Slashdot Mirror
Should All Government IT Systems Be Using Open Source Software? (linuxjournal.com)
Writing at Linux Journal, Glyn Moody reports that dozens of government IT systems are switching to open source software.
"The fact that this approach is not already the norm is something of a failure on the part of the Free Software community..." One factor driving this uptake by innovative government departments is the potential to cut costs by avoiding constant upgrade fees. But it's important not to overstate the "free as in beer" element here. All major software projects have associated costs of implementation and support. Departments choosing free software simply because they believe it will save lots of money in obvious ways are likely to be disappointed, and that will be bad for open source's reputation and future projects.
Arguably as important as any cost savings is the use of open standards. This ensures that there is no lock-in to a proprietary solution, and it makes the long-term access and preservation of files much easier. For governments with a broader responsibility to society than simply saving money, that should be a key consideration, even if it hasn't been in the past.... Another is transparency. Recently it emerged that Microsoft has been gathering personal information from 300,000 government users of Microsoft Office ProPlus in the Netherlands, without permission and without documentation.
He includes an inspiring quote from the Free Software Foundation Europe about code produced by the government: "If it is public money, it should be public code as well. But when it comes to the larger issue about the general usage of proprietary vs. non-proprietary software -- what do Slashdot's readers think?
Should all government IT systems be using open source software?
"The fact that this approach is not already the norm is something of a failure on the part of the Free Software community..." One factor driving this uptake by innovative government departments is the potential to cut costs by avoiding constant upgrade fees. But it's important not to overstate the "free as in beer" element here. All major software projects have associated costs of implementation and support. Departments choosing free software simply because they believe it will save lots of money in obvious ways are likely to be disappointed, and that will be bad for open source's reputation and future projects.
Arguably as important as any cost savings is the use of open standards. This ensures that there is no lock-in to a proprietary solution, and it makes the long-term access and preservation of files much easier. For governments with a broader responsibility to society than simply saving money, that should be a key consideration, even if it hasn't been in the past.... Another is transparency. Recently it emerged that Microsoft has been gathering personal information from 300,000 government users of Microsoft Office ProPlus in the Netherlands, without permission and without documentation.
He includes an inspiring quote from the Free Software Foundation Europe about code produced by the government: "If it is public money, it should be public code as well. But when it comes to the larger issue about the general usage of proprietary vs. non-proprietary software -- what do Slashdot's readers think?
Should all government IT systems be using open source software?
"Should All Government IT Systems Be Using Open Source Software? " where it makes sense sure. The primary thing I want government to do is spend intelligently, Open Source is definitely part of that, but don't use open source just because it is open source. I would rather them buy what is most efficient as the primary factor as those public servants are the costly inefficient piece and anything that makes there job slower is really bad for all of us.
Just having the sourcecode of software doesn't mean much. Quite some governments have access to source code of proprietary software. What is more important is the freedom of software to be used and changed by anybody for their own purposes.
Has become, I’m surprised the switch hasn’t happened earlierly.
It seems most proprietary software preempts the end-user or administrator in a myriad of ways, knowing “better” at best (I grew up luckily in an era where computers still took direction) or is just malware/spyware/adware at worst.
Which is why I loathe smartphones so. Such great potential. So utterly wasted. It’s a shame what the net turned into as well though.
Recently a Gartner report on open source in The Netherlands made an interesting case why with the current legislation the Dutch (and likely European) governments could not contribute to open source software. Governments may use it, but a software developer disguised as civil servant must never be provide patches or features back to the open source project, nor is the government allowed to publish their work in public, publication should be strictly limited to other governments. This would be prohibited due to unfair competition with software suppliers that build closed source software not having the advantage of government support. Now the case of no-vender-lockin still remains, but unless we first change these kind of laws, harnessing the true power of open source: collaboration, is legally not possible.
Support Eachother, Copy Dutch Property!
>> Should all government IT systems be using open source software?
All IT systems should be using open source software.
aaaaaaa
..."If it is public money, it should be public code as well..."
