Except they didn't. It's just a campaign to strongly consider open source software in future contracts. Currently, there is a contract with Microsoft that isn't ending just yet. So they will have a mix of LiMux and Windows for the foreseeable future.
This is a bluff that almost every government and municipality with huge contracts with MS does whenever they come up for renewal as part of their negotiations, nobody in MS is impressed.
In other news: 2020 is not the year of the linux desktop.
Do you assume it's bluff or is there more behind that statement?
I understand that it has been in the news that MS spent a lot of energy on preventing people from moving away from Windows, and I think I remember steep discounts but it has been a while since I read about it, but while the government could be going "haha let's put out another statement and get more discounts" I would say it's more likely that there are people in power in Europe that genuinely want to lessen our dependency on the benevolent big brothers from the USA. But that's just how the news reports it, perhaps I'm not cynical enough. Anyway, if you know of a statement by anyone in a relevant government that claims this was a negotiation tactic, that would be interesting.
The statement is in the language of the agreement as reported: "Where it is technologically and financially possible, the city will put emphasis on open standards and free open-source licensed software,"
Which in practice binds absolutely no one to absolutely anything.
They're already on Linux currently. A government change and Microsoft Germany moving their HQ to Munich made them reconsider. Now the government changed again.
The actual migration to Linux went mostly fine and was largely a success as far as government software programs go, FWIW.
From my personal experience most times the hardest part is to convince the users. You can throw a lot of resources at the migration, money, good technical experts, and get it done right. But making the regular users happy? That's a whole different subject. Truth is most users still run Windows at home and are familiar with it. Getting used to something else will meet some resistance. Then you have the minor compatibility issues especially in a people facing institution.
To this day eventually this attrition pushes many back to the more compatible and familiar solution.
I've seen (usually very small) companies going to Linux and other FOSS solutions without actually understanding this beyond "they're free". Migration went great. Later on I came back to do the migration the other way around once they faced that attrition and realized that support costs were dwarfing the license cost.
> From my personal experience most times the hardest part is to convince the users.
Former actual LiMux team member here, so I can speak from experience (almost a decade ago though). The hardest part was convince users to be happy on utterly inadequate and aged hardware. No, 1 GB of RAM won't cut it when you have to have Firefox, Thunderbird, OpenOffice and some fucking Java crap open at the same time, and 1024x768 screens won't do so either.
Yeah, that must have been a huge problem. These days however an OKayish notebook can be had for $400 or an i3 NUC with 8GB RAM and 256GB SSD with an IPS 1080p monitor for $500, and those would run any Linux well for foreseeable future without killing city's budget.
Unfortunately that isn't easy with the way government procuring works. Basically, you make a specsheet what you want (e.g. spare components must be available for a duration of X years) and then have to publish tenders, and usually the vendors offering bids have to fulfill certain conditions (e.g. must have a reliable history working with governments).
The last part alone excludes many cheap vendors as they don't have government contracting experience, and the spare parts requirements kills the rest as they can only be so cheap because they don't have to build much of a spare parts infrastructure. (Note: this is not specifically relating to LiMux in any way, I have no knowledge about the contracts/tender process that happened back then)
Also, the scale matters: 15.500 PCs are even at 500$ apiece 7.75 million €.
>>> Getting used to something else will meet some resistance.
It's too gentle to say that users are struggling or resisting.
The fact on the ground is that a significant part of users are simply incapable of learning new computer software, they can't even use effectively the software they've been working on for years. This is glaring in a giant government organization with tens of thousands of employees. Thus any attempt to move OS is doomed to fail.
Classic example is the popup when closing a word document. "Do you want to save the document before exiting? Save/Quit/Cancel". The typical user clicks the little red cross in the top right corner of the dialog.
IMO this is UX fail. If the question starts with "Do you...", only reasonable button choices should be "Yes" and "No", not "Save", "Quit" and "Cancel".
It's not reasonable. This goes against all major HIG. I encourage you to read them so that you come to an understanding why verbs must be used instead of yes/no.
It would be more helpful if you included a link or two. I don't have time to read "all major HIG". It might not be reasonable from your POV, but from mine it is - if I ask you "Do you know Michael Jordan?" and you answer "Cancel" I would've thought you're nuts.
