France's homegrown open source online office suite
github.com
github.com
I'm fortunate to be collaborating with them as their Docs product is built on top of our open source BlockNote text editor (https://www.blocknotejs.org).
Docs specifically started as an international collaboration with Germany [2] to explore how different EU countries can collaborate in building sovereign workplace solutions (several other countries including NL have shown interest as well).
They're actively supporting us, and related projects like Yjs (https://yjs.dev) by sponsoring feature development.
I'm sure many of the team members will follow along here as well! Happy to answer any questions.
[1] https://news.ycombinator.com/item?id=46873294 [2] https://www.zendis.de/en
Anyone think what they might about La Suite, but blocknote is a solid product!
"A collaborative note taking, wiki and documentation platform that scales. Built with Django and React."
An office suite's 'docs' component is usually a word processor and people sometimes try to (mis)use it for the functions you actually list - i.e., you can try to use Word as a wiki, linking pages somehow, but it's not nearly as efficient as a purpose-built wiki.
Based on the quote description, it looks like your project inverted the thinking: Is word processing not a/the primary function? Are the other functions truly prioritized - e.g., is the wiki somewhat as efficient as MediaWiki?
In addition, there are some advanced integrations with other products in La Suite. For example, video calls made in Visio can be automatically AI-transcribed and presented in a Docs document, etc.
I much prefer Google Docs over word for this reason too.
I was writing a datasheet really and it’s really surprising how there isn’t ia straightforward solution. Confluence wasn’t expressive enough, while getting Word to apply consistent styles across tables, margins, headings etc is such a pain.
“Content over form” so you don’t really need all the formatting options of something like Word when you are just trying to write meeting notes.
They are definitely trending more towards a wiki, but it is still early days for this whole experiment. Though, many of the municipalities in the French gov are using it for their day to day work so it is clearly useful in some capacity. I don’t have numbers, but it’s definitely respectable
Even in this example, the French are building this in-house, but the Germans are repackaging this into their suite. And the Netherlands is on their way to do the same.
So the approach would be different depending on which country you approached.
My advice to you would be to follow government events like Hackdays to get yourself in front of people who can point you in the right direction
- I think administrations in the EU are (slowly but steadily) adopting "Public Money, Public Code" policies and looking more seriously at open source
- Note that policy / strategy on this depends a lot per country / local administration / project etc. I think most governments don't actively develop in house - France is quite the exception in this
- There are a number of conferences that might be relevant (FOSDEM for example just finished)
- We also benefitted from EU grants (e.g.: NLNet) to bootstrap our work and the early research phases
For those who want to know more: https://publiccode.eu
I don't know the current salary ranges,but they offer other values like vacation days, Work-Life-Balance (proper time tracking to avoid extra hours etc), part-time.offera, child care options and some other benefits, which most corporations won't give in addition to being the state, which means they won't go bankrupt, won't do reductions in force in the way companies do it, ...
I could see there being.. biases that make it much harder in practice. Not particular to France, but anywhere in Europe.
My friend might just start brushing up on their French.
From my experience using open-source collaboration groupware like Nextcloud, their solutions written in dynamic programming languages like PHP and Python are always woefully slow. Only thing that got somewhere near of the commercial offering is OwnCloud's Infinity Scale (OCIS) which is written in Go. It is no surprise since OwnCloud is indeed running an open-core business and you cannot use their binaries in businesses. OpenCloud is the "open-source" fork but they are already in legal trouble with OwnCloud due to industrial espionage claims.
If European governments are serious, the amount of money they _guarantee_ should be in the degree of tens of billions of Euros. Not fun 10k hackaton projects. The money should be secured immediately that cannot be touched by the upcoming governments. It should increase taxes. Independence has a price. We as Europeans should be ready to pay it. And yes it will probably cause whatever current party to lose elections, independence has a price. It is high.
True as it may be that they are slow, I doubt it's caused by the use of dynamic programming languages.
> The money should be secured immediately that cannot be touched by the upcoming governments. It should increase taxes. Independence has a price. We as Europeans should be ready to pay it.
You do you, but increasing taxes to build products to replace products built by private enterprise sounds like a 180 degree opposite of what Europe needs to prosper.
Depends which ones. Python? Definitely a source of slowness.
I would first blame the programmers, the design and lack of specialty offloading before blaming any programming language. Well designed web calls scale nearly linearly with usage and usually poor design or programming is the source of slowness. You can always trade language complexity for speed but assuming it is the cause of all perceived slowness is a poor man's view.
It is the same story every time again, first it was java, which has so many large scale projects most people won't even know it's running things they use, now it's apparently python who is to blame for all slowness on the web. When the next JIT or scripted language comes along which is not someone's favourite pet that will get the blame.
I also denounce the notion that trading language complexity for slowness is the case. Python is already complex, and there's some language and frameworks that are actually quite a bit easier to use for web backends. Like java, or dotnet. It just makes no sense to use python for this usecase, even if you ignore the slowness.