No, dude...
"If it is public money, it should be public code as well only if it works and does work well..."
But I am almost embarrassed to say that in my little world, apart from the browser, open source desktop software sucks big-time. It just does not cut it.
One has to "fight" with a situation where you have the same library named differently, installed in different locations, installed with older versions of the same depending on distribution...The arrogance in the open source world simply makes matters worse. Who has the time for all this nonsense?
Next question please.
Quisque verborum suorum optimus interpres...
The software has been more than good enough for a decade, or more if you have actually competent admins.
Not admins and users that are mentally stifled by having been treated like morons and unable to adapt their software to their actual needs for decades. Who had to settle for the dumbest common denominator, and eat whatever is put down their throat. (Yes, Windows 10 and macOS, I'm talking about you. Oh and don't think I forgot you, Gnome. You too.)
E.g. writing a shell script that gets triggered by a shortcut or udev or cron etc, should come naturally at least to the admins (who should be able to do it in their sleep), if not to the users. IMHO, current GUIs (but not GUIs per se) are considered harmful.
The failure has been, as always, in curbing the treason (aka "lobbyism") that drives deciders towards wasting money on for-profit imaginary "property" organizations instead of getting a fair deal for something made efficiently.
Also, closed-source software is a huge security risk, as security is incalculable by definition. And the constant drive to keep adding things to half-assedly justify making further money only makes it worse. Especially when combined with the death spiral of dumbing down that happens, when companies always want to make it "simpler" for users, but the dumbest users are the most vocal that they listen to, and if it's made easier, will just slack off even more and become even dumber, demanding to be spoon-fed even more... until you end up with today's UIs that are so "simple" that they are horribly painfully cumbersome. (E.g. the lack of being able to script/automate some repetitive task away forever, which would actually save time.)
The advantages of teamwork over a dog-eat-dog anarchy is the entire point of having a state and a government. That is also the key advantage of open-source over closed-source software. It's a human thing, dear lizard brains.
Yeah, the commercial offers sucked. And the market decided. For a better product and a better deal. Made by the "corporation" called "government", which is the "corporation" that we're all shareholders, employers and employees of.
The commercial suppliers simply hated an actual free market (and especially it balancing itself out). Like apparently all corporations and businesses without exception always do. Because they prefer unfair competition, but only if it's them doing it, e.g. in the form of a monopoly (even imaginary ones on imaginary property).
I think in the long run, FLOSS will win over all closed-source software. As an egoistical sole company simply cannot compete with everyone teaming up to make something free and libre. It's why social species succeed over everyone-for-himself species. And the imaginary property delusion won't last forever. People are gonna want to only pay for actual work, not for mere copies or mere profit, since they had to actually work for their money too. They only don't right now, because they have no choice, and because those who steal their money wrote laws and propaganda that became the cultural norm in some sad parts of this planet.
Nope, Windows is not open source, but users and developers are cheaper. I'd rather not pay the taxes needed to support all OSS.
In an ideal world where faries get you off daily? Sure. But in reality, no.
I've not had this problem. But I have not used anything other than Windows for most of 26 years. Every attempt, no library issues.
Of course I gave up each time so it was not long lived. So what are these libraries?
Sure, everyday insanity that is prevalent in software selection, but insanity nonetheless. The waste of money and the sheer dependency on a single or small number of companies is not acceptable.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
That is nonsense. Nonsense often repeated, but still untrue.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
And this "unfair competition" doctrine is the result of years (decennia!) of neoliberal lobbying. Why should be a government be prohibited to do what's best for its citizens and cater first to corporations which, in return try to avoid taxes as "cleverly" as they can?
I mean: corporations /can/ be the government's allies in fostering the citizen's well-being, but they can be also its enemies. It should be up to the government to decide when and how.
Lobbyists should be scrutinized much more closely. IMO half of them should be in jail, along with the politicians listening to them (the latter are worse).
