1. 25
  1.  

  2. 3

    I wonder why they don’t provide any hosted solutions or services like 99% of the other web servers do. That has shown to be incredible viable and sustainable. But I don’t really know anything about business practices and SaaS

    1. 41

      Many times when the OpenBSD group asks for donations, people come back with: why not offer some subscription or some more tangible goods like USB keys or whatnot? The answer is that it’s just more work to be done, and we’re not asking for money in exchange for more work selling more goods, we’re asking for money in exchange for what we’ve already done. The software we’ve written is the thing you are getting, and you are already getting it, and we would like to be supported financially to continue doing that specific thing.

      Offering hosted solutions and support contacts just means less developer time on the code that everyone uses, and much more time on the hosted solutions and support contracts for those few that gave money.

      1. 5

        Have companies been reluctant to donate rather than purchase things? I’ve noticed patio11 gripe against it but not investigated his legal concerns.

        1. 2

          whoa. that’s an extremely interesting point. it’s like the linux distributions that used to sell cds - it was implicitly understood that you weren’t really paying for the cd, or even all that much for the convenience (since iirc the model survived well into the time that home bandwidth was good enough to download an iso), but to support the project. nonetheless the cd set both a clear figure for a “reasonable amount” of financial support by an individual to the project, and gave you something tangible to point to and think “i paid for that”.

          1. 2

            At least for OpenBSD, it has been a lot easier for larger companies to swallow sending a check to “OpenBSD Foundation” instead of “Theo de Raadt”, but in either case it is still considered a (taxable) donation.

            1. 2

              On the smaller side of budgeting I’ve definitely seen that. A lot of companies (even some universities) allow employees to buy software, books, etc. related to their job, up to a budget limit. But it’s structured as reimbursement for purchases, not as a pool of money you can do whatever you want with. So, for example, you can buy Sublime Text for $70 if that’s your preferred text editor, and they’ll reimburse you the $70. But if Sublime Text were donationware, you wouldn’t be able to donate $70 of the company’s money to them.

            2. 4

              But people…. just don’t work that way, do they? Without a tangible carrot in exchange for their money, most will not give money. If they already have the carrot, most do not feel compelled to give the money. You need some sort of, uh, “coercion”. Also, a lot of companies just can’t donate money. Their accountants need to be able to say, “this is what we got in return for the money”, and charity is not something.

              For Octave, I have been trying to find a way to get free carrot to sell, because donations barely cover a few people’s plane tickets to our yearly meeting.

              1. 3

                You are 100% nailed it with the “this is what we got in return for the money”. Additionally, while support is a very solid something, it tends to be a fungible thing. “Well, did you shop around getting support for product X from other companies?”. That is what is great about the consortium model – there is no other source for consortium membership. It is a great line item “Consortium Membership”. It is a combination of support AND influence (read: direct contact), and companies might only pay a little for support, but will pay a good deal more for piece of mind and influence.

            3. 23

              I struggled with this myself (author of Sidekiq here) as a OSS developer and project lead.

              1. I didn’t want to start a business with all the admin overhead that implies.
              2. I didn’t want to sell support, that incents you to provide poor documentation/community.
              3. I didn’t want to provide a SaaS, that requires 24/7 support, hiring a team, etc. See 1.

              In the end, I felt that selling additional features was the best option for me personally, thus Sidekiq Pro and Sidekiq Enterprise. Of course I had to start a business to make it legit but I don’t have 90% of the admin overhead of a “real” business since I have no employees and provide no benefits to myself (aside from a basic paycheck).

              I actually spoke with Matt privately a few days ago and offered what advice I could. Everyone’s situation is different so different strokes for different folks.

              1. 7

                I found you speaking on https://changelog.com/92/ and https://changelog.com/159/ useful on this topic, as I’ve started my own business supporting http://prometheus.io.

                1. 5

                  Good luck, I hope it works out for you. I’m glad to see not only large VC-funded startups but also boutique startups like us focused on OSS.

                2. 5

                  Incidental support seems nightmarish, dealing with many individuals. But what about the SQLite Consortium model of focusing on a few large “users” and allowing them to basically fund it. https://www.sqlite.org/consortium.html

                  I really think the consortium model is very sane and applicable to a lot of infrastructure projects (like Caddy or Sidekiq). It asks those who can easily pay to pay, and they get something in exchange for it.

                  1. 2

                    That’s very intersting, thanks! Compared to other OSS projects, such as those run by the ASF, consortium members essentially get a combination of benefits a non-profit foundation would provide in terms of project oversight, and what 3rd party companies with developers on staff would offer in terms of support contracts.

                    However, this only works if some of your core developers have consulting skills they wish to sell, and especially this promise is a bit rough on developers: “Consortium members can call any developer at any time, day or night, and expect to get their full and immediate attention.” I’m not sure I would be up for that. But that’s just a detail of terms. The model itself is quite sound.

                    1. 5

                      “Consortium members can call any developer at any time, day or night, and expect to get their full and immediate attention.” I’m not sure I would be up for that. But that’s just a detail of terms.

                      Yeah, and remember, the “floor” is $75000 yearly, and most major companies contribute far more and very seldom call. It is a security blanket. See the agreement: https://www.sqlite.org/consortium_agreement-20071201.html for more information.

                      It seems that more infrastructure based open-source projects should look to SQLite (arguably the most successful infrastructure product ever: https://www.sqlite.org/famous.html) for guidance on how to manage funding.

                      1. 1

                        $75000 o_O

                        Now, will anyone say again that it’s impossible to make a living while giving software away for free? You just have to be very good at what you’re doing…

                        1. 3

                          I am not sure “very good” is even required. I think you just have to be something companies depend on. They have a vested interest in your success at that point, you are in it together. If you vanish they have to switch (non-trivial cost) or take up maintenance themselves (super-non-trivial cost). From Sidekiq to Caddy there are DOZENS of projects in open source that I could see easily being funded this way. Companies are happy to lend support for like $6k a month in exchange for support and more important, influence (non-binding influence, but influence nevertheless).

                3. 9

                  It doesn’t seem like they are looking to start a business, but a way to compensate developers for their time and effort. Assuming they generate some cashflow, I wonder what the distribution will actually look like. I guess we will have to wait for a follow up.