to the next so as not to confuse or completly derail the workers. Customizing the desktop is one thing, but installing any random app is a bad idea.
In that light why the hell should the place even expect the users to need to know stuff like that. ? In terms of apps being available the usual course is to go through IT management and request it. And, if you have a user who's knowledgeable enough to want a specific linux app by name, why not consider moving 'em into the tech department?
This is not a "special case," but one that could easily be duplicated in almost any government or business office environment that runs enough desktops to have its own sysadmins -- or at least a contract with a Linux-hip outside contractor -- to take care of the network.
My wife has worked in more than a few government and commercial office environments that ran Windows, and they *always* had a separate IT or network support staff to take care of the computers. She wasn't supposed to add software or even mess with the things at all. In fact, in her last "real" job, doing customer support for a pager company, the biggest office computer network problem they had was employees bringing in software (especially games) from home and installing them on their own. Often the self-installed software screwed things up like mad.
Running a client/server network completely eliminates this problem. It also makes updating productivity apps a lot easier for the sysadmins.
This is not a "special case," but one that could easily be duplicated in almost any government or business office environment that runs enough desktops to have its own sysadmins -- or at least a contract with a Linux-hip outside contractor -- to take care of the network.
The actual "special case" is probably the single machine senario (and certainly the "home" computer senario).
My wife has worked in more than a few government and commercial office environments that ran Windows, and they *always* had a separate IT or network support staff to take care of the computers. She wasn't supposed to add software or even mess with the things at all. In fact, in her last "real" job, doing customer support for a pager company, the biggest office computer network problem they had was employees bringing in software (especially games) from home and installing them on their own. Often the self-installed software screwed things up like mad.
Not only that it probably helps turn the IT staff into BOFH clones.
Man what type of crack are you on? I've been a sysadmin here for 2 years, and tried to push out linux to users who wanted it on several occasions. And guess what? Linux is great -- that is if, and only if, you have perfect hardware and perfect setup and a standalone system. The minute that something is out-of-spec, linux goes AWOL and the poor desktop user is SOL because they don't know the difference between KDE and dd. OTOH, I've plopped in the install CD for W2K, filled in a few simple details, walked away, came back 30 minutes later and had a system up-and-running without any problems. Sure back in the day of WinNT4 (I humbly agree that Win9X was a POS, but don't get me started on why those existed and why customers demanded Win98SE and WinME...) there were a bunch of problems, but I have had very few problems installing windows 2000 systems (and *zero* on reputable machines (i.e. Dell), well there was that 1 problem, traced to a defective HD)... And secondly, as soon as the user wants to do something new with their system, they're SOL again, not only because installing isn't as simple as "click here to download, run setup.exe and you're installed" (albiet lately in linux it has gotten a LOT better), but also because the apps simply aren't out there.
Oh BTW -- check out my W2K Server, up for 131 days without a reboot, and also survived CodeRed without a scratch because I set up security properly from the day that I installed the server. There's a cool realtime stats program up on the CodeRed attacks and other neat things: here [ohmygodmyarmfelloff.com].
To recap, simply, I like linux. I think that it has a lot of potential but it simply isn't anywhere close enough yet to be a mainstream system. Remember this [slashdot.org], and the fact that it's manhours spent with linux as well. I hate to be the harbinger of news here, but windows is much easier to use, period. There's no debate about that one, and with XP it just gets easier. Try putting your mother down in front of a linux machine, and then do the same with a windows machine. There is a reason why windows is used on 90% of desktops, and why Microsoft is the software giant that they are (reasons beyond the typical slashdottery about squishing competition and cheating and crap), more than just "being in the right place at the right time".. It's because, for better or worse, they have the best set of software products out there. Office and Windows are extremely successful because they're good, and people like them and use them a lot. That's a fact, hard to dispute.
I don't think that linux will make it as a mainstream OS anytime soon, or at least until most of the linux users (BTW, I think that part of the reason of linux non-acceptability is because of the typical i-love-linux-and-hate-windoze attitude and immaturity, not everyone, but just enough are immature and slander and swear and yell and scream and kick and fuss and act like children to give linux a bad name. Don't believe me? read our very own CmdrTaco [slashdot.org]. I think that he made some really good points there, the thermostat in hell must have broken that day...:> ) are part of the reason.
Oh well. I think that until linux users give up the I-want-everything-for-free-as-in-beer-as-my-god-gi ven-right (what IS UP with that slashdot auto-spacing-long-lines-because-they-must-be-evil- and-would-confuse-the-reader-so-we-have-to-insert- spaces??-) mentality and start paying for things again (yes, I know a lot of linux users pay for distros, but VERY FEW do compared to the manhours put in to develop and sustain it, and VERY FEW do compared to the number of people who USE it.. Actually, it costs more to put out a distro because of the costs of the bandwidth that you must provide to allow people to download the thing) so that businesses will start to develop seriously for linux, things will not change. With the yelling from business about wanting a "profitable internet", do you think they're going to sink developmental funds into something that they're not going to earn any money at, much less turn a profit? Sure, volunteers and the occasional corporate sponsorship does work occasionally, but look at how long it has taken to get here, and how slowly it works.
Ahh... I'd seriously like to see a competetor to Windows and Microsoft products, but unfortunately right now I just don't see it, and... maybe ? is that it, that dot on the horizon? dunno... only time will tell.
That's twisting thet truth significantly. The major reasons cited in the Wired article for the failure of Linux adoption were:
Problems with hardware compatibility with theexisting computers in use. This would not apply if compatible hardware had been initially targeted upon purchase of the system, as it had been done with Windows.
A lack of local Linux expertise among teachers. Just because the teachers are not experienced with using Linux as opposed to Windows does not draw the conclusion that Linux is more difficult to use - just that the knowledge base is not there yet.
Political movements from the Government. Of course proprietary vendors weren't going to just sit back and lose out on a contract this large.
Furthermore, it appears the ScholarNet initiative is not over yet. The current progress has not achieved the penetration desired, but future iterations hold more promise, as hardware compatibility improves and the "seeded" Linux knowledge from the successful installations trickles down.
Anyone interested in the attractions of Linux implementations in developing countries might want to have a look at a paper I wrote for a final year Engineering unit: postscript version [uwa.edu.au]. It has some mistakes in it I haven't corrected yet, but I'm open to revising it:)
That raises an interesting point, which I find is often overlooked - documentation.
I'm not talking about man pages/how-tos etc (which I personally find are usually of a much greater standard in terms of usability in an open source arena), but outlining:
* the exact steps taken to build the environment in the first place,
* maintenance steps (the where/why and how of upgrades, bugfixes, sec pathces etc)
* new user tasks - from both the admin and user perspectives
* general usability guidelines
If these tasks are done at the beginning, and tailored for whatever particular business environment you are in then I've found it is much easier to get it accepted by management and perhaps even spread to other departments.
One common stumbling block I've come across in trying to build these types of office environments is management fear that if I (or the admins with experience with open source setups) were to leave then they'd be stuck - after all it is much easier from a corporate view to find someone familiar with setting up/running a wintel / nt backend environ than linux.
Providing detailed outlines covering conception through to delivery/expansion really helps make a choice like this seem much less risky from the management view. After all if you are the only open source 'evangelist' in the company it is going to be a tough sell no matter how convincing your arguement.
When it is time to move on the steps you leave behind can help less experienced admins really get a grasp on not only the how, but the why of 'open offices'. Its how I really started, and I've now left many converts in my wake.
by Anonymous Coward
on Monday August 13 2001, @09:42AM (#2138064)
"Ummm...yeeaaahhh...if you could come in on saturday and ummmm...recompile the kernel with the latest USB patches so that your scanner will work...ummmmm...yeahh...that'd be great. Ummm...I'm also gonna need you to come into the office on Sunday too...yeeahh...you gotta get Linux printing Pantone colours to that Winprinter over there...yeaahh...that'd be great too. Oh...and don't forget those TPC coversheets! When you've figured out how to make Linux print as well as Windows, go print yourself out a dozen copies. Grrreeeaattt!"