But that's not completely true, there is one very good reason to use python. Your devs know it. But, that doesnt say anything about the language itself.
Shhh, don't tell them.
(Kidding, of course.)
The best solution is skin-in-the-game, for-profit enterprise coupled with rigorous antitrust enforcement.
Companies will go a million times faster than open source. They're greedy and will tear the skin off of inefficiencies and eat them for lunch. That's what they do. Let the system of capitalism work for you. It's an optimization algorithm. One of the very best.
But when companies get too big and start starving off competition, that's when you need to declaw them and restore evolutionary pressure. Even lions should have to work hard to hunt, and they should starve and die with old age to keep the ecosystem thriving.
Don't we have enough examples showing that this simply cannot work long-term, because the for-profit enterprises will _inevitably_ grow larger than the government can handle through antitrust? And once they reach that size, they become impossible to rein in. Just look at all the stupid large american corporations who can't be broken up anymore because the corporation has the lobbying power and media budget to make any attempt to enforce antitrust a carrier killer for a politician.
I think it's very myopic to say that corporate structure is the "best solution".
Any example of a politician carrier killed by an attempt to enforce antitrust?
Him putting Lisa Khan in charge of antitrust enraged the tech oligarchs, who then all went MAGA and bought Trump the election.
Didn’t Harris actually raise and spend more than Trump on that election?
And the fact that a 3rd party supports an opponent does not kill any politician's career. Biden retired by himself, following his own party's pressure. And Harris is still around, I believe.
Yes it is. It's the same reason desktop GUI apps are now slower than Windows 95-era apps that were written in C.
This is not a situation where you'll have thousands of people editing the same document, that'd be insane with Django for sure - but at general collaboration tooling with <100 (random number I made up) editing, Django is unlikely gonna be the bottleneck
What part of your document editor needs to be backed by a relational database?
Why use an MVT system if you don't need the Model part of it?
I see, in your broad and experienced mind, document editors don't have users, permissions, and the whole document management itself, comments on lines/threads, reactions on comments
Seriously, theyre all as cookie cutter perfect usecase for Django as you can get, but I guess you haven't actually thought about the domain and just wanted to take a dumb on other devs with intern-to-junior level insights
And in this case, as would be obvious from thinking about it, the only part it's not suited for is the live syncing of the text edit on the frontend, which is one one small part of the whole.
Now go check if they're doing it or you're just suggesting from the dunning kruger symptom.
TLDR: C++, WASM, and some form of GRPC with C++ on the server side as well. Because you need a language that's fast, can contain high complexity and large programs without collapsing (which is a short list of languages) and can work fast for the bits that need speed.
But thanks for answering honestly.
There's new languages where it's a bit of an open question still where they lie on the spectrum. Go would be one of them. I'd guess somewhere between Python and Java. Javascript I would argue is between perl and python. And Rust ... well ... good question.
An office suite is a gigantic application, which will need feature upon feature upon feature upon feature. If you want it working on the web, I'd propose something like C++ and WASM.
Edit: also appears to be based in the eu, how fitting for this thread.
This is all a fact.
And yes, I'm talking about teachers and medics. We don't have enough of either, because we don't pay them enough compared to their workload. Those things we will always need, in great quantities to support our population. Greater quantities than engineers, architects, researchers, etc. but guess where everyone flocks because it pays more?
- https://iris.who.int/bitstream/handle/10665/250330/978924151...
- https://ipsnoticias.net/2022/10/el-mundo-necesita-69-millone...
For instance, when every employer (including those that may be only marginally successful to begin with) is expected as a matter of law to extend onerous labor protections against firing and laying off to each and every worker,[0] this results in marginal workers (who may have been socially marginalized originally for reasons of ethnic heritage and the like) being completely excluded from the market, which makes their plight even worse. (Except for forms of "gig work" or informal employment, of course - which in practice function to sidestep the most onerous regulations to some extent.) A very relevant issue in present-day France.
[0] And to fund those costly welfare programs through payroll contributions that are levied on employees and employers alike - which is its own issue and often amounts to exploitive, confiscatory taxation for the most marginal workers.
The last sentence is doing a lot of heavy lifting for a post that collapses if poked gently with a stick.
"The welfare state cannot exist without exponential population growth."
Sounds mathy, but is wrong. Welfare states do not require exponential population growth. They require a sufficient ratio of contributors to dependents, plus productivity. Those are not the same thing.
Exponential curves + limited resources = ecological faceplant. No serious economist argues that infinite demographic growth is a prerequisite for social insurance. What they talk about instead are levers: labor participation, productivity, retirement age, automation, taxation structure, and yes, migration.
"Government policy makes 4+ child families rare."
Prosperity itself lowers fertility. Governments can nudge at the margins, but they are not mind-controlling people out of large families. Most people stop at one or two kids because time, money, energy, housing, and sanity are finite.