Easy, just create a software ecosystem that is superior to or at least on-par with the commercial offerings (being open source is not by itself valuable for the users), then grow a generation of non-technical users into that ecosystem
That should be "offering" singular, not "offerings" plural.
And it's totally feasible to produce a superior ecosystem. The hard part is absorbing years of financial loss while building the new ecosystem from the ground up while it isn't compatible with the existing proprietary system.
> That should be "offering" singular, not "offerings" plural.
Why ? An ecosystem consists of multiple offerings and there are also multiple commercial ecosystems.
> And it's totally feasible to produce a superior ecosystem. The hard part is absorbing years of financial loss while building the new ecosystem from the ground up while it isn't compatible with the existing proprietary system.
Being able to finance something is an inherent part of its feasibility.
I think that this is not a charitable interpretation of the situation.
While I generally agree that FOSS software (Libreoffice calc) is inferior for business needs, a _lot_ of the issues actually stem from compatibility.
Not just compatibility of the document system and interpretation, though that is a large part. But compatibility of functionality too, along with UI design.
People are more likely to re-train for a new UI on a new edition of their previous software because it is seen to them as a progression, but they are not so open to changing UI paradigms with different software that does the same thing.
In order for something "better" to be accepted it has to not just be better- it has to be _so_ much better that it trumps everything else.
in the very best of cases it has to be 100% compatible in all areas and _still_ at least substantially better.
That's without getting in to opensource financing, which is a difficult topic. Though I can imagine if Munich put their resources into a FOSS foundation which was providing productivity software it would help tremendously. I don't think they're going to do that because Linux (and ecosystem) is seen as cheaper, so why negate the cost saving?
You could make it illegal to price discriminate on software sales. I.e, if you sell a piece of software at one unit price to someone, you are obliged to sell with that price to anyone who wants it.
Obviously that would disrupt quite a few existing business models in software industry, so do not take that as a completely serious proposal.
Year ago I remember learning about appliance stores who advertise "lowest prices guaranteed!"
More specifically, if you can find another place which sells the same model for less, then you get it for the other price.
Thing is, they are the only seller of that model. The vendor has a model X123 for that shop, a model X124 for another place, etc. And that shop is the only one who can sell that specific model. So they of course have the lowest price.
Something similar exists for software: that's why there's "home" and "office" and "professional" and "enterprise" and so many other variants.
So, your "home" version only supports 2 connections at a time, your "office" one doesn't implement a certain function that "professionals" are more likely to use, and your "enterprise" defaults to logging. Even better, with software you can have a unique product for each sale.
All different pieces of software, so all different prices.
Then on top of that, you don't sell the software but software + support. And the support costs can vary.
Which means I think it's easy to bypass your hypothetical restriction.
> More specifically, if you can find another place which sells the same model for less, then you get it for the other price.
> Thing is, they are the only seller of that model. The vendor has a model X123 for that shop, a model X124 for another place, etc. And that shop is the only one who can sell that specific model. So they of course have the lowest price.
This is endemic in the mattress industry. Products that are identical and manufactured by the same company will be relabelled based on the company that buys them.
That's why your "dream-air luxury liner" could not get a discount because the other company has a "dream-air luxury comfort", which is the same exact product.
Open document standards are a great place to start. Any standard format needs to have multiple possible vendors. And I agree it goes further than just document formats. This goes for all important formats, including how software interacts with the operating system.
There needs to be other vendors who can supply document editing software that is 100% compatible with MS Office. But more than this, there needs to be other vendors who can supply operating systems that are capable of running Windows software.
As long as a single company has the leverage that Microsoft has over the enterprise ecosystem we won't have a free market. Governments can start by mandating that new software deployments have several possible vendors, including at the operating system level.
Key quote: "So these kinds of decisions must come down to political philosophies about the marketplace as well as increasingly important issues like digital sovereignty. This issue is one reason why this time Munich is unlikely – or any other German city or state – to flip-flop on open-source software decisions again. "
> As yet, none of the big players appears to have lost significant business to the free software movement in Germany. But it seems likely that commercial vendors will have a tougher time here in the near future.
