One topic dominated the recent 2025 OpenInfra Summit Europe, and it wasn't AI:
Unlike any tech conference I've attended in the last few years, the top issue at the 2025 OpenInfra Summit Europe at the École Polytechnique Paris was not AI. Shocking, I know. Indeed, OpenInfra Foundation general manager Thierry Carrez commented, "Did you notice what I didn't talk about in my keynote? I made no mention of AI." But one issue that did appear -- and would show up over and over again in the keynotes, the halls, and the vendor booths -- was digital sovereignty.
Digital sovereignty is the ability of a country, organization, or individual to control its own digital infrastructure, technologies, data, and online processes without undue external dependency on foreign entities or large technology companies. In other words, Europeans are tired of relying on what they see as increasingly unreliable American companies and the US government.
Carrez explained: "We've seen old alliances between the US and the EU being questioned or leveraged for immediate gains. We have seen the very terms of exchange of goods changing almost every day. And as a response to that, in Europe, we're moving to digital sovereignty." That shift, in turn, means open-source software.
"The world needs sovereign, high-performance and sustainable infrastructure," continued Carrez, "that remains interoperable and secure, while collaborating tightly with AI, containers and trusted execution environments. Open infrastructure allows nations and organizations to maintain control over their applications, their data, and their destiny while benefiting from global collaboration."
Carrez thinks a better word for what Europe wants is not isolation from the US: "What we're really looking for is resilience. What we want for our countries, for our companies, for ourselves, is resilience. Resilience in the face of unforeseen events in a fast-changing world. Open source," he concluded, "allows us to be sovereign without being isolated."
[...] To make life easier for users -- and to turn a profit, naturally -- many European companies are now offering technology programs to help users achieve digital sovereignty. These programs include Deutsche Telekom, with its Open Telekom Cloud, and OVH, STACKIT, and VanillaCore. Each of these companies relies on OpenStack to power its European-based cloud offerings for individuals, companies, and governments. In addition, other European open-source-based tech businesses, such as SUSE and NextCloud, offer digital sovereignty solutions using other programs.
In conversations at the conference, it became clear that while the changes in American government policy have been worrying Europeans, it's not just politics that has them concerned. People are also upset about Microsoft's 365 price increases. Another tech business issue that's unnerved them is Broadcom's acquisition of VMware and its subsequent massive price increases. This has led to a rise in the use of open-source office software, such as LibreOffice, and its web-based brother, Collabora Online, and the migration of VMware customers to OpenStack-based services.
The sovereignty issue is not going to go away. As Carrez said in a press conference: "It's extremely top of mind in the EU right now, it's what everyone is just talking about, and it's what everybody is doing." Open source is essential to this movement. As Mike McDonough, head of software product management for Catchengo, a "sovereign by design" cloud company, said: "No one can lock you up; no one can take it away from you, and if someone decides to fork the code, you can continue adopting it anywhere in the world."
All in all, participants agreed that Europe's sovereign cloud movement is reaching critical mass as governments and enterprises move data back from the US-based hyperscalers. European organizations are realizing they need more private infrastructure capacity and local talent to run big cloud initiatives. So, they're turning to open source because, as Carrez concluded, "what makes us resilient is our open-source community."
Related Stories
The quite famous FOSS developer Poul-Henning Kamp (aka PHK) has posted his feedback to the EU regarding European Open Digital Ecosystems [Intro in Danish, article in English] and their call for evidence. In it he brings their attention to open standards in points 2 and 3:
At the most fundamental level, the EU has three options:
1. Pick and bless a set of winners, consisting of:
a) Operating system, portable to any reasonable computer architecture.
b) Text-processing, suitable for tasks up to a book.
c) Spreadsheet
d) Email client.
e) Web Browser
f) Accounting software, suitable for small organizations.and fund organizations to maintain, develop and support the software for the future as open source, turning that software into infrastructure like water, power and electricity, free for all, individuals, startups and established companies alike, to use and benefit from.