One forgotten cost when using open source software is support. Every time an open source project adds or removes features it prompts a surge in support requests from users. Firefox is one example. When Firefox removed support for legacy add-ons everyone wanted to know how to replace their lost functionality. The removal of bookmark descriptions instead of just limiting their size caused another rash of questions. The removal of the Never Check for Updates means that every user is nagged to update to the newest version before it can be tested and rolled out in a controlled manner. Multiply these kind of problems to other OSS products for document processing, PDF, compression, graphic editing, multimedia playback, etc. and the support costs grow greatly.
Another problem with OSS is who do you call for tech support. Most OSS products have limited support for enterprise level problems. Many software packages STILL require a user to run in administrator mode to work properly. Saving user preferences in the Program Files area still happens in some software. Every software package that displays the infamous UAC warning will cause support problems in a managed system. Software packages that use the Windows Temp folder for some intermediate file use will be blocked by some anti-malware software. Who does a company contact to fix these types of problems? To be fair, some of these problems are still present in proprietary software.
Part of the appeal of OSS is the price; however, most people forget that part of the cost of retail software is the built-in cost of maintaining a support center, normally with a 1-800 number for question, or at least a knowledge base system to reduce the cost of support phone calls.
--
Who would deal with the inevitable liability suits? What about integration with vendor systems which are often proprietary or under NDA? What about vendor-derived systems full stop (not shrink-wrap, more thinking vendor has a core product which they then customise for each client)....
It's too blanket a rule.
is security, then that would be just an example of security hy obscurity.
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
... you have a piece of software that doesn't work. You call in the highly expensive support from the vendor and they won't be able to do much more than shrug at it. It's something I have seen at large companies and very large vendors.
"Free Software" means that you can change the software if you please. That implies that the software is simple enough for you to make meaningful changes to it. The simpler the software the more reliable and secure it usually becomes, that's why when hardening a system you throw out stuff you don't need. If you don't have your own staff understanding vital systems, you have done something severely wrong.
How will I easily find exploitable flaws if they use closed source software?
No.
Public/government IT systems should use open data standards and open APIs so that data is not tied to one vendors system.
Having that you can use whatever licensed software that does the job and is economically viable.
It's whether they're able to or not. There will be custom and proprietary software and hardware running on a variety of Unix, Windows and posiibly even mainframe systems. There will no doubt be plenty of OSS in there as well but until there's an easy and cheap migration path then the proprietary software isn't going anywhere.
I guess you haven't ever looked into it, and just swallowed it whole.
No, for-profit is, by its very definition, never cheaper. Since it's the cost of doing the work, plus the profit, plus the training that you have to pay.
And even non-profit closed-source is also not cheaper, since it's effectively still a (imaginary) monopoly combined with artificial scarcity. You know... those things that are major crimes in any non-imaginary-property industry.
Finally, even training is easier for open-source software, as you can see every time your beloved Microsoft alters their damn UI for the sake of justifying paying money for a "new" version again.
Also, listen here, lizard brain: Sure, you can refuse to chip in, and keep all your things for yourself. But how do you not realize that we won't share any of ours with you either? Even crows and squirrels realize that! Wasn't the whole point of the invention of commerce, that you can exchange things you don't need that much for things you need more? Isn't humanity so successful due to, among other things, using the advantages of teaming up?
I think your chances in natural selection look pretty bad, compared to social humans.
I do live is this ideal world. My OS has been the same for the last 15 years. The system is still clean as a whistle, yet I've got all the new features unless I didn't want them. Thanks to it being open source, I grew a host of little scripts and patches that make it fit me more snugly than a perfect glove. My computer does its actual job: Automate my work away, unless it really needs my input. ... What's so hard about it anyway? It’s all menus and bars of icons and property/settings widget blocks and input fields. You look for the word or image that's closest to what you need. Her old printer even works again under Linux, so she doesn't need to buy a new one. Thanks to some contributor.
While my girlfriend transitioned from Windows 7 and MS Office to Linux Mint and LibreOffice without any hassle whatsoever.
And we haven't paid a cent.
