I think they meant to say the maintainers for redis-py, lettuce, and jedis have already relinquished control to Redis Inc. The libraries are still online, but their repositories have been moved to the official Redis organization and are "controlled" and "owned by Redis" according to an email quoted in the linked Github Issue.
I worked in a huge docs website a while back and I’ll never forget the day I saw a request from a Redis lawyer that insisted we called it:
Redis™*
*Redis is a trademark of Redis Labs
The dumb thing is they insisted the footnote be on the first webpage Redis™ was displayed, as-if I had control over the order of the pages people viewed on the website.
Since then I’ve started removing Redis as a dependency from as many projects as I can.
Yup, technologically illiterate and developer hostile behaviour like that is your canary in the coalmine going belly up. If you stick around after that, you have only yourself to blame for what follows.
I'm obviously not a lawyer, but from what I can see, a project like redis_py is way older than any trademark registration from Redis lab.
If I'm not wrong looking at the trademark, database, they have registered them recently (2018,2019, and a lot very recently 2023) in a very predatory move.
In my opinion, this usage should be considered legitimate based on prior art or something like that.
Anyway, in all case I hope that projects will not give up to this asshole move from redis lab.
Maybe it might be possible to rename the project, but also forbid redis lab to use the name of the projects it had targeted. Like a poison pill. I think that it is legitimate for them to consider that an usage of name s like redis_py by Redis would be counterfeiting of well known brand of these projects.
In the end, it might be assumed that it is the fault of Redis to have not enforced any copyright or branding on these projects for more than 10 years despite the fact that they couldn't have ignored their existence and success.
I'm even quite sure that we can find a publication of Redis lab with how-to use any of these open source projects...
Probably something up to 50$. Redis lab is not that big as a company so they might not be able to hire top notch lawyers.
But regarding the history of free software that is made of individuals fighting the overreach of predatory corporations, I will be very saddened if everyone of all these projects would give up...
Maybe the fsf or software conservancy can assist some of the projects if they were to ask it?
$50 might get you 5 minutes of an attorneys time.
For someone to fight this in court will in the hundreds of hours of billable time. None of these open source client library maintainers is likely going to be able to fund that.
> Redis lab is not that big as a company so they might not be able to hire top notch lawyers.
Redis is a company with a billion dollar valuation that has raised hundreds of millions of dollars, they can afford good lawyers.
Prior art for redis_py clearly belongs to Redis, regardless of when the trademark is actually filed. This is because Redis had to have existed before redis_py.
There are organisations like the Software Freedom Conservancy who, even if they themselves can't help, can probably help you try and find somebody who can.
I'm not suggesting any particular maintainer *should* decide to fight, or even consider whether they want to try - that's very much up to them - but assuming that fighting is untenable in advance without exploring options is something I'd consider an unnecessarily unfortunate outcome.
Is this going to become a trend? Project starts making trademark claims years after a downstream library has been using it without issue. Bit rich, given how accessible libraries can bootstrap an initial product ecosystem.
Is it ever safe to use a trademark name? Will all packages now have to be generic enough to imply the product for which they connect? It”s not a Java/AWS/Redis/Wordpress/Shopify extension, but a thingbob, which just so happens to connect to That-Which-Shall-Not-Be-Named.
You can mention a trademark. You can't use the trademark in the name of your own product. Even in cases where you're legally OK (e.g. mentioning your product is better than <trademark>, or is made for <trademark>) they can still sue you, though it'll be much easier to win the case.
However, I'm not sure where courts would land on a name like "redis-rs". There are arguments on both sides, and I'm not sure if there's enough case law to make the legal outcome clear.
I suspect Redis is going to pull in the various popular Redis libraries under their umbrella and make them incompatible with Valkey. It seems like a defensive move against Valkey.
If I was in charge of any of these libraries, I would replace any mention of Redis with Valkey and move on.
Sure. You can do that. But the community needs a single standard. Open standard that all in-memory data stores like Redis, ValKey, DragonflyDB, KeyDB, Kvrocks will support. Otherwise you will end up in a vendor lock.
Kvrocks (under Apache foundation umbrella) is another interesting Redis-compatible alternative, this one is targeting memory and SSD/NVME storage since it is implemented on top of RocksDB. So depending on your needs it may be a more efficient choice compared to Valkey.
That's a cool one, thanks, I find RockDB-like solutions a very nice niche for performance computing (e.g. ClickHouse, erigon), but I think with all the backing by huge corporations, Valkey will be the replacement for Redis.
This serves as a lesson when naming projects -- call it something different to the core product. If you must have a package called ${name}-rs make it a wrapper around your repo.
Trademark claims are also defended by the otherwise useless PSF. Here GvR objects to someone continuing Python-2 uńder the name "python-2.8" as well as "py28". Informally, bringing in lawyers was threatened at one point:
So, after reading the Github[1] comments, Redis needs a rust client library. It's hilarious to watch the legal and PM figure this out. No strategy. Just trying to solve "enterprise" needs because I'm sure someone asked for a rust client they'd pay over the moon for.
I would love for someone to fight out these bullshit trademark claims. Are you seriously going to win against a crate author for using your business name after letting them do it for 10 years and even featuring and recommending them on your own website?
Who is going to pay to defend? The rust author is making it clear they want no part of this legal mess. Which is going to be true for nearly any open source contributor. The only people with legal representation are going to be the VCs looking for value extraction.
Yes, that’s the problem. I understand the perspective of the maintainers but it leads to a situation where the large corporation can do illegal stuff and bully people around because it’s too much effort to fight back.
I've sometimes wondered if it would be possible to have a voluntary organisation funded by developers and other tech folks whose sole purpose is to fight legal battles against companies and other organisations looking to bully contributors on dodgy legal grounds.
I'm sure a lot of these legal threats would go away very quickly if the possibility of a large scale legal battle was on the horizon every single time. Or at least, they'd be a lot more willing to settle quickly if their opponent was fairly evenly matched resources wise.
Let them fail. Does anyone today remember SCO chasing virtually everyone on Earth after unix licensing violation and copyright infringement claims? There is no shortage in KV stores with similar properties so if redisinc. wants to continue their path without delivering actual value to the customers especially to enterprise customers willing to spend money they know exactly where they will land in several years from now.
I have to thank Redis Inc. for bringing on this shitshow. I had no idea about Valkey and now I do. Corporate hats should consider the loss of general trust and negative marketing their actions produce.
And yeah, I wish FSF or some other, strong and well-funded org would be more "broken glass" and take issue with even the tiniest of legal fights over OSS.
Do ctrl+f for „trademark“, there are currently 12 occurrences. Here’s what the Redis representative wrote:
> These are not my words nor the message I want to convey. In a private call, I said that it feels strange to use the brand name for a library that keeps supporting all the forks indefinitely and does not grant compatibility for the product itself. I also said that it does not make a lot of sense to pretend to rename a crate and replace it with the same software or an alternative library, thus creating confusion and frustration.
However, companies do consider protecting their trademarks where their reputation is challenged. This does not need any explanation. In my private call with Armin, I highlighted that. Of course, there is no plan to take action regarding the trademark, but to find a reasoned and agreed-upon way forward to avoid introducing another Redis Rust client library where changes, improvements, and innovation can be introduced.
So basically the point is to make thinly veiled threats of copyright enforcement to „agree to a solution“.
The Mafia also agrees with you on a payment plan after highlighting that a fire would be a shame in your establishment.
He published the emails where it says they want to own the library so they can set the road map for it, they're offering to pay for it.That quote even says they're not looking to take action over the trademark.
I just quoted the part where the Redis representative describes a private call. How does the content of the emails negate that?
> That quote even says they're not looking to take action over the trademark.
Yeah, he’s basically saying „also, we think you’re infringing but we won’t have to enforce that, right? I’m sure we can come to an agreement, wink wink“.
Maybe I’m reading this completely wrong but this reminder is basically a threat to sue you to death if you don’t comply.
IMO you are because your attempting to "read between the lines" and add things that aren't there while ignoring things that contradictory to what this hypothetical lawsuit would include.
He says they don't want to cause a mess of renaming, or forking because it would be worse off and confusing for everyone
For users of Node.js, Java, Python and very soon Go -
Valkey-Glide:
https://github.com/valkey-io/valkey-glide
It will stay out of the hand of Redis, can promise that, and if you use another key-val db and want to contribute to its compatibility with it, we will be happy, come talk to us.
While it's difficult to compare the pettiness, this is definitely sillier than the WordPress thing as far as trademark enforcement goes.
Trademark-as-a-Weapon is becoming too common now. Are we going to be forced to use legally distinct (and confusingly different) names for all our libraries now to defend against this nonsense? Pick projects based on the brand protection posture of the licensors? What a waste of time.
The risk here is they intentionally make it incompatible with Valkey, similar to how Elasticsearch client libraries complain when a newly added slogan header is missing
Consolidation of control and brand recognition. As soon as Redis revealed their hostile intentions, forks split off. Valkey seems to be the favored one, and in my direct experience is what projects are using moving forward. If we don't intend to use Redis' commercial services we have no reason to use them given the license fuckery.
Like IBM with Redhat, they tried to overstep their position and exert undue control, so the community rejected their status as BDFL. Now IBM has Rocky and Liberty to deal with, and Redis has Valkey rapidly gaining credibility as the go-to solution. The social contract of FOSS goes both ways, and greed is righteously punished with the loss of power.
