> González Waite said that all of the large proprietary software companies ""are big bullies"". He has been called into the US embassy and been threatened because Mexico was using technology that was not from the US; those threats were dialed back when he explained that the government also used software and services from Amazon, Google, and Microsoft. Various companies use the US government to bully other countries, but they also use license audits as a reaction to projects that move to open-source software. Every time a successful switch happened, ""six months later there was an audit""; having the right legal team helps defend against those tactics, he said.
It matches also what I heard from someone working for the Dutch government. He said that whenever they needed a new software system, that Microsoft would send multiple consultants for "free" which all could "help" the transition to a new service from Microsoft.
tl;dr: LiMux didn't just fail due to politics (although politics did play a large role and I will forever dislike Dieter Reiter for a multitude of reasons, LiMux being among them), it was set up for failure from the beginning, mostly budget related.
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
The issue was propaganda. There are even rumors that Dieter Reiter told someone that he regretted moving away from LiMux after he had to pay so much money for IT.
Depends on how one judges the state of equipment - I remember my private laptop having 4x the RAM than the city's computer... or my school's MuSiN computers running on win2k (!) until at the very least 2012. If that isn't a budget problem, I really really don't know what else could be one.
> There are even rumors that Dieter Reiter told someone that he regretted moving away from LiMux
Reiter regretting something, now that might be a first for him
Anyway nice to see (or rather: read) you again!
and musin is not the city of munich, that's federated governments for you
> Anyway nice to see (or rather: read) you again!
Dito, i hope you are fine :)
Then like a year later they doubled down on MS products (right after a new IT head came in). The IT people I spoke to had no idea why this happened but no one seemed to think going back to MS was a good idea.
Discussed more here https://news.ycombinator.com/item?id=41717607
It seems like Hypernews was only turned off in 2021, much much later than planned, but they did do it.
The experiment SW is a lot like the buildings and labs in every physics department I've been to: everyone is on a separate 3-5 year grant cycle and as such is contractually obliged not to push for newer infrastructure. People stick to the same software/lab until it is discontinued/condemned.
Scientific Linux is discontinued, though. A few experiments went to CentOS and (when that was moved to CentOS Stream) to AlmaLinux. But practically speaking the OS the experiments are using is a RHEL-like base with almost everything important overwritten via LD_LIBRARY_PATH, PATH, etc. and pointing to a fuse-mounted file system called cvmfs
https://github.com/cvmfs/cvmfs
For better or worse this allows O(weekly) releases that change what would normally be core components of the OS.
It's kind of weird how all the interesting stuff at CERN is linux and open source, and then all the IT infrastructure is outdated MS services and Windows.
I was one of the few Linux heads at the time on my group, and took part on the Scientific Linux rollout experiment.
On my group, most papers were either authored on FrameMaker or Word with a LaTeX like template.
The reason is the same as on the last agencies I have worked for since then, IT doesn't want to support Linux hardware, just like OEMs don't want to sell Linux computers (yes there are a few exceptions like System76 and TUXEDO), and most folks that really need it, get by with UNIX on macOS, or running guest VMs on Windows.
The large majority of CERN researchers aren't using Linux directly, and writing code for LHC experiments, hence why.
Within my experiment (which has a few thousand collaborators) every paper is written in LaTeX and has to build in a linux environment to be submitted for publication. All the central software work is done in one of three ways:
- ssh to a cluster running linux, develop there
- work in a linux VM on your machine (usually docker, lima, or other container management products)
- with a linux laptop or desktop
Of course the engineers and hardware developers are often using a whole Windows-dependent stack of FPGA design software, LabView, etc.
They are out to lock you in to ensure sustained cashflow, not solve your problem in the most impartial and cost efficient way for you.
https://m.youtube.com/watch?v=duaYLW7LQvg
Yt-dlp while it's hot. Microsoft sure sends DMCA takedowns after this one.
The only "Linux Desktop" ready desktops and laptops to find at local shopping mall for normies are Android and ChromeOS devices, likewise in out-of-the-box experience for hardware support for peripherals.
