Hacker News new | past | comments | ask | show | jobs | submit login
Uncurled: A book by Daniel Stenberg, creator of curl (curl.dev)
96 points by tapanjk on March 1, 2023 | hide | past | favorite | 13 comments



> In more situations than not, the paying company would then ask for and prefer that I maintain all copyrights for all the changes and that I not make it publicly noticeable that this was in fact paid for by this company. For the casual outsider, it appears as if I just had a period of extra motivation and energy and one fine day delivered this feature.

> Why companies would not want to use their sponsoring an Open Source project for marketing purposes and good will always baffle me. I have heard it explained by things like that the company does not want their competitors to so obviously spot that they use this project, but that seems like an odd and weak argument.

Thoughts? My guess is because if it becomes known they will pay for feature requests on open source, they might worry other open source projects will ask for money for features they would have added for "free" otherwise.


Basically this. Once you establish a track record of paying for things, you will be expected to pay for everything.

It’s not only feature requests. I’ve filed bug reports on open source projects complete with example reproduction cases and summaries of my debug knowledge, only to have the maintainers demand money for it. In the last case, the bug was a show-stopper that led us to switch to a different library, but I thought I’d do them a favor and share my knowledge as we switched. The response was a snarky reply about how we need to hire and pay them if we want it fixed (even though we didn’t need it any more).

Lately I’ll just submit big reports from my personal email/account to avoid this whole back and forth.


Hm. Open source is really not thriving.


Or maybe it is a way to hide a weak or extremely busy engineering team and they prefer to delegate the tasks.


Daniel is worth following on the Fediverse if you have an account. He shares interesting stories from maintaining curl such as one where a user raised an issue with curl but it turned out that curl was correctly identifying that the users antivirus was acting as a MITM on their network: https://mastodon.social/@bagder/109880784239388087


Ah, I've been on that conversation a few times. My favorite has been a couple of times the conversation went some variant like this:

"Why can't I download this thing from you?"

You have a web filter that is forbidding you.

"No, I don't have a web filter."

Yes, you do.

"No, I definitely don't. You and/or your product suck if they can't download a simple file like this."

Here's the exact page & headers the download process is getting; you have $BRAND web filter, it is complaining that you're violating $POLICY, and it says it's located on this IP/domain address.

"Stop making excuses and just fix it."

Sir, if I was able to just reach out and disable your web filter, you would not actually be happy about that.

"I DON'T. HAVE. A. WEB. FILTER."

(Some hours pass)

"I have spoken with IT and it turns out we do have a web filter. A local blame-finding and finger-pointing meeting has been scheduled for who snuck that on to the network without telling anyone else in the organization. I have now redirected all my quite substantial and more-than-a-little profanity-laced anger towards my IT team, which as I am the VP of $DEPT is going to be a bad time for them; heads may roll. In the meantime, we've gotten the whitelist entry added for this and your stuff is now working fine."

---

I interpolate a little, obviously, and this is a composite of several incidents I've had over the years, but I do sometimes enjoy the little subtle hints dropped in the last communication we receive on the matter that mean that my last interpolation there is likely quite accurate.


To start reading, go here: https://un.curl.dev/


Probably worth changing the submission to point here, as the source should be readily findable if needed.


Ok, we've changed to that from https://github.com/bagder/uncurled. Thanks!


Unfortunately, no way to change the URL of the post. Only admins can do this.


I found the source more interesting. For example, I spent some time looking at badwords.txt and how the project used it. I took a look at the make file to understand how they are doing spell check, etc.

I say leave the link alone for this platform.


Discussed within the last year:

Uncurled – running and maintaining Open Source projects for three decades - https://news.ycombinator.com/item?id=31410042 - May 2022 (9 comments)


another interesting : https://polite.technology/preview

// polite.technology is a free text under the CC0 license originally penned by Tom MacWright

"This is a book about the social interactions and situations that are the underpinnings of the open source world. It's about how humans collaboratively create software."




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: