5 ms·

I just wonder about the process and the results. I mean, it doesn't look to me, as if there is a management behind this, that actually has a goal. It looks mor

I just wonder about the process and the results. I mean, it doesn't look to me, as if there is a management behind this, that actually has a goal. It looks more like they are giving funds to projects who apply for them. IMO, they should state 3 clear goals and sponsor specific projects which reach those goals. To give some example how those could look like: - create a decentralized, federated instant messaging platform, that is build on public standards - create an e-learning platform that is usable with already established devices - establish a market for the created software with partners Naturally, all result would have to be open-source products and the goals would need some details/numbers to measure them. They could even invest into already established projects, but please, with easy to understand goals.

input_sh 6y ago
I disagree. I believe the governments should fund existing open source software that are considered to be "critical" infrastructure (as in lots and lots of people rely on it) instead of chasing some random goals and adding bureaucracy on top that would slow down lead developers. Just give them money and trust them that they'll do whatever it is they've done so far that many people recognised and started relying on their solution to the problem. Without looking them up, there's exactly three pieces of software in the list above your comment that I don't recognize: FLUX TL, WSO2, midPoint. I'm happy to see all the other names on it, and I'm pretty sure I'll feel the same way about these three after I look them up.
- arendtio 6y ago
  I am not so sure that we disagree. I would consider instant messaging services 'critical infrastructure'. The difference seems to be, that you want to support projects that are large already, while I would certainly put more focus on smaller projects as those are the ones which can initiate change. I find it heartbreaking, that we still depend on WhatsApp and Zoom. Neither service owner is particular trustworthy. Communication is definitely critical infrastructure and yet, the open-source alternatives are very limited (in quality, not in quantity). So investing in this kind of functionality is key.
  - tortasaur 6y ago
    In terms of software, Signal and Jitsi are great alternatives to WhatsApp and Zoom. People don't use them because of network effects, not because the incumbents are inherently superior.
    - eecc 6y ago
      Yeah,and they crumble to bits like Signal is doing right now, breaking the wave of mass adoption. Cynical, profit driven and well funded operations have enough capital to weather these downpours, indeed they’re actually planning for them. I hope Signal sorts their capacity quickly, I’m terrified that my circles switch back to WhatsApp. And then I’m fucked, I either surrender myself to Facebook or cut off from society
    - arendtio 6y ago
      For many use-cases they are indeed proper alternatives, but Jitsi has a much lower limit when it comes to numbers of participants. So if someone wants to organize a group call (the typical use-case) you come a with a lot more scenarios into trouble using Jitsi. And it hurts me to write this, because I hate how much growth Zoom had in the last year while being dishonest about critical encryption features. I would love to promoted Jisti and have tried so in the past, but I can't recommend something when it doesn't fit the use-case. Signal, on the other hand, probably comes close to the WhatsApp features (I use neither one) and while I encourage everyone to to switch, I am missing the federation aspect. IMO, communication should be federated by law (which would also solve the network effect problem). Imagine a world where you could not call someone who has a phone number from a different provider? The current state of instant messaging is exactly this. XMPP solved these problems decades ago and just because the standard didn't catch up with the speed of the mobile revolution, we don't have to reinvent everything from scratch. Properly implemented modern clients work very well (including reliability and battery consumption), the big issue though is that many traditional clients don't support all features and all companies in the business try, to build walled gardens as those tend to driver stock prices.
- burrows 6y ago
  > I believe the governments should fund existing open source software Governments don’t fund things, taxpayers do.
throwaway098237 6y ago
I work in EU research funding programs. The process is very bureaucratic, making it difficult for any small company to apply. Not to mention projects without a company. It's also quite ineffective at giving money to the projects that are meaningful. Large companies and large university projects rake in plenty.
- secfirstmd 6y ago
  Yeh so true. One of fhe great things about the Open Tech Fund was that it makes it easier for people with open source ideas, like Signal or our own Umbrella security training app, to get funding. Doing that through European funds would be impossible.
  - pabs3 6y ago
    NLnet uses EU funds to help open source projects, in a similar way to OTF: https://nlnet.nl/ https://nlnet.nl/
    - setbackreset 6y ago
      NLnet seems to have been doing the good fund for quite some time!
    - oever 6y ago
      Not just EU funds, but those are currently the largest funds at NLnet. NLnet does 'funding as a service'. If you want to fund a particular area of FOSS, we (I work there) can administer that. We provide low overhead funding to FOSS projects and pay particular attention to how the ecosystem benefits from projects. So 'boring' projects that improve adherence to standards, increase deployments of standards compliant software, test standards compliance are very welcome. Redecentralizing the web requires this work, but few people think that this work is fundable. It is and in our opinion is has a high ROI for society.
      - pabs3 6y ago
        I have found that NLnet's funding is very competitive and pretty hard to get.

