In the EU the size of the state is often bigger than 50% of GDP. What the government buys is very important and means a lot of $$ for projects, consultants and the rest of the open source ecosystem.
I like the publicly funded open source funding in theory, in practice I suspect these guys had to pay consultants to create a funding project application, that went through some arbitrary agency, and the money that got to the developers is probably less than half of the money that was spent in the process. And then if this becomes more widespread an the existing software companies that do business with government will start sucking money out of such grants and the government quality code.
If our governments had a way of funding quality software development we would not get the software that we get.
Every now and then they will strike gold with stuff like Blender funding, but even that is peanuts comparably, and only passes through the art/culture channels probably.
Actually, the grant process at NLnet is supposed super light weight. It consists of a single short form (https://nlnet.nl/propose) with very little boilerplate. No consultants needed...
Nice, I didn't see this is small scale grants, this is great, like Blender case. Unfortunately I don't know that this scales to serious budgets.
My experience being involved in applying on a "digital transformation" funded project was that it was basically pointless to do it without an agency because it will cost you more to figure out everything on your own and you'll likely fail anyway at some random step - and that the people applying to these kind of calls are basically there to gobble government money with appalling delivery history, but the only thing that gets reviewed is credentials.
That is very true. Budget is something that most organisations are fairly bad at. So it does make a lot of sense for e.g. the European Commission to work with organisations like NLnet that do get it.
The application process is pretty easy. I applied a couple of years ago. I did take it seriously, but I probably should have put more time and effort into my message and presentation.
What I don't really get about NLNet is their page titles are all about the Public Nature of the Internet, but the granted projects are all over the place. Not a bad thing, and being overly vague is a necessity to not push projects a certain way, but it hinders clearer communication, I think.
Yes, I checked it out afterwards, seems like a decent program. My comment was more about EU investing in OSS large scale. I've seen how EU projects get awarded and I doubt anything of value will come out of that, especially once cost is accounted for.
> I like the publicly funded open source funding in theory, in practice I suspect these guys had to pay consultants to create a funding project application, that went through some arbitrary agency, and the money that got to the developers is probably less than half of the money that was spent in the process. And then if this becomes more widespread an the existing software companies that do business with government will start sucking money out of such grants and the government quality code.
I agree that systems are far from perfect at the time.
I also think that governments have been putting money into digital tech for much less time than private enterprise.
> And then if this becomes more widespread an the existing software companies that do business with government will start sucking money out of such grants and the government quality code.
Agreed, but keep in mind this must be compared to the phenomenal inefficiency of the developing monopolist oligarchy to the end user. I find it interesting how much harsher people are when some fraction of government money is siphoned, when the current alternative is *most* of it being siphoned, albeit without passing through government first.
The implicit claim often seems to be that people see less value per dollar when money passes through the government. But this is pure nonsense, particularly when you compare things like the cost of healthcare in the US vs everywhere else. Or generally most cases where government managed services can be compared to consolidated markets.
Just because some government money is wasted in ways that are less applicable to private industry doesn't at all mean the private industry creates more value for the consumer.
In the US it's 40%. The 10% difference is explained by the US having limited to no public healthcare or pensions. Those are largely paid for privately, which pretty much balances it out.
In certain spaces we have seen OpenSource become a runaway train: linux kernel, sqlite, postgres and many other projects.
All of these projects have one thing in common, it's not where the money is. The money is in the direct contact with end users.
As far as I know, there aren't any open source projects which are ubiquitous in the consumer facing space. I think that there is a chance that NextCloud could become a runaway train vs. Microsoft Teams, Google workspace etc.
Some reasons...
## Simple to use & signup
Through providers like Storage Share by Hetzner, it's a simple click, provide card info, start using experience for NC. One could argue even simpler than getting Google Workspace to work!
Also, it's about 5% of the price of Google Workspace for 8 users.
## Big open source improvements
Companies like Deutchetelekom have engineers submitting pull requests to improve their code since NC is now an integral part of MagentaCloud (over 2M users). Thousands of universities are using NC, they too have engineers submitting pull requests.
*I wonder if there are any SF startups using NC and potentially interested in improving the code.*
Is it possible that eventually, there are more high quality engineers interested in improving the NC code than there are at Microsoft or Google? I think there is a chance.
I wrote a little cli[1] that is a bit easier than curl for getting data to and from google sheets as csv. I do think this sqlite extension would be even more convenient if it supported writing data with INSERT/UPDATE.
For one, there's no CSV standard and often csv based data exchange methods have lurking (and/or obvious) correctness problems. Maybe gsheets and sqlite are lucky there, maybe not.
This project is also well documented including the gsheets side (which is quite non-obvious).
> Maybe gsheets and sqlite are lucky there, maybe not
Seems weird to to pooh-pooh the idea without any reason to think it isn't good. Also, it's not really a question of luck, is it? It's not that mysterious how to validate an integration like this.
Well yes, there are actually many many CSV standards, my mistake for imprecise language. What I meant to say that there's no single CSV format that programs agree on. Starting with Excel...
Apparently it's quite common in Gsheet land to base your CSV export around things like this copypasta javascript code and customize as needed (follow forks, in parent direction too, for more of the story) https://gist.github.com/mrkrndvs/a2c8ff518b16e9188338cb809e0...
The problem is with the cloud integration not necessarily with the Microsoft part so if Microsoft wanted to they could sell an version of office to the school with the cloud functionally disabled or changed to point to an on-prem SharePoint installation but this is the smallest part of why school administrators like o365 and google apps.
The problem is that when you look into the email/groupware market and particularly how to manage devices/accounts, something both o365 and google apps(with Chromebooks) offer as a part of the cloud subscription, so now your kind of dealing with implementing authentication, file hosting, email, chat and a whole lot of associated problems based on picking individual solutions for each part of the puzzle.
What the regulators expect will happen is that smaller local companies will start selling integration services to the schools because that's how it worked when IT was introduced to the marketplace. But that market have been decimated over the last few decades so it's not obvious who in the local markets can still do it as the big vendors have been positioning their training offerings and best practice guides towards less locally controlled infrastructure.
If governments can collaborate to engineer and build airplanes for example, why can't they do the same for edu software? The economic benefits (of edu) and the number of people directly and indirectly affected has to be on the order of airplanes / airtravel.
> If governments can collaborate to engineer and build airplanes for example, why can't they do the same for edu software?
Airplanes are a good example on just how utterly bad this can become - we have a lot of issues there. Both the EU and US sued each other over unfair subsidies, both Boeing and Airbus have massive structural issues because they are also expected to distribute pork over election districts...
At the core, the problem is anti-competition laws. Governments have enough firepower to simply muscle every free market competition to its knees - the private competitors would simply sue any too effective effort by the government away, and I cannot even claim that this would be without reason.
Nextcloud is the prevailing solution being adopted along with Collabora office suite integration. I have it running on a RPI here at home, and it is great. It is all free software, but companies and institutions pay subscription for support.
In the EU the size of the state is often bigger than 50% of GDP. What the government buys is very important and means a lot of $$ for projects, consultants and the rest of the open source ecosystem.