Free software and open source doesn't mean non-commercial. SUSE in Germany provides commercial support to their distro for example. Such a weird article, zdnet doesn't getting any better...
I can't imagine any US party echoing the FSF's statement, even if just in a statement of political intent. It's impressive that the movement has enough public interest to make platitudes worthwhile in Germany.
The FOSS movement in Germany always seemed to me a tiny bit more mainstream than the one in the US. Tech publications report more on Linux happenings (as opposed to the sometimes intense Apple focus of US tech publications).
Linux also has more market share in Germany than any other western country, I think.
According to statcounter, Germany has the highest usage of Firefox and fourth highest usage of Linux (behind Ireland, Greece and Luxembourg) among European and North American countries. Additionally, you have the Chaos Computer Club, which is somewhat prominent and could influence the thinking about FOSS.
Why would the US government try to avoid Microsoft? It's a well-known local company. And the government can get a good chunk of the license costs back throughout company and employment taxes.
In the eyes of the FSF and people with similar ideals, the benefits gained by spending that money on free software dwarfs the tax revenue gained. The German people seem to have enough support for free software to at least make the government notice, which does not seem true in the US.
There is another interesting section in the document[1]. The city introduces the "Munich Open Source Sabatical". The city will pay professional programmers for 3 till 6 months to work on open source projects that benefit the community.
> Die Stadt München unterstützt die Entwicklung von Open Source-Projekten mit einem "Munich Open Source Sabbatical". Professionelle Programmierer*innen, die sich für drei oder sechs Monate ganz auf die Fortentwicklung eines Open Source-Projektes kon-zentrieren möchten, können sich dafür auf ein städtisches bezahltes Stipendium bewerben. Die Projekte müssen einen kommunalen Nutzen haben.
The document doesn’t say how much. I guess, it’s just an umbrella document. More detailed regulations will be eventually adopted. That said, my estimate would be 500€ max per month.
that would not work. 500€ is not enough to pay rent in most places.
it would have to something that covers your current living expenses, no matter how high they are, because unless you are young and single with a lot of flexibility, you are not going to be able to put a hold on all those expenses.
therefore the two reasonable options are: the government continues to pay your current salary, or it pays what you'd get if they'd hire you. the latter would disadvantage high earners, but that is somewhat reasonable.
anything less would unlikely get any people interested.
Governments should develop and run opensource systems. That way different governments can collaborate to build better systems through open innovation. Not wasting tax payers money on closed source.
I bet tax and pension systems have the same basic requirements cross countries with some local specifics.
While I completely agree with you, there is a compelling argument for governments spending tax payer money on goods from local companies rather than competing with them. Of course, I don't think I would class MS as a "local company" for Munich ;-). I'm just saying that there are lots of reasonable points of view on this topic.
how is hiring local developers as employees any different from outsourcing to a local company? if the city is developing software then it is surely something that doesn't already exist, and when they do they are also not selling it, so i see no competition there.
if the goal is to create jobs, then that also happens either way. i don't see the problem.
I'm working on some projects for the city of Amsterdam, by default all we develop is open source.
But whether this is a legal issue is actually an open debate, in some specific cases companies could potentially sue Amsterdam for sharing solutions with other governments and thereby "competing" with the market. How a judge would rule is not clear. But this is something they are aware of and it's a risk. It seems to also be connected to the way the project is shared.
As far as I know they are looking into changing the law on the national level so this stops being a problem. Also strangely enough, I've been told, as soon as the project gets funded by the EU, this law is not applicable anymore.
Open source means published on github, not many projects are actually properly documented.
Same governmemt can spend the same money within the same local economy by paying programmers. There is no valid reason and not that many "reasonable points of view" about paying a private company 5000 times over for the same piece of software, and still not owning it or being able to get full functionality out of it or being able to see into it or your taxpayers (who are subject to it's operation) being able to see into it, vs paying programmers to write the same thing 1 times over and then you DO own it and can get full functionality out of it and can see into it, and every town in your whole country can use it with no legal burden, no audits, no broken license mangers... And tomorrow, instead of paying for another 100 copies of the same license yet again, you can pay those programmers to make what you have even better or build the next new thing.