I just had to shove Windows 2000 down the throats of my sales force. Transition from NT4.0. You should have heard the bitching/whining/crying/demands/etc... It doesnt matter what you put in front of them, they will piss-and-moan if it is different in any way. (Hell 2 years ago when I got here and made NT4.0 erase the login-name from the login screen for security they tied to hang me!) Changeing them to KDE or Gnome is no harder than any other change. In fact the Linux change will make the admin's life a dream. (No more idiots installing software that makes the system unstable!!! Damned AOL on NT!)
and apply policies. In win2k you can tune down a system as much as you want, from one single machine for your complete network, per user, per system, doesn't matter, you can.
So if you don't want AOL or other crap installed, apply a policy that the user can't install it. You can cry all you want that OS A is better because on OS B you can't prohibit a certain action, but all you have to do is read a couple of docs and get your butt in gear.
That would be nice, and in your world it probably is true. But in the corperate world it dont work that way...
we run 3 seperate vertical apps that require, (that's right REQUIRE) administrator access to the machine. Now this is the Traffic and Billing software (also requires Admin access to the SQL server!) which by the way is the largest T&B software package out there... it is the de-facto standard in media, you use it, discussion over.
Second we have an AVID. everything MUST be run as administrator. Dont log in and use as admin? too bad.. you don't get to work.
finally. I have a nice self updating Software package for the sales software suite. Now I am 1 IS/IT guy that supports 3 offices spread about 2 hours apart and over 100 machines. If I were to do it your way I need to spend every thursday and friday installing software via VNC or by drivig there.
It may work for you in your small computing environment, but in a large scale corperate environment NT cannot be configured to keep the cluebies from demolshing the hardware....
Oh, and management responds to my request to reprimand users that trash their desktop pc's?
"What did we hire you for? go fix it and shut up."
so... I am doing my job... better than any MCSE ever has here (awarded 3 times for productivity and excellence) and NT cannot do what I need. Linux can. Hell Linux can force the user to drop everything in their user directory (documents and files) instead of spreading documents all over the machine. NT? not possible.
As I read the articles I started out very skeptical, went on to be very impressed, and ended up pretty skeptical again.
It sounds like the system definately meets the City's needs, but it also seems like the Newsforge article is trying to overstate what those needs are. Lets look at the system specs.
400 Clients (800 Users)
Dual 933MHz system
3GB of Memory
18GB Hard Drive Space
Peak of about 230 concurrent users (from the first article)
Each concurrent user gets 11 to 12MB of system memory in which to work. They stated that they designed the system so they didn't have to hit the swap space. The 18 GB of hard drive space needs to be split between the OS application software and user storage space. Some users will need less space than others, so lets just ignore the space for the software and divide the total by the number of users. 18 GB/800 users = roughly 22.5 MB. How many people can here can honestly say that they don't have more space than that used for their email. The numbers tell me that the secretary they interviewed that was using Word Perfect, and email at the same time was one of their power users. I wouldn't be surprised if several hundred of their users don't even know how to access their email.
The big question is, so what? It's still a real system, that's meeting real users needs. The problem I have is that the article goes on to make tons of apples to oranges comparrisons.
It compares the cost of a thin client system in which users have very limited needs to a system with Windows desktops for everyone. How about Windows Terminal Server or other solutions that are more similar. I just don't buy the $300,000 a year hardware savings either. THese users have very limited needs, they don't need a new computer every year and a half, and $300,000 / 400 = $750 a year. Even if your buying new systems with monitors, that's way too much. $400,000 or $500,000 to run Exchange for their user base? Bullshit. I'm not saying that a Microsoft solution is cost effective, or even better excluding costs. It sounds like they found an exelent solution to their needs. The Sourceforge article however was too full of FUD to have much credibility.
While the article definately has a Linux vs Windows angle to it (and you have to admit, it's hard to talk about one being deployed without considering the effects the other one may have had, had it been deployed), I'm simply more interested in hearing secretaries extoll the virtues of KDE.
While this issue has been in full-fledged war mode for years, I think *nix proponants such as myself would have far more success focusing on the suitability and usability of KDE and Gnome than always boiling it down to a Xwindows vs Windows debate. Sure, Windows does the job, I run it at home; but if this article proves that End User X, dumb as a post, doesn't mind KDE (I'd use it daily if my audio-apps ran in *nix), force it on em! Well, at least in situations where it's my tax dollar...
Of course, the long term upside is that newbies 'n average users would finally have some variety in their computing experience before they blindly pledge allegience to the only OS they see commercials for; thus helping solidify *nix and KDE/Gnome as a viable platform for the Everyday Joe in the minds of the consumer.
The key here is Total Cost of Ownership: Windows boosters and shills like to point out that a free-purchase-price does not mean cheaper overall cost. But this article points out that the savings Largo is looking for are not the licensing costs, but the support costs. In Largo, the network is the computer. The idea that you could see "your" desktop from another workstation is just a dream in the Windows world.
This is not a victory for Linux. This is a victory for one old variant of Windows: yes, X-Windows.
Seriously though, I was administrating a company which all users had roaming profile - and it's a PAIN in the ass if a user move from one machine to another - you simple have to wait until of his desktop will be copied from the server - sometimes is few megabyes, but some heavy users have gigabytes of data to move...
On Linux/Unix it's different - nfs mount, finished...
Roblimo got a chance to see the system in action to find out how ordinary office workers are proving that the old "Linux is tough to use" shibboleth is nothing but FUD
We all know that the truth is, linux is hard to use for novices, and a good portion of linux users are not inclined to help newbies out. Read a few usenet posts to see this. "Linux is tough to use" is not FUD, it's the ugly truth. The people who realize this, and don't shrug it off as "FUD" are the ones who are positioned to correct this flaw.
Don't get me wrong, there's been tremendous progress made in linux usability, but the majority of it has been in the initial install area. There are still a lot of problems with UI consitancy, and any usability [slashdot.org] expert will tell you that this isn't a minor flaw. There's also the problem with installing software, because there's not the same one-click method for every program that Mac and Windows have.
Progress has been made, but we are certainly in no position to dismiss problems with linux' usability as FUD. When we do that, the progress will stop.
We all know that the truth is, linux is hard to use for novices, and a good portion of linux users are not inclined to help newbies out. Read a few usenet posts to see this. "Linux is tough to use" is not FUD, it's the ugly truth.
This simply isn't true, and my mother is a perfect case in point. Not terribly computer literate and with no desire to be, she simply wants her email, her web browser, and her word processor. Oh, she was delighted by the Bach, Beethove, and Mozart ogg's I ripped from her CD collection and made available on her hard drive... with xmms she can now listen to hours of music without changing CDs, and some of the other toys in her KDE menu she enjoys playing around with, but in truth her desires were relatively simple.
I bought her a $50 copy of applixware so she could read and write word documents, and guess what? She prefers her GNU/Linux box over her windows box at work by orders of magnitude. In fact, she has become much more zealous in advocating GNU/Linux and disparaging Windows than I ever was. Why? Because she, as a user, has found GNU/Linux to be much easier to use, much more stable, and much faster than her old windows install (to which she has never returned and which now provides additional storage for her burgeoning ogg-vorbis collection as she, herself, rips her own CDs using grip). Indeed, her discovery that it wasn't her, or her "stupidity" that was the root of nearly all of her computer mishaps, but the underlying instability of the operating system itself, has made her positively scathing when speaking of Microsoft. I guess she took Microsoft blaming the shortcomings of their products on her, and the denigration of making her feel stupid in the process, a little personally... not that any rational human being could blame her.
