Slashdot Mirror
Open Source More Expensive Says MS Report
doperative writes "Much conventional wisdom about programs written by volunteers is wrong. The authors took money for research from Microsoft, long the archenemy of the open-source movement — although they assure readers that the funds came with no strings attached. Free programs are not always cheaper. To be sure, the upfront cost of proprietary software is higher (although open-source programs are not always free). But companies that use such programs spend more on such things as learning to use them and making them work with other software"
I predict that this report will be met with much skepticism on /.
I also predict that I will make the argument that open source really *isn't* always all it's cracked up to be--and be shouted down by many, many voices
SJW: Someone who has run out of real oppression, and has to fake it.
The authors took money for research from Microsoft, long the arch- enemy of the open-source movement-- although they assure readers that the funds came with no strings attached
Pay no attention to the man behind the curtain! THERE IS ONLY ZUUL!
Living With a Nerd
Maybe that is because software by certain companies deliberately ignore standards and try to maki it as hard as possible to work with other peoples software.
Come on, Bill gates more popular than the pope, Total Cost of ownership bullshit... I agree this is news for nerds but, is it stuff that matters? No.
In soviet russia the government regulates the companies.
So using an MS or MS-compatible (thanks to years of aggressive marketing by MS) stack is less expensive in terms of training time than inserting a piece of open-source software into that stack and trying to make everything work? Interesting...next up, replacing my car's wheels with motorcycle wheels makes it take longer for me to get to places. Perhaps I should just get the entire motorcycle instead?
An old-timer with old-timey ideas.
Am I the only one who see's this summary as picking the most incendiary portion of this article, and elevating it by taking out of context? The latter part of the article discusses choosing carefully, and promoting open standards to allow for more compatibility in open source software. Plus, this is a partial book review...what's up with that?
{sarcasm}Yeah, having a company around that maintains and tests their products for compatibility is always better than having to do it yourself.{/sarcasm}
I do software development for a small company, we have a mix of tools in our environment.
Recently, my development workstation was upgraded from an old Windows XP desktop to a late model Windows 7 desktop.
Microsoft Visual Studio versions from a few years ago complain of compatibility issues and some need to be run in "XP compatibility" mode to function. "Would you like to check for compatibility updates online?" - Yes, I would. Fancy that, there aren't any.
ActiveState Perl and Python development environments and my HTML editor-of-choice VIM all function with no oddness at all.
THIS is why the first paragraph gets sarcasm tags.
"Flame away, I wear asbestos underwear"
It's a good thing Microsoft places so much value on keeping the user experience the same across its various upgrades. Certainly a user of Microsoft Office didn't have to change their mannerisms when they switched from Office 2003 to Office 2007's "ribbon".
Certainly, all of my XP habits still apply to Windows 7's Aero. None of the functionality has moved around in the slightest.
And it's also a good thing Microsoft places a lot of value on interoperability. Certainly they have never seen an incompatibility that prevents Active Directory from working with other LDAP solutions. Or certain Windows code that creates spurious error messages when run on a competitor's version of DOS.
I give Microsoft all the credit it deserves for making reports like the possible.
--Joe
Chickens taste better, say panel of cows.
If Slashdot were chemistry it would look like this:Cadaverine
Funny. My wife's office recently upgraded from Office 2003 to a more recent release.
How do I know?
The first day it was on her computer, the conversation at home went something like this:
Her: "What the FUCK! The fuckheads in IT gave some new bullshit version of Word on my fucking computer and it is completely fucking different. I spent like a fucking hour trying to find how to do "X". Where the fuck are my fucking toolbars? There is this new bullshit toolbar that is completely useless." ... continue 15 minute profanity laced tirade...
Me: "It's called the 'ribbon'."
Her: "Whatever the fuck it is called it is fucking stupid. And what the fuck is this 'docx' bullshit?"
Companies spend more money on learning how to use open source? The three-year quota on profanity that my wife used up in a day says otherwise.
The closest the article comes to saying this is "that free programs are not always cheaper". Headlining it as "Open Source More Expensive Says MS" is pretty disingenuous.
One of my favorite quotes of all time summed it up nicely. I forgot it exactly and I can't find it now, but it was something along the lines of
If they're going to put scare quotes around "free" they should do the same for commercial software because you don't really "own" it.
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
Right.
You only have three options when there isn't a ready made solution for your requirements.
Conform your requirements to the available proprietary software, develop it from scratch, or build on the work of an existing near solution.