I think the only ones who still argue like you, are the ones who have never actually used a computer, but only used software like a fixed-function appliance that happens to use a computer internally. And that still treat Linux, if they ever tried it, like Windows. (Hint: If you run across a repetitive task... like always placing a window a certain way, or always executing a certain task at a certain event... find the setting to do it, and if you can't, for the love of cod, at least learn to write yourself a small shell script. Even Windows can partially do that sort of thing nowadays.)
It's not hard! If you can write a recipe, you can write a shell script.
OpenSSL.
node.js last year
PEAR this year
Open Source also has some fairly substantial supply chain security problems. The delivery model, and update cadence can also be pretty terrible.
The requirements of using something at home are vastly different than for the government, and scale becomes an issue. Your either paying a closed source vendor to manage this, or your bloating the size of your IT team and paying for it that way.
Using open source to save money is a myth.
I've not had this problem. But I have not used anything other than Windows for most of 26 years. Every attempt, no library issues.
Of course I gave up each time so it was not long lived. So what are these libraries?
That kind of depends on the distribution you are using, some of them are crap when it comes to this but there are enterprise distributions that do some good and proper quality control. However, if you pick some thing like the Ubuntu or Fedora community distributions you are going to have this problem because those people have no issues with backwards compatibility, a lot of them just don't understand what all the fuss is about. The people running the enterprise distributions do understand it because they get angry phone calls and e-mails from customers every time, for example, the Python team decides to break backwards compatibility because they came up with a more elegant way to structure their API. You could also make the case that Windows is better because of QA and they do good QA these days but keep in mind that there you are limited to one distribution and no tech support worth mentioning unless you pay through the nose. I used to work for a telco that had a gold plated support agreement with Microsoft but apparently that didn't even include a provision for Microsoft to get off their ass and fix bugs. All the local MS dealer seemed to do was collect extra payments for marginally better support. For proper support from MS you needed a solid gold, platinum plated diamond encrusted support agreement that ships in an unobtanium case and that we could not afford. With FOSS you can at least either change distributions or hire a mercenary coder to fix your issue because you have the source.
In a number of cases no, no it should not. FedWire being one.
Nonsense? Point me to the code in open source that can move wire transfers, both Fed and SWIFT.
Yes, universities need student worker jobs for experience, research grant funding to try out new ideas in support software, longer term planning which requires investing instead of short term cloud fees.
But governments which exist as a representation of the collective... is deeply aligned with the shared public work that open source is; with the biggest difference being it has an organized management with funding, power and the overhead of safe guards. That power and funding are what brings about most it's political problems... Sadly, the corruption and failing to fight against marketing/lobbying but in the USA, the increasingly anti-social culture is the main reason we do not collectively take on any new pubic works.
Open source projects are so unorganized, volatile, unpredictable it deters adoption and isn't enough to counter the close-minded thinking it is wrong for collective works to replace privatized services.
I do not think a national highway system could be built today. Obvious new public work projects that in the past would have easily been done have had trouble getting serious consideration. Such as, an information super highway... public health insurance, public healthcare, public car insurance, legalized co-operative insurance (illegal in some places...like public ISP are illegal too,) free college (high school wasn't free either until everybody needed it.) public recycling, trash, electricity.... or what everybody would lke: automatic TAX preparation by the IRS... which was proven cheaper but lobbyists killed that off.
I've worked with local governments. They do have plenty of lazy workers. I've worked consulting too; they have just as many lazy workers but those are forced a bit more in my view. It comes down to management in each. The main difference is that the public employees care MORE than the private employee (especially now with the lack of loyalty to workers.) Public workers have at least tiny bit more loyalty to their community/country if not a lot more. Many of the poor ones I run into and explore out of curiosity actually cared too much and the dysfunction of the system crushed their spirit too much. This one is most easy to see in the ones who quit their careers as cops/teachers etc. and the ones who are still plugging along are in the middle ground. If we stopped hating on our public institutions (like Russia wants and has been doing since the cold war... you ignorant Americans haven't got a clue! ) these people would be far more productive and happy.