GNU/Linux is as easy, if not easier, to use than any version of Windows out there, and as others have pointed out, many GNU/Linux distributions are easier than Windows to set up and install as well.
Yes GNU/Linux is different, and yes, users must be willing to take an hour or two to learn those differences (ie "something new"), but new isn't the same as "difficult" or "tough to use." I spent an hour with my mother showing her the basics of navigating the KDE desktop and the differences between it and Windows, as well as the differences between Applix and MS Office. Again, this wasn't because GNU/Linux is "tough to use," this merely because it was a little different, and therefor new to her. Indeed, according to my mother, Linux is actually easier than Windows to use, so yes, saying GNU/Linux is "tough to use" is FUD in no uncertain terms. Saying "we all know it is tough to use" is adds a whole new level of dishonesty to the discussion, indeed it could be said that such as claim is FUD to the second power.
Now my mom's non-computer savvy friends are bugging me to come and set them up with GNU/Linux as well, so it looks like Microsoft's worst nightmare is in fact slowly coming true: regular, non-savvy Microsoft users are defecting to GNU/Linux in increasing numbers despite all the FUD Microsoft and its shills can possibly muster. Sometimes justice can be poetic.
Linux is hard to use as a hobby system with no *nix experience.
Ok, I'll accept that.
Linux isn't hard to use. It's a pain in the ass if you screw with it for fun. But it isn't hard for most users. They just need Email, a Word processor and solitair.
Just in case it works, try "CTRL-ALT-KEY_PAD_PLUS". That will switch between the currently configured resolutions, if there are more than one. The resolutions (and many other things related to hardware and graphics) are configured in/etc/X11/XF86Config. There is some documentation floating around on how to write your own video modes (getting a little bit more resolution, getting the image to line up nicely on your monitor, etc), but that's not terribly necessary with new monitors. Recent XFree86 also seems to know internally about a number of resolutions.
Note that changing resolution while running only changes how much appears on the screen-- the server will scroll the screen when you move the mouse to the edge so that you can get to the whole thing. The total size of the screen is set by the largest video mode listed for the depth (== bits/pixel), and can only be changed by editing the config file and restarting X.
If you want to change the color depth, either start X with the -depth <n> option or change the DefaultDepth parameter in the Screen section (where the resolutions are).
This information is from a number of different versions of XFree86, so not all of it will be right for your particular configuration. If all else fails, read the XFree86 man page (man XFree86) and possibly some of the other documentation it refers to.
As a more general point, there's a lot of good documentation for Linux stuff. The main problem is that it takes a certain amount of sophistication to know which thing you're dealing with, so you can read the applicable documentation. E.g., resolution is determined by XFree86, but background color/pattern is done by xsetroot, which may be called from a number of places, and what happens when you click on the background depends on your window manager.
When I saw this post I thought "something's up" because I have a hard time believing that people accustomed to running Windows could switch to KDE and not absolutely HATE it.
That's when I read they've been running Unixware for the last several years. Hell, they're accustomed to clunky interfaces! Moving from one clunky interface to another is no big deal. They simply don't know what they're missing.
I've seen secretaries and the like jumping through hoops trying to use poorly designed character/terminal interfaces in corporate environments who were PLEASED as PUNCH! Why? They didn't have a better system to compare it to.
So before you all start patting yourselves on the back, I think you need to give those secretaries some credit: Anybody can learn how to do most anything given time. There's no specific reason why a secretary can't learn to deal with clunky interfaces like KDE or Gnome just as easily as they can learn to deal with some hideous 1980's character based interface.
This says nothing about KDE's usability. It's still clunky. These folks simply don't know what they're missing.
Being both CTO of a small company (100+ employees on 3 continents) and husband of a non-technical wife:), I am desperately trying to do the same in our company, and at home. Seeing the roadblocks I am hitting may be interesting to some of you.
I see two types of objection to switching.
The "Necessary Condition" objections are mainly "Office", "Outlook", and "IE". Which is, alas, what everyone spends all day using. And until MS gets spilt up, this will not change. But also "that new accounting package", "my scanner", "our new CRM software", "our ERP project", and so on. And these are actually much harder to overcome. I think maybe we can identify a small group of users who do not use accounting, ERP, CRM etc. If we have to change all those, implementing Linux would actually cost us a lot of money.
Eh, before you say it:
StarOffice etc do not work well enough. Always some problems converting Word and Excel files.
VMware is slow, but it also defies the entire object (you still have to pay for an MS license)
Anyway, then there's the...
"Usability" objections. These are easy to fix in time - or they should be. But we are not there yet! I just spent a whole weekend setting up a new desktop machine for myself - Athlon 1 GHz, 512 MB RAM, RedHat 7.1. I had to do a kernel upgrade before it would see my Envidia graphics card. I still cannot print to my samba printer. And having installed machines ([pre-]CP/M, DOS, Win, Novell, Linux) for 20 years, I am not new to PCs or to Linux, but I still cannot figure out how to rewrite the Gnome/Ximian menus! And the config tool core-dumps: I have had 20-odd core dumps in the first day alone. And the lack of "OLE" drives me mad - an experienced PC user spends his life cutting and pasting, and the lack of this in common Linux desktop environments are a real obstacle.
So now I am looking for small groups of "expert users". Our (mainly hardware-) engineers come to mind first. But I am looking hard for real interoperability so we can roll out across the company. My estimate: 2 years out. I hope I am wrong.
but to get there, I had to take the difficult step of getting people to stop sending me Word and Excel files. Sending around insecure, application-specific files is a bad idea anyway unless it is absolutely necessary.
At least 95+% of the time someone sends me a Word file or Excel sheet, it is something that I only need to read, not edit, modify, and send back.
I'm an independent contractor, and whenever I receive a Word or Excel file that I cannot read in StarOffice, I politely reply back that I don't have Microsoft Office and I cannot read their files. I suggest that they resend the document either in PDF format, or RTF if they can't generate PDF. For Excel files, If they can't save as PDF, I suggest saving to an older version of Excel that StarOffice can read, albeit with some loss of formatting.
I have been able to change the file sending habits of a surprising number of people, especially when they realize that PDF files actually look more consistent on other people's systems, especially if they use non-standard fonts.
I do have one system with VMWare and one copy of Office for those very rare occasions that I receive a Word or Excel file that I actually have to modify and send back, or if the sender absolutely refuses to send another format, but this option doesn't get used very often, so I don't need it on all my systems.
I'm not going to comment on the relative costs of the hardware/software, because it's all true: Linux vs. Windows will win out, pure and simple. But the story (about Roblimo's take) is comparing apples and oranges.
Look at how they talk about backups: it sounds as though their concept of backups in the Windows world is to have users saving documents on their local hard disk, rather than to a server. The users have become accustomed to system crashes and network failures. I'll address at least part of the former complaint in a moment. The latter is the fault of either poor network administrators (as opposed to systems administrators, or a flaky server that hasn't been set up correctly. One of the biggest reasons people think NT is unstable is because the pretty GUI encourages rank amateurs to call themselves systems engineers. Blame this on the paper-MCSE syndrome, or on Microsoft's psychology, or whatever: but let's at least be honest and admit that, should the quality of admins increase, so would the quality of experience.
The other problem I have with this, and what really prompted my subject line, is that the comparison is between a Linux-based thin client network and a MS-based fat client network. Hello? If you took away all the Windows desktops and put in something like Citrix MetaFrame, then guess what? You'd realise several of the same benefits that the article touts or implies as being advantages unable to be put forth in a Windows-based system.