2. Continuously develop/pick, bless and meticulously enforce open standards of interoperability, and then "let the competition loose".
3. Both. By providing a free baseline and de-facto reference implementations for the open standards, "the market" will be free to innovate, improve and compete, but cannot (re)create walled gardens.
Indeed, if the protocols and file formats are not publicly documented, freely available, and royalty-free, then what benefit would there be to implement them, FOSS or not?
There is an unreproducable javascript link on the EC page which goes to a relevant PDF document. It is labeled, "Call for evidence - Ares(2026)69111". It is worth checking before sending in feedback. Although English is the main language, the other official languages of EU member states can be used. The deadline for feedback is 03 February 2026.
Previously:
(2025) Why People Keep Flocking to Linux in 2025 (and It's Not Just to Escape Windows)
(2025) Europe's Plan to Ditch US Tech Giants is Built on Open Source - and It's Gaining Steam
(2025) Euro Techies Call for Sovereign Fund to Escape US Dependency
(2025) Petition on EU Linux Operating System in Public Administrations
(Score: 5, Informative) by VLM on Thursday October 23, @02:11PM (2 children)
I have a lot of experience with OpenStack over the years as an admin of the cluster and as a mere sysadmin with sysadmin rights on the cluster and all I can say is "good luck"
The only upgrade path for the tightly coupled Nova/Cinder/Neutron/Swift/Keystone/Glance was to forklift upgrade, migrate everything off the entire cluster to another cluster, build the new cluster on new versions move everything back. This shows the value of IaaS where you can just run Ansible which runs Heat templates.
This brings up migration. I've had Nova nuke a VM trying to migrate it. That's the only time in all my years of vmware and proxmox that I've lost a VM while migrating.
Another hilarious one: Masakari would nuke HA VMs if it got jammed which it often did. You literally had higher availability if you didn't use Masakari because Masakari caused more damage than it prevented. Again, the only experience in ALL my cluster admin work ever where a HA system destroyed a VM.
Speaking of destruction, again, Openstack Nova is the only system I've ever used where a UPS/power failure (UPS are less reliable than wall power where I live, obviously a red state LOL) would jam up or damage a VM making it unable to restart on power on. Yeah sure fsck is normal, I'm talking about being jammed up at the VM level. Very exciting in an unwanted way.
Most of this stuff is a thin reskin of "real technologies". The best way to run Docker on OpenStack in my experience is to set up a Nova VM of Ubuntu or something and run docker like its real metal instead of a VM. There is Zun but you're better off (in my years of experience with it) just running real docker in a VM. Yes, its cool that Zun is basically like K8S in that imagine it has a driver to talk directly to Cinder/Swift/Neutron but it's generally not worth it. When it doesn't work, good luck figuring it out because all errors and logging is utterly totally opaque. Also the OpenStack "product" for log gathering was uninstallable and didn't work (again, just install an ELK stack in VMs...)
Likewise Glance (Swift?) is just an object store like S3. It actually works well. Its not optional this is where your ISO files for VM installs live. Its a PITA to use and you're better off scripting the loading of ISO install media. From what I remember from 3 years ago Glance is an app for ISO images that stores raw data into Swift the object store. I messed around with Swift, the idea of an object store is nifty, if essentially useless if you can't think of any system-wide applications. It would be useful at the application layer if it wasn't working at the infrastructure level.
I never got into Ironic I was more of a Kolla-Ansible installer guy. Its a BIT complicated to install the moving parts its not like pop in the install ISO usb key for proxmox or vmware and hit enter like twice and everything just works, oh heck no. Installing OpenStack will put hair on your chest or gray hair on your head at least.