"Mass immigration of uneducated people doesn’t cover the gap."
Ah, bundling multiple claims into a single blur. Efficient, but sloppy. Refugees are not permanently "uneducated"; education and skills are state-dependent, not genetic properties. (Except if you are one of those right-wing grifters that think only white people are capable of intelligence, and maybe east asians. Those people get a hearty fuck you from me, that is not worth discussing at all). Early years cost money; later years often don’t. But you know what, the same is true for children.
Fourth argument: "Extending welfare to immigrants makes it worse."
This assumes welfare is a static pot rather than a system designed to convert non-participants into participants. Welfare states don’t exist just to reward contributors; they exist to stabilize societies over time. Cutting people off doesn’t magically turn them into productive workers. Quite often it does the opposite.
Now, let's zoom out a bit for the real category error here. Modern welfare system are intergenerational risk-sharing mechanisms, not growth cults.
"This is all a fact."
Sure thing buddy
This is not true. Women entering the workforce instead of having babies earlier in life lowers prosperity. In our society women working during those early years creates more prosperity (two incomes) but those who are very rich like Musk has no issue producing a big stable of kids.
But why import uneducated immigrants when you could import educated ones instead? The Canadian model has been a resounding success on that front and European countries should copy it. (And no, the "brain drain" argument doesn't really hold water. The successful migrants/expats tend to go back to their homelands after a while and become a much needed force for progress there, if there's even the slightest scope for actual improvement.)
Refugees are not "imported." They are people fleeing war, persecution, or state collapse under international law obligations that Europe helped write. You don't get to say "we'll take the engineers, but not the bombed-out schoolteachers." Treating asylum like a points-based talent visa is a category error, not a policy preference.
The brain drain argument absolutely does hold water. Systematically pulling scarce doctors, engineers, and academics out of low-income or fragile states weakens those societies. Some people return and contribute, yes, but many don't, and many return to systems too damaged to absorb their skills. That's not controversial. It's well documented in development economics.
What's being presented as "common sense" here is really a value judgement: that human worth should be ranked by immediate economic utility to the receiving country. That's not a fact, and it's not how real migration systems actually work.
If the goal is serious policy discussion, collapsing refugees, migrants, education, and prosperity into a single slogan doesn't get you there. It just makes the world simpler than it is.
One more point about the word "import," because language matters in how we think about policy.
Describing people as being "imported" frames migration as a centrally planned, top-down process, rather than as a response to war, persecution, economic collapse, or climate pressure. It shifts attention away from those underlying causes and toward the idea that governments are deliberately "bringing people in" as if they were interchangeable inputs.
That framing makes it easier to talk about migrants in abstract, instrumental terms, sorted by usefulness rather than understood as people reacting to circumstances, and it tends to oversimplify how migration actually works in practice, which is far more reactive and constrained than intentional or engineered.
Being precise about language helps keep the discussion grounded in reality rather than drifting into metaphors that flatten complex human movement into something it isn't.
Want to admit more refugees without endangering social cohesion? Then you should make sure that you're also carefully selecting your economic migrants as best you can. It's not a matter of assigning different human worth to each, but of simultaneously abiding by legal obligations towards actual refugees that are binding for the country, and also trying to do the absolute best you can for the highest amount of people who might be wanting to expatriate to it for different but nonetheless valid reasons - without unduly burdening that country and society in the process.
I'm curios what do you mean by this. Could you provide some examples of such policies or propaganda campaigns?
But when you think about it, their survival depends on it, so it makes perfect sense. Most of those making those arguments have cushy bullshit jobs, completely dependent on stealing the work of others to live. Funnily enough, you would pay them to do nothing; it would be preferable for society because it would cost less money, and they wouldn't be able to create insane bureaucracy to satisfy their power trip.
But it doesn't matter; reality has a way to always catch up and expose the liars. The system is clearly unsustainable, and enemies have been probing for weakness for a while now. It's unclear how long we have left until a full-strength attack happens but it seems hard to avoid now.
Same with welfare state. Which countries do you count as non-welfare states? And when do they stop being a welfare state? Let’s take Poland as an example. When do they stop being a welfare state? If they lower the unemployment payments, will they stop being a welfare state?
And at what timescale do you think Poland will stop existing because of demographics?
The welfare state for corporate interests is alive and well though, and costs much more.
(2025) "Corporate Welfare in the Federal Budget" -- https://www.cato.org/policy-analysis/corporate-welfare-feder...
(2024) https://corpgov.law.harvard.edu/2024/07/16/100-years-of-risi...
> There is nothing to disagree about. With current demographics projection no amount of taxes can cover welfare states
Okay? Let's get rid of that much more expensive type of welfare then!
As if we have "real capitalism" - not even on a scale of local bakeries any more. Even the small businesses often are just a shop owned by a corporation. Not that I'm against some level of concentration, a lot of economic activity requires it. A lot of products are too expensive and require a certain scale to be viable at all.
What is the goal of economic activity anyway? For the few to live well, while the majority struggles? By "struggle" I don't mean that the majority already lives in the streets, to me it is enough that they have to be afraid. Of getting sick, of losing the job, of anything bad happening. I saw myself how a single unfortunate event could spiral out of control, and a guy making a lot of money in enterprise sales ended up alone, broken, and sick in the streets. I count all those having to fear such a development as part of the "losers", even if they are still making money and living in their house now. That fear, suppressed or not, should not be necessary, and it influences stress levels and decisions, consciously or not.
I mean, you are also right with your message, and I actually agree.
The flow of money around and away from too many people should not be happening. Being part of the economy should be easy for the majority, and real "welfare" should only be necessary for the sick and otherwise temporarily or fully disabled.
If a lot of normal people need welfare, something is not right.
But then you need an economy that provides those easy options to participate and get enough of a share.
You also need a system where an unfortunate event (or some) does not put you into an unescapable downward spiral, and provide a way back into the economy.
Why are you keep repeating this myth?
The only relevant player who might break up oligarchies before they become to powerful is the state they operate in.
Why do you write like someone's crazy uncle on Facebook? The caps are inappropriate here, as is the hysteria and hyperbole.
> Oligarchs buy up competition
You realise that those advocating for free markets are against oligarchy, right? That they say that in most, if not all cases, regulatory capture and monopolies are the causes of lack of competition, right?
Even California billionaires would rather leave the state than pay the 5% wealth tax. All to provide “services” that are generally superfluous or tied to corrupt kickbacks.
They do not need to be linked, they generally aren't, in the EU at least.
The real bottleneck lies on the database side, but it is rare for an average organization to actually hit its limits. Don't think at Microsoft scale if you aren't them.
so likely a decade or more of double spending in the meanwhile.
that's 2 election terms in France for context. Good luck making the political parties agree to this.
When any random company makes a Build vs. Buy decision the question is “is this a core competency?” Most companies use a package from MS or GOOG because it’s unlikely that they’ll be so good at productivity software that it’s (A) worth distracting themselves from their actual job and (B) good enough. The same caveats apply here.
>Microsoft Used China-Based Engineers to Support Product Recently Hacked by China
https://www.propublica.org/article/microsoft-sharepoint-hack...
Open source alternatives, audited by international teams could be much more trust worthy than closed sourced black boxes.
I suppose the assumption is that every HN reader follows headline hacking news, which is valid but still I did a double-take at the comment.
Android has a higher bounty than Iphone. Anyway, sorry about your easily hackable phone. Hope you arent a VIP.
Minor powers still bend to the great powers.
People always want more and it will never be finished.
My personal setup is Emacs/org-mode, using babel for the rest; I use Python with Polars, Plotly, and very occasionally SymPy just to avoid using Maxima if I'm already in Python. I see no reason at all to use LibreOffice, MS Office, or anything similar. This is what's actually needed. Billions should be invested in IT training, not in copies of software from another era designed to let untrained secretarial staff use a desktop.
I used to use it all day every day and now i use it once a year maybe (often for government related things, coz theyre often the only ones still asking me to fill out and sign PDF forms).
Most office functions are better supplanted with a decent cms, spreadsheet, email and something to let you create forms for people to fill in.
- the logic is hidden; until you click on a cell, you don't know "what it actually contains"
- there is no entry point, no main(), so there is no way to read it other than keeping 100% of what the sheet does in your head, or ignoring parts of it and risking breaking them while working on others
- the logic tends to be coded in single lines rather than multi-line with proper indentation, which makes reading it very difficult
This is just the conceptual basis, without even counting the improper auto-formatting of cells that has even led to renaming genes to prevent them from being considered dates https://www.theverge.com/2020/8/6/21355674/human-genes-renam... or the absurdities regarding date calculations.
They can be convenient when you have to play around with a few pieces of tabular data, yes, but the price you pay is much higher than working with high-level languages that would be easy for even the average user to understand if only they had started with them.
[1] https://en.wikipedia.org/wiki/Productivity_software#Office_s...
Docs and Drive aren't the only products in this suite: they also provide alternative for Meet, Chat, GMail or Sheets. I have no doubt that Microsoft and Google products offer more features but my point still stands: a lot of employees (like myself) need productivity tools but only need the core features.
[1] https://lasuite.numerique.gouv.fr/produits/fichiers
[2] https://en.wikipedia.org/wiki/OpenDesk
[3] https://github.com/suitenumerique/drive/blob/46c9730d1b6d5c4...
Even though many formal processes still require printable PDFs, we are slowly migrating to something paperless, or at least not paper-centric.
Google docs is still great when you need to make something you mean to print, it just tends to not be that often anymore.
I even use markdown shortcuts to format in google docs nowadays.