There is no valid reason to just pay rent to anyone for nothing, even if they are a local company. That's not a good enough reason.
which risk? a software company is going to make sure that their risk is covered financially. so if the software turns out to be useless, they will sure find a way to shift the blame and get paid anyways.
Unless it's paid-to-build, the software company's risk is creating the software and then not being able to sell it because a different software is better.
Communal software isn't always custom-built, the majority is normal off-the-shelf stuff, just focused on that niche. My involvement is limited to a small city a few years ago, that may be different in a large city as Munich if they have lots of custom requirements they don't share with other cities.
if the software already exists, then it's not a risk question, but one of cost.
the buyer can not shift the risk to the seller. how would that go? put out a tender for a completed solution and then wait until someone builds the software on their own risk, hoping to win the tender? that's like a designer competition, and those have been considered unfair.
Software doesn't just "exist", it's created. And spending time and/or money on creating means taking on the risk to fail. Again, they're not putting out contracts for bidding for every software they want, just like most enterprise software isn't custom made. They buy lots of software from companies. Those companies have competitors, so cities can choose which software to buy. Those companies shoulder the risk of not being chosen and having wasted the money.
If the cities built it themselves, they would not be able to choose who to buy from. And if their self-built software sucks, they've wasted money and have a shitty software. One that they may not be able to replace, because there's no longer a market for that kind of software. Public endeavors into building things (airports, train stations, rapid train lines etc) in the recent decades in Germany does not predict a bright future for these kinds of things.
i get your point. what i am trying to express is that if things are done well, then a government will not develop custom software if a suitable product already exists.
of course software is created. but the risk a company takes in creating is irrelevant to the decision a government makes. that risk is from the past.
you said that the government can either choose to take a risk by developing their own, or offload that risk.
but that is not the decision it needs to make. when it takes an existing product it is not offloading that risk, because the risk that vendor took is independent of the risk of the government.
the situation should be rather like this: either a solution exists, then we use that solution with a small or no risk because the vendor already took that risk in the past. or the solution doesn't exist, then we don't have a choice but to custom develop with all the risks that that includes.
at no point is there an option to offload an existing risk.
the only way to offload an existing risk would be to say: we wait until vendors come up with solutions.
(it may be that i am getting hung up on little details here. i believe that we are actually mostly in agreement in how a government is handling custom development)
>Governments should develop and run opensource systems.
I'm for this proposal too, but for different reason.
opensource system can create much needed transparency in government project and organization effort. In my home country the unemployment agency spent atrocious amount of money to create an unusable job-posting app; something private companies do for fraction of the cost. More surprisingly, they doubled down on it; instead of scrapping it when it wasn't delivering. My guess it that some nepotism/favoritism was at play, and if the project was at least open source it could've been examined by others.
> Governments should develop and run opensource...
Happy to say that I am working on such a project. Heavy (and unnecessary NDA) so I cannot say in cleartext but it is in use in one Nordic country already, it is opensource and already I hear that the Dutch found it to be a good starting point and probably will be using it too.
Some more details, the company I work for is owned by one country but doesn't compete as much as cooperate - I am a consultant from another company. And we both use open source software and libraries and publish as open source most of what we create.
IANAL, but any NDA i have seen, even those overly broad make a clear distinction between talking about what has already been made public and what hasn't.
Absolutely. Where I live, the city administration sent a survey to people, asking what projects to tackle, where to save money/increase income etc etc. One of the options in the "saving money" category was going open source. You can bet it was one of the things I chose and in the end, it ranked 3rd which is not bad. Will be interesting to see how the administration will vote on it.
Is there evidence that governments are saving that much more with open source? You'd still have to pay developers...and some cities are going to freeload off others to avoid overhead costs.
It's a complex issue. There are some calculations and they can go both ways, depending in what and how you calculate. E.g. you could consider training costs - but training costs are one-off for a change, not eternal. Or you can consider coding costs, but they might vary massively from region to region and depending on how (in)competent the procurement goes and how the local market works. Estonia is a country that seems to have figures this out to an immense degree, creating most of the infrastructure themselves and then tendering small-scale procurements for small parts of the system and sometimes allow private competition - but always with government-set data standards and often more on interfaces etc.