One of these things is not like the others.
One of these things is not the same.
One of these things means you don't meet your original requirements and you have to deliver something that is less than what is asked wanted.
They assume that you've already invested in training employees to use MS products. On top of that MS saves a lot of money when interfacing with other systems and software as you pay them to do all that for you instead of you having to figure out how to do it on your own. All have to do is pay for all your licenses and support instead of having to install free software and hire people that have brains to set it up for you. Using their figures this save you tens of thousands of dollars.
The benefits of vendor lock-in and proprietary file formats cost me way less money every year. The ROI of adopting the latest and greatest version of my proprietary software gets faster every year too. I really like the way my choices become more and more limited, and dictated by a governing body focused mainly on capital and politics. Not to mention that secure feeling of having a digital noose around my neck, dragging my head towards a grinding wheel with each revision of my server software. The benefits of meeting new and exciting people is a big plus as well. Just last month, I upgraded my proprietary mail server software only to find out there was some sort of misconfiguration error on my part which was causing my users to be unable to log in. I was on the phone with so many people from so many third world countries that I actually managed to learn a new language! We didn't fix the mail server issue, but for now, we use a Swingline stapler balanced on the spacebar to automatically close the error message dialogs to keep them from taking all available memory over night. What a creative solution! And it only took two weeks to figure it out! The vendor of our proprietary system promised us they will have it fixed in the next release. You can't get that kind of commitment with open source.
boycott slashdot February 10th - 17th check out: altSlashdot.org
The increase in cost is in the switching. Well that's nothing new. That's pretty much a one-time cost. After that training for new versions is going to be much cheaper, and with closed software you would also have to pay that. Difference is that once you have switched, you software costs drop to nothing, and you can choose whoever you want to do support/training as there is no lock in. That switching cost is something MS and others rely on, but it's a false economy to keep avoiding it.
I agree with one of the authors points. We just ran into a problem where OpenOffice would not properly read a docx file. The problem of course is the maker of the priority (guess who) software constantly changing file formats just enough to break everyone's conversion code. The old MS moto - "It's not done until Lotus won't run."
Conservative, mod down for violating
And yet the same applies to many of Microsoft's products. I have yet to do a major upgrade in server software without some pretty big hangups and issues, uninstalling certain software just to make sure the upgrade succeeds. These problems do not go away just because you paid $1000 for the base software and $5000 in CALs.
The problem I so often have with these claims is they seem to mistake ease of installation with the actual maintenance of the software. You betcha that Exchange is a lot easier to install than any potential open source solution. But maintenance over time? I've had more issues over the last five years with my Exchange server than I've had with the Postfix server I run, which only really gets kicked when I need to do upgrades. The thing just keeps humming, chewing up significantly less resources, thus requiring less expensive hardware, licensing-wise being much much much much much much cheaper, so I'd say that while I had my upsets getting Postfix going, if we're going on man hours, it's been far cheaper than Exchange.
Another good example is VPN. I reviewed a number of VPN solutions; Cisco, 3Com, Microsoft, and had assorted issues from stability to cost, and at the end of the day picked OpenVPN, which is secure, rock solid, and all in all pretty damned easy to configure. The people who, I suspect, would have problems with something like OpenVPN would be admins who sit completely within the Microsoft ecology, taking the courses, getting their MCSEs (or whatever its called these days), and thus having avoided as much as possible notions like text editors and command line tools, having drunken the kool-aid on "ease of use" to the point where the ease makes them almost useless at working on other platforms.
I'm not saying that things don't have their places. I'm fully cognizant of the fact that there are are areas where Microsoft is superior, certainly other groupware solutions, both closed and open source, are not the equal of Exchange+Outlook, though to be honest that's more a user ease-of-use issue than an admins. I think one needs to be pragmatic, but saying "open source is inferior" is just plain inaccurate. It certainly is true in some cases, but not in all. The worst part of living in a Microsoft world to my mind is that they've produced a legion of intellectually crippled sysadmins, who view competing products like Unix with either derision or fear, often times not realizing how inferior Microsoft is in some areas. This is clearly Microsoft's intent, to control the kinds of admins that get into any shop.
The world's burning. Moped Jesus spotted on I50. Details at 11.
TFA is even worse than that with their usage of weasel-words.
Pay particular attention to the "not always" in that statement. If only ONE "open-source" app is more expensive than a SINGLE closed source app then their statement is true.