Not being Microsoft Office®-compatible does not make something not an office suite. In that case, there is (by design) only one Office® suite in the world
> not a wiki/markdown editor
I was wondering if you meant WYSIWYG editing as opposed to markdown editing, but then you say
> La Suite Docs seems a product more similar to Atlassian Confluence
which is WYSIWYG (the best web-based wysiwyg editor I've ever used, in fact; even if I'd never choose it for being a vendor lock-in that has shown they want to own your data by removing the self-hosted options, maybe with exceptions for giant enterprises idk but at least we had to migrate and it wasn't fun)
so then what are you saying? What makes an 'office suite' an office suite to you?
In the last 10 years I've been spending much more time at the office consulting and editing confluence and web pages (sharepoint / mkdocs / readme and other markdown based resources) than the cumulative time spent on word, excel, powerpoint and pdf documents. I imagine it is the same for a significant portion of the population.
Also, libreoffice is already a thing and nobody edits office365 documents using the web versions except when their employer can't/don't want to pay the license for the full version or the client is not vailable on their OS (linux users). Libreoffice doesn't have that problem, you only really need storage with sharing facilities, not featurefull web clients for your docs.
whatever doesn't map 1:1, imo just trash it.
if you can't do your work using markdown, you should be fired.
if i'm downvoted it is by people who deserve to be fired.
If not, they can adapt.
Doing layout is not easy. Programming layouts well requires real expertise, which is why most layout engines expose a gui and let you deal with larger text components graphically. Maybe someone else will come up with a usability innovation here, but I'm not aware of it, and markdown certainly doesn't have that capability.
There are plenty of UI editors where they give you the basics for formatting and inserting stuff.
Markdown is fairly portable, and with AI it is easy to generate and share as well.
- a word processor - a spreadsheet application - presentation software
This doesn't look like it has any of these
- chat
- video calls
- notion
Might now be more important to an office than word processing or presentation software?
- Word processor: https://github.com/suitenumerique/docs
- Spreadsheet: https://github.com/suitenumerique/calc
All of this goes out the window when you're dealing with a government bureaucracy that has hyper specific document formatting requirements.
This is a real foundational need of nearly every business at some point. Every court system and government agency has their own rules and they need to be tracked and followed perfectly. There are whole sub-industries around dealing with this for legal documents in MS Word.
By a "decent layout" engine, you'd like the ability to change fonts, add spacing between paragraphs, segment the page into regions, e.g. by changing the number of columns on a page, insert images and diagrams and choose how text wraps around them, create captions for embedded media, set page numbering policies (or navigational policies for web generation), generate table of contents, set table headers, make more complex tables with merged cells and different types of boundaries, generate table of references, generate tabs, then export to a web page, to a pdf, or to whatever other format you want, hopefully all from the same source. Then when you send it out for review, people can attach their comments to portions of the document, you can accept or reject their changes, there is a document revision history, etc.
So for example, you could write up a quarterly report by importing summary financial data into your spreadsheet, doing some basic analysis, export tables to your word processor, generate some bar charts or graphs, draw some boxes and arrows with your drawing program, stick that into your word processor, add some footnotes and hyperlinks, put the same info into a slide presentation. Then if you want, you could save that as a pdf or turn into a webpage, etc.
One time I had to write up documentation for various security certifications, and that was my introduction to the world of Microsoft office. Learning office made the project more fun, I had never used it before and was impressed with the functionality.
That said, I don't think most people who have office suites actually need a full powered office suite. Probably markdown + slack is enough for a lot of people, but I always like a good drawing or diagram.
Most people I know have shifted to google suite, it really has everything you need and is OS independent. But apart from the convenience of being browser-based, it is just another MS Office clone. The google drawing functionality is very basic, I often wish you could integrate something like excalidraw into it.
I am not suggesting this project needs to keep adding clones of MS Office functionality until it turns into Libre Office, but funded by the French government. That would be a waste. Instead, why not make it better? Reimagine it? Look, for example, at excalidraw. It's fantastic. It was really a fresh take on the stuff people were doing with visio.
I think there is lots of room to make a truly modern office suite instead of another MS Office rehash, so I would encourage this project to go its own way and do something interesting if they intend to be a replacement for proprietary office suites like google suite or MS Office.
But after some thought, I feel a cloud collaboration suite makes more sense as big orgs often run on online-first solutions like Sharepoint. So they can tick the essential boxes by being an online collaboration suite, and fill in formatting features later.
Though your points on speed and architecture do make me wonder if Python was their best choice...
LibreOffice has a cloud version :). From what they presented at T-Dose like 10 years ago, it's basically an instance of the software running on the server, cut up into tiles and displayed on a webpage as zoomable image using Leafletjs, the same way that google maps worked before switching to vector graphics 15 years ago. Clicks and other input events are presumably emulated on the server and the resulting display update is sent back to the client, a bit like VNC but using a map library
"The BND stated that these expenditures are being used not only to sustain the ongoing war in Ukraine but also to expand Russia’s sub-threshold operations against countries on NATO’s eastern flank."
https://defence24.com/armed-forces/half-of-the-defence-budge...
good for the french, they made the right choice.