Using open source to save money is a myth.
OSS doesn't always save money, but may. Open standards may allow you to change supplier, though, and allows you to shop around the market better if others are using open standards. If the software you are using does not use open standards then you may have issues with integration, which can cost money.
Most of the IBM hardware supports Red Hat and SUSE, but you still have a good point because I couldn't see anyone buying a pseries machine and not putting AIX on it. You would be losing so many capabilities such as being able to dynamically resize partitions etc.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Government has an obligation to make our data as safe as possible for as cheaply as possible and it ends there. If an open source solution fits those qualifications than use open source; but it's usually going to be a bad idea.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
IN practice, open source may not be compatible with legacy systems, or missing critical functionality. And support can be a nightmare, with no vendor to provide updates or respond to bug support.
And before you say do it yourself, that adds more cost than the licences, for programmers, managers, testers, etc.
(Note: This applies to most U.S. Government agencies, but not all.)
O.k., here is some "inside baseball" stuff. Every bit of software, from major applications, application helpers, plugins, drivers, etc. must be tested and accredited and supported. In a number of agencies, there are U.S. origin requirements.
The large corporations, for example, Microsoft, host government employees, to include DOD civilian and uniformed, to be part of the testing process. A few years ago, Microsoft implemented changes to Windows 7 authentication directly as a result of the DOD move to smartcard (CAC/PIV).
Support is another area of concern for the USG. All hardware and software must have continuing support, enterprise licensing, and continuing maintenance. The major corporations and some opensource do provide this, complete with published support and maintenance plans. They also participate in vulnerability assessment and reporting.
If you want an open source project to be considered, you need invite the government in, and understand the software/hardware acquisition process and requirements. Simply tossing your source to the government saying, "Here, check it out for yourself" doesn't work.
I worked on a large program (that you probably heard about) with a lot of embedded and command & control software. We made extensive use of both COTS products and open source.
Here are some of the impediments to using OSS we observed
1. The plethora of licenses! We kept 2 lawyers (one government, one prime contractor) busy nearly full-time for several years evaluating open source licenses. Each project had a different license, that needed to be understood for its impacts on procurement, use, distribution and maintenance, and how the licenses work together in a deployed system.
2. There was a big fight on the GPL. Many believed GPL would require the government to reveal all of its source code for this (weapon system) project. We never really did resolve this, and some GPL projects were disqualified from consideration due to license issues.
3. Maintenance was a key concern. For a commercial product, you can negotiate maintenance with the vendor. For OSS, you -might- be able to negotiate a support contract with a vendor (e.g. RedHat). But the government also might need to assume the maintenance burden if it couldn't buy support.
4. Related to #3: control of the evolution. With COTS products, there's a commercial entity that you can influence (including pay) to get the changes you need. With OSS, there's no guarantee the OSS product would migrate the direction you needed.
5. Related to #4: Complexity of integration. If you have N products, you have N! ways those could fail to integrate :-)
That being said, we used a lot of OSS in the project. We also took advantage of government site licenses on COTS, negotiated specific COTS contracts, and in some cases ended up writing our own code where we couldn't find an alternative. The project had a formal process for each significant component that required government and prime contractor concurrence. OSS tended to win in cases where there was a solid user community, some options for support (including training, by the way), and we understood the life-cycle risks. COTS won where there was an established product with clear maintenance costs (and things that the government already had site licenses for were obviously at a significant advantage.)
And I still remember the one government group that showed up with a 1.2m line application written in Visual Basic, who were totally pissed when we told them "We have no provision for Microsoft Windows in our computing environment. If you want to use a Windows application, your group will be responsible for the life-cycle costs to buy WIndows licenses where you need them, install/provision Windows and the associated software such as Anti-Virus, pay for the support costs including software maintenance and the people costs to maintain a Windows environment, and the training for the users and administrators for Windows applications."
