LWN: Comments on "The sustainability of open source for the long term"

The sustainability of open source for the long term

karim — Wed, 08 May 2019 15:12:40 +0000

Over the past ~25 years, I've seen the following work (really quick list, I'm missing quite a few things probably):

- Provide services around an OSS package, use your work on the package as marketing. Several projects and companies work very well this way.
- Work for BNC (big name corp) that depends on mainstream OSS you contribute to. Works great for many kernel developers.
- Big corp commoditizing non-differentiating functionality by open-sourcing it. There's a ton of OSS coming from this, from very large companies. Most players now understand that this is a great way to keep the cost of non-differentiating software down and generating "positive" marketing from developers.
- Gaining geek cred by open sourcing toy project you worked on. Lots of that of this sort of abandware on github, but sometimes they contain some nuggets that are hard to come by.

I've seen the following not work or work very unevenly (forget RedHat, it's the exception):

- Open core. This looks good on paper, but doesn't work in real life, except for very niche products ... which end up not too far from the project+services model mentioned earlier.
- "Opt-out" dual-licensing models with commercial license available. This almost invariable ends up being bait-and-switch if it's led by a company. Affero-licensed software is almost always of this sort.
- Commercial offering with "toy"/crippled OSS version available. Slightly different from open core, more prevalent in the early days when some companies who didn't do OSS tried to get into OSS.

In short, I don't see VC'able OSS as viable -- though I know some VCs say they "specialize" in this. But you can live "well" on working around OSS.

The sustainability of open source for the long term

mathstuf — Thu, 25 Apr 2019 14:11:45 +0000

The transcript reference has been fixed for those checking out the transcript now.

The sustainability of open source for the long term

mathstuf — Thu, 25 Apr 2019 13:48:36 +0000

There was an episode of The Changelog recently which mentioned a VC firm focusing on just funding FOSS(-based) ventures. Roaming the comments, I found this[1] which mentions something called "Oasis Capital", but that looks to be something completely different at first glance. Further searching finds OSS Capital[2] which is likely what was meant.

Now whether Tidelift is backed by them or not, I don't know. But there are FOSS-aware VC firms out there.

[1]https://changelog.com/podcast/341#transcript-76
[2]https://oss.capital/

The sustainability of open source for the long term

rgmoore — Wed, 24 Apr 2019 21:42:17 +0000

When leftpad can become a separate project, project number inflates quickly.

I would guess there's a question about direct vs. indirect reuse. The number of people who directly include something like leftpad in their projects is probably much smaller than the number who use it indirectly by using a library that uses it. If you count "users" as only those people who use it directly, the numbers would get a bit more manageable.

There's also the question about how people would change their usage patterns if they were paying for the code they reused. The chances are people would be a lot more reluctant to reuse a tiny function like leftpad if it were costing them $1 every month. Instead, they'd rewrite little things like that, or somebody would make a bigger utility library that people could include as a big chunk instead of including dozens of functions separately. Programming languages with large standard libraries would have a big competitive advantage because they'd save their users real money compared to ones that require large numbers of external libraries to get anything done.

The sustainability of open source for the long term

nilsmeyer — Wed, 24 Apr 2019 20:49:47 +0000

What is worrying me is that this is a VC funded venture, how are they going to produce the gains a VC requires while also serving their customers? Sustainability and the profit needs a VC fund needs to achieve aren't really well aligned.

The sustainability of open source for the long term

jake — Wed, 24 Apr 2019 20:20:36 +0000

> The bulk of the issues the article complains about are opened by other devs.

I'm sorry you see it that way. The article was not meant to *complain* about the issues being filed, but to instead report what Luis said about how that increases the demands on the maintainers of these projects. He was not complaining about it either, just noting that GitHub has made it easier so people can more easily file issues, so they do. Apparently I worded my replay of what he was saying poorly.

jake

The sustainability of open source for the long term

nim-nim — Wed, 24 Apr 2019 20:08:59 +0000

The average user is not on github.

The bulk of the issues the article complains about are opened by other devs.

The sustainability of open source for the long term

halla — Wed, 24 Apr 2019 18:25:07 +0000

The article as is, is a bit weird, but your reply is weirder. And honestly, once you start getting a dozen bug reports a day, the value of a bug report is about that of an idea proffered by a fan to a novelist.

And yes, I've been maintaining a software project since 2004, and I've had to cut down on user support in order to stay sane, and I've had to make a clear separation between the place where users report issues, and the place where my developers look for work, mostly.

The sustainability of open source for the long term

pizza — Wed, 24 Apr 2019 16:27:21 +0000

> Sad to see the average dev still does not understand what maintenance means.

You may be correct, but the average developer still has a far better understanding than the average user. (The phrase "entitled prat" comes to mind...)

The sustainability of open source for the long term

nim-nim — Wed, 24 Apr 2019 16:13:26 +0000

> if these maintainers could, say, get $1 per month from each of their users

What a completely unrealistic assessment. Don’t people have any idea of the scale of software reuse nowadays? Anyone on lwn is using thousands of different projects every month (sometimes tens of thousands). Probably quickly approaching the tens of thousands. When leftpad can become a separate project, project number inflates quickly.

> When you have 1000 users who needed to sign up with a Bugzilla instance somewhere to file a bug, the average response was to file no bugs

And that was good, exactly, why?

> many developers would rather have help with bug triage than receive money for their work; they just want to get their time back, he said

I'm quite sure a lot many issue reporters would want their time back too, since a lot of those are (auto) closed (sometimes quite rudely). Sad to see the average dev still does not understand what maintenance means.

If you don't want to maintain your software just do not open an issue tracker. If you open an issue tracker don't complain people use it. Writing a report is work too. Just try to spend a day forcing yourself to report the problems you hit in the software you use, you'll see how far you will last.

The sustainability of open source for the long term

mjthayer — Wed, 24 Apr 2019 12:46:49 +0000

The Tidelift web page says that it leaves projects to manage things as they see best, not putting any support requirements on them or similar. I would think that there ought to be some way for projects to provide extra benefit to Tidelift supporters to prevent a tragedy of the commons. Extra support would be an obvious place. I wonder if the Tidelift people have or are considering making that easy.