This was the obvious play when the original Garantia Data-now-Redis Labs went on their acquisition spree and subsequent buttoning up of the ecosystem via their cloud/hosted enterprise offering even before 2013 and then taking the Redis name for themselves. I believe it was only slowed down because of the community aspect pushing back against the stupid licensing changes of it, otherwise they would have gone full steam into it to get even more control. It's so sad to see a good product like that be hung out on the whims of some scummy leeches.
You can enforce by requiring an agreement, and making the terms reasonable enough that an OSS project can agree to them (this isn't entirely easy, but it's not that hard either).
Also, there's the thing where these other projects have been using the name since forever, so there's already a history of non-enforcement.
The best time to start enforcing their trademark (from a legal perspective) was ten years ago. The second best time is now. Negotiating many agreements with many OSS projects is a lot of legal work which they may not want to pay for. And none of this undermines my original point that their primary motivation is probably just trademark preservation. That they did a poor job previously doesn’t mean they aren’t trying to do better now, and that they could use less aggressive methods doesn’t mean they aren’t primarily focused on trademark preservation. It’s the simple explanation. Occam’s razor.
What does a car do that you can't trivially do with something else ?
Like a boat or an airplane ?
You simply cannot compare an in-memory Key-Value store, to a RDBS. And that's on a conceptual level. If you then consider performance, network latency and the guarantees, we're absolutely comparing airplanes to donkeys: You wouldn't use an airplane to climb up a rough mountain, and you certainly wouldn't prefer a donkey to travel to the other side of the world.
That is a very bold claim. A message queue can have many different requirements: Some message queues tolerate lost messages, others don't. Some require a notification mechanism (i.e. a push model instead of a polling model). Other require precise management of dead letters.
All this to say that, yes, you theoretically CAN use SQLite as a message queue. But atomic operations are usually WAY lower on the "must have" list for a message queue than other parameters.
The easiest way you could verify this is by looking at the guarantees provided by some of the most known MsgQueues.
Looks like Mysqlization tragedy, the best would be when FOSS maintainers join their corporate agile sprints to meet roadmap demands, with naming rights gun aimed under table lol.
First time I hear about Valkey, it acts as MariaDB in this drama. /s
Provide a payment infrastructure to OS developers, otherwise you are left without options.
Anyone can fork your hard work,
anyone with more resources than you can push you out of your own work and expropriate it,
noone pays for your hard work.
When someone realizes this and changes the license the hardhat OS advocates turn against you and fork your hard work, essentially trying to deprive you of any income.
Who on fucking earth thinks this is a good model? And who benefits, anyone dared to ask the question?
For a starter: Force github to implement proper monetization options, that is visible and allows the creators to ask a sum for their work.
Your work will remain open source, buy you must pay a low sum. If anyone is against this, then go, create your own project and don't use publicly available code.
Plenty of OSS developers/companies have survived and make a living out of their work. You just have to provide enough value on top of your project with your closed-source services that people are inclined to pay for them, instead of hacking up their own solutions. Or consultation.
No one forces you to make your project OSS. It's a trade-off you should consider carefully before you do. But because people keep doing it, it's clear there are massive benefits to it.
And many open source projects are just modules extracted from the company code base to gather developer mindset and to reduce costs by harnessing outside developers. It's, as such, a fine model as it is. Rug-pulling and exertion of power through stupid lawyer BS is what kills those projects if anything. If people are not paying for your OSS addons, perhaps they aren't that good after all.
My theory is that open source was never intended to be based on free work.
Rather, it is an open collaboration between companies, which realize that it will lower their cost, while not being a competitive advantage for either party. It improves the profit margin for everyone.
What you're proposing is to convert Github into a small-business platform that sells "open source" software products, which is entirely different.
Twitter essentially hasn't loaded for me since Musk took over, so: https://archive.is/c5lnW or,
> Looks like Redis is trying to take over the all the OSS Redis libraries.
> Jedis, Lettuce, and redis-py are down, they are now threatening redis-rs (link in reply).
There's no link in the post, and I don't have a link to the referenced reply, but I'm guessing the link is https://github.com/redis-rs/redis-rs/issues/1419
Use xcancel instead of archive.is and you will see the replies.
https://xcancel.com/TomHacohen/status/1861137484249252093
Or nitter
https://nitter.poast.org/TomHacohen/status/18611374842492520...
Isn't xcancel.com a Nitter instance?
It is.
Damn it, thank you for the link, sir.
Can we please make this SNS disappear?
What does it mean by "redis-py" is down? It's still on Pypi?
I think they meant to say the maintainers for redis-py, lettuce, and jedis have already relinquished control to Redis Inc. The libraries are still online, but their repositories have been moved to the official Redis organization and are "controlled" and "owned by Redis" according to an email quoted in the linked Github Issue.
Thanks, that makes more sense of the "down" thing.
I didn't realize Jedis was an official Redis client now, looks like it has been since 2020 without any major impact to the development.
I guess they are unlikely to add any features to support alternative servers.
I worked in a huge docs website a while back and I’ll never forget the day I saw a request from a