If you take the article as being a good example of how simple it is to migrate users over from Windows to Linux, then fine. But the system level comparisons are obfuscatory at best, and dishonest at worst. Yes, there's no way you could get the same level of performance out of the hardware they use if you went with a Windows implementation; but an article that compares a 10-person IT staff supporting Linux (or any OS) on 400 thin-client devices with supporting that many devices all running Windows on individual desktops is simply not a valid comparison. Is that really fair? By all means, let's point out the advantages for Linux in terms of ROI, open-source, and so on -- there are plenty of valid bases here -- but let's also be intellectually honest. Pretty please?
It is fair. why? Because the alternatives are all screwy too. The alternatives for 400 full windows machines is to have Terminal Services clients on all the desktops. So, compare it to that. Go out and buy 400 TSC capable systems and support them on however many machines are required.
You can compare these apples and oranges because, simply put, in the end the job they are to accomplish is the same. Supply the required office/administrative capabilities to 400 people in a work environment for the least amount of support headaches and cost.
There are a dozen of ways to try to accomplish it. This just happens to be one way that works well.
It is still automation. Of course humans are still writing the software, they are just doing it for free. Kind of like how robots dont go on strike or demand wages. Free software also takes advantage of the automation of replication. I can copy my piece of software and give it to a million people. I cant do that with windows software, or other proprietary systems.
This form of automation is the basis for the entire digital age. Microsoft takes advantage of it, but doesn't pass that advantage onto the customer. Only pirates get replication automation. Dont get left behind. Dont work for companies that license software as-is, with no responsability. Use freesoftware and pay for service contracts and insurance. Much wiser, I think. Plus the strange twist (paradoxical?) that freesoftware is usually more robust and stable than proprietary.
The future is away from proprietary commodities in software. There will always be proprietary specialty software, I grant that, but operating systems, web servers, web browsers, music players, on down the line, are all simple commodities and should be free, or cost a small sum (say your computer retail would charge to bundle them with your system). I'm glad I have staked my mental future on proprietary API's (win32 for most of y'all). I am free from manipulation, or so I think.
No. What kind of logic is that. Where do you think that M$ got its start? In the office of course! People will start using Linux at work and then want it for home too. As demand grows so does demand for games and MP3 players.
On a side note for the most part games and MP3 players are sub-culture of teenagers. Joe User doesn't play Quake III
I have to say this to people all the time. If you think Linux is hard to install, try installing Windows sometime. I have to reinstall Windows a lot at work and at home (mostly due to hard drive / processor upgrades), and it's a very laborious process. Even once you do get it running, you have to grab drivers (and reboot), reinstall all your apps (and rebooot, and reboot...). Linux, on the other hand, I just answer a few questions, take a 30 minute break and do something else, and come back to a ready-to-go box. Depending on the distro's age, I might also run LiveUpdate for Mandrake or apt-get for Debian. The best part is that none of these things requires rebooting, which means I don't have to sit in front of the machine while it works, wasting my time.
I recently let my girlfriend loose on my Linux
box at home. Setting up an account for her was
all I had to do - she's playing around with it
and has had no complaints so far. And she doesn't
even like computers (but goes out with a geek like me - go figure:). Yes, sometimes she does ask
where she kind find some application or other, but
comparing that to the problems the people at work
have getting their Windows boxes to behave, I
was surprised at how quickly she adapted to Linux.
Did you notice that the entire system runs the X protocol? I doubt they are even using XHOST auth on the NCD terminals (Of course I don't actually know anything). With the entirety of the X sessions going over the network in the clear I doubt that they are as worried about the security of their big beefy servers. Those can be kept on their own switches and under tight administrative control. I suppose that they could use SSH and maintain keyfiles for each user across all hosts serving X apps, but this wouldn't help with hosts that are serving Citrix apps and any user who can break root on one of the servers has access to all accounts on all machines anyway. It might give you warm fuzzies but I think that it would only add overhead.
Of course, I could be wrong. I'm just playing devil's advocate.
we removed all floppy drives over a year ago. It solved tons of problems IS wise and it eliminated the fired-employee taking files syndrome.
Work is for work, it stays there. If you want to work at home, then apply for a laptop.. if your super approves it then we give you laptop+docking station. (no CD drive or floppy but 2 extra batteries.)
Even though you admit its flamebait I still have to bite.
If you can change their views and switch them over to Linux, then and only then can you claim some sort of victory.
I thought the point of articles like this were to show those people that it can be done. In order reach that point other companies have to show they can do it first.
I remember reading that message when it was referenced at linuxtoday. The reason for it was to inform the KDE developers on how the deployment went and to let them know about various issues he encountered. Like scalability, control of desktop themes, etc.. You totally misconstrued why that post was made.
And I don't think that the rest of the articles make for a mutual admiration society. I can use information like this when I discuss things like licensing terms, alternative solutions to problems with my co-workers. These are pertinent stories that can be used to advocate linux.
I would mod you down. But not for violating slashdot groupthink. I didn't bother reading most of your comment. I'd mod you down for bitching about how you expected to be modded down. That sort of thing is really lame. If you have something to say, then say it. Don't be a whiny little bitch about it.
Er, who cares what *you* think about this? The company I work for used to be a windows shop, until they went through a BSA audit. Then I went to management and proposed the switch to linux (which they knew about because of the great job it did as a server). Guess what, after paying through the nose in wasted time and resources for the BSA shit, Linux started looking pretty good to them. Now we're a Linux shop, running RedHat/KDE/StarOffice and users are being re-trained for the new environment. Apparently, the BSA audit was very efficient in convincing management that it's worth re-training your employees in exchange for not hearing about Microsoft ever again. And the users? Who cares. They're being paid to do their jobs, training is prvided to them, so they'd better learn to do it the way the company wants them to, since this particular way has been deemed the most efficient. They can have all the eye-candy disguised as ease of use they want on their home computers.
So to asnwer your point, yes, I've been there, I converted Windows shops to Linux, I keep hearing from my friends that their companies use Linux more and more and yes, I do think that this kind of articles are useful and inspirational. to those of us who actually go out and convince management guys that Linux is a better choice. For those like you on the other hand, who can only bitch and moan about how Linux is not ever going to makie it in what your head tells you from inside your ass it's the "real world", they're probably useless. You'll never get it anyway.
That's great... (Score:2)
Like most businesses our size, we use a variety of custom, semi-custom, and prepackaged applications. While, yes, we could use free alternatives to our operating systems, office software, and email/scheduling software, there's no way in heck we're going to find a free replacement to our inventory, financial reporting, or human resource management software, for example, any time soon.
Even if someone came up with one, the cost of switching to the new packages would be enormous, given the complexity of such software and its impact on day-to-day business.
What's more, the specialized software we use requires that MS-Office be installed or it can't do simple things like generate reports. So, if we have to stay with these inventory, HR, and financial reporting packages, that means staying with MS-Office and MS-Windows.
Oh, how I'd love to move the whole corporation to free software. But there's just no way that's going to happen until all the mid-size apps are moved to Linux.
Until then, I'll continue to use FreeBSD, OpenBSD, and Linux on the server side as much as possible. But that's where they'll stay for the forseeable future.
Configuring it is the snag. (Score:5, Insightful)
Easily, my non-tech-savvy friends could get used to KDE and become comfortable in it, but I don't think they could set it up to be usable (nor do they care, and rightly so, they shouldn't have to). They could actually install easily--the Redhat install was exquisitley easy--but as far as installing programs, setting things up the way they like, etc., I don't think they could.
Then again, many people can't do the same with Windows (installing programs and configuring it to their liking).
Re:Configuring it is the snag. (Score:2)
And that's a good reason to prefer Mandrake over its Red Hat cousin (or Caldera or Corel over either) in that situation. Half the bitching here and elsewhere about Linux could be eliminated if people would pick appropriate distros for the job at hand, rather than becoming too attached to a particular one, often for ideological reasons, which are often the wrong reasons. Try several - cheapbytes is your friend. When you find one you like, buy a boxed copy.