It seriously offends me when I download something from a government Web site and discover that I cannot read it without buying a copy of Microsoft Word or some other proprietary software. It is not my government's job to guarantee Microsoft a market for their products.
4. Related to #3: control of the evolution. With COTS products, there's a commercial entity that you can influence (including pay) to get the changes you need. With OSS, there's no guarantee the OSS product would migrate the direction you needed.
The idea of OSS is: you hire people to make the changes/evolution you want. So you actually have much more influence over an OSS project than over a closed source project. However you rather pay the $130/h to a company which might make some changes in time instead of the $100/h to a freelancer.
Hint: if the software you want to be changed is Java, C++ or Python, you find hundreds of people here on /. who jump into it directly. Probably even a few dozen C# fans ...
As I mostly live in Thailand no, I probably would even lower my price to $90 :P
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
That depends, of course, on finding competent workers and companies (even body shops) to contract with. For my project, that included all the overhead and pain of doing contract work for the US government. Usually, defense work requires be performed in the US by US citizens, so that rules you out :-(
One has to "fight" with a situation where you have the same library named differently, installed in different locations, installed with older versions of the same depending on distribution...
Unix supports that scenario just fine. It was only Windows where it was ever a problem (DLL hell) though even Microsoft has largely solved it now.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Windows is not open source, but users and developers are cheaper.
You're ignoring the cost of running Windows. Not just the up front costs, but the maintenance costs, and the lost opportunity costs when closed source makes something difficult or impractical.
I'd rather not pay the taxes needed to support all OSS.
OSS supports YOU at the same time you support IT. It's not all outlay, you get the software back, and you get improvements from others.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
No, they should not exclusively use Free Software (sorry, "Open Source" guys, I never hopped on that bandwaggon) but they should have a strong preference for it.
Sadly, there are many areas where no Free Software of adequate quality exists. Areas that are vital for government work, and a government should not restrict itself. However, if an adequate Free Software exists, the government should strongly prefer it.
Security? Let's not forget two things: a) Free Software isn't bug-free, either, and especially tricky parts with security implications regularily don't get enough eyes on them. And b) we're talking about governments here. Unless you're the government of some tiny island, you can probably pressure big software vendors into giving you their source code for inspection. I mean, you seriously think the NSA (which is tasked with keeping the US government IT infrastructure secure) doesn't have access to the Windows, Office and whatever other source code they want? For large enough governments, every software is open source.
Assorted stuff I do sometimes: Lemuria.org
Apparently the submitter - and editors - fail to realize that many IT systems in the government are not PCs.
The non-PC systems are waning, though. These days, the government is more likely to use cloud services, or otherwise employ a cluster of PCs.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Perhaps I can masquerade as one :D
Anyway, such jobs I would do remote, so it rules me out, as I don't plan to live in a mayour US city. Country side would probably be ok. But honestly I'm to old to do this green card shit and follow all the regulations, I would not even work for Apple or something like that. Oki, Space X ... that I probably could not resist.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
An exotic example does not make a valid argument here. Incidentally, this will often be interbank agent owned software that they developed in-house and that is a trade secret. You only get the client side or the interface spec and that you may not even be able to buy.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
It's not always feasable. However every government contract for non open source should include a provision for data export in an open format.
They souldn't only be using Open Source, they should be using Free Software, preferably under some GPL or BSD license, with the weighing tilted towards GPL. And if they can't find it available, they should build it themselves (and publish it).
There may be a very few small instances where they shouldn't publish it, but in those cases the software shouldn't be distributed in object form either.
I think we've pushed this "anyone can grow up to be president" thing too far.
>Using open source to save money is a myth.
Whether or not FLOSS save money, depends upon what the comparison is with.
License Fees: FLOSS is usually gratis. Non-FLOSS is usually non-gratis.