I never got into Manila which seems to boil down to a wrapper for a FreeNAS (Well TrueNAS now) vm installer to create a NFS partition. Just skip Manila and install TrueNAS in a VM. About the same deal for Octavia (WTF) and Designate. Designate is a cool idea if you've never used K8S, imagine a cluster wide distributed DNS server (if you've used K8S you don't need to imagine this) its cool and works but a bit tedious to use. Hope you use IaaS tools to automate or enjoy nightmarishly complicated API calls from scripts, its not as simple as "just edit a bind file" LOL. I never got into Barbican IIRC for most of the couple years I was actively admin-ing two OpenStack clusters it was uninstallable although theoretically a feature. This situation was a common occurrence in OpenStack. I never got Freezer to work the entire time I was an OpenStack admin, Freezer was literally uninstallable, yeah I kid you not there was no internal backup system (your wildest dream of one is basically Proxmox's backup server which works and integrates beautifully). Aodh was also like that.
Some of it is WTF. Zaqar: Why? Mistral is like Cronicle, I guess, I don't see the point of installing Mistral if you can install Cronicle. Blazar: Why? Magnum: Why do I need something to get in the way of installing K8S, its easier to just install RKE2 in a VM. Trove: Why, just "docker run" a mysql container like a normal person. Kuryr: This never worked for me and was essentially uninstallable and seems to be what Zun wanted to be when it grows up although Zun actually worked whereas Kuryr did not. Setting up Neutron to work with load balanced failover trunks having multiple VLANs is (or was...) challenging.
Probably the only thing I used that I was impressed with was Heat which is a NIH reimplementation of something like Ansible. I know you guys will laugh but I used Ansible to automate Heat templates because Heat was just SO nice and all the internal tools that Ansible also automated are obviously unavailable in Heat. If you know/get Ansible you'll pick up Heat templates in about half a day. They are genuinely pretty nice.
My advice if you use OpenStack is you better have your IaaS up to date with EVERYTHING scripted and you better have your backups working. Also I have about 1000 billable hours on OpenStack as an admin across a couple years of clock time and OpenStack is EXACTLY like owning a sailboat the two best days of your life are they day you get it (get it installed) and the day you get rid of it (the day you say F it, it would be easier to install Proxmox, so you start installing proxmox and moving workload to the new proxmox cluster)
(Score: 4, Insightful) by krishnoid on Thursday October 23, @08:38PM (1 child)
Considering it's politicians making these decisions, I suspect their considerations run along the lines of "Probably better luck than we'll have with US companies under Trump."
(Score: 5, Insightful) by c0lo on Thursday October 23, @10:53PM
And there is a chance the OpenStack will become less awful under the pressure of the users.
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 4, Informative) by VLM on Thursday October 23, @02:35PM (1 child)
I will continue my OpenStack rant as I'm really happy I don't have an OpenStack cluster to admin anymore (woo hoo proxmox rocks) and the caffeine is hitting hard along with the up half the night on a maintenance window (thank god I'm paid hourly 1099 for life F W-2 employment)
The usual experience with an OS or a product is its smooth to an equal-ish level and internally compatible with itself. OpenStack is not a product, its a meme.
So you think of something cool that you had on VMware that would cost roughly a house mortgage to buy so we ain't using vmware at this employer (and its 10x worse now) and lets see what openstack has.
OH I see there's a project on the web page that sounds just like either an external non-integrated product (lets say, an ELK stack for logging) or a hyper expensive vmware product (lets say, LogInsight (c)(tm) from VMware)
Well OK I will install that on the dev cluster (we had four clusters for obvious reasons). If it works this will be added to the test cluster, which will become the prod someday, and someday after that, the prod will become the "old" as in "oldprod". So thats why we had 4 clusters. Wait there's little to no docs this is turning into more of a struggle than you'd think. WTF I don't think this is even installable. Hmm I research online and the last time this worked was on an OpenStack "release" that hasn't had security patches provided since 2017. OK send a flaming everything to the usual OpenStack forums/mailing lists/irc (did they have a discord/slack? I don't remember) and ask why this is still advertised as a project if it's obviously dead. "Well this is a vanity project by a guy who ragequit five years ago and he likes having it on his resume and we have no process to depricate a dead project... would you like to take the project over?" Naaah fuck you guys I just wasted three days trying to make an abandoned project work. I'll install, I donno, 10 VMs to run a large ELK stack (like 7 ES and a 3 server "Kibana" cluster) and call it good. Then, I kid you not, I fall for it again in a couple months. Oh hey why install K8S using RKE on VMs (back when there was no RKE2 IIRC it was a long time ago), how about this Magnum project. Oh bleep they got me again, see above.
