Slashdot Mirror
InfoWorld on Switching to Linux
brentlaminack writes "The latest Infoworld is running a lengthy piece about
The Real Cost of Switching to Linux, where it makes sense and where it doesn't. As one of their columnists points out, the debate has switched from "if" to "where". One of the big wins for Linux was in the area of remote administration. Specifically noted was ssh. Also of note is the shift in calculating cost from TCO (Total Cost of Ownership) as has been calculated in the past, to ROI (Return on Investment) that focuses more on what you can do with the technology to get work done."
Thing that I've noticed is that if a large organization gets into Linux, even if they buy it, it's theirs for the duration and all of the upgrades that they can work into it, instead of requiring either yearly site license fees or massive payouts every so many years for new versions of software to do essentially the same thing. Even paying a consulting company or services company to deploy Debian would make sense in a way, as long as the apt server were the organization's, versus a public server, so that as long as someone is maintaining the package database on the local apt server, they can keep updating the workstations.
Large organizations usually have some form of IS department, so instead of paying them to run around and fix Windows Millennium or XP problems, pay them to keep the network deployed OS current, and fix the bulk of the problems from their desks.
Do not look into laser with remaining eye.
will put the flags in temp.text on your local machine, but -will put it on the remote machine instead.> One of the big wins for Linux was in the area of remote administration. Specifically noted was ssh.
I admin ~25 machines remotely, most of them in a room that I don't even have access to without special arrangements. With SSH I can do that without ever having to make those arrangements, except in the case of a major upgrade or a hardware failure.
You can write scripts that will take a shell command as an argument and then step through all your machines executing it on each in turn, greatly simplifying remote management.
You can also use pipes and redirects to channel information between processes on the remote machine and your local machine, e.g. -
Or, if you want to do all the work on the remote machine and only redirect the output to your local machine, use -and the grep will actually execute on remotehost.
The example is trivial, but you can do some powerful sysadmin stuff that way. However, there are a few gottchas: a few services crap out if you try to restart them with -so you do have to be careful about some things. (Sure wish someone would figure out what causes that and fix it!)
Sheesh, evil *and* a jerk. -- Jade
"The jury is in. After years of experimentation with Linux in the enterprise, customers, analysts, and vendors are starting to sing a consistent tune about where Linux makes financial sense and where it doesn't."
They still don' t get it. Even though the article is moderately positive, any article about Linux that starts with "the Jury is in" was written by someone who does not fully understand the dynamics of Open Source. How can "the jury" be "in" on an environment that changes so rapidly as Linux does? How can you say for certain where Linux has a role and where it doesn't? A move in the right direction, but the hacks still need some educating.....
People who think they know everything are a great annoyance to those of us who do.
I have to admit, this is definitely one of the better write-ups that I have seen on the subject. Most, as the article states, base all decisions on the TCO (amount spent) rather than the ROI, which allows organizations to determine how much they would save in the long run if they were to switch.
Main thrust seems to be that the savings increase with the amount of technical resources converted to Linux systems. Perhaps this could be a deciding factor for many companies and organizations considering taking the plunge.
Favorite Quote:
"Discount retailing's a tight business, and we're wicked cheap," explains Burlington Coat Factory CIO Mike Prince..."Instead of having a superhorse you have a team of horses -- you don't have to have this genetic [RISC] wonder."
-CSA
My colleagues and I had several discussions about switching to linux costs during the past years. I am not going to report everything we talked about (especially when we got "hot" and yelled using not very fair terms), but just the essentials. The cost, both for just server or even for workstations, depends a lot upon whether there's at least a professional employed there actively using linux (a geek almost necessarily) and the kind of applications needed to be "ported". In my case, a switch not only would be very expensive (30 workstations using Windows and -gosh- MS Access), but almost impossibile without thinking about an almost complete rewrite of the applications. In many other cases though the switch is not only possibile (email, wordprocessing, spreadsheet) but even very very inexpensive.