> With Docs our job is not to replace Microsoft Office
https://docs.numerique.gouv.fr/docs/ed2e1dbf-07a2-43bb-ae1e-...
Also: like when switching from AWS to EU provider, the goal is not feature parity. Not only it is costly to implement, but also a reason why so many features are in AWS or Office is to ensure vendor lock-in due to feature comparisons.
Learning to do more with less is a feature, not a bug.
Can you expand on this or source this? I'm quite interested in OpenCloud, and haven't heard anything about this. I searched for a few keywords (espionage, legal, lawsuit), which only lands your comment on top.
https://github.com/orgs/opencloud-eu/discussions/262#discuss...
They seem to avoid openly discussing and comparing products to avoid further action. Apparently some of the former members of OwnCloud have switched to Heinlein (the maker of OpenCloud) and Kiteworks isn't happy about this.
The closest thing to an alternative office suite from an European company is Proton, and even that is barely a replacement.
?? this is not true, please provide a source
I always laugh my ass off when people cry about Microsoft and Office. Well, the thing is that there are no real competitors, and we can't blame them if everyone else is more incompetent than they are.
Apple has been pretending to work on an alternative for years, and it is still nowhere near as powerful/good.
But we live in a feminized world, where it is profitable to virtue signal by siding with the pretend victims even though they are not any better than the winners and, in fact, just as bad, as Apple routinely demonstrates.
It is a French organization that offers plenty of alternatives to Google and other big tech products. A lot of them are just rebranded and hosted open source software, but they also develop their own, such as PeerTube and Framaprout (the last one is a joke, but PeerTube isn't).
Where they can make a difference is for fellow organizations and maybe small companies. A lot of them go to Google because that's the most convenient, even if it sometimes against their principles, they are proposing an alternative.
One minor criticism I have is that while they are not hiding the fact that they are rebranding off-the-shelf free software, they could give them a bit more visibility, should users want to self host at some point.
I’ve compiled a bunch of answers in an FAQ on this doc https://docs.numerique.gouv.fr/docs/ed2e1dbf-07a2-43bb-ae1e-...
Cheers!
Love the fact that your AI customer service bot takes a question and then asks for an email address for a reply.
Love your incoherently bolded statements: "The Docs app is a note taking and knowledge management software.
Love how the vast amount of emojis really clarify things and help the reader.
s/love/hate/g
Though they also seem to be on github https://github.com/cryptpad/cryptpad
- an office suite, where La Suite is at least partly a coherent package bundling existing software which has documents, chat, video calls, etc but wouldn't really play the role of an office suite IIUC - they serve different purposes mostly
- E2EE, which comes with its unique set of benefits and drawbacks
(and yes, sadly at XWiki SAS we host our code on GitHub too, I wish it wasn't like this)
- Germany‘s OpenDesk: https://www.opendesk.eu/en
- Netherland‘s MijnBureau: https://minbzk.github.io/mijn-bureau-infra/
I cannot send messages to people on another homeserver (such as the obscure matrix.org one) whereas other people can, as well as some other issues which I forgot by now. Not at all usable. It was a short-lived stint and I didn't even try to enable encryption this time :(
Groups work btw, it's just private chats that cannot be initiated from either side (I have like five pending 'rooms' or something now), so it's not that one server banned the other. I just took it for typical matrix experience
Given JD Vance obviously hates UK/EU way more than Trump, and he may be next US president, he may in fact threaten Microsoft to do it against UK and EU.
On the other, we shouldn’t take the opinions of the sort of fan who hangs around on a corporation’s subreddit too seriously.
In that world, France betting on Microsoft is not only benign, it’s a positive. That’s also the world of Davos and Jeffrey Epstein.
We’re experiencing a global shift toward nationalism which has pushed back hard on that trend. There’s things to like about that and things to dislike, but those things differ wildly depending on your politics.
The title should be changed.
*Lieu de Rencontre Français Pour le Contrôle de Version
When MS pulls services you are largely screwed.
When GitHub pulls services its a few hours downtime and a new provider.
The word is not ironic it is pragmatic.
Literally the "gotcha" fool from the Matt Bors cartoon: https://thenib.com/mister-gotcha/
That does not sound very sovereign by the US to me.
You can take some open source software made in some other country and use it or fork it no strings attached to its country of origin. No leader from that country can decide to abruptly cut you off your usage of the software because they feel like it.
Don't let perfect be the enemy of good.
Terms used like; “European hyperscale cloud” “Sovereign infrastructure” “Strategic autonomy” “European data centers for critical workloads”
Which ended up in various efforts and projects
Digital Europe Programme, Recovery and Resilience Facility, IPCE
(I am not deeply familiar with EU projects)
I believe funding was around low hundreds of millions (€) total
To build one hyperscaler region might cost around €10 billion.