So yeah, OpenStack, no good memories. Some software you have good memories even if it was a PITA sometimes. EMACS. Clojure. But something like OpenStack? The only good memory I have was installing Proxmox on top of those hosts. I vaguely remember shutting down the last server running Keystone before repurposing the hardware and breathing a sigh of relief.
(Score: 2) by FunkyLich on Friday October 24, @11:26PM
There is not much I can add to this OpenStack piece. But more or less it was the same thing for me as well. It felt like things would stop working if you spoke too loud, everything was glued with Python and a myriad of libraries it needed to do what it did. When you wanted to do something 6 months later after setting up finally what seemed to be working, it was like preparing to enter the Colosseum as a gladiator against 4 tigers and and a Cerberus while all the spectators were running away to safety... Everything about it felt like it was not worth the trouble to deal with it, like setting up a Space Shuttle to travel a couple of hundred meters distance and rejoicing that you did it, damned be if it was useful or not.
(Score: 4, Informative) by VLM on Thursday October 23, @03:03PM (1 child)
They need a solution to who controls the DNS system and similar very low level problems, but they're worried about user level stuff like "LibreOffice" hmm.
Next level up is infrastructure like dockerhub.
It is a top to bottom problem not a user application software level problem.
Next problem: The easy way to install windows software is systems like chocolatey. Ooops now you need chocolatey.eu not chocolatey.org. Ooops now the problem is "choco install libreoffice" gets you something last updated in 2018. Well OK maybe if you're a bit of an insider or you google around you'll find out you want "choco install libreoffice-fresh" which was last updated to 25.8.1 on Sept 4th. Of course the latest upstream release branch is 25.8.2 which isn't on chocolatey yet whereas the most recent previous release is 25.2.6 which is available as "choco install libreoffice-still" although official security patch support ends in ... eight days. A bit of being stuck between a rock and a hard place.
At least if you check out https://www.libreoffice.org/imprint [libreoffice.org] you'll see they're based in Berlin so no need to set up libreoffice.eu to replace libreoffice.org.
All of this is rather moot, its 2025, nobody runs stuff locally on easily stolen laptops etc. We going to run libreoffice in the cluster and connect in the browser. So every user at work gets a linuxserver.io docker image of their own on the K8S cluster with an assortment of personal and shared cluster storage using https://docs.linuxserver.io/images/docker-libreoffice/ [linuxserver.io]
1) Oh bleep the latest release is 25.2.5 not old-release 25.2.6 not release 25.8.2 ugh ugh ugh. Well, welcome to Fing 2025 I'm making a Dockerfile to manually install the latest libreoffice to an ... Alpine or Ubuntu linuxserver.io webtop.
2) Oh bleep squared (cubed?) now we need to NIH and replicate linuxserver.io as linuxserver.eu for political reasons. Oh great such fun. Wait is it more EU-friendly to use the Alpine or the Ubuntu webtop docker container? Or "arch-xfce" because ARCH is (was?) Canadian (is Canada aka "New North India" euro enough to be considered EU?) although now Arch is mostly developed in Germany (or not)?
(Score: -1, Offtopic) by Anonymous Coward on Friday October 24, @12:48AM
There there grandad, time for your meds.