Support:
* Tier 0: Gratis for both FLOSS and non-FLOSS;
* Tier 1: For non-FLOSS, when offered by the developer, it usually is gratis for a short period of time --- 90 days from date of purchase, or date of registration, is typical. When offered by a Third Party, as oft as not the cost of the first year or two is included in the price charged by the retailer, and the developer being completely out of the picture. Usually not available for FLOSS;
* Tier 2: FLOSS is usually more expensive than non-FLOSS. FLOSS support packages tend to be per incident. Non-FLOSS support packages tend to be per seat, per year;
* Tier 3: Non-FLOSS Tier 3 support generally requires a minimum number of seats per year. FLOSS Tier 3 generally looks at incidents per year.
Training:
* Non-FLOSS: First Party Training: Typically available, albeit at a high price;
* Non-FLOSS: Third Party Training: Typically available, pricing is all over the map. Quality of training is also all over the map;
* FLOSS: First Party Training: Typically not available;
* FLOSS: Third Party Training: Can be difficult to find. When available, the cost is usually higher than the equivalent non-FLOSS Third Party Training;
For Joe Sixpack, LibreOffice is going to be less expensive than Microsoft Office, simply because Joe Sixpack will purchase neither training nor support.
For MySmallCompany, INC. Microsoft Office, with genuine Tier 2 or Tier 3 support, will save money, when compared to LibreOffice with Tier 2 or Tier 3 support. Unfortunately, most Third Party Tier 2 and Tier 3 non-FLOSS support is run by scam artists;
For LargeEnterprise, INC. the cost for Tier 3 support is roughly the same, regardless of FLOSS or non-FLOSS status. The cost of training users for FLOSS is higher than training for non-FLOSS.
For VeryLargeEnterprise, INC, FLOSS with Tier 3 support is cheaper than non-FLOSS with Tier 3 support. The additional cost involved in training for FLOSS may or may not equal the reduced costs of Tier 3 support.
Remember, Sun purchased StarOffice, GMBH, because it was cheaper to do so, than purchase the same number of licenses for Microsoft Office.
If you're a SOHO looking to migrate from Microsoft Office to LibreOffice, your best course of action is to retain a migration expert, to guide your organization in how to store and archive your existing data. Budget for per incident Tier 2 and Tier 3 support, for at least five years after migrating.
If you're an SMB looking to migrate from Microsoft Office to LibreOffice, your best course of action is to retain the services of a company that has experience in migrations, for the migration period. Then retain an individual to do Tier 3 support. This individual can be either an independent contractor, or an in-house employee.
Large organizations should have at least one individual, either an independent contractor, or in-house employee, whose sole function is to provide Tier 3 support for the FLOSS software that it uses.
A possible example is in federally sponsored bioresearch. If money from the Feds are used, the data needs to be made public. Why not software? The fact that some is bad could be an opportunity to fund it to make it better. I don't really buy the idea that only FOSS software can be bad, while all paid one is worth. The former can be held accountable for its quality but not the latter...
I work in tech support at the IRS. Billions each year are thrown in fire for Microsoft software that is unreliable, and broken worse by every "fix" they send out. The Windows 10 Upgrade is a disaster. The ticketing system from HP is a waste of billions that gets in the way of doing our work. Adobe Acrobat is an unjustifiable expense now the PDF is no longer a patented technology. I could go on forever about the awful software billions have been wasted on, and how tech support is stretched way to thin trying to babysit all the junk. Instead of that, I will talk about something good. The VA's Vista system. It is the only electronic medical records system developed hand in hand with the doctors and nurses who had to use it, and the only one in the industry medical professionals don't hate. It was developed in house by the VA on the sly, as the bureaucrats never would have authorized its development. Government should stop wasting taxpayer dollars on commercial software, period.
I feel your pain.
The Uncoveror: It's the real news.
For example,Biowulf, 100th fastest supercomputer on the planet, at the NIH, mostly runs Linux. And many peopel use R, rather than paying the licensing for Matlab.
Now, whether management wants to support Linux and OSS, or repeats in their sleep "THE WORLD BELONGS TO M$" is another story... but it's heavily used.
Just for fun, slashdotters, look up https://www.spi.dod.mil/lipose... - a lightweight secure distro of Linux, can run from a flash drive.
