Hacker News Reader

Comment by wcerfgba | Hacker News Reader

Thank you for committing to release it all as open source, instead of hoarding it or putting it in the bin like so many other services do when they fail to achieve sustainability.

View on HN

"Just open source it" is not always a viable option. IP can be locked for one or the other reason, and it is not always in the power of founders to release it.

Since major of startups fail, isn't it a good idea for a startup to commit to this at the beginning? Even privately to not disalign the success goal. If they don't commit to this it is zillion times worst than technical debt.

Colin Percival's Tarsnap famously has such mechanic (though not a startup). It ensures you wouldn't fall into some kind of bus problem. However, it creates an incentive where the general public has an incentive to see the product fail. I word it like that because from my PoV the general public has a net benefit from F(L)OSS. I suppose that is why people prefer to keep such secret.

But hopefully not before it is too late. We should all think about situations where we are no longer around. Like having your keys/passwords at a notary.

>I word it like that because from my PoV the general public has a net benefit from F(L)OSS.

I'm not so sure about that. An actively developed/supported project is often more useful than an abandoned open source one.

If you're prepared to pay market rate for someone to maintain something you should always be able to find someone willing to accept market rate (by definition). This separates the cost of maintenance from costs flowing from rent-seeking behavior.

Arguably industry is always willing to pay, in aggregate, the cost for free software to be maintained, but is not willing to pay, despite the optimism of those who sell it, for the marginal utility it gives them per-customer (which should be much higher). When asked to do so they will simply support free-as-in-beer or cheaper alternatives until one of those becomes the dominant player. This is why successful and heavily commercialized free software projects often seem to be only just clinging on to profitability (eg Docker).

>If you're prepared to pay market rate for someone to maintain something you should always be able to find someone willing to accept market rate (by definition).

Are you saying that if you can pay someone to develop their proprietary code for a price you can necessarily find someone to accept the same amount of money for the same work but open sourced? That doesn't seem true. Those are not equivalent offers.

Your comment sounds like you think open-source work would be _more_ expensive than closed-source work, or did I misunderstand? That would seem surprising to me.

I would assume some people (clearly a minority) would be willing to take a small pay cut when working on FLOSS.

E.G. could someone be paid slightly less at GitLab versus the same role at GitHub, if they believe in GitLab's principles

There are people who value open source and there are people who value assets. The point is that there isn't a single market rate for the two offers since they aren't the same thing.

No, I am saying that if a project is open sourced, and there is sufficient demand from people willing to pay for support, maintenance, etc, someone will come along to meet that demand.

If there isn't then the original commercial company was obviously never viable either.

Since an entity getting payment for maintenance can neither expect to collect a monopoly rent for the code, nor is required to pay one to anyone else, the cost of maintenance becomes the market price.

I think your view is too strongly assuming that everyone in the market is perfectly rational and perfectly replaceable with zero transaction costs..

Especially when you start talking about a single person company, it could be as simple as the founder being sentimentally attached to the company and continuing it despite making less than they could with other opportunities, but that doesn't mean that if the founder gets hit by a bus someone else will also be willing to take a sub-optimal gig. Or a company might currently be barely worth running with $20k a month in revenue, but if there is a period of turmoil and lost customers in the process of open-sourcing/founder getting hit by a bus/whatever that now the customer base would only bring in $10k a month, it's no longer profitable, and the market fails to find someone to run the company.

If the code is open sourced, there doesn't need to be a dedicated company behind supporting it. Assuming a compatible license, any party that wants to use can fork it and pay people to maintain it for their own needs without advertising that they "support" the code.

But just because something is open source doesn't mean it's easily maintainable by anyone. It's one thing to install and run it somewhere, but if it needs fixes or customization it quickly becomes necessary to learn and understand the architecture and organization of the code, the data models and structures, etc. and if it was previously the product of a "one man shop" then you're going to have to pay someone to do that learning.

Some products are viable as opensource projects but not as a commercial venture, and the overhead of a business entity might be enough to get out of viability.

> However, it creates an incentive where the general public has an incentive to see the product fail.

Maybe Colin should make tarsnap free for bus drivers? ;)

> However, it creates an incentive where the general public has an incentive to see the product fail

That's not really the problem. The problem is that it lowers the company value: why should I, potential acquirer, shell out $$$ to get your IP, when I can just sit down and wait for it to come down the river?

Most of the value of a company is in the people, structure and brand.

And there's no guarantee the other acquirers will cooperate. "Why should I wait for the company to collapse and open source it's stuff forcing me to compete with other acquirers when I can buy them now and get a monopoly?"

Presumably if the technology is good, there is some competitive advantage in retaining sole control over it, or licensing it out.