The second problem is that systems that were suggested out of it still relied on US software stack, US computers, etc.
It is not like the EU member states could not fund it, some estimates say aggregated EU and member states have spent €350 billion in Ukraine.
That is not to say they should not do that, nor to suggest you have to chose one or the other but it is demonstration that EU+Member states can fund massive efforts, If deemed important enough.
and EU+Memberstates so far have not felt an urgency or will to really invest in digital sovereignty.
API & Third-party Integrations: Does La Suite expose APIs that allow integration with external tools and services?
What's the current roadmap for pre-built connectors or integrations with commonly-used government systems?
Cross-Tool Workflows: Beyond the "same interface" access, how seamlessly can users automate workflows across different La Suite tools?
For example, can a Visio meeting recording automatically trigger document creation in Docs, or data collection in Grist?
Does La Suite support webhooks, automation rules, or workflow orchestration (IFTTT-style logic) to reduce manual repetitive tasks?
Identity & Access Management Integration: Beyond ProConnect, can La Suite integrate with existing government LDAP/Active Directory systems for organizations with different identity providers?
Data Synchronization: Are there automated sync capabilities between Grist databases and other data sources, or between Fichiers and external file storage systems?
Export Format Coverage: The site mentions reversibility in standard formats (.ppt, .xls, .odt). Does this apply equally to all tools?
Specifically, what export formats are supported from Docs, Grist, and other collaborative tools?
Import Capabilities: Can users import content from competing tools?
For instance, can users migrate Grist tables from Excel with full formula preservation, or import documents from other collaborative platforms?
When collaborating in Docs or Grist, can users work with non-native file formats (e.g., editing .docx files directly without conversion loss)?
Metadata Preservation: Does La Suite preserve document metadata, formatting, comments, and revision history when exporting and re-importing files?
Interoperability with Open Standards: Beyond exporting to common formats, does La Suite use open file standards internally (e.g., ODF for documents, OpenDocument Spreadsheet for Grist)?
Lock-in Prevention: Are there documented procedures and guarantees for bulk data export in case an organization decides to migrate away from La Suite?
You (at least I) would not think of France as having a good Open Source presence, but they do. Over the years I have heard of many good Open Source Projects coming out of France.
I sometimes wonder if it is because of French vs English Language were you hardly hear of their projects in English speaking Countries.
https://en.wikipedia.org/wiki/OCaml
https://en.wikipedia.org/wiki/Alain_Colmerauer
https://en.wikipedia.org/wiki/French_Institute_for_Research_... (INRIA)
Also Docker.
Also Lichess.
Also Scikit Learn.
France has always been super heavy on open source. They even used to host Les Trophées du Libre, international open-source software competition. FramaSoft (i.e. PeerTube) and VLC are also French.
You have not been paying attention.
They thought they could support trump because they were upset with the democrats policies on crypto and AI cautiousness
But instead they got someone willing to break the world order and our alliances which will harm tech growth
Of course its SDK has components choosen with care to maximize alternative (present and future) availability and its code is not stored on microsoft github.com.
It's neither pro or anti business. This or "creating a more business friendly environment" policies is a false dichotomy. That could be done too via other means. It is unrelated. Speaking about this "business friendly" only is either misdirection or myopic.
If the aim is indeed sovereignty, data and software (and this is software not data), and in general, then they need an effective and comprehensive plan. I think taxpayer-funded state-developed open-source software brings very little at a high cost and can even be counter-productive. Frankly I think it is apolitical move internal to the French state to keep the gavy train coming to government agencies.
Rather I think the US, and also China that does it even more, are much more effective at this by throwing money at the marketplace to develop a whole ecosystem competitively that can also compete globally. An important thing to note here is that EU rules prevent a lot of state action (for instance they would not be allowed to buy only French cars or do things seen as direct subsidies, etc)
France will continue to fall further behind unless it really gets it act together, which is unlikely TBH.
You're not engaging with the sovereignty aspect - "compete globally" isn't the main goal at all as I said above, you're just restating you misconception. And so this part comes across as pure projection:
> You seem to be arguing for the sake of argument while avoiding the substance of my point
Yes in the most basic sense it does since they build their own tool instead of getting a foreign one. To be a little bit provocative I could say that Warsaw Pact countries used to do the same and built plenty of uncompetitive products themselves...
But beyond that it isn't a plan because it does not scale, it does not help the country develop its own industry and economy, it does not help competitiveness, and it is a huge cost for very little. Again, the sovereignty aspect means all of this must be addressed otherwise it is just a stunt and waste of taxpayers' money.
You've got to have a competitive industry to achieve and maintain 'sovereignty' in a broader and positive sense otherwise you end up like the Warsaw Pact or China before it realised that. You might survive but look more and more like North Korea (poor and obsolete but, yes, sovereign).
So yes, taxpayer-funded and state-developed internally in isolation for the sake of it is pretty much universally bad.
Taxpayer-funded is not bad per se, as already said in my previous comment, but here it is indeed more than that, it's the government building random stuff internally for frankly no good reason. Maybe next the government will manufacture its own 'sovereign' pencils as well?
They could have spent the same amount of money supporting small companies to develop similar products and that would have helped creating a competitive 'sovereign' ecosystem and commercial products to sell to everyone at home and also abroad. Much more bang for their buck and long term virtuous circle.
So, again have you got a point to discuss beyond just wanting to argue against me?
Edit:
An example of why competitiveness is important: Arianespace. It's great, Europe has the 'sovereign' ability to launch satelittes. That's useful for government agencies, a niche use-case. But everyone else in Europe who wants to launch a satelitte uses SpaceX because Arianespace is not competitive and is obsolete at this point.
"It's not as plan because it doesn't.." (describes thing that isn't a goal of the plan). This is nonsensical. Not even wrong. This is the entire point.
> Again, the sovereignty aspect means all of this (develop its own industry and economy) must be addressed
No, you simply are not understanding "data sovereignty and security" at all.
> a huge cost for very little.
Citation needed. You assume the costs are "huge" and I am not convinced at all - as others have noted there are existing open-source libraries to underpin this now. Likewise the small size of benefits is your unfounded assumption. Security benefits are not measured in money on a stock market.
Many government functions such as standing armies in peacetime are actual "huge costs". But they have benefits not measured in money on a stock market. It's an incorrect framing in that case too.
You're just repeating misconceptions.
Don’t get me wrong…something is hard…I still use Microsoft Word because I feel like I have to. But what is keeping the industry from building a word processor that doesn’t suck and is capable of interfacing with .docx files?
Mimicking Excel - woof. This one is used by so many people in different ways, that unless you offer 1:1 bug compatibility, it would be challenging to get 100% of people to meet everyone's current use case.
Seems like a common approach is something like using yjs for sync with a temporary LSM storage like rocksdb for updates and then periodically snapshot to postgres for full text search and compaction.
Given that the only step necessary to host git on the internet is making port 22 publicly accessible, I fail to see why so many projects are hosted on this malware site...
I'm a very light user and only moved to onlyoffice because it was freezing[0] on my then new laptop, but at least on mac, I feel like it needs a UI refresh, icons that are not blurry, a look at the performance when doing basic tasks, etc.
It's free and opensource, which is good, but it's not as polished as other paid alternatives.
This is very surprising to me considering some of the largest sites in the world are built on Django. Instagram, Pinterest, for instance. Large parts of stripe and Robinhood are implemented with Django. Eventbrite, bitbucket. I believe even Sentry is.
All commercial products.
Stripe has always been a Ruby-heavy company.
It’s a pretty large app (I think >1M lines of Python). I ran a self hosted version for several years and found it performant and pretty easy to keep running and updated.
It's great that Django's API design allowed them to move this way easily, but they aren't actually using Django in the traditional sense because it can't handle their scale.
I've found that with the Django ORM and DRF especially, it's very easy to create a poorly performing app by following the established patterns (N+1 queries being a huge problem created by DRF serializers). You need to be extremely diligent to create something performant in this ecosystem. Not every dev team has Armin Ronacher :P
Where I work we found this exhausting, and moved on to FastAPI and ASP.NET. We make our queries much more explicit using tools like Dapper, and now a senior engineer can have a much better idea how a particular route will perform just by reading the code (obviously, we still do some profiling).
[1]: https://djangochat.com/episodes/django-instagram-carl-meyer
Also, I doubt solving Instagram-level scale issues is on the top ten list of concerns for this project. Just getting something out there and gaining users is way more important than solving far future scaling issues.
I’m afraid i am one of those people :)
It has also been acquired by Anthropic recently.
Does not look like a great choice.
Honestly, at that point, it's indistinguishable from LLM slop
I'd have thought that Vue or Svelte would be a slam dunk choice. Do project managers love bloat and lag or something?
It has a weird learning curve, where you can ship something somewhat working, fairly fast, but to write it properly, with no bugs, you need to understand a lot of niche React-specific things, and their solutions (and those solutions are never useEffect https://react.dev/learn/you-might-not-need-an-effect).
At that point, I wouldn't recommend it to anyone who isn't already experienced with React. It's been an uphill battle, trying to work with anyone that is using React, without understanding how to write properly.
Previously:
This week: https://news.ycombinator.com/item?id=46873294
2 weeks ago: https://news.ycombinator.com/item?id=46767668
But, what would be your stack of choice? Or, what stack gives you the most confidence?
Also means that the tooling to make collaborative work in this suite possible already exists because it's a common use case on the web and less so on native software (see Microsoft Office vs. Microsoft 365 online).
These are the same reasons Google Docs took off, and they are real advantages.
That one explivitly cites the ICC judge incident as one of the reasons, even zo the motion to reduce dependence on American big tech was voted before that happened