"There were a lot of costs I didn't expect-- hidden migration costs," says Cedars-Sinai's Duncan. During the migration from NT to Linux, his staff insisted that because they had been running RAID disk mirroring and striping on NT they should buy SCSI RAID controllers for the Linux servers. "It was like $1,000 per box extra that I hadn't planned on."
That wasn't a hidden cost. Linux could have easily handled RAID disk mirroring and striping without the special controllers.
This was an example of the IT staff knowing they have a much larger than normal project budget and milking it for all it was worth.
Learning HOW to think is more important than learning WHAT to think.
If you're developing on it. If you're using it for regular users who need email and web and word processing, it doesn't matter what the licensing is. Your memo written in ABIWord doesn't have to contain the GPL.
And if you're developing, there are commercial libraries available to you. There are BSD-licensed libraries too. You don't have to use Stallman's libraries, you can get them elsewhere. Hell, IBM even builds compilers, as does Intel. The entire point of GPLed stuff is for it to remain for everyone. If you don't like that, build it yourself, buy it, or find another non-GPL one.
It's not impossible to do this. It just takes brains and research. I'd rather sink my money into that than into a mindless purchase of a product that goes "BOOM!" far too frequently and forces one into paid upgrades.
Do not look into laser with remaining eye.
What you failed to understand about the parents post is that he is talking about site license fees, renewal fees, etc ... You don't have to spend $500/seat with Linux every five years as you do with Microsoft. "It's theirs for the duration" means, simply, that they don't need to pay out the nose. Itdoes not refer to the GPL.
A note about the GPL, which you also missed, is that if you make a change to somones GPL'd software, you must also make your code GPL, or a compatible license. However, providing source code is a provision only when you are distributing. If you don't distribute that work outside of the company, your GPL'd work doesn't see the light of day. Once you distribute it, however, you need to provide a way for the person who obtained the binary to obtain the source code.
Also, you need not provide everyone with the source code: you only need to provide those who have obtained, through you, the source code in question. Of course, they are also given the right to use, modify and distribute that source code. However, that doesn't mean you need to put it on a public FTP server in a tarball for every person in the world to download. Which, actually, destorys your argument in it's entierty really. You can profit, you just can't have a stranglehold on the world with your technology.
--LordKaT
I haven't yet seen a TCO study that includes the risk of a BSA audit in a Windows shop. The TCO for running Windows should include the cost of insurance against the disruption of a BSA audit and the penalties paid for apparently unlicensed software.
Mike O'Donnell http://people.cs.uchicago.edu/~odonnell/
After the MSBlaster worm and SoBig virus activity of the last few weeks, you'd think that there'd be a little more than a passing reference on page 3 of the article saying that Linux is "virtually virus-free".
I'll bet that none of these expensive studies ever include the cost of cleaning up after the virus/worm of the week that comes with running Microsoft NT/2000/XP. Having everyone in your company having 2 or 3 days a year when their desktop/laptop/server/whatever is unavailable because of cleanup activity should have a definite negative impact on TCO or ROI.
Yet one more reason to use Linux, *BSD or OS X.
- Remote Desktop/Terminal Services (you don't even need a RD client, just a browser, which nearly every modern machine has, unlike ssh
.NET development tools, to quickly code up anything you need that can't be covered by the above.
.Net is hardly quick to learn or code in. It's a bloated OOP framework in the Java tradition. Fine for applications, sux for writing sysadmin tools.
Bzzt. Every modern Linux machine comes with ssh.
- VBScript (horrid, but gets the job done most of the time)
VBScript compared to perl/bash etc.? lol.
- WMI (Windows Management Instrumentation, do damn near anything remotely, but be sure to properly secure your network)
Compared to UNIX instrumentation tools like SNMP? lol x2.
- The MMC tools (ADUC, etc etc), which fully operate remotely, as well.
MMC tools vs. UNIX remote admin ? hahaha
- The
$1000 IDE license for the above vs what you get for free in Linux? You don't have a clue as to what you are talking about. And
I really love watching Windows admins paging through dialog boxes looking for incorrect settings. It's hilarious.
Try remote admining your Windows box from a PDA on a train on your way to work, fella.