(Score: 4, Insightful) by ichthus on Thursday October 23, @03:14PM (10 children)
This is definitely about digital sovereignty -- no longer being held hostage to the likes of Microsoft is a good thing. But, make no mistake, this is also about censorship and surveillance. US tech companies aren't unwilling to put backdoors in their products at the request of the EU. Additionally, companies like X and, more recently in the news, 4chan aren't willing to bow to their censorship whims.
(Score: 4, Touché) by Thexalon on Thursday October 23, @05:27PM
Please explain how backdoors can be more easily inserted into open source software than proprietary software, or be more easily used to introduce censorship against the wishes of their users. I'm not saying it absolutely can't be done, but it would seem like it's harder to pull off than a phone call to a company and a negotiation about the price (either monetary or non-monetary like ignoring regulatory violations), because lots of people could see it happening and treat it as a bug to be fixed, and then have all the tools they'd need to fix the bug.
Also, X still censors stuff and people. It's just not people whose viewpoints you approve of, I suspect.
"Think of how stupid the average person is. Then realize half of 'em are stupider than that." - George Carlin
(Score: 3, Touché) by c0lo on Thursday October 23, @11:03PM (8 children)
[Citation needed]
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by aafcac on Thursday October 23, @11:20PM (1 child)
Trust me bro
(Score: 2) by c0lo on Thursday October 23, @11:30PM
achhh... ah... big joke [youtu.be] :)
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by ichthus on Friday October 24, @12:44AM (5 children)
(Score: 5, Insightful) by c0lo on Friday October 24, @12:56AM (4 children)
Letting aside that "the burden of evidence shall forever stay with the claimer" and in regards with the citation you provided: did you forget or were you never aware that UK Brexited completely from E(uropean) U(nion) in Dec 2020?
Would you mind trying again to provide a citation?
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Informative) by ichthus on Friday October 24, @01:12AM (3 children)
Oh, yes, I forgot about the convention that we all abide by when posting, that we provide links to back up every statement made -- even when the statements are easily [cyberscoop.com] substantiated [bankinfosecurity.com] through a simple search. [google.com]
(Score: 3, Informative) by c0lo on Friday October 24, @01:59AM (2 children)
And you think US is a safer bet (bill introduced, not yet voted on) [congress.gov]?
It's only a matter of time, all the govts will try [theconversation.com] until they'll succeed
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by ichthus on Friday October 24, @02:04AM (1 child)
I don't recall making any comparison with the US. But, you ought to note that the link you provided has nothing to do with encryption backdoors.
(Score: 2) by c0lo on Friday October 24, @02:56AM
But you did mention US companies. Which will be pressured by the US politicians/laws.
EFF says otherwise [eff.org]
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Informative) by pTamok on Thursday October 23, @03:25PM (2 children)
VLMs criticisms are illustrative of the challenge that awaits. Bert Hubert writes at a higher level, and points out that moving from dependence on USA-supplied infrastructure is not going to be easy:
Bert Hubert 2025-07-11: Cloud Overview [berthub.eu]
If you want to know his background and experience, read his intro on https://berthub.eu/ [berthub.eu]
It's going to be a bumpy and exhausting ride.
(Score: 3, Insightful) by VLM on Thursday October 23, @04:54PM (1 child)
I read the guy's "cloud overview". He seems to be a good dude in general although I have three critiques of his overview:
1) He wrote that with the intent of selling cloud. Its a very journalistic idea to determine the conclusion then do the research and writing to support the predetermined conclusion. Was NOT impressed with that. All of his essays seem to be like that.