To add one other angle: what is cheaper, paying 5 million to a company in Redmond yp provide licenses for opersting systems and officd software, or 10 million to a number of local companies to set up and adapt a slightly less refined open source tool for the local needs? In one scenario the money leaves the local economy, in other the city might spend more money but the money stays local and is used to buy houses or sandwiches, pay handymen or bus tickets or taxes. Where is the long term cost/benefit/ratio higher?
Maybe the difference to last time is that the Greens are now the strongest party? I guess they are not as easily "lobby-able" by Microsoft as the others.
No, they were implementing it at scale for 10+ years. There was an issue with lots of incompatible legacy software so they were still using windows as well but >50% of desks were Linux only.
Then in a very short timespan (1) a new conservative city government came in, (2) Microsoft announced a big local office and (3) the new government killed the open source project without warning and anlunced they'd switch everything to windows 10.
This transition back to windows is now still ongoing and as contracts are signed will not really be feasible to stop. The now new-new Green+social democratic government makes noe a pledge to take open source seriously for all future needs (and I suppose switch back to a Linux asap).
I still don't understand why major nations don't create or help create national OSes for their own countries/industries. You would think an OS is national security issue at the very least. A municipality isn't going to cut it. It's going to require a major national effort to stave off a monopoly like microsoft.
I suspect this time they will be able to make it stick, even if a new government comes into power in following years. The options, particular for desktop linux, have gotten better since the last time they tried it.
In case of government spend it is hard for me to comprehend why almost no place in the world has fully open source gov technology. Governments everywhere spend these insane amounts on really bad software :/
The sad reality is that there just isn't enough awareness or understanding of technology within Government to make it onto the agenda. The exceptions of this are little pockets of like-minded individuals such as GDS [1] who have done some exemplary work on open sourcing tech.
The downside is that much of the 'trunk' of IT (workstations, Business Software, internal systems) are still contracted out to monolithic, old-school contractors who have no incentive to innovate by non-technical procurement teams. This leads to a culture of proprietary software where the established/expensive solutions are seen as the most trustworthy and there isn't any cost to vendor lock-in or non-interoperability of systems until the next tranche of officials have taken over and all of those issues can be written off as 'legacy' and as a sunk cost.
As a state government worker, this jives with my experience. The ultimate beneficiaries of any budget savings or improved processes are citizens who have no idea, let alone input, about what we're doing behind the scenes. It's very hard to convince public employees to work with those people in mind. You're right, the people buying the software are even harder to convince. With no good measures of effect, they rely on what's easiest for them. So Microsoft and PDFs everywhere.
Our software licenses are paid through a grant from the federal government. If we replace them with open source software, we don't get to keep the difference. So now I have licensed software like SAS and ArcGIS gathering dust on my machine while I use R and Python.
And sharing software we develop is even harder. I've pushed to release code for my statistical reports. When I said researchers and other states could benefit by reusing parts, my supervisors only saw it as a liability with no benefit to us. I might try to slip it in as part of the technical notes.
Well, in US, that probably is because of several reasons:
1. Lack of central organization, it's easier to work with and pay a single private company or a contractor than hiring folks and working with open source projects that usually has no legal body, no billing, etc. Hiring and working with various companies to update, maintain, and so on with many projects isn't as easy as it sounds. (True, you could hire a single company to maintain a single distro for the whole gov't to use but is it cheaper than a private solution that 80% of the market already uses?).
2. Lobbyists, they'll explain it's cheaper to hire private companies to produce and cheaper maintenance over time especially since ^.
3. A share of citizens that doesn't believe in a bigger federal govt.
4. States want to do it their own ways instead of trusting the fed govt.
89 comments
[ 2.7 ms ] story [ 139 ms ] threadIn other news: 2020 is not the year of the linux desktop.
I understand that it has been in the news that MS spent a lot of energy on preventing people from moving away from Windows, and I think I remember steep discounts but it has been a while since I read about it, but while the government could be going "haha let's put out another statement and get more discounts" I would say it's more likely that there are people in power in Europe that genuinely want to lessen our dependency on the benevolent big brothers from the USA. But that's just how the news reports it, perhaps I'm not cynical enough. Anyway, if you know of a statement by anyone in a relevant government that claims this was a negotiation tactic, that would be interesting.