This seems to make the company more valuable toward the end of life. If you don't buy out your competitor, your customers will have an Open Source alternative.

it depends, many startups whole purpose for getting founded is getting acquihired, if this the case, then clearly the answer is no

It’s a lot of work to set something up for being a sustainable, usable OSS project. It certainly doesn’t make sense to invest time in it upfront when your goal is to sell access to it. Afterwards, founders have usually poured everything they have into trying to make the business survive and the work is just unappealing.

Perhaps, but many startups do leverage paid products and components, leaving the open-source product nonfunctional (without a user paying those same licenses). To suggest that the company has an obligation to only use OSS is not reasonable.

What's the incentive? Startups are created to earn founders money, at least in the form of wages off funding or ideally when acquired/listed. Pledging to open source doesn't work towards that goal.

A commitment is probably not worth much when the entity who committed literally doesn't exist anymore.

> isn't it a good idea for a startup to commit to this at the beginning?

The source code is part of a startup’s IP. When a startup fails the IP goes to the investors and debtors who can try to recoup some of their investment by selling the IP.

By the time a company fails, the source code isn’t really theirs to give away any more. They have to hand it over to investors. It’s the same reason a failing startup can’t just give away all of their office furniture or other assets.

Even if we ignore that, releasing source code doesn’t actually do anything for most customers of a SaaS platform. They were paying for the service, not the source code. The source code wasn’t designed for individual use, so self-hosting isn’t going to make sense. Maybe some other group would try to run a company around it, but if the first company failed to become sustainable with the source code it’s less likely that a second company would.

How many open-sourced codebases from failed startups have succeeded as open-source projects? Can you name a single one?

Blender

Netscape Navigator, which became Mozilla, which became Firefox.

OpenOffice

aka LibreOffice aka StarOffice

https://en.wikipedia.org/wiki/StarOffice

Amazingly, there is still a really cool 90's website at https://www.staroffice.com ! You can even buy a neolithic version of StarOffice!

darn, that _is_ a cool site

Parse (https://parseplatform.org)

Would Godot count?

Actual Budget (https://actualbudget.org) - I'm not connected to them but I use their software on a daily basis and it's still well maintained.

https://www.chatwoot.com/

OTOH this whole "I didn't value your service enough to pay for it, but now that you failed and open sourced it I'll jump on it" attitude is kind of annoying.

I can absolutely understand startup founders who are a bit spiteful and don't want to give away for free what nobody was interested in buying from them.

It's better for your mental health to instead think "well, at least my loyal, paying customers who believed in me have a path".

True, that's a good counter point. I'll take it :)

Perhaps another option would be to give paying customers an invitation to a private git repo

Maybe the pricing is off?

There's lots of things I'd use at $n but wouldn't use at $10n

And the corollary of this - if open source didn't exist, the result wouldn't be me paying for lots more products - it would be using lots less products.

Certainly - the advent of paywalls has meant that I mainly don't read those sources. Even the more affordable ones come with a mental tax of deciding to subscribe/cancel, remembering to renew/cancel etc...

The first time I've heard of it is today so I can't really have ever given them money to begin with.

But I'd be curious to see the source once it's released, for sure.

Open source code is inherently more dependable than proprietary code, in that if the maintainer quits and this tool is crucial for your business, if it's open source you can maintain it yourself (or if it's used by multiple parties, they can pool together to keep maintaining it)

I myself am not interested on any proprietary offerings at this point; this has no relation to any qualities of the proprietary product. It could be awesome for all I know

I think there’s a lot of people that’ll happily install it if it’s open source, and then do absolutely nothing with it. The same people wouldn’t pay $20/month for the same privilege.

The people to whom it is valuable are already paying.

PSA - the HN article submitter ("Tomte") is not Placemark's author (Tom MacWright) despite the three matching letters.

Tom M seems to not like HN, so this makes sense.

https://macwright.com/2022/09/15/hacker-news

He does have an account here, tho, and is responding to comments.

https://news.ycombinator.com/user?id=tmcw

Either the snippet has a bug, or he changed his mind and removed it, but it’s not working.

Hacker News puts `noreferrer` on the link (because apparently Internet Points and Arguments are more important than the courtesy of respecting the wishes of the author.)

>I’ve learned some bad habits from Hacker News. I added Caveats sections to articles to make sure that nobody would take my points too broadly. I edited away asides and comments that were fun but would make articles less focused. I came to expect pedantic, judgmental feedback on everything I wrote, regardless of what it was.

>Writing for the Hacker News audience makes my writing worse.

Yeah, understandable.

Reply on news.ycombinator.com