"Normies" from the 80s do not represent the dissemination of personal computing we experienced in the last 10-15 years. So far we have one or two generations whose experience with personal computing is limited to downloading apps from app stores and ,at best, check webpages. That is very far from what people used DOS for.
I'm saying that there's a far larger portion of the population using computers, and they all benefit extensively from R&D going into UX design that allow people to use computers without having to "learn".
What kind of dissociation from reality leads anyone to believe your regular joe will even have the motivation to waste their time sitting in front a computer, open a terminal, and type commands? Some software engineers don't even want to touch a computer when they clock out, and you expect others to push themselves to "learn" something the have no interest in?
A big part of being a CIO or CTO is having and maintaining relationships with key suppliers. This is especially true with SaaS/IaaS, where your business is valued based on whatever bullshit churn metrics the company cooked up. Your $2M deal may have way bigger impact on a Sales VP bonus than you think. You have to be a different kind of asshole to maintain control of these guys than in the old software world.
"Economists who have studied the software industry concluded that the value of a software business is about equal to the total costs of its customers switching out to the competition; both are equal to the net present value of future payments from the customers to the software vendor. This means that an incumbent in a maturing market, such as Microsoft with its Office product, can grow faster than the market only if it can find ways to lock in its customers more tightly. There are some ifs and buts that hedge this theory around, but the basic idea is well known to software industry executives. This explains Bill G's comment that `We came at this thinking about music, but then we realized that e-mail and documents were far more interesting domains'."
Also it strikes me that as a developer my investment of learning is more in code d So it’s easier to switch
Quote from here and the idea is from this https://www.researchgate.net/publication/200167344_Informati...
Does anyone know if <https://www.lightbluetouchpaper.org/> is permanently gone or just temporarily down?
License costs are a factor, yes, but they are not the only cost, and in most cases not the significant one.
In some cases the offerings are similar enough that it moves the needle. PostgreSQL for example is a good candidate- Oracle is expensive, and the number of people interacting with it is limited. Plus the fundamentals of Oracle and PostgreSQL are more-or-less the same.
On the other extreme the cost of training and support dwarfs the license cost. If all staff come with knowing how to use say Windows and Excel,but require training and support for say Linux Desktop and Libre Office, then the "free" thing costs more.
It's no accident that OSS has done better on the backend than the front end.
Success for OSS means putting the right product in the right place, taking all things (not just license cost) into account.
(Aside: corruption is a red herring, corrupt officials and companies can be corrupt regardless of software license.)
If you think Microsoft makes it sales because its buyers are putting the right product in the right place, then you haven't seen a lot of Microsoft sales. This is absolutely not how it works. And it doesn't have a lot to do with corruption either.
The people making the decision to buy an IT product are often not its users, and often not that much concerned with making the best short/mid/long term deal in the interest of the business. They are very much concerned in making the best deal for their own careers, and as these are the people who buy the thing some companies have competently specialized in optimizing sales given that fact. Oracle has a reputation for this, and Microsoft as well, but all big tech does it (just some do it better than others). Of course, there's some nuance, you can't get away with it if your product doesn't work.
If the end user says: I have no training in X, that mostly means they don't want to work with X or don't want to accept extra workload related to X. The company then provides training in X, another magic spell meaning money was spent so the company officially did something, and the excuse about not being trained won't work anymore. Blame has now been shifted from the company to the worker or even the end user. So training being expensive has better optics.
Big tools like MSOffice have the ability to be put in hiring contracts: Everyone is assumed to know how to use it, so the company is allowed to deny the no training excuse without spending money.
Actual training for the whole company is more like a networking event, and if you ignore the trainer and read the manual or watch some youtube, you may actually learn what you're supposed to. Once in a while, you get a trainer who knows what they are talking about. But all that is secondary.
This has largely been my experience working in IT as well. I think folks in the tech world take for granted how "normal" continuous learning is for us, and how undesirable it is for others in the workplace. Your average office drone very much does not want to learn something new. They want to use what they always have, and keep the same workload.
Even just attempting a switch from Windows to macOS, when their workflow is almost entirely a web browser, damn near caused a full on worker revolt at one organization I worked at. Training didn't fix it, because the desire to learn wasn't there in the first place.
That's the kind of inertia that needs to be overcome for something like open source adoption at the end-user level. It's comparatively simple for the back-end to transition, but without an already willing user-base, the front-end average office worker is not going to achieve the same level of productivity for a really long time.
I would claim that if you haven't worked in the finance or insurance industry (or some related industry) for quite some time, you very likely don't know how to use Excel (I have a feeling that a similar points holds for Word and Powerpoint with respect to some industries, but I think for these applications this phenomenon is a little bit less pronounced).
Indeed, I'd claim that most books about how to use Excel are simply crap. To just give you a glimpse how to use Excel, here some internet classic on this topic:
You Suck at Excel - Joel Spolsky
https://www.youtube.com/watch?v=JxBg4sMusIg
Most people don’t know much about the OS or Office, so you have to train them anyway for the companies use cases and non standard programs.
MS just seems convenient, but isn’t. In large companies every update brings problems.
Things like Teams change the UI quite often which leads to support questions.
So for me the training costs don’t differ that much.
And why? Because the designers need to prove that they so something?
I don't think this is a realistic assessment of the problem. Windows 11 slremains very much recognizable since the Windows 7 days, and the transition from Windows 10 to Windows 11 was seamless. Moreso with walled gardens like macOS.
In the meantime you can't sneeze in the direction of a mainstream windows manager for Linux without it introducing radical changes, not to mention how all distros are heavily fragmented and sometimes even customized.
In my experience tons of office jobz require you to have 20-30 windows open. And you switch through them constantly.
So this streamlining murders productivity.
You can try to change it withan external program, ut good luck doing that on a corporate computer.
Corruption implies that somebody is making enough money to pay bribes. Therefore, corruption will naturally be to the advantage of those that make the most money from the least quality offering.
In other words "give me this govt contract and I'll route n% of it back to you". n% has to be "reasonable" or there's no commercial point in offering the bribe to begin with.
If there's a cash-flow issue (you gotta pay the bribe before getting the contract) then I guess it favors deep pockets.
In a developing country, you really want to be careful with your dollars/imports. Local labor is almost certainly cheaper and more abundant. The question is, can they do it all. If they can you save dollars and you build local skills that transfer, a double benefit. If they can't, yes you might be forced to buy and import the tech.
The structure of the phrase is [[no brain]er], not [no [brainer]].
See https://en.wiktionary.org/wiki/-er , sense 7/8.
> While open source seems like a "no-brainer", it turns out that governments can be surprisingly resistant to using FOSS for a variety of reasons.
Postgres developers, as the license says, have NO OBLIGATIONS TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS. This means you have to do one of three options:
- hire Postgres experts into every ministry. Not very efficient.
- create a single government agency that provides support to the rest of the government. Might easily lose efficiency, as any other bureaucracy
- create a commercial support provider that has to earn money by selling Postgres support to private enterprises. Again, there's a risk that it will start charging the highest possible price for its services.The whole process took multiple months, reading all contracts I couldget my hands on, answering the phone at 3AM, a lot of patience, and treating Oracle as a student who was 2 weeks late with their homework. After that, all oraclies at the organization were completely awed because I was the first person ever who managed to get a usefull new patch out of oracle's service.
If you want results, get Postgres. If you want someone to blame, get Oracle.
I've twice offered fixes to Oracle (didn't even get a thank you); countless times I've had to find workarounds; and I've lost endless hours working with support to get things fixed.
I have no respect for Oracle software. It gives me nothing but headaches. I can't really just blame Oracle, because I have to keep my organisation working. I just wish I could get rid of it.
It's the someone to blame part that's important in big corporate bureaucracy. It's being able to tell your boss "We've done everything we can, it's in Oracle's hands now" limits your own liability vs. "We need to fix this ourselves, and haven't solved the problem yet."
Have you received reasonable help in reasonable time that way? How big was the investment on your company's side to make that happen?
It was a meme on every DBA team I worked with.
It's not just about fixing the problem, but about protecting your own career. "No one ever got fired for buying IBM".
