Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Redis is open source again

antirez.com

Post reply on HN
Print view View original

711–720 of 815 posts

Re: Redis is open source again

#711

Post by bravetraveler »

bravetraveler wrote: post #688
fastball wrote: post #683

Earlier quoted context omitted.

I'm not really sure what your point is. Salvatore Sanfilippo (aka antirez and the author of this blog post) has never been the CTO or CEO of Redis Ltd. He didn't start it and only briefly worked there (relative to how long it has been around as a company).

antirez isn't the focus of my ire, stop. They simply have the misfortune of proximity. The actual CTO told us 'always BSD'. Now Salvatore is telling us it didn't totally expire. Great! Except... the CEO forgot. Here's a lesson about AGPL and the Big Bad Cloud. Give me a break or I'll take one. At risk of confusing things further: YouTube is leaking, is kind of my point. Apologies again for not exhaustively phrasing m…

    ~ $ curl -s 'https://news.ycombinator.com/user?id=fastball'
    [...]
    I used to argue on the internet too much.
Co-founder and tinkerer at supernotes.app
Ah, it all makes sense now. Another point of criticism. Or, a Supernote, if you will: s/used to//

Re: Redis is open source again

#712

Post by bravetraveler »

fastball wrote: post #460
Spivak wrote: post #422

Earlier quoted context omitted.

Yeah this take kind of surprised me, you really wanted Valkey to be the default option for cloud customers ensuring they'll have no migration path to your own offering? I just don't get it. You were getting $0 and now you're getting $0.

How often were people migrating from AWS Redis to Redis Cloud in the first place? I'm guessing not a lot.

+1. Not a lot, I'd guess. Every time I've used Redis (or ValKey in more recent times), latency more than 20ms or so would have been terrible.

I've seen workloads perform better without caching at all at 30ms

Re: Redis is open source again

#713

Post by Brian_K_White »

amrocha wrote: post #591
Brian_K_White wrote: post #519

Earlier quoted context omitted.

Then don't try to sell it. Or make something else that people can't live without and are willing to pay for and can't just pay themselves to write an equivalent instead of paying rent to you forever, whatever, what's it to me or anyone else? Do whatever you want but it's no one else's problem if you can't figure out what. Ya got no argument here.

The point is that if there was no business behind projects like Redis and ElasticSearch you wouldn’t be using them either. You’d be using some random Microsoft or Oracle product. I’m guessing you don’t want that either, so come up with a way to make sure Redis has the financial support it needs to hire engineers to do the full time work that part time contributors like us don’t want to.

The point is I don't have to come up with any such thing.

It is not true that if redis didn't exist then I'd be using some MS or Oracle product. I might, if it was practical. Or someone might have invented redis, or I might if that was a space that still needed filling and somehow no one else did it.

It's like saying if linux didn't exist we'd all be using NT.

That's completely ridiculous. No we would not. BSD already existed and if not that then a minix clone or someone else would have started some other unix clone. There were several small unix clones and other full OS's made by completely small developer teams by then, even single people, commercially. If a single guy can do it at all (regardles that they were doing it to sell), it means the job is not infinitely big and so perfectly doable by a few self-motivated volunteers, especially given how that kind of work has no deadline.

Except "volunteer" is the wrong word because they aren't some kind of weird saint doing something just for you or me. They are doing it for themselves, and you get to have it too.

Everything is like that. Redis is no different. There is nothing magic about redis. Things exactly like that get created when the need for them arises every day. If redis didn't exist is a nonsense invalid premis because 12 redis-alikes will always exist any time the need for it exists. It doesn't matter that I didn't already do it muyself, and I don't have to now either, and that is not just beacause redis happens to exist.

Re: Redis is open source again

#714

Post by jenadine »

Macha wrote: post #686
jenadine wrote: post #507

Earlier quoted context omitted.

> AGPL+CLA is not a license I'd contribute under, Why not? Is it because only one company can make proprietary fork? Would you rather contribute to MIT software where everyone can make proprietary fork? Or AGPL without CLA where nobody can make proprietary fork?

Yes, the asymmetry is a problem. If the terms the company has to use my contributions is more favorable to them than the terms under which I can use their contributions, then I'm not into playing that game.

There is an asymetry anyway as the company is probably contributing more than 90% of the code, maintain the infrastructure for the CI and website, do the promotion of the software, and more.

Depending of the motivation for your contribution, you get what you want, eg, the feeling of contributing to an open source project presumably used by many people, or having that entity to maintain your patch "for free".

You can keep your change in a fork if you like, but the likelihood that your fork is getting used is not that big, and mean more work from your side to rebase the change.

Re: Redis is open source again

#716

Post by theamk »

palata wrote: post #531
theamk wrote: post #484

Earlier quoted context omitted.

All companies should do this.. if anything so we know what not to use. Any sort of proprietary product (be it fully closed source, source-available, or "open source with limitations") is always a risk. You were happily using database X, but they got acquired by Broadcom and now their product costs 100x what it was before. What do you do now? That's why it is much safer to adopt open source - worst case is the company…

> worst case is the company goes under, but you still keep using last released version indefinitely, and hope new entity (maybe even hyperscaler!) forks the code But it's not always the worst case, is it? If the licence is permissive, an hyperscaler can just take is and make a proprietary product out of it. Hence copyleft licences without a CLA . Nobody could make Linux proprietary because the copyright is shared bet…

Who exactly is suffering in this case?

The original company went under, they no longer exist, so they obviously don't care.

The software user does not care either. OK, someone made a proprietary fork, and that "someone" is a hyperscaler... so what? The last released version is still out there and still can be used-as is / maintained by you / maintained by other 3rd party.