The "right" distro depends on what I'm trying to do: I prefer Caldera or Mandrake on the desktop, Red Hat on servers (it reduces support problems), e-smith for garden-variety multi-purpose office servers, Turbo or Red Hat on mainframes, and Lineo or BSD for embedded devices. Choosing wisely (and widely) keeps you from the IT equivalent of hammering in screws with a pair of vise grips...
Oh, and be flexible - the costs for switching between distros are low and becoming less as we move to LSB-land, so don't be afraid to switch when it makes sense.
Not "varius combinations ..." (Score:4, Insightful)
I am all for being nice to "the other side" on these things but what I see is people strugling to use Gnome for ideological reasons and other people getting work done with KDE for financial reasons.
You know Finacial reasons like "Less money spent on Asperin", "fewer monitors shot at" and best of all you can fix the problems that do come up for less than it costs to fix the stuff you pay a grand more per seat for.
Re:Not "varius combinations ..." (Score:4, Insightful)
What are you basing this on? As a consultant in St. Louis, MO, I have had the pleasure of working with a few firms who provide Red Hat/GNOME solutions for corporate desktops and workstations. And GNOME use is rapidly expanding from Red Hat (and other distributions such as Debian and Turbo Linux) to other Unix variants. With Solaris switching to GNOME 2.0 in place of CDE as the default desktop environment, and HP-UX likewise embracing GNOME, you're going to see even more validation for GNOME on the desktop in the next 6 months to a year.
All this is to say, where are the facts that support your statement?
Parent
Re:Not "varius combinations ..." (Score:2)
I didn't make the initial statement, but I will respond. All my GNOME systems are up to date as of today.
The GNOME 1.4 release significantly diminshed speed and functionality to achieve only a baic file manager / web browser type app. Its very unusual that someone would create a new version that does LESS things than then old did, but I still can't create a new application launcher from the desktop or edit an existing one as with KDE
* The current GNOME control center (yes there's a new one on the way, no it isn't here yet) is confusign with its `test' `OK' implementation. This isn't consistent with many other GNOME apps
* 48 x 48 icons that are `supposed to look good' at 20 x 20 often don't. One size does not fit all.
* AFAICT there's no MacOS / Windows / KDE type style guide which can be used to define consistency between applications
* Defining a filetype -> program mapping is difficult in KDE but especially more so in GNOME.
* GNOME still has many programmerisms within it. Sawfish and GNOME might be seperate apps but from and end user viewpoint they should work seamlessly. Having a `meta' button under the GNOMECC which only defines options avaliable for the Sawfish branch is one such programmerism. And what does `meta' mean to a non tech?
ot: just on time (Score:3, Funny)
Cheers,
-N.
Hmmm (Score:3, Insightful)
The complaint that Linux is hard to use is not one necissaryly of GUI per se, it's set up. Basic quality assurance questions are hard to answer: What hardware are you going to use? What software are you going to use? Where is that software located? How do you install that software? Where is the software located on the Hardrive after you instal it? How do you get OpenGL to install? Why did it take me 15 minutes to find the PPP dialer? How do you set up a network where hard drives are shared? How do you put things into the menu?
If a sys-admin takes the time to form a planned approach for Linux installs, makes a custom CD for installing the "supported apps", then Linux can be usable. Problem is the Linux approach of shipping not only with the kitchen sink but 10 different kitchen sinks to choose from, all of which are disassembled and in their boxes is not one for newbies.
This reminds me of... (Score:3, Troll)
Followed up a year later by another story stating that never happened because Linux was too hard to use.
I say
Re:This reminds me of... (Score:5, Insightful)
2) This is a done deal, not a "someday" or "we plan to" thing. I wandered around Largo city hall and talked to actual, everyday users.
3) I'd like to go back and speak to Dave and Mike in a year, yes -- to see how their plans to use OpenOffice pan out. The biggest holdup (as I wrote in the NewsForge story [newsforge.com] linked to above) is the lack of a good OpenOffice filter for WordPerfect files.
- Robin
Parent
Re:So Robin, I gotta ask (Score:4, Funny)
Gee, how unusual. I suppose the average corporate drone is handed a Windows CD and an Optiplex?
Parent
Re:So Robin, I gotta ask (Score:3, Funny)
<sarcasm> Oh, right. Let me guess: where you work, all the secretaries installed NT on their own, and as soon as those W2K boxes arrive, they're going to upgrade. </sarcasm>
Re:So Robin, I gotta ask (Score:3, Insightful)
Where the network admins get involved is usually when it comes to interfacing with the main databases. With 400 seats, you _need_ a database system that makes MS Access look like a toy. (And I speak as one who has attempted to take Access right up to the limits MS admits to...) So unless you've somehow managed to put your whole database onto a IExplorer-compatible website, you are going to have to install a database client on each machine. The front-end one I'm familiar with (for a Progress database running in Unix) is a one-floppy disk 5-minute install, which is a heck of a lot easier than installing Win or Linux.
IMHO, the best way to provide corporate desktop computers is to buy hundreds of identical boxes, find the optimal installation for Linux and required applications, and then clone that configuration. But if they won't give you the budget to buy computers and put them in the closet until needed, then you wind up ordering one at a time for new hires or to replace broken down machines; almost every !@#$% machine is different (at least from lowest-bidder sources), so in Linux you'll spend hours on each one downloading drivers, setting up the configuration, etc. Windows pretty much requires the same time if you install it yourself and bother doing it right, but when you buy pre-configured boxen, it certainly looks like you've saved all that labor -- except for the six times that box is going to crash or catch a virus and have to be re-installed, but that cost comes later, and management isn't going to add it up and realize that buying the alleged industry standard was really boneheaded...
Re:So Robin, I gotta ask (Score:3, Interesting)
One thing that many of the people on these forums forget is how much of a special case we are. Your average office work is never permitted to install their own software. People have been arguing the case for thin clients for year because the average office user has no need to do their own installations and should be discouraged from doing so. I mean, we don't 'em wasting time with net Quake or anything. Not to mention ungodly mess a user administered machine can become. And, if you're running the type of place where people share work-stations like in a call center or clerk desk then you need to preserve uniformity from one session to the next so as not to confuse or completly derail the workers. Customizing the desktop is one thing, but installing any random app is a bad idea.
In that light why the hell should the place even expect the users to need to know stuff like that. ? In terms of apps being available the usual course is to go through IT management and request it. And, if you have a user who's knowledgeable enough to want a specific linux app by name, why not consider moving 'em into the tech department?
Re:So Robin, I gotta ask (Score:3, Interesting)
My wife has worked in more than a few government and commercial office environments that ran Windows, and they *always* had a separate IT or network support staff to take care of the computers. She wasn't supposed to add software or even mess with the things at all. In fact, in her last "real" job, doing customer support for a pager company, the biggest office computer network problem they had was employees bringing in software (especially games) from home and installing them on their own. Often the self-installed software screwed things up like mad.
Running a client/server network completely eliminates this problem. It also makes updating productivity apps a lot easier for the sysadmins.
- Robin
Re:So Robin, I gotta ask (Score:2)
The actual "special case" is probably the single machine senario (and certainly the "home" computer senario).
My wife has worked in more than a few government and commercial office environments that ran Windows, and they *always* had a separate IT or network support staff to take care of the computers. She wasn't supposed to add software or even mess with the things at all. In fact, in her last "real" job, doing customer support for a pager company, the biggest office computer network problem they had was employees bringing in software (especially games) from home and installing them on their own. Often the self-installed software screwed things up like mad.
Not only that it probably helps turn the IT staff into BOFH clones.