Useless, but true.
If I add the lost productivity using TFS vs any of the FOSS SDLC or even simple version control software. The only statement we have been hearing from our MS tickets is "It shouldnt do that" Or "That was by design"... So far I a loosing about 1 man day every 4. And now that we have an "expert" on site, I am full time trying to debug this for MS.
The worst part of living in a Microsoft world to my mind is that they've produced a legion of intellectually crippled sysadmins, who view competing products like Unix with either derision or fear, often times not realizing how inferior Microsoft is in some areas.
Indeed, i have to deal with people like this every day... The problem is that smart competent people will evaluate multiple options and pick the one that best suits their needs, but people doing this would result in lost business for MS. So they would rather have an army of clueless drones that will accept what they're given and not try to think for themselves.
The problem I so often have with these claims is they seem to mistake ease of installation with the actual maintenance of the software.
Indeed, and monkey can get a windows based network up and basically hobbling along, and this is what their marketing concentrates on...
In many cases, a monkey could get many linux based setups hobbling along in the same way, its just that there is no marketing trying to convince people of this.
However, you don't want a system that is barely functional being managed by someone who has no clue how it really works... You want someone who knows the system inside out, who will configure the system to run optimally and ensure that necessary updates are applied in a timely manner etc.
However, running a system optimally is actually much harder with windows than it is with unix:
Under the interface, the windows system is actually far more complicated than any unix system... As soon as you try to do anything advanced which isn't catered for by the gui, or have to fix a serious problem it becomes extremely complex.
Windows suffers from security related design flaws which either have to be accepted as security risks, or kludgily worked around... Things like authentication with uncracked hashes, weak password hashing, file typing (including execute ability) by filename, broken apps which only run as admin, disorganised filesystem etc...
Windows lacks centralised package management, you will probably have to spend extra to buy a patch management system.
Windows requires anti virus, you will probably have to pay a lot of money for this too.
So sure, if you want a horrendously insecure and flakey network, hire the cheapest MCSEs you can and then you have money left over to buy the software... Such a setup will probably seem ok for a while too, until something goes badly wrong.
If you want a network thats reasonably secure and functional you will need to hire decent competent (ie more expensive) staff in any case. So why not run unix instead of windows, you will need less staff which will be a significant saving for people at this competence level, and you will obviously save on all the software (and perhaps be able to use less/cheaper hardware).
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
And knowing open source also means that you know how to fix it when it breaks.
If Microsoft software breaks all you can do is pray and hope that it will be resolved in a future bug fix. A call to M$ support renders you a long wait on the phone where you can't do anything and finally a question if you have tried to reinstall, and if you have done it and have any kind of custom software in the vicinity then they can't help you.
So either you are putting in some hours to get in control or you give up control to Microsoft.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
I'm not sure why this keeps coming up. In some cases proprietary software is better. In other cases open source software might be better. You can site a hundred cases for each that involve the initial environment, costs, goals, etc. Each company should look at open source to see if it can meet their needs. Sometimes it's better to pay Microsoft than it is to hire a contract programmer to fire-and-forget customize your product. Perhaps you have some great in-house talent already and they can customize and maintain your projects.
Anyway, to say one is cheaper than the other doesn't mean anything as a blanket statement.
Oh, no...they have to CONSTANTLY train people on Windows stuff, contrary to popular belief to the otherwise. MS changes up their stuff regularly enough to keep people that might draw a bead on their stuff at bay and to give reasons for people to "upgrade" to the newest stuff (If you think the ribbon interface to things is "easier", didn't need training, was needed to improve their products, etc. I have some nice swampland to sell you...). It amazes me to no end that people keep believing that it's "easy" and they don't have to spend tons of money on retraining with Windows and other Microsoft products- all the while spending scads of money on it all the same.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
Envision an alternative reality in which development of FOS software, let's say Linux, OpenOffice, etc., predated Microsoft's rise to power. In that alternate universe, the FOS tools got widespread adoption. Services sprung up to support them, in line with the idea that you can make money from open source by providing quality support and services for a fee. Because of its widespread adoption, documents are exchanged as a norm in OpenOffice format.
Now, along comes a commercial entity that says, "Our $500 office software and $350 operating system is the way to go."
They would get nowhere, for reasons like lack of widespread support, lack of de facto standardization, etc. And of course cost would argue against them.
But in our universe, the commercial software got traction first and, in a sort of positive feedback loop, as they made more money they wielded more power and increased their influence and made more money.