Which in practice binds absolutely no one to absolutely anything.
The actual migration to Linux went mostly fine and was largely a success as far as government software programs go, FWIW.
To this day eventually this attrition pushes many back to the more compatible and familiar solution.
I've seen (usually very small) companies going to Linux and other FOSS solutions without actually understanding this beyond "they're free". Migration went great. Later on I came back to do the migration the other way around once they faced that attrition and realized that support costs were dwarfing the license cost.
Former actual LiMux team member here, so I can speak from experience (almost a decade ago though). The hardest part was convince users to be happy on utterly inadequate and aged hardware. No, 1 GB of RAM won't cut it when you have to have Firefox, Thunderbird, OpenOffice and some fucking Java crap open at the same time, and 1024x768 screens won't do so either.
The last part alone excludes many cheap vendors as they don't have government contracting experience, and the spare parts requirements kills the rest as they can only be so cheap because they don't have to build much of a spare parts infrastructure. (Note: this is not specifically relating to LiMux in any way, I have no knowledge about the contracts/tender process that happened back then)
Also, the scale matters: 15.500 PCs are even at 500$ apiece 7.75 million €.
It's too gentle to say that users are struggling or resisting.
The fact on the ground is that a significant part of users are simply incapable of learning new computer software, they can't even use effectively the software they've been working on for years. This is glaring in a giant government organization with tens of thousands of employees. Thus any attempt to move OS is doomed to fail.
Classic example is the popup when closing a word document. "Do you want to save the document before exiting? Save/Quit/Cancel". The typical user clicks the little red cross in the top right corner of the dialog.
But that lends the question; what does it take to end the Microsoft monopoly on our governments?
It can’t just be open document standards. I think it goes further than that; into our schools maybe.
I don’t have answers, but it’s a serious question I have.
That should be "offering" singular, not "offerings" plural.
And it's totally feasible to produce a superior ecosystem. The hard part is absorbing years of financial loss while building the new ecosystem from the ground up while it isn't compatible with the existing proprietary system.
Why ? An ecosystem consists of multiple offerings and there are also multiple commercial ecosystems.
> And it's totally feasible to produce a superior ecosystem. The hard part is absorbing years of financial loss while building the new ecosystem from the ground up while it isn't compatible with the existing proprietary system.
Being able to finance something is an inherent part of its feasibility.
While I generally agree that FOSS software (Libreoffice calc) is inferior for business needs, a _lot_ of the issues actually stem from compatibility.
Not just compatibility of the document system and interpretation, though that is a large part. But compatibility of functionality too, along with UI design.
People are more likely to re-train for a new UI on a new edition of their previous software because it is seen to them as a progression, but they are not so open to changing UI paradigms with different software that does the same thing.
In order for something "better" to be accepted it has to not just be better- it has to be _so_ much better that it trumps everything else.
in the very best of cases it has to be 100% compatible in all areas and _still_ at least substantially better.
That's without getting in to opensource financing, which is a difficult topic. Though I can imagine if Munich put their resources into a FOSS foundation which was providing productivity software it would help tremendously. I don't think they're going to do that because Linux (and ecosystem) is seen as cheaper, so why negate the cost saving?
Obviously that would disrupt quite a few existing business models in software industry, so do not take that as a completely serious proposal.
More specifically, if you can find another place which sells the same model for less, then you get it for the other price.
Thing is, they are the only seller of that model. The vendor has a model X123 for that shop, a model X124 for another place, etc. And that shop is the only one who can sell that specific model. So they of course have the lowest price.
Something similar exists for software: that's why there's "home" and "office" and "professional" and "enterprise" and so many other variants.
So, your "home" version only supports 2 connections at a time, your "office" one doesn't implement a certain function that "professionals" are more likely to use, and your "enterprise" defaults to logging. Even better, with software you can have a unique product for each sale.
All different pieces of software, so all different prices.
Then on top of that, you don't sell the software but software + support. And the support costs can vary.
Which means I think it's easy to bypass your hypothetical restriction.
> Thing is, they are the only seller of that model. The vendor has a model X123 for that shop, a model X124 for another place, etc. And that shop is the only one who can sell that specific model. So they of course have the lowest price.