Re:So Robin, I gotta ask (Score:5, Insightful)
Man what type of crack are you on? I've been a sysadmin here for 2 years, and tried to push out linux to users who wanted it on several occasions. And guess what? Linux is great -- that is if, and only if, you have perfect hardware and perfect setup and a standalone system. The minute that something is out-of-spec, linux goes AWOL and the poor desktop user is SOL because they don't know the difference between KDE and dd. OTOH, I've plopped in the install CD for W2K, filled in a few simple details, walked away, came back 30 minutes later and had a system up-and-running without any problems. Sure back in the day of WinNT4 (I humbly agree that Win9X was a POS, but don't get me started on why those existed and why customers demanded Win98SE and WinME...) there were a bunch of problems, but I have had very few problems installing windows 2000 systems (and *zero* on reputable machines (i.e. Dell), well there was that 1 problem, traced to a defective HD)... And secondly, as soon as the user wants to do something new with their system, they're SOL again, not only because installing isn't as simple as "click here to download, run setup.exe and you're installed" (albiet lately in linux it has gotten a LOT better), but also because the apps simply aren't out there.
Oh BTW -- check out my W2K Server, up for 131 days without a reboot, and also survived CodeRed without a scratch because I set up security properly from the day that I installed the server. There's a cool realtime stats program up on the CodeRed attacks and other neat things: here [ohmygodmyarmfelloff.com].
To recap, simply, I like linux. I think that it has a lot of potential but it simply isn't anywhere close enough yet to be a mainstream system. Remember this [slashdot.org], and the fact that it's manhours spent with linux as well. I hate to be the harbinger of news here, but windows is much easier to use, period. There's no debate about that one, and with XP it just gets easier. Try putting your mother down in front of a linux machine, and then do the same with a windows machine. There is a reason why windows is used on 90% of desktops, and why Microsoft is the software giant that they are (reasons beyond the typical slashdottery about squishing competition and cheating and crap), more than just "being in the right place at the right time".. It's because, for better or worse, they have the best set of software products out there. Office and Windows are extremely successful because they're good, and people like them and use them a lot. That's a fact, hard to dispute.
I don't think that linux will make it as a mainstream OS anytime soon, or at least until most of the linux users (BTW, I think that part of the reason of linux non-acceptability is because of the typical i-love-linux-and-hate-windoze attitude and immaturity, not everyone, but just enough are immature and slander and swear and yell and scream and kick and fuss and act like children to give linux a bad name. Don't believe me? read our very own CmdrTaco [slashdot.org]. I think that he made some really good points there, the thermostat in hell must have broken that day...:> ) are part of the reason.
Oh well. I think that until linux users give up the I-want-everything-for-free-as-in-beer-as-my-god-g
Ahh... I'd seriously like to see a competetor to Windows and Microsoft products, but unfortunately right now I just don't see it, and
{/end rant}
Parent
Re:This reminds me of... (Score:5, Interesting)
That's twisting thet truth significantly. The major reasons cited in the Wired article for the failure of Linux adoption were:
Furthermore, it appears the ScholarNet initiative is not over yet. The current progress has not achieved the penetration desired, but future iterations hold more promise, as hardware compatibility improves and the "seeded" Linux knowledge from the successful installations trickles down.
Anyone interested in the attractions of Linux implementations in developing countries might want to have a look at a paper I wrote for a final year Engineering unit: postscript version [uwa.edu.au]. It has some mistakes in it I haven't corrected yet, but I'm open to revising it :)
Parent
Re:This reminds me of... (Score:5, Interesting)
I'm not talking about man pages/how-tos etc (which I personally find are usually of a much greater standard in terms of usability in an open source arena), but outlining:
* the exact steps taken to build the environment in the first place,
* maintenance steps (the where/why and how of upgrades, bugfixes, sec pathces etc)
* new user tasks - from both the admin and user perspectives
* general usability guidelines
If these tasks are done at the beginning, and tailored for whatever particular business environment you are in then I've found it is much easier to get it accepted by management and perhaps even spread to other departments.
One common stumbling block I've come across in trying to build these types of office environments is management fear that if I (or the admins with experience with open source setups) were to leave then they'd be stuck - after all it is much easier from a corporate view to find someone familiar with setting up/running a wintel / nt backend environ than linux.
Providing detailed outlines covering conception through to delivery/expansion really helps make a choice like this seem much less risky from the management view. After all if you are the only open source 'evangelist' in the company it is going to be a tough sell no matter how convincing your arguement.
When it is time to move on the steps you leave behind can help less experienced admins really get a grasp on not only the how, but the why of 'open offices'. Its how I really started, and I've now left many converts in my wake.
Parent
OfficeSpace Linux (Score:5, Funny)
talk about the strange interface (Score:4, Offtopic)
Then do your job for once (Score:2)
So if you don't want AOL or other crap installed, apply a policy that the user can't install it. You can cry all you want that OS A is better because on OS B you can't prohibit a certain action, but all you have to do is read a couple of docs and get your butt in gear.
Re:Then do your job for once (Score:3, Informative)
we run 3 seperate vertical apps that require, (that's right REQUIRE) administrator access to the machine. Now this is the Traffic and Billing software (also requires Admin access to the SQL server!) which by the way is the largest T&B software package out there... it is the de-facto standard in media, you use it, discussion over.
Second we have an AVID. everything MUST be run as administrator. Dont log in and use as admin? too bad.. you don't get to work.
finally. I have a nice self updating Software package for the sales software suite. Now I am 1 IS/IT guy that supports 3 offices spread about 2 hours apart and over 100 machines. If I were to do it your way I need to spend every thursday and friday installing software via VNC or by drivig there.
It may work for you in your small computing environment, but in a large scale corperate environment NT cannot be configured to keep the cluebies from demolshing the hardware....
Oh, and management responds to my request to reprimand users that trash their desktop pc's?
"What did we hire you for? go fix it and shut up."
so... I am doing my job... better than any MCSE ever has here (awarded 3 times for productivity and excellence) and NT cannot do what I need. Linux can. Hell Linux can force the user to drop everything in their user directory (documents and files) instead of spreading documents all over the machine. NT? not possible.
skeptical (Score:3, Insightful)
It sounds like the system definately meets the City's needs, but it also seems like the Newsforge article is trying to overstate what those needs are. Lets look at the system specs.
400 Clients (800 Users)
Dual 933MHz system
3GB of Memory
18GB Hard Drive Space
Peak of about 230 concurrent users (from the first article)
Each concurrent user gets 11 to 12MB of system memory in which to work. They stated that they designed the system so they didn't have to hit the swap space. The 18 GB of hard drive space needs to be split between the OS application software and user storage space. Some users will need less space than others, so lets just ignore the space for the software and divide the total by the number of users. 18 GB/800 users = roughly 22.5 MB. How many people can here can honestly say that they don't have more space than that used for their email. The numbers tell me that the secretary they interviewed that was using Word Perfect, and email at the same time was one of their power users. I wouldn't be surprised if several hundred of their users don't even know how to access their email.
The big question is, so what? It's still a real system, that's meeting real users needs. The problem I have is that the article goes on to make tons of apples to oranges comparrisons.
It compares the cost of a thin client system in which users have very limited needs to a system with Windows desktops for everyone. How about Windows Terminal Server or other solutions that are more similar. I just don't buy the $300,000 a year hardware savings either. THese users have very limited needs, they don't need a new computer every year and a half, and $300,000 / 400 = $750 a year. Even if your buying new systems with monitors, that's way too much. $400,000 or $500,000 to run Exchange for their user base? Bullshit. I'm not saying that a Microsoft solution is cost effective, or even better excluding costs. It sounds like they found an exelent solution to their needs. The Sourceforge article however was too full of FUD to have much credibility.