So FOS fights an uphill battle and much of that battle isn't really based on cost so much as power and influence, especially in high government and corporate places. When the IT shop suggests open source, MS doesn't drop in to visit with the programmers; they go right to the top, suits talking to suits, and before you know it we have things like Australia mandating MS "standards" for government.
As an individual, I use Linux, OpenOffice, etc., exclusively. As an individual, proprietary software suits have no interest in buying me off. I can provide my own support and to me up-front cost is all the cost, and makes a big difference. But the corporate and government worlds are hardly the same. I sincerely hope FOS will make more and more corporate inroads. But it's a tough battle against powerful foes.
Assuming that you do have the staff and infrastructure in place to fix, test and compile the code, you and your organization are subsequently forked from the official release. A scenario which is can be just as bad than waiting for MS to fix.
Once you code a solution to your defect, the proposed fix still has to be submitted, and may not be accepted by the developers. This could continue on for months, or even years of never implementing your fix. The onus is on your org, then, to repeatedly merge in changes, recompile, retest, and re-release your forked code base.
Again, this is all on the assumption you have the resources/knowledge to actually code a solution. Which many organziations do not.
And knowing open source also means that you know how to fix it when it breaks.
First off, I am a professional software developer/entrepreneur and a big supporter of open source software and freedom. But the argument you bring here is the worst argument *for* open source. I work with Evolution in my professional life and it crashes on me quite often. (Not often enough to replace it, so I will continue using it.) Even that I am a programmer, I do not know how to fix it. Getting to know a moderately complex piece of software takes a lot of time and effort (and thus money), that I rather spend on working for my customers. They actually pay me for my work.
Otoh, I also purchased Novell Groupwise, combined with SuSE Linux Enterprise Desktop (I thought it was a viable commercial Linux solution), and it is a lot worse than any do-it-yourself packages, so I guess closed source sucks too. Just in a different way.
-- The Internet is a too slow way of doing things, you'd never do without it.
That's how it works for us. We're a 300 million dollar company, maybe not big enough to get their attention? And yes we've gone through CDW et al, and done the whole licensing review and all that. They always come back with some absolutely astronomical figure for us to upgrade to new products -- when we're pretty happy with what we're running and only want to upgrade a couple of things.
For instance, our Exchange 2000 enterprise upgrade is something like $50K, based on 300+ users. We pay $300+ for retail copies of Office rather than deal with their crazy licensing schemes. We still run XP and have only rolled out Windows 7 machines recently (as machines are replaced they get 7 now instead of XP).
It's the same way when we call HP for EVA4400 support, or support for our BL460c's. HP won't even allow us to integrate all the computers we've bought from them throughout the years under a single support contract. It's a mess.
Even that I am a programmer, I do not know how to fix it.
I too spend much of my working life programming. It's really not that hard.
1) Download the source.
2) Compile with all debugging symbols and perhaps -fmudflaps
3) Run the program (with valgrind or mudflaps)
4) Go to the line number of the first error, and have a look.
Most of the crash causing errors are simple things, like uninitialised pointers. Some require some digging. But I have successfully fixed bugs and added fearures to a few projects and it's often not as hard as you might expect.
The modern tools available on any decent linux system are feally fantastic. Evil, nasty bugs like subtle memory corruption can be caught much, much more easily than before and therefore require much less in-depth knowledge to fix.
But for some reason, a number of high profile distributions don't have a package with the mudflaps helper files, even though they enable it in gcc.
SJW n. One who posts facts.
On a real IT planet, you have a problem, you solve the problem, you deploy the solution, tweak it until it works forever, and then you move to the next cycle. File servers are yesterday's news. Should there be any cost there over and above electricity and depreciation? Yes, I know I'm exaggerating a bit, there also has to be a massive restructuring of middle management every time disk drive technology breaks through another BIOS barrier. On a real IT planet, BIOS barriers are not revisited in living memory.
In economics, there is this problem about the reference basket for measuring the inflation rate. Sometimes you have to update the reference basket.
I think Microsoft is partly pulling off the funny math by ignoring the fact that if you stick with open source, your reference basket updates more quickly as things you used to pay for become to cheap to meter.
For the high churn technology that isn't yet too cheap to meter (and Microsoft dearly hopes this day never arrives) the cost of integration within an open source culture is non trivial, but it comes along with the agenda of eliminating the problem forever, not just persisting with the bleed and weep status quo, turning it over with one low low low small-bite-out-of-your-ass monthly payment until the end of time.