input_sh 6y ago

I disagree. I believe the governments should fund existing open source software that are considered to be "critical" infrastructure (as in lots and lots of people rely on it) instead of chasing some random goals and adding bureaucracy on top that would slow down lead developers. Just give them money and trust them that they'll do whatever it is they've done so far that many people recognised and started relying on their solution to the problem. Without looking them up, there's exactly three pieces of software in the list above your comment that I don't recognize: FLUX TL, WSO2, midPoint. I'm happy to see all the other names on it, and I'm pretty sure I'll feel the same way about these three after I look them up.

arendtio 6y ago
I am not so sure that we disagree. I would consider instant messaging services 'critical infrastructure'. The difference seems to be, that you want to support projects that are large already, while I would certainly put more focus on smaller projects as those are the ones which can initiate change. I find it heartbreaking, that we still depend on WhatsApp and Zoom. Neither service owner is particular trustworthy. Communication is definitely critical infrastructure and yet, the open-source alternatives are very limited (in quality, not in quantity). So investing in this kind of functionality is key.
- tortasaur 6y ago
  In terms of software, Signal and Jitsi are great alternatives to WhatsApp and Zoom. People don't use them because of network effects, not because the incumbents are inherently superior.
  - eecc 6y ago
    Yeah,and they crumble to bits like Signal is doing right now, breaking the wave of mass adoption. Cynical, profit driven and well funded operations have enough capital to weather these downpours, indeed they’re actually planning for them. I hope Signal sorts their capacity quickly, I’m terrified that my circles switch back to WhatsApp. And then I’m fucked, I either surrender myself to Facebook or cut off from society
  - arendtio 6y ago
    For many use-cases they are indeed proper alternatives, but Jitsi has a much lower limit when it comes to numbers of participants. So if someone wants to organize a group call (the typical use-case) you come a with a lot more scenarios into trouble using Jitsi. And it hurts me to write this, because I hate how much growth Zoom had in the last year while being dishonest about critical encryption features. I would love to promoted Jisti and have tried so in the past, but I can't recommend something when it doesn't fit the use-case. Signal, on the other hand, probably comes close to the WhatsApp features (I use neither one) and while I encourage everyone to to switch, I am missing the federation aspect. IMO, communication should be federated by law (which would also solve the network effect problem). Imagine a world where you could not call someone who has a phone number from a different provider? The current state of instant messaging is exactly this. XMPP solved these problems decades ago and just because the standard didn't catch up with the speed of the mobile revolution, we don't have to reinvent everything from scratch. Properly implemented modern clients work very well (including reliability and battery consumption), the big issue though is that many traditional clients don't support all features and all companies in the business try, to build walled gardens as those tend to driver stock prices.
burrows 6y ago
> I believe the governments should fund existing open source software Governments don’t fund things, taxpayers do.

throwaway098237 6y ago

I work in EU research funding programs. The process is very bureaucratic, making it difficult for any small company to apply. Not to mention projects without a company. It's also quite ineffective at giving money to the projects that are meaningful. Large companies and large university projects rake in plenty.

secfirstmd 6y ago
Yeh so true. One of fhe great things about the Open Tech Fund was that it makes it easier for people with open source ideas, like Signal or our own Umbrella security training app, to get funding. Doing that through European funds would be impossible.
- pabs3 6y ago
  NLnet uses EU funds to help open source projects, in a similar way to OTF: https://nlnet.nl/ https://nlnet.nl/
  - setbackreset 6y ago
    NLnet seems to have been doing the good fund for quite some time!
  - oever 6y ago
    Not just EU funds, but those are currently the largest funds at NLnet. NLnet does 'funding as a service'. If you want to fund a particular area of FOSS, we (I work there) can administer that. We provide low overhead funding to FOSS projects and pay particular attention to how the ecosystem benefits from projects. So 'boring' projects that improve adherence to standards, increase deployments of standards compliant software, test standards compliance are very welcome. Redecentralizing the web requires this work, but few people think that this work is fundable. It is and in our opinion is has a high ROI for society.
    - pabs3 6y ago
      I have found that NLnet's funding is very competitive and pretty hard to get.