2) In the short term, cloud is bad for EU. They have WAY too much supply of "computer peeps" and WAY too little demand of "computer jobs" so pay is stunningly low. I vaguely recall a sci fi author named Stross who moved into the famously low paying genre of small time amateur book authors because that pays far more than being a sysadmin in the UK. AFAIK working as a sysadmin outside of London in the UK pays less than working at Panda Express fast food right down the road from my house in the USA. EU needs fewer uni grads and/or more jobs for uni grads, at least WRT "computer stuff". My point is it makes sense for Google to cloud everything up if their two options are pay a guy $600K without cloud or pay a new guy a mere $400K with supply demand pressure from the cloud, and thats OK because the cloud middlemen can make $100K of profit, Google sysadmin dept can save $100K overall, of course the employee gets screwed by making $200K less, but if he's happy with a "mere" $400K/yr thats seen as "OK". This does not extend to EU because then UK sysadmins would go from minimal pay to a large negative pay rate, the social turmoil of annihilating the entire IT industry in the UK is not worth it as a top down policy. Furthermore the UK companies can't afford to pay the greedy inefficient cloud middlemen $100K because they don't even pay their staff that much in total right now, they literally can't afford the higher costs of cloud its outside their economic class. Similar sized companies scratch the back of similar sized companies. What makes sense for giant Ford or giant GM does not linear scale to VW. It might stab a knife in VW and destroy the company to try to emulate Ford. What helps big companies get bigger and further ahead of smaller companies... does not help smaller EU companies... Why pay Azure thousands of dollars per month for a simple managed MSSQL server if you could hire an entire consulting company in Poland for less money and get 10x better service? EU should be competing against the cloud, not trying to be the cloud. They should be trying to be outsourced to, not be outsourced from. Its a supply demand relative salary thing. Cloud is dumb if labor is cheap and middlemen are expensive.
3) In the long term, cloud is bad for EU. They as a continent seem hell bent on having their legacy civilization invaded and replaced by a stronger, growing civilization. In the long run, only half the UK will be literate, that only in Arabic, and the other half will be doing stuff they're doing now like toss the gays off rooftops and similar social views and activities that they presently have. They like rape gangs in the UK, for example, and that will be a big thing in the future in the EU in general. A cloud system is complicated and expensive because of the middlemen. A smaller, shrinking, less competent, far less educated, lower IQ violent and unsafe EU of the future cannot afford a wildly over complicated and expensive cloud strategy that makes sense for hyperinflated high-trust somewhat uni-cultural growing areas like Silicon Valley and not much else. The system-wide solution to problems in the EU is likely to match the systems and processes to the population's demographics and skills, maybe roll back to paper forms, typewriters, that sort of thing. The EU of the future will quite literally be something like the Zimbabwe of today because it will literally be those people, living in the EU, living day to day just as they live today outside the EU. There is no need for a cloud based digitally sovereign AI processor if the primary business process in the EU in, say, 2050, is some gang leader rolls up with ten guys with rifles and machetes and takes what he wants and thats how accounting works in 2050 in the EU. EU of 2050 is not going to be using REST APIs and blockchains if their primary tools and protocols post-demographic replacement involve slashing knives and starvation. Does Haiti in 2025 need a cloud, LOL heck no. They need clean drinking water, sewers, and food and they do not have the ability and competence as a population to provide that, they have no need for "a cloud", that's for other more advanced civilizations. The guy preselected a solution that does not match any of the EU's long term problems.
(Score: 5, Insightful) by pTamok on Thursday October 23, @07:28PM
An interesting proposition. I look forward to reading the argument.
What? I live in Europe, and I don't recognise this characterisation. Where has this come from?
I didn't think the PISA statistics [wikipedia.org] were that bad, and the USA is worse.
No, not at all. No idea what your source is, but it is wrong.
The average educational level of the EU is very high. As for public safety, just contrast the homicide statistics per head of population in the USA compared to anywhere in Europe. As for high trust: the canonical 'high trust' societies are the Nordic countries, in Europe.
You could be right: there is a lot of academic work on the problem of the missing productivity [wikipedia.org] - despite huge investment in IT, the productivity of knowledge workers has scarcely budged.
That's a bold statement, but I don't see any data backing it.
That's a huge 'if' doing a lot of work there.
How does one provide "clean drinking water, sewers, and food" efficiently and effectively? IT systems of some sort might be involved somewhere along the line. Centralised IT has a habit of being used, which is basically 'cloud' (computing on somebody-else's computer)
Great statement, but I see no evidence to back it.