Now circle back to the main thread - you invest some multi-million budget yearly to have support, but you get just an excuse.
People are expensive, but not "oracle expensive". Plus there are many smaller shops that sell support for opensource databases, I'd you don't want to hire senior people.
It is not bureaucracy vs magical efficiency. There is the same issue in any large organisation, even corporations. In practice I don’t think it would be as bad as you suggest. Having an agency focused on providing support instead of sending money to shareholders also limits other kinds of inefficiencies.
> Again, there's a risk that it will start charging the highest possible price for its services.
That is exactly the situation we’re in, where governments are tied and dependent on single providers (be it Microsoft, Oracle, SAP, or others). This solution would create competition opportunities by opening a market, it does not need to be a monopoly.
So you switch to a different commercial service provider, problem solved. The software is open source, they can't lock you down in a predatory contract. Worst-case scenario, you can always choose to fork it yourself.
I can't comment on how well they perform in practice compared to the conventional monopolised support model that closed-source software tends to offer - perhaps better, perhaps worse - but at least in principle, the solution is there.
[0] https://www.postgresql.org/support/professional_support/
If Oracle support isn't all the way there the alternative is basically to kick rocks
Assuming decent providers exist, yes, that's an advantage.
Larry Ellison, never change. It'll be interesting to see how they ruin TikTok if their bid succeeds.
I do think that open source is slowly growing in traditional enterprises, although I think recent interest in cloud computing and artificial intelligence has pushed a lot of software contracting out of the company. Open source migrations might become harder in the future when the enterprise no longer controls their own databases and models.
At one time, I anticipated the rise of FOSS consortiums. Jurisdictions with similar needs would join together to share the cost and risks.
Canada, Mexico, USA each have 1,000s of juridictions. Surely at any one time there's a handful planning a technology refresh of some domain.
One easy example I know of is property tax administration. There's a bunch of counties of similar size all doing the same thing, but all running off in separate directions. Vendor options are complicated, expensive, and have lock-in. Surely it'd be beneficial to pool their resources and own their stack?
Another is election administration. US counties used to do all it themselves. Candidate filings, voter registration, poll books, yadda, yadda. Now it's all outsourced. Lower service for higher prices. (The "certification" process was captured, serving to protect incumbents. Natch.)
Any way.
I was a grunt for a member of a consortium FOSS project. It was awful. "The Logic of Collective Action" explained a great deal of the pathology. Also, Byran Cantrill's quote (wrt Open Solaris) about "having the freedom but not the power to fork" was spot on for our project.
Any way.
Does any one have examples or game plan or vision for realizing more FOSS in govt? I'm not quite ready to give up on the dream.
This is a dirty trick I ran into in [US] state procurement, in a state known for widespread corruption. It's basically a no-show job where you just hold a bunch of long-term contracts for the state, and claim a monthly "support" fee for doing it. Even worse, you got to negotiate those contracts (and set up your kickbacks or self-dealing.) Bonus points for needing to call the contractor in order to have the contractor call support, and the contractor taking a fee for doing it.
Endless avenues for corruption with a setup like this.
This is the worst kind of graft and should result in criminal charges. The software development industry is still in a nascent stage and our tools are great but professional standards are still undeveloped.
> Technology is often seen as the problem, he said, but he generally found that the problems were due to using obsolete technology and a lack of knowledge about the data being handled. There is often no documentation of the data and its structure, coupled with no understanding of that by the people in charge of it. Poor leadership in the agencies is another barrier; there needs to be a champion for a change of this sort, who understands what needs to be done and properly assigns people to work on it.
Oh. Well. Precisely.
> storing the data three or four times in order to charge more money.
Given that they’re not disputing that it was being stored that many times, then it’s plausible the vendor was using replication as their error recovery strategy which isn’t an invalid choice. Erasure coding is a more difficult alternative to implement. This is too short a sentence with too little detail to draw any actual conclusion from. Heck, maybe the software even had a configuration option but there was no in-house expertise to configure it properly.
Uh... Without knowing the exact details, my first thought that "a little knowledge is a dangerous thing"