Joe Shmoe doesn't mind KDE?! Run with it! (Score:5, Insightful)
While this issue has been in full-fledged war mode for years, I think *nix proponants such as myself would have far more success focusing on the suitability and usability of KDE and Gnome than always boiling it down to a Xwindows vs Windows debate. Sure, Windows does the job, I run it at home; but if this article proves that End User X, dumb as a post, doesn't mind KDE (I'd use it daily if my audio-apps ran in *nix), force it on em! Well, at least in situations where it's my tax dollar
Of course, the long term upside is that newbies 'n average users would finally have some variety in their computing experience before they blindly pledge allegience to the only OS they see commercials for; thus helping solidify *nix and KDE/Gnome as a viable platform for the Everyday Joe in the minds of the consumer.
total cost of X-Windows (Score:4, Interesting)
This is not a victory for Linux. This is a victory for one old variant of Windows: yes, X-Windows.
Re:total cost of X-Windows (Score:5, Insightful)
Seriously though, I was administrating a company which all users had roaming profile - and it's a PAIN in the ass if a user move from one machine to another - you simple have to wait until of his desktop will be copied from the server - sometimes is few megabyes, but some heavy users have gigabytes of data to move...
On Linux/Unix it's different - nfs mount, finished...
Parent
That's not FUD Ti-MAY (Score:5, Interesting)
We all know that the truth is, linux is hard to use for novices, and a good portion of linux users are not inclined to help newbies out. Read a few usenet posts to see this. "Linux is tough to use" is not FUD, it's the ugly truth. The people who realize this, and don't shrug it off as "FUD" are the ones who are positioned to correct this flaw.
Don't get me wrong, there's been tremendous progress made in linux usability, but the majority of it has been in the initial install area. There are still a lot of problems with UI consitancy, and any usability [slashdot.org] expert will tell you that this isn't a minor flaw. There's also the problem with installing software, because there's not the same one-click method for every program that Mac and Windows have.
Progress has been made, but we are certainly in no position to dismiss problems with linux' usability as FUD. When we do that, the progress will stop.
Adding FUD to FUD doesn't make it any less FUDdy (Score:3, Interesting)
This simply isn't true, and my mother is a perfect case in point. Not terribly computer literate and with no desire to be, she simply wants her email, her web browser, and her word processor. Oh, she was delighted by the Bach, Beethove, and Mozart ogg's I ripped from her CD collection and made available on her hard drive
I bought her a $50 copy of applixware so she could read and write word documents, and guess what? She prefers her GNU/Linux box over her windows box at work by orders of magnitude. In fact, she has become much more zealous in advocating GNU/Linux and disparaging Windows than I ever was. Why? Because she, as a user, has found GNU/Linux to be much easier to use, much more stable, and much faster than her old windows install (to which she has never returned and which now provides additional storage for her burgeoning ogg-vorbis collection as she, herself, rips her own CDs using grip). Indeed, her discovery that it wasn't her, or her "stupidity" that was the root of nearly all of her computer mishaps, but the underlying instability of the operating system itself, has made her positively scathing when speaking of Microsoft. I guess she took Microsoft blaming the shortcomings of their products on her, and the denigration of making her feel stupid in the process, a little personally
GNU/Linux is as easy, if not easier, to use than any version of Windows out there, and as others have pointed out, many GNU/Linux distributions are easier than Windows to set up and install as well.
Yes GNU/Linux is different, and yes, users must be willing to take an hour or two to learn those differences (ie "something new"), but new isn't the same as "difficult" or "tough to use." I spent an hour with my mother showing her the basics of navigating the KDE desktop and the differences between it and Windows, as well as the differences between Applix and MS Office. Again, this wasn't because GNU/Linux is "tough to use," this merely because it was a little different, and therefor new to her. Indeed, according to my mother, Linux is actually easier than Windows to use, so yes, saying GNU/Linux is "tough to use" is FUD in no uncertain terms. Saying "we all know it is tough to use" is adds a whole new level of dishonesty to the discussion, indeed it could be said that such as claim is FUD to the second power.
Now my mom's non-computer savvy friends are bugging me to come and set them up with GNU/Linux as well, so it looks like Microsoft's worst nightmare is in fact slowly coming true: regular, non-savvy Microsoft users are defecting to GNU/Linux in increasing numbers despite all the FUD Microsoft and its shills can possibly muster. Sometimes justice can be poetic.
Re:That's not FUD Ti-MAY (Score:4, Insightful)
Ok, I'll accept that.
Linux isn't hard to use. It's a pain in the ass if you screw with it for fun. But it isn't hard for most users. They just need Email, a Word processor and solitair.
And those run just fine.
Parent
Re:That's not FUD Ti-MAY (Score:2)
Note that changing resolution while running only changes how much appears on the screen-- the server will scroll the screen when you move the mouse to the edge so that you can get to the whole thing. The total size of the screen is set by the largest video mode listed for the depth (== bits/pixel), and can only be changed by editing the config file and restarting X.
If you want to change the color depth, either start X with the -depth <n> option or change the DefaultDepth parameter in the Screen section (where the resolutions are).
This information is from a number of different versions of XFree86, so not all of it will be right for your particular configuration. If all else fails, read the XFree86 man page (man XFree86) and possibly some of the other documentation it refers to.
As a more general point, there's a lot of good documentation for Linux stuff. The main problem is that it takes a certain amount of sophistication to know which thing you're dealing with, so you can read the applicable documentation. E.g., resolution is determined by XFree86, but background color/pattern is done by xsetroot, which may be called from a number of places, and what happens when you click on the background depends on your window manager.
They're quite accustomed to clunky interfaces. (Score:4, Interesting)
That's when I read they've been running Unixware for the last several years. Hell, they're accustomed to clunky interfaces! Moving from one clunky interface to another is no big deal. They simply don't know what they're missing.
I've seen secretaries and the like jumping through hoops trying to use poorly designed character/terminal interfaces in corporate environments who were PLEASED as PUNCH! Why? They didn't have a better system to compare it to.
So before you all start patting yourselves on the back, I think you need to give those secretaries some credit: Anybody can learn how to do most anything given time. There's no specific reason why a secretary can't learn to deal with clunky interfaces like KDE or Gnome just as easily as they can learn to deal with some hideous 1980's character based interface.
This says nothing about KDE's usability. It's still clunky. These folks simply don't know what they're missing.
I am trying to do the same, but... (Score:5, Informative)
I see two types of objection to switching.
The "Necessary Condition" objections are mainly "Office", "Outlook", and "IE". Which is, alas, what everyone spends all day using. And until MS gets spilt up, this will not change. But also "that new accounting package", "my scanner", "our new CRM software", "our ERP project", and so on. And these are actually much harder to overcome. I think maybe we can identify a small group of users who do not use accounting, ERP, CRM etc. If we have to change all those, implementing Linux would actually cost us a lot of money.
Eh, before you say it:
StarOffice etc do not work well enough. Always some problems converting Word and Excel files.
VMware is slow, but it also defies the entire object (you still have to pay for an MS license)
Anyway, then there's the...
"Usability" objections. These are easy to fix in time - or they should be. But we are not there yet! I just spent a whole weekend setting up a new desktop machine for myself - Athlon 1 GHz, 512 MB RAM, RedHat 7.1. I had to do a kernel upgrade before it would see my Envidia graphics card. I still cannot print to my samba printer. And having installed machines ([pre-]CP/M, DOS, Win, Novell, Linux) for 20 years, I am not new to PCs or to Linux, but I still cannot figure out how to rewrite the Gnome/Ximian menus! And the config tool core-dumps: I have had 20-odd core dumps in the first day alone. And the lack of "OLE" drives me mad - an experienced PC user spends his life cutting and pasting, and the lack of this in common Linux desktop environments are a real obstacle.