Numbers. Simply numbers. It's the same reason that nobody reports on any of the 100s of fringe OSs with user bases measured in the thousands. Linux has more users and therefore gets the most attention. FreeBSD had its chance to have the biggest user base but it lost to Linux. This was despite a significant headstart in the form of 386BSD. There are at least six reasons I can fathom as to why this happened.
First, the AT&T lawsuit against Berkeley (1992) scared a number of developers away from 386BSD at a very critical time in its evolution. Why invest time into developing 386BSD if AT&T was just going to steal your hard work? And "steal" is the right word here; it really would have been theft if AT&T had won because the 386BSD developers would have lost ownership of code they'd written themselves. Developers were scared away from 386BSD and towards Linux, which was seen as being "litigation-free". The parallels with the claims made by SCO today are frightening.
Second, the Jolitzes. They were custodians of 386BSD and Bill was notorious for being slow to accept patches (1 year of unapplied patches). The formation of FreeBSD was essentially the "Gang of Three" getting frustrated with the slow pace of 386BSD development. They combined 386BSD plus the existing "patch kit" and sold the result as a CD-ROM. This was unfortunately too little, too late. Linux had a 2 year headstart on FreeBSD by this stage. Also the splintering pissed off a number of developers who stopped contributing to both 386BSD and FreeBSD. Instead they started contributing to Linux.
Third, the license. FreeBSD advocates say that the BSD license is "more free" than the GPL but to some people (myself included) the BSD license is offensive. Nothing stops a commercial company leeching off your hard work if you use the BSD license. BSD advocates say this isn't a problem: "you wanted it to be free and now it is". The problem is I don't really want companies getting rich off my code. I want them to contribute back with more code. The GPL enforces this. The BSD license does not. In 1991, when Linux was still very much in its infancy, it managed to get more attention from more programmers than 386BSD ever managed. This was despite Linux being technically inferior to 386BSD. The license simply appeals to certain people. If Linus had used a BSD license then I don't think Linux could have ever wrested the #1 spot away from 386BSD.
Fourth, the Internet. Linux development began at a time when Internet access was appearing in homes. The early adopters of home-Internet access were (of course) technology enthusiasts. The percentage of potential Linux developers in this group was relatively high. This meant from the start Linux had a huge base of developers to draw upon. And isn't it more fun to contribute to a brand new project than an existing project? Linux attracted the developers simply because it wasn't finished.
Fifth, the installers. Back in 1992 (1991?) I was using Interactive UNIX at home. The software was showing its age so I was looking to get into one of the "Free UNIX" that was floating around the Internet. I'd already used (and dismissed) Minix because it was incredibly limited. I had a choice between 386BSD and Linux. The 386BSD installer required a 40MB download, a SCSI hard drive, and required me to destroy my existing Interactive installation. The Linux distribution came on 2x 5.25" floppies, supported IDE hard disks, and could coexist with existing operating systems. It was a no-brainer. Linux won because it cared about the newbie, even back then when I admittedly needed all my UNIX experience to get the damn thing installed. The FreeBSD distro didn't come until late-1993 but by then it was too late; I'd already deleted my Intera
- Risk of the 'Software Police'
- Timewasting 'licence audits'
- Microsoft business practices
- Paying again every 5 years
- Viruses, worms
- Staff timewasting on Ebay etc
- Overworked, frustrated tech staff
All these are avoidable... as Sternie Ball of guitar string maker Ernie Ball explains here.What makes Slashdot readers think Linux will take over the desktop and server markets when FreeBSD didn't?
Hate to say it, but it's the GPL which will enable Linux to gain in marketshare beyond what BSD has.
The BSD License allows companies to take the work of the BSD developers, make changes, and not share those changes back with the original developers.
You could say that BSD codebase has been adopted widely throughout the industry, but it has been through other companies adopting (read: "Embrace and Extend") BSD code into thier own propoietary products without compensating the developers or community that made that code possible.
In this sense the BSD License is "more Free" than the GPL, but the BSD license does not ensure that that changes to the code will be Free as well.
Read, L