This is endemic in the mattress industry. Products that are identical and manufactured by the same company will be relabelled based on the company that buys them.
That's why your "dream-air luxury liner" could not get a discount because the other company has a "dream-air luxury comfort", which is the same exact product.
There needs to be other vendors who can supply document editing software that is 100% compatible with MS Office. But more than this, there needs to be other vendors who can supply operating systems that are capable of running Windows software.
As long as a single company has the leverage that Microsoft has over the enterprise ecosystem we won't have a free market. Governments can start by mandating that new software deployments have several possible vendors, including at the operating system level.
This is all politics. Nothing more...
Free software and open source doesn't mean non-commercial. SUSE in Germany provides commercial support to their distro for example. Such a weird article, zdnet doesn't getting any better...
Linux also has more market share in Germany than any other western country, I think.
> Die Stadt München unterstützt die Entwicklung von Open Source-Projekten mit einem "Munich Open Source Sabbatical". Professionelle Programmierer*innen, die sich für drei oder sechs Monate ganz auf die Fortentwicklung eines Open Source-Projektes kon-zentrieren möchten, können sich dafür auf ein städtisches bezahltes Stipendium bewerben. Die Projekte müssen einen kommunalen Nutzen haben.
[1]: https://www.gruene-muenchen.de/wp-content/uploads/2020/04/Dr...
it would have to something that covers your current living expenses, no matter how high they are, because unless you are young and single with a lot of flexibility, you are not going to be able to put a hold on all those expenses.
therefore the two reasonable options are: the government continues to pay your current salary, or it pays what you'd get if they'd hire you. the latter would disadvantage high earners, but that is somewhat reasonable.
anything less would unlikely get any people interested.
I'm sure Canonical would happily hier some devs from Munich to make this look local.
I bet tax and pension systems have the same basic requirements cross countries with some local specifics.
if the goal is to create jobs, then that also happens either way. i don't see the problem.
But whether this is a legal issue is actually an open debate, in some specific cases companies could potentially sue Amsterdam for sharing solutions with other governments and thereby "competing" with the market. How a judge would rule is not clear. But this is something they are aware of and it's a risk. It seems to also be connected to the way the project is shared.
As far as I know they are looking into changing the law on the national level so this stops being a problem. Also strangely enough, I've been told, as soon as the project gets funded by the EU, this law is not applicable anymore.
Open source means published on github, not many projects are actually properly documented.
There is no valid reason to just pay rent to anyone for nothing, even if they are a local company. That's not a good enough reason.
Buying software has the benefit that the manufacturer carries the risk.
Imho, they should set standards, APIs, file formats etc to enable competition and keep vendor lock-in to a minimum.
Communal software isn't always custom-built, the majority is normal off-the-shelf stuff, just focused on that niche. My involvement is limited to a small city a few years ago, that may be different in a large city as Munich if they have lots of custom requirements they don't share with other cities.
the buyer can not shift the risk to the seller. how would that go? put out a tender for a completed solution and then wait until someone builds the software on their own risk, hoping to win the tender? that's like a designer competition, and those have been considered unfair.
If the cities built it themselves, they would not be able to choose who to buy from. And if their self-built software sucks, they've wasted money and have a shitty software. One that they may not be able to replace, because there's no longer a market for that kind of software. Public endeavors into building things (airports, train stations, rapid train lines etc) in the recent decades in Germany does not predict a bright future for these kinds of things.
of course software is created. but the risk a company takes in creating is irrelevant to the decision a government makes. that risk is from the past.
you said that the government can either choose to take a risk by developing their own, or offload that risk.
but that is not the decision it needs to make. when it takes an existing product it is not offloading that risk, because the risk that vendor took is independent of the risk of the government.
the situation should be rather like this: either a solution exists, then we use that solution with a small or no risk because the vendor already took that risk in the past. or the solution doesn't exist, then we don't have a choice but to custom develop with all the risks that that includes.
at no point is there an option to offload an existing risk.
the only way to offload an existing risk would be to say: we wait until vendors come up with solutions.