So now I am looking for small groups of "expert users". Our (mainly hardware-) engineers come to mind first. But I am looking hard for real interoperability so we can roll out across the company. My estimate: 2 years out. I hope I am wrong.
StarOffice Worked for Me, but... (Score:3, Insightful)
At least 95+% of the time someone sends me a Word file or Excel sheet, it is something that I only need to read, not edit, modify, and send back.
I'm an independent contractor, and whenever I receive a Word or Excel file that I cannot read in StarOffice, I politely reply back that I don't have Microsoft Office and I cannot read their files. I suggest that they resend the document either in PDF format, or RTF if they can't generate PDF. For Excel files, If they can't save as PDF, I suggest saving to an older version of Excel that StarOffice can read, albeit with some loss of formatting.
I have been able to change the file sending habits of a surprising number of people, especially when they realize that PDF files actually look more consistent on other people's systems, especially if they use non-standard fonts.
I do have one system with VMWare and one copy of Office for those very rare occasions that I receive a Word or Excel file that I actually have to modify and send back, or if the sender absolutely refuses to send another format, but this option doesn't get used very often, so I don't need it on all my systems.
Invalid comparisons (Score:5, Insightful)
Look at how they talk about backups: it sounds as though their concept of backups in the Windows world is to have users saving documents on their local hard disk, rather than to a server. The users have become accustomed to system crashes and network failures. I'll address at least part of the former complaint in a moment. The latter is the fault of either poor network administrators (as opposed to systems administrators, or a flaky server that hasn't been set up correctly. One of the biggest reasons people think NT is unstable is because the pretty GUI encourages rank amateurs to call themselves systems engineers. Blame this on the paper-MCSE syndrome, or on Microsoft's psychology, or whatever: but let's at least be honest and admit that, should the quality of admins increase, so would the quality of experience.
The other problem I have with this, and what really prompted my subject line, is that the comparison is between a Linux-based thin client network and a MS-based fat client network. Hello? If you took away all the Windows desktops and put in something like Citrix MetaFrame, then guess what? You'd realise several of the same benefits that the article touts or implies as being advantages unable to be put forth in a Windows-based system.
If you take the article as being a good example of how simple it is to migrate users over from Windows to Linux, then fine. But the system level comparisons are obfuscatory at best, and dishonest at worst. Yes, there's no way you could get the same level of performance out of the hardware they use if you went with a Windows implementation; but an article that compares a 10-person IT staff supporting Linux (or any OS) on 400 thin-client devices with supporting that many devices all running Windows on individual desktops is simply not a valid comparison. Is that really fair? By all means, let's point out the advantages for Linux in terms of ROI, open-source, and so on -- there are plenty of valid bases here -- but let's also be intellectually honest. Pretty please?
Re:Invalid comparisons (Score:4, Insightful)
You can compare these apples and oranges because, simply put, in the end the job they are to accomplish is the same. Supply the required office/administrative capabilities to 400 people in a work environment for the least amount of support headaches and cost.
There are a dozen of ways to try to accomplish it. This just happens to be one way that works well.
Parent
Re:Not a fair test of ease of use (Score:2)
What is a valid usability test?
Re:Say OpenSource if you're for more IT unemployme (Score:2)
This form of automation is the basis for the entire digital age. Microsoft takes advantage of it, but doesn't pass that advantage onto the customer. Only pirates get replication automation. Dont get left behind. Dont work for companies that license software as-is, with no responsability. Use freesoftware and pay for service contracts and insurance. Much wiser, I think. Plus the strange twist (paradoxical?) that freesoftware is usually more robust and stable than proprietary.
The future is away from proprietary commodities in software. There will always be proprietary specialty software, I grant that, but operating systems, web servers, web browsers, music players, on down the line, are all simple commodities and should be free, or cost a small sum (say your computer retail would charge to bundle them with your system). I'm glad I have staked my mental future on proprietary API's (win32 for most of y'all). I am free from manipulation, or so I think.
Re:Won't office working kill Linux? (Score:2)
No. What kind of logic is that. Where do you think that M$ got its start? In the office of course! People will start using Linux at work and then want it for home too. As demand grows so does demand for games and MP3 players.
On a side note for the most part games and MP3 players are sub-culture of teenagers. Joe User doesn't play Quake III
hahahahahaha (Score:5, Informative)
From first POST to "installed":
Linux: 35 min
Win2k: 45 min
Time to get drivers up to speed.
Linux: 0 min (had all my stuff)
Win2k: 25 min (nvidia, creative)
Time to get Quake3 running
Linux: 5 hours (still doesn't work right)
Win2k: 10 min
Time to get my RAID ATA-100 card working
Linux: 0 (it doesn't work)
Win2k: did it at boot, only took 2 min
-Jon
Parent
Re:Linux surpassed W$ in ease of use long ago (Score:3, Insightful)
I have to say this to people all the time. If you think Linux is hard to install, try installing Windows sometime. I have to reinstall Windows a lot at work and at home (mostly due to hard drive / processor upgrades), and it's a very laborious process. Even once you do get it running, you have to grab drivers (and reboot), reinstall all your apps (and rebooot, and reboot...). Linux, on the other hand, I just answer a few questions, take a 30 minute break and do something else, and come back to a ready-to-go box. Depending on the distro's age, I might also run LiveUpdate for Mandrake or apt-get for Debian. The best part is that none of these things requires rebooting, which means I don't have to sit in front of the machine while it works, wasting my time.
Re:Linux surpassed W$ in ease of use long ago (Score:2)
Re:Linux surpassed W$ in ease of use long ago (Score:2)
Also adding additional hardware profiles onto the same CD is easier...
Faster too in case one computer has different hardware (I.E. a scsi card for a scanner or capture device)
Still, having to download a driver that should have been there? (the network,video and sound were out before 2000 was made and RH6.2 even found them)
Re:FUD? (Score:2)
Re:For crying out loud... (Score:2)
Did you notice that the entire system runs the X protocol? I doubt they are even using XHOST auth on the NCD terminals (Of course I don't actually know anything). With the entirety of the X sessions going over the network in the clear I doubt that they are as worried about the security of their big beefy servers. Those can be kept on their own switches and under tight administrative control. I suppose that they could use SSH and maintain keyfiles for each user across all hosts serving X apps, but this wouldn't help with hosts that are serving Citrix apps and any user who can break root on one of the servers has access to all accounts on all machines anyway. It might give you warm fuzzies but I think that it would only add overhead.
Of course, I could be wrong. I'm just playing devil's advocate.
Re:I submit to you (Score:3, Insightful)
we removed all floppy drives over a year ago. It solved tons of problems IS wise and it eliminated the fired-employee taking files syndrome.
Work is for work, it stays there. If you want to work at home, then apply for a laptop.. if your super approves it then we give you laptop+docking station. (no CD drive or floppy but 2 extra batteries.)
Re:Masturbation (Score:5, Insightful)
If you can change their views and switch them over to Linux, then and only then can you claim some sort of victory.
I thought the point of articles like this were to show those people that it can be done. In order reach that point other companies have to show they can do it first.
Parent
Re:Masturbation (Score:4, Insightful)
And I don't think that the rest of the articles make for a mutual admiration society. I can use information like this when I discuss things like licensing terms, alternative solutions to problems with my co-workers. These are pertinent stories that can be used to advocate linux.
Parent
If I had mod points... (Score:3, Insightful)
who cares (Score:5, Interesting)
So to asnwer your point, yes, I've been there, I converted Windows shops to Linux, I keep hearing from my friends that their companies use Linux more and more and yes, I do think that this kind of articles are useful and inspirational. to those of us who actually go out and convince management guys that Linux is a better choice. For those like you on the other hand, who can only bitch and moan about how Linux is not ever going to makie it in what your head tells you from inside your ass it's the "real world", they're probably useless. You'll never get it anyway.
Parent