With the basket of goods thing, an idiot can mount a persuasive case that the cost of living in 2010 exceeds the cost of living in real terms in 1970, by placing zero value on any of the goods that couldn't be bought back then for any price. This would be done by focusing on the cost of energy which has gone up (maybe back date this to 1968), rather than what you can now do with the same unit of energy (talk on the phone to Asia for a whole tank of gas, and not owe the phone company a kilo of coke).
It seems that the researchers really didn't find anything, only confirming what many here have probably already seen. In the real world, open source and proprietary solutions work side by side in many if not most large organizations. It simply isn't practical to 100% standardize on a Microsoft or open source solution. We IT folks have to get our money the old fashioned way! Only the smallest organizations would find going all one way or the other an attractive and workable option. I think that what Microsoft is worried about is that small businesses can more easily cut them out of the picture and have a strong incentive, very good free open source applications, to do so. And with the global economy not being so great, perhaps MS is feeling the pinch. In any case, anyone trying to sell software or services has to market them, so I'd expect another such report in a year or two.
To the making of books there is no end, so let's get started
I am sure I would be able to figure it out. And I have made some additions and tips on open source projects. But, in general, fixing bugs that you encounter, while you are at work is not an option. It takes way too much time to figure things out and fix things.
The quote that the previous poster said, to wait for Microsoft to fix bugs, is just as true for open source projects. I am waiting for a bug to be fixed in Firefox, that is there since 2001 (!). I did try a shot on fixing it myself, but it would need a significant amount of my time, which I do not have available for this.
The only advantage on open source projects, is that you can hire a programmer to fix your bugs. But that would make the free software a lot more expensive.
-- The Internet is a too slow way of doing things, you'd never do without it.
Just because you get paid to write code doesn't mean you crap daisies and unicorns. I've been in the industry for 21 years now, I've seen the code that professional programmers write. In general, open source is at least as good and quite often better than what the professionals are writing. That's not to say OSS doesn't have its problems, but they are problems that are fixable if you're so inclined. Retaining a programmer to fix them might be expensive, but is it more expensive than modifying your business processes and just living with something a closed source company is unwilling to change? I don't think Microsoft is qualified to make that judgment.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
But, in general, fixing bugs that you encounter, while you are at work is not an option. It takes way too much time to figure things out and fix things.
From a business perspective of total cost of ownership, those bugs aren't the expensive ones. It's the ones in critical paths. Where entire projects or systems are at the mercy of a third party vendor. I've seen it happen a few times. In all cases the corporations survived, but spent $M of dollars rewriting/reporting existing production systems.
Yes, having found a problem, and then found a solution you have to maintain or share that solution. The horrors!
I non-open code you have the choice of paying potentially millions of dollars to get a fix from the vendor, and having paid that sum, you receive one fix once, with no promise that your fix will become part of the product line's subsequent release. So when that subsequent release is made and it _doesn't_ have your fix, you get to pay all that money _again_ even though they already know the problem and solution. Lather. Rinse. Repeat.
With open source you don't have to "fork" just to retain the patch and reapply it, usually with virtually no effort since, if someone is working on the code you patched, they likely used your fix, something like your fix, or didn't touch the lines you patched in any meaningful way.
I have had a kernel patch to "smarten up" termios for years. I submitted it and it was rejected for reasons like "we are about to change that code anyway" and "someone might have written code that _uses_ the fact that you can end up blocking on a one-byte read, waiting for one byte to be received, despite the fact that there is more than one byte in the buffer".
With every subsequent release of the kernel I just apply the patch and move on. I didn't "fork the kernel" etc. Nothing ever so daunting.
It is an obvious truth that exploring an option and making use of an opportunity is _always_ more effort and "clear expense" then just throwing up ones hands and living with no choice in the matter.
The costs of surrender are always hidden, prorated, long term. [Ask the French, their defense against Germany was sabatoged, as it always was, by Belgium's habit of buckling like a belt when threatened no matter how much the promised to do their part as a key point in the defense of europe. Nobody blames Belgium for being the useless twits they always are, but to this day France takes a load of shit for their surrender once their entire north flank went for strudel.]
Agree with Microsoft? I suggest you read up on "Plays4Sure"... and every single "microsoft preferred partner" in history.
Innocent people shouldn't be forced to pay for inferior software development.
--"Code Complete" Microsoft Press