(it may be that i am getting hung up on little details here. i believe that we are actually mostly in agreement in how a government is handling custom development)
opensource system can create much needed transparency in government project and organization effort. In my home country the unemployment agency spent atrocious amount of money to create an unusable job-posting app; something private companies do for fraction of the cost. More surprisingly, they doubled down on it; instead of scrapping it when it wasn't delivering. My guess it that some nepotism/favoritism was at play, and if the project was at least open source it could've been examined by others.
Happy to say that I am working on such a project. Heavy (and unnecessary NDA) so I cannot say in cleartext but it is in use in one Nordic country already, it is opensource and already I hear that the Dutch found it to be a good starting point and probably will be using it too.
Some more details, the company I work for is owned by one country but doesn't compete as much as cooperate - I am a consultant from another company. And we both use open source software and libraries and publish as open source most of what we create.
I hate it that I can't talk about it.
- capacity planning for renewables (hydro power)
- streamlining mass transit
- large scale collection and recycling of bottles
- search and rescue operations in the North Sea
- education
as for why the NDA, my best guess is because "they could". Someone should tell them they are losing out on some great publisity and goodwill.
Which is why I think the NDA (or my reading of it) is way too wide.
https://international.gov-design.com/
To add one other angle: what is cheaper, paying 5 million to a company in Redmond yp provide licenses for opersting systems and officd software, or 10 million to a number of local companies to set up and adapt a slightly less refined open source tool for the local needs? In one scenario the money leaves the local economy, in other the city might spend more money but the money stays local and is used to buy houses or sandwiches, pay handymen or bus tickets or taxes. Where is the long term cost/benefit/ratio higher?
Then in a very short timespan (1) a new conservative city government came in, (2) Microsoft announced a big local office and (3) the new government killed the open source project without warning and anlunced they'd switch everything to windows 10.
This transition back to windows is now still ongoing and as contracts are signed will not really be feasible to stop. The now new-new Green+social democratic government makes noe a pledge to take open source seriously for all future needs (and I suppose switch back to a Linux asap).
https://en.wikipedia.org/wiki/LiMux
I still don't understand why major nations don't create or help create national OSes for their own countries/industries. You would think an OS is national security issue at the very least. A municipality isn't going to cut it. It's going to require a major national effort to stave off a monopoly like microsoft.
https://www.zdnet.com/article/linux-champion-munich-will-swi... , https://www.techrepublic.com/article/end-of-an-open-source-e...
The downside is that much of the 'trunk' of IT (workstations, Business Software, internal systems) are still contracted out to monolithic, old-school contractors who have no incentive to innovate by non-technical procurement teams. This leads to a culture of proprietary software where the established/expensive solutions are seen as the most trustworthy and there isn't any cost to vendor lock-in or non-interoperability of systems until the next tranche of officials have taken over and all of those issues can be written off as 'legacy' and as a sunk cost.
[1] https://en.wikipedia.org/wiki/Government_Digital_Service
Our software licenses are paid through a grant from the federal government. If we replace them with open source software, we don't get to keep the difference. So now I have licensed software like SAS and ArcGIS gathering dust on my machine while I use R and Python.
And sharing software we develop is even harder. I've pushed to release code for my statistical reports. When I said researchers and other states could benefit by reusing parts, my supervisors only saw it as a liability with no benefit to us. I might try to slip it in as part of the technical notes.
1. Lack of central organization, it's easier to work with and pay a single private company or a contractor than hiring folks and working with open source projects that usually has no legal body, no billing, etc. Hiring and working with various companies to update, maintain, and so on with many projects isn't as easy as it sounds. (True, you could hire a single company to maintain a single distro for the whole gov't to use but is it cheaper than a private solution that 80% of the market already uses?).
2. Lobbyists, they'll explain it's cheaper to hire private companies to produce and cheaper maintenance over time especially since ^.
3. A share of citizens that doesn't believe in a bigger federal govt.
4. States want to do it their own ways instead of trusting the fed govt.
5. Distrust in govt in general on all fronts.
This is the same ol' carousel we've been going round and round on now for more than 2 decades.
Some people 'get' free and open software, the other people run our world.
This reminds me of a quote by the character Humphrey Appleby on "Yes, Minister" ("The Office"-like sitcom about government):
"Almost anything can be attacked as a loss of amenity. And almost anything can be defended as not a significant loss of amenity."