The new maintainers / fork authors should not care too. They base off the last released version, and their fork is not affected in any shape by what some hyperscalers do.

(I guess you can make an argument that proprietary fork might pull users/resources from the open-source one... but I don't think it's a big issue. If the users chose open-source version to begin with, why would they switch to proprietary fork?)

Re: Redis is open source again

#717

Post by yjftsjthsd-h »

immibis wrote: post #532
yjftsjthsd-h wrote: post #426

Earlier quoted context omitted.

I'm sympathetic to the OSI not being the only authority. So, have the DFSG, FSF, any BSD, or literally anyone credible endorsed it as Free and/or Open Source?

None of them have bothered to evaluate it or have explicitly decided it's not worth evaluating for now. They haven't endorsed it as Proprietary and/or Closed Source either. And notice the OSI's objection actually makes no sense and could apply to any copyleft license! By the same reasoning, AGPL isn't open source. Yet they say it is.

> None of them have bothered to evaluate it or have explicitly decided it's not worth evaluating for now. They haven't endorsed it as Proprietary and/or Closed Source either.

That's not quite true. Per https://en.wikipedia.org/wiki/Server_Side_Public_License , Debian did explicitly reject it (in addition to listing OSI and Red Hat likewise), citing https://web.archive.org/web/20240226231216/https://bugs.debi... -

  However, the SSPL is clearly not in the sprit of the DFSG, yet alone
  complimentary to the Debian's goals of promoting software or user
  freedom.

  In light of this, the Project does not consider that software licensed
  under the SSPL to be suitable for inclusion in the Debian archive.

Re: Redis is open source again

#718

Post by palata »

immibis wrote: post #703
palata wrote: post #680

Earlier quoted context omitted.

I don't understand what you are saying. What has a "supplier" to do in this? Copyleft is always the best for the customer. Copyleft says "the customer has rights ". Permissive says "do whatever you want, as long as you keep my name somewhere in the attributions".

A product - in this case, software - is supplied from a supplier, or vendor, to their customer. Is this business 101? The usual worst case for the supplier is that they give the product but don't get paid, and the usual worst case for the customer is that they give the money but don't get the product. With copyleft specifically, the supplier is much more likely to not get paid, but the customer receives the benefit o…

> Is this business 101?

I understand what a supplier is. I don't understand how it relates to what I said.

> With copyleft specifically, the supplier is much more likely to not get paid

That doesn't make any sense. Whether the supplier open sources their code as permissive or copyleft doesn't have any impact on the likelihood of getting paid.

But if they distribute it as permissive, a competitor can just make a proprietary fork. Possibly continuously importing the new improvements from upstream and focusing on differentiating. Whereas if they distribute it as copyleft, a competitor has to share their changes, that upstream can benefit from.

So for the supplier, if that's how you want to call it, it's better to licence code as copyleft. Except if the whole idea is to be adopted by corporations, but in that case don't whine when they take your code and build a proprietary product without contributing anything back.

Re: Redis is open source again

#719

Post by palata »

motorest wrote: post #689
palata wrote: post #681

Earlier quoted context omitted.

> I mean, to believe that you first need to believe that having access to the source code is somehow a problem. How in the world did you get there from what I said? Open source code has a licence that says what the copyright owner allows or not. LLMs are laundering machine in the sense that they allow anybody to just ignore licences and copyright in all code (even proprietary code: if you manage to train on the code…

> LLMs are laundering machine in the sense that they allow anybody to just ignore licences and copyright in all code (...) No. Having access to the code does that. You only need a single determined engineer to do that. I mean, do you believe that until the inception of LLMs the world was completely unaware of the whole concept of reverse engineering stuff? > Once it's been used to train an LLM, there is no right anym…

> do you believe that until the inception of LLMs the world was completely unaware of the whole concept of reverse engineering stuff?

The scale makes all the difference! A single determined engineer, in their whole life, cannot remotely read all the code that goes into the training phase. How in the world can you believe it is the same thing?

> Nonsense. You do not lose your rights to your work just because [...]

It is only nonsense if you don't try to understand what I'm saying. What I am saying is that if it is impossible to prove that the LLM was trained with copyrighted material, then the copyright doesn't matter.

But maybe your single determined engineer can reverse engineer any trained LLM and extract the copyright code that was used in the training?

Re: Redis is open source again

#720

Post by palata »

growse wrote: post #706
palata wrote: post #682

Earlier quoted context omitted.

I see that argument over and over, and I don't understand how people can consider it makes sense. "My clipboard learned the code, just like a human would. So it should be fine to copy-paste anything and call it my own". "How is killing a human any different to killing a computer ?" "If humans can vote, why couldn't computers vote as well?" Can we start at "humans are not computers", maybe?

> Can we start at "humans are not computers", maybe? Sure. So it stands to reason that "computers" are not bound by human laws. So an LLM that finds a piece of copyright data out there on the internet, downloads it, and republishes it has not broken any law? It certainly can't be prosecuted. My original point was that copyright protections are about (amongst other things) protecting distribution and derivative works…

> So an LLM that finds a piece of copyright data out there on the internet, downloads it, and republishes it has not broken any law?

Are you even trying? A gun that kills a person has not broken any law? It certainly can't be prosecuted.

> I'm not seeing a coherent argument that feeding a copyrighted work (that you obtained legally) into a machine is breaching anyone's copyright.

So you don't see how having an automated blackbox that takes copyrighted material as an input and provides a competing alternative that can't be proven to come from the input goes against the idea of copyright protections?

Post reply on HN