Slashdot Mirror
How To Save $1 Trillion a Year With Open Source
ChiefMonkeyGrinder writes "Cygnus founder Michael Tiemann estimates IT customers globally could save a trillion a year with open source or free source software." Not that a guy with a title like "VP of Open Source Affairs" at Red Hat would have a reason to be biased, but it's an interesting little read about a guy who's been doing this longer than you. Well, most of you anyway.
If you don't get it from Red Hat.
Uhh, who's gonna pay to hire a trillion dollars' worth of architects, developers, testers, trainers, managers, distributors, support personnel, human resource departments, etc etc etc?
Or is all that functionality supposed to spring from the ether in a perfect steady-state universe of human perfection & utopia?
Assuming that losing license fees directly means profit gain is somewhat dubious logic to say the least. Sometimes it pays to invest in paid solutions; and rarely is any one software stack purely OSS or propriety.
throw new NoSignatureException();
"Suddenly pulling a trillion dollars out of the economy would have a pretty severe effect."
Since companies would take the Trillion dollars they save and throw them into a furnace to heat their buildings...
Georgia Tech, the leader in Chia(tm) technology.
You're forgetting that the savings would be immediately put back into executive salaries.
RedHat (current owner of Cygnus) has made a successful business providing high quality support for FOSS software, and I think that's great! However, the $1T estimate seems like it might just be a tad biased and perhaps ignoring some hidden costs, but I can't tell from the FA because it just references the figure without any details for the estimate.
I'd be pretty pissed to see folks in big offices making real nice livings off of software that I designed and developed and tested.
I guess that's why I'm not a F/OSS developer.
It's NOT me! It's the meds! I'm on 1000mg of Fukitol.
Though he makes good points in the very brief article, turning it into a $1Trillion USD figure just comes off as shock tactics, and probably comes off as more open-source ranting to anyone just reading the headlines, or to anyone with a bias against open source proselytizing.
I don't have strong opinions about the matter, myself. I've seen some open source disasters where the proprietary solution is the industry standard for very good reasons, and I've also seen open source projects that are amazing, and amazingly practical.
I spend more on support than I do on software and there's almost no support even purchasable for opens source so I'd save a bundle!
I'm sure I'll get modded troll or something but I'm being serious. Some software is really expensive like matlab. But it always works. But a couple times a year I have to swicth from Fink to macports or vica versa because one or the other won't build the dependencies I need for matplot lib or octave. that costs me a lot of money in time.
Open source is not a cheap replacement if your time has value. But I still use it a lot none the less. it may not be cheap but sometimes it is better or has features you can't replicate easily in a single computing environment outside open source.
The biggest advantage and problem with open source is portability. I use open source so that I gan give my code written on top of it to someone else. I can't do that if I write in matlab and use exotic toolkits. But on the flip side it's also why code written in open source rather than a homogenous environment is so fragile and may not work in a few years (because say some critical library is gone). (Take for example the disappearance of whythelucky stiff and thus the demise of all SYCK based YAML bindings.)
SO it's true that you'd save a bundle on open source. You'd wish you could pay to have it maintained. You will pay with your own time instead.
Some drink at the fountain of knowledge. Others just gargle.
While I can appreciate the appeal of open/free source for IT guys like myself, I can't help but think that some of us push this ideal a bit too far. I currently make a living writing software, as millions of others do, and I'd like to continue making a living for the foreseeable future. Developers need to eat, too. The normal reply to a comment like this is that customers will pay for the support, rather than the software itself. Okay with me, but then how are customers going to save one trillion dollars?
What other industry consists of so many people that argue that the products they develop (or services they provide, if you prefer) should be free? Do doctors or lawyers or engineers ever argue that their service should be free? Construction workers? Accountants? Anyone? We're shooting ourselves in the foot.
I got a chuckle from this gem:
The reality of the situation was that we couldn't find any names that were not previously registered. When I lamented this fact to a couple of my Net friends, one of them searched the dictionary for words that contained "GNU". And "Cygnus" seemed the one that was least obscene
Free Martian Whores!
While I found this read interesting, I was a little disappointed to find much of his evidence random strings of numerical data. I'm sure anyone here can infer the cost savings and increased support in moving from an MS office to OpenOffice suite scheme within their enterprise, or transitioning from [Microsoft Product X] to [Opensource Magic Y]. On the other hand though, there's no insight as to how to deal with the seemingly obvious problem of our interdependency on these licensed products. I'm a database developer where I work, so speaking from where this impacts me the most, I can appreciate simple things like leveraging MySQL or other free source apps where appropriate. On the same vein, I don't see how reading this article immediately makes me jump up and go "Oh! Let's transition off of oracle for our company wide HR system." There's a reason all of these products have kept themselves going over the past 10, 15, etc years - and its more than just marketing and capitalism at work. Saying you can completely replace all or most of your IT resources with open source initiatives is ambitious at best, and completely ignorant at worst.
You're forgetting that the savings would be immediately put back into executive salaries.
And it's us, the computer janitors, who would have to retrain everybody and have to work around the quirks of all new systems that were replaced by working ones. Wonderful!
Posts not to be taken literally. Almost everything is sarcasm.