Sorry, I don't think your text supports the initial proposition.
(Score: 0, Disagree) by Anonymous Coward on Thursday October 23, @05:43PM
They will only squeeze harder [politico.com], and EU leaders are soft and mushy and will only collapse under the pressure
(Score: 5, Funny) by krishnoid on Thursday October 23, @05:56PM
I mean, if their open source ecosystem gains Steam, that'll get a lot of the middle-aged and younger generations on board right away. Good strategy.
(Score: 1, Interesting) by Anonymous Coward on Friday October 24, @12:46AM (1 child)
OpenStack sounds a bit too exciting but hopefully they can get it to be boring.
AWS, Azure and other cloud vendors have their own excitements plus the vendor's urgent need to make things more exciting for $$$$$$$$$$.
(Score: 2, Interesting) by pTamok on Friday October 24, @08:16AM
I want an O/S that is capable of lasting longer than mere decades.
If you want a long-lasting foundation, then a dry-stone foundation with a slate cap is pretty good. The problem with most kinds of concrete is that unless you are obsessive about keeping groundwater out, almost all groundwater will leach out the alkalinity. If reinforced with rebar, it'll rust, expand, and crack the foundation, and if not reinforced, it'll crumble into its components.
https://www.amphoraconsulting.co.uk/concrete-edu/detail/concrete-leaching [amphoraconsulting.co.uk]
Journal of Building Engineering, Volume 100, 15 April 2025: Effect of leaching on mechanical properties and durability of concrete: A review [sciencedirect.com]
You can use marine concrete ( https://civilgurukul.com/marine-concrete-a-comprehensive-guide/ [civilgurukul.com] ), but that's expensive.
When people buy structures made of concrete, they often don't realise that the service life of concrete can be substantially shorter than the leasehold they are signing, and definitely shorter than the duration of a freehold.
(Score: 2, Interesting) by cyberthanasis on Saturday October 25, @01:54PM
Disregarding the rants of American FOSS supporters*, who should really be happy if Europe or Australia or China converted to FOSS software, but evidently they are not (see previous comments), I have not seen a push for FOSS software in my EU country.
On the contrary, at the university that I work for, I see is a push for MS Windoze, MS Office, even MS outlook. And a push for Matlab, AutoCAD and even Photoshop.
At the university we had an IT manager who was a FOSS supporter, and used LTSP with Ubuntu for the campus PC-laboratories, not only because it is FOSS, but because it is far more manageable than a Windoze solution. We never had a problem, outside some hardware failures on the client PCs. Unfortunately the manager passed away, and immediately began the pressure to install Windoze. In only one semester, Windoze was installed in all the PCs of the laboratories, along with MS-office and AutoCAD and Matlab. Now, less than half of the PCs work, due to the lack of personnel to do extensive maintenance needed by Windoze and like.
There is even pressure to use MS-Email services, when we have had our own mail server for decades.
When I tried to tell the students to use LibreOffice and that I will not accept reports in docx/xlsx/pptx format (as a matter of principle), I was told to reverse my decision. When I tried to tell them that the students are poor and LibreOffice is free (as in beer – most professors do nor really understand what Free software means or care about it), I was told that the university pays MS for a campus license, but magnanimously I could personally use LibreOffice if I wish so. The same about Matlab and the same about AutoCAD.
So please me count me skeptical about FOSS in EU.
* I have not installed openstack, and certainly I have not admined it, but I am the user of VM built/maintained by openstack at the university. I run a small HTTP service which grades student’s exercises. I never had a problem in 3 years. YMMV.
* I also worry about DNS services in EU, but I think the money is in the user applications, and I think that using FOSS applications (in Linux) is a huge win.
* I strongly believe that the users’ applications should run locally in the users’ computers, and not in in someone else’s computer in the cloud (obviously except web pages and the like). Likewise about the users’ data.