upvote
You'd be surprised to learn this about free and open source software, but if a maintainer is unavailable, you have both full rights and full source code to... wait for it... fix it yourself (or pay someone to)!

There is something unhealthy in this relationship only if you project "no warranty" into unrealistic expectations.

reply
This is true for the majority of open-source projects, but the most serious ones, on which a lot of software/businesses/infrastructure depends, are controlled by foundations or some kind of other management entity.

cURL also offers paid support and also paid access to the rock-solid (LTS) version, with guaranteed response times, and the blog post states that there's still people to respond to these.

reply
You don't really though. Sure you can fork it and fix your issue, but then what? Are you going to maintain your fork in perpetuity? Are you going to patch all the software that depends on the code you fixed to use your version instead of upstream? Are you going to get your users to do that too?

In most cases this is extremely impractical.

reply
> but then what?

Then you send the patch upstream, they incorporate and maintain it for you. Congratulations, you just FOSSed.

reply
> Then you send the patch upstream, they incorporate and maintain it for you

Firing patches upstream is still adding burden to the (likely already over-burdened) maintainers.

In an ideal world, if you want a patch upstreamed, you would be contributing to upstream maintenance (or at least donating to the upstream maintainers)...

reply
Fair, but it is less of a burden than just submitting a report with no proposed fix. Also, submitting quality patches regularly seems to be a good way to eventually become a maintainer, provided that both sides are interested (cURL generally is – at least that seemed to be the vibe at the last year's cURL Up event I attended).
reply
They do, he said at the end if you have a support contract then they will respond and deal with security issues.

I guess the whole point of the article is to show that people should buy a support contract if they need support.

reply
They do.

> Everyone with a paid support contracts will of course still get full and appropriate service even during this period.

reply
It does. The article clearly says that if you have a paid support contract they will be on-call as per usual.
reply
> And it surprises - and saddens - me that not even friggin curl has the financial muscles to have somebody on-call for one month...

Is it that they can't or don't want to. I'm sure curl is popular enough that it could attract a co-maintainer if it wanted to. Of course there is a cost to that. Software projects done effectively by a single person are often more focused and designed more coherently. I'm not sure curl would be as good a product if there were multiple maintainers with potentially conflicting visions.

reply
And I'm assuming you're not going to pay for them to have that someone on-call, even though you're worried about this scenario
reply
deleted
reply
Consumers, not customers
reply
Reminder: ‘the software is provided “as is”…’.

It’s not their problem that you, or anybody else, think you are owed 24/7/365 emergency support.

reply
They do. You just seem to expect that it will somehow be free.
reply
I wonder how far we are from the agents just maintaining the packages
reply
The thing which bugs me is that OpenAI (which is an unprofitable company) is spending around what 100k$ per month for an completely AI generated slop called Openclaw. (All because of Hype)

I have seen there to be an more influx of open source software as people are starting to create more software with vibe-coding and other things and just open-sourcing it, which while good in OSS'ing it but its mostly less valuable as compared to the curl codebase which was created by hand and over the years improved itself.

Yet the funding is going towards making more and more (OSS/non-OSS) AI slop by people, companies and dare I say countries yet we are unable to take the same wealth and money into, say, the curl project (and the likes)

There is also an visibility issue. We all know curl and this is the state of curl. Imagine all the projects which we all don't know that much about or aware about going through same issues.

reply
>The thing which bugs me is that OpenAI (which is an unprofitable company) is spending around what 100k$ per month for an completely AI generated slop called Openclaw. (All because of Hype)

For whatever reason, real people seem to desperately want Openclaw regardless of it being AI generated slop.

OpenAI is certainly not wasting the money they're spending on Openclaw, even if I personally wouldn't want to touch that particular piece of software.

reply
> For whatever reason, real people seem to desperately want Openclaw regardless of it being AI generated slop.

I can agree with it but I am unsure how much the desperation is out of FOMO or out of real use-cases.

Surely curl has more use-cases and projects relying on it than OpenClaw.

The demand seems to be generated out of hype rather than sustainability. Openclaw project isn't even an year old and from my time hearing about it, it isn't safe or sustainable in any fashion and it seems that the hype around Openclaw has now started to slow down as I hear less about it (which to me is actually a good thing imo) but it shows what the market reality of these tools currently are (at the moment).

reply
>I can agree with it but I am unsure how much the desperation is out of FOMO or out of real use-cases.

I frequently run into people using it, they seem happy with it. I remain highly skeptical about this being a good idea, but I'm quite convinced that many people genuinely really like it and find it useful.

reply
> I frequently run into people using it, they seem happy with it. I remain highly skeptical about this being a good idea, but I'm quite convinced that many people genuinely really like it and find it useful.

That can be the case and good for them, at the very least its open source software that they are using and it raises more awareness about them.

But I think that we have strayed a bit afar from my main premise that I think we both agree on that although the value of an project is always subjective and its up to the companies on how they direct the funds to. It's Okay for OpenAI to sponsor Openclaw if they absolutely want to.

But the question is if its entirely reasonable as to a project like Curl getting less funding overall, simply because everyone is using curl underneath but the tech is boring (as I think it should be), but this makes everyone think that curl is well-funded when it isn't.

I think that its a reasonable decision for a company to give a very small chunk if it has massive profits to curl to sponsor the project to be more sustainable, but I am not the one at the decision-making involved in that said company, so I don't know what is the rationale behind blocking or not sponsoring Curl.

Is the rationale that they can get away with not sponsoring curl in the first place and use it with its permissive licenses in its code so why invest/donate the money in first place, but this practise doesn't seem sustainable to me!?

reply
>But the question is if its entirely reasonable as to a project like Curl getting less funding overall, simply because everyone is using curl underneath but the tech is boring (as I think it should be), but this makes everyone think that curl is well-funded when it isn't.

I think the returns fall off really really quickly when you increase investment in a boring, mature project like this.

It might be nice if people sponsored curl more, but the software isn't going to significantly improve because of it.

reply