Most proprietary software companies spend little money on software development. The big players have margins close to 80% with a significant portion of their expenses in marketing and sales. Open Source companies are conduits of money and support to FOSS projects, making money off support and add on features. Generally low margins and small marketing and sales budgets (mostly word of mouth and try before you buy). Now, a massive movement to open source software will cause less total employment in the software industry, but the vast majority of those losses will be in non-technical fields. The economic issue is software is worth only ~25% of what people pay for it today. As performance gains from software purchases decline, the ROI is less compelling, and thus cost of software more critical. The critical shift now is convincing software consuming companies to shift from buying prepackaged software to contributing to the development of open source software. That could be co-ops of like minded software consumers, or some other innovative way.
Normally the windows in the Broken Window Fallacy are glass windows, not Windows OS.
Support Right To Repair Legislation.
SHUT UP! As soon as your company converts to FOSS all will be rainbows and unicorns! Your secretary will automagically know Thunderbird. Your graphics team will pick up GIMP in a day. Your sys admins experience and training on Windows servers will make him an instant Red Hat server guru. There's no learning curve, just awesomeness and freedom to change the software as you please!
Utilizing the synergization of benchmark e-solutions to pre-workaround action items!
They should, my company is trying to save money by dialing the heat down to 65 in the offices but on the other end, one of the IT departments spent a lot of money on a license for a syslog server. Not kidding, the company sold them a virtual appliance with a configured syslog-ng daemon and they are paying a license based on the events/minute.
The company I work at spends literally millions in closed source licenses for all types of crap that can be easily done using open source alternatives. Sometimes I wonder if there is nobody that actually checks what other software there is available on the market that would fulfill their needs.
Custom electronics and digital signage for your business: www.evcircuits.com
And, the heat generated by the burnig of this trillion will accelerate global warming. Open source is bad for the planet.
The ever-useful Google/WikiPedia combo pointed to a research report estimating the global size of the software industry at $308B in 2008. Saving $1T by not paying licensing fees to an industry worth 1/3rd as much would be a neat trick. Especially given how even $0 Open Source software is not free to support.
SirWired
It does provide a link to the paper that is the source of the estimate, but I suppose clicking the link would be too much to ask.
Wait, when would an end-user need to use the command for any of his daily tasks? Office apps? Email? Browsing porn on company time? Okay, maybe if he's looking at ASCII porn in elinks. Sure, the IT guys will spend some time in the command line, but you see the same thing in Windows administration.
And what work environment are we talking about where end-users are permitted to install *anything* on their machines? That's a huge best practice violation. If they want something installed, they submit a request to the helpdesk, who install it for them. The EXACT SAME way it is done in any decent Windows environment.
Time adjustment is a non-issue. How long did it take people to adjust from XP to Vista, or 98 to XP? Hell, the #1 most common help ticket we get at work is people who can't figure out how to do something in Office 2007 because of the retarded ribbon. They knew where the command was in the menus in 2003. Linux takes time to adjust, but once done, you don't have to keep readjusting every time a new release comes out. The typical adjustment is one of interface issues, and with the exception of KDE 3.5->4.X, you just don't see the major UI changes in Linux that you do in Windows.
Uh, I believe the correct answer is that "FOSS is easier to use anyway" and "People only can use [insert Windows product here] because they grew up with it" and "FOSS is better designed and more intuitive because of the clearly superior development model."
Of course, it's the developers that have been working on the product for 15 years that say this.
hehe. :) it's entertaining. But, all that said, I actually like using FOSS...
[that last sentence was primarily to try to mitigate the "troll" mod.]
I don't want to fuel this old thread, but I upgraded my computer, which was a dual boot machine. Windows and Linux. I changed everything but the hardrive. After the upgrade, Linux just booted at the new hardware without a blink. But iIt took me three hours to set up windows properly.
I couldn't find the setup CD for my wall-in-one printer. It was impossible to print on Windows before I downloaded the 200mb(!) setup utility for windows. Guess wich OS just printed and scanned out of the box?
Oh, BTW, back when I installed Linux, I didn't used a single line on the shell. All the setup was done with some kind of GUI.
Welcome to the 21th century, where Linux evolved quite a bit while you were whining about it
-- dnl
And I'd like to know why it took someone 3 hours to get Windows working. I can backup, do a clean install, and apply patches and install software in three hours.
This guy's anecdote is not about the ease of use for Linux or inferiority of Windows. It's about how much more he knows about Linux than Windows. His Linux upgrade went fine, but he messed up the Windows upgrade. He complains lost a disc he needed for Windows and had to download one which was 200 MB. Nevermind the fact that it's his fault he lost the CD, he also would have had to download the Linux ISO.
In my Linux experience, for every system that worked fine out of the box, I've dealt with one that needed drivers not in the distro, configuration the installer did not perform but should have, and on more than one occasion getting a system that would not boot because of kernel or bootloader issues ("noapic nolapic" anyone?). I've had problems installing Windows, too, but I've done that so many times I basically don't even need to think anymore and it still comes out right or I'm familiar with how to fix all the common and uncommon problems which arise. HDD not detecting? No SATA drivers. I have a USB floppy or I can use nLite. I'm to the point now that I know that if I'm going to install XP that I should look for the driver disk before I even begin.
This isn't the fault of Windows or Linux. You just need more experience.
The road to tyranny has always been paved with claims of necessity.
I can play. We spend $5000 a cpu for Ciber Fusion file transfer software, which uses the SFTP native on the machine, and sends passwords in the clear, and requires world write on everything. Because "it's our standard". We spend $23,000 a machine for SQL Server Enterprise, when either SQL Server Standard Edition ($1800) or PostGresQL or MYSQL would do the job. Because, "it's our standard".
I was taught to respect my elders. The trouble is, it's getting harder and harder to find some.