Put out by the US Air Force.
Windows is not open source, but users and developers are cheaper.
You're ignoring the cost of running Windows. Not just the up front costs, but the maintenance costs, and the lost opportunity costs when closed source makes something difficult or impractical.
These also apply for running OSS. I'm sure it's possible to ultimately replace Active Directory with some implementation of LDAP on CentOS, but a virtually any sysadmin with a pulse can go from bare metal to multiple domain controllers with checkbox-compliant GPOs, DHCP, DNS, shared folder permissions, and server clustering in an afternoon or two. I've yet to come across a drop-in replacement for that sort of core functionality in an OSS package. Additionally, a whole lot of closed source software only runs on Windows; moving to not-Windows yields lost opportunity costs on that end as well.
I find myself as a software pragmatist. I would love nothing more than the Department of Developers (DoD?) whose job is to write OSS software that is compliant enough to replace closed source titles in use by the federal/state/local government. However, it would be a matter of principle, not a matter of cost savings...and it's been a very, very long time since we've had a political climate where such a department could be effectively founded and funded.
If one of those tools is Windows and one of those tools is Linux, who cares? As long as it's the right tool.
First off; In a world without Windows, why would you need AD?
I'm not asking to be mean, but IMO this is one of the bigger problems with switching out proprietary software, specifically Microsoft's offerings. People are so indoctrinated, that they keep trying to solve Microsoft problems, the Microsoft way, which invariably leads to anything different being deemed "inferior". If you look at it that way, your question is the perfect example.
Let's look at a handful of things AD does that would likely apply to Linux clients:
1.) Centralized authentication. Users should be able to have their password apply to any computer in the environment. LDAP does this particular part pretty well.
2.) Failover/Replication. LDAP supports this. LDAP does not support this in less than an hour from a bare metal install unless you have a bunch of scripts already written.
3.) Group policies. How do you ensure different departments can only print to their own printers (Linux users print, right?)? How do you make sure profile folders are transparently redirected to the server (Linux users store data, right?)? How do you schedule patching intervals (Linux users want patches applied after hours, right?)? How do you specify proxy settings, especially when adding a trusted certificate for HTTPS filtering (Companies don't allow free-for-all internet access for Linux users, right)? Now, the answer may well be "shell scripts at logon", but do you have different scripts for different user/computer combinations? All of this is done via group policy.
That's just off the top of my head.
Secondly, your version of a DoD sounds like a good idea, but it wouldn't just be a matter of principle. It would be a matter of trust and control too. One can only ever have one master, and as long as we (as in we, the people of the state) rely on commercial actors, who ultimately have a completely different agenda and set of desires from what a state has, there will be conflicts of interests. It's crazy to have a state beholden to the whims and desires external entities!
I'd love there to be a DoD, but I also fear that government developers would be hamstrung in some of the very worst ways. infinite scope creep, "why are we funding this finished project; we don't need no stinkin' patches?", "Your EMR connector needs to be able to understand data from $STATE_A and $STATE_B, each of whom use different systems built by direct competitors to be as incompatible as possible", constant subservience to the political and budget wind, standoffs regarding who gets to make the standard and who gets to conform to it (exacerbated if a state who has opted out of a new system still has to get their current one into compliance), incumbent systems dating back to the 80's, kowtowing to requests of different states if they're willing to directly fund projects, secondary effects from/to the private sector, and even the fundamentals - do they assume you're running GovSys from the BIOS up, do we assume Windows and GovLinux versions of everything, can they write a program with a depedency on Oracle? Could they do so if Oracle was compelled to release a version of their software that could be utilized to fill that requirement without expenditure, and if so, do we now reopen the can of worms that was the San Bernadino iPhone case?
A new country starting today could probably make that one of their enumerated departments and require conformity from the very first computer purchased might have a fighting chance. China and DPRK who own the major software houses anyway could have one; it'd basically be a standards body at that point - one of the silver linings of an absolutist government. The USA...sadly...would be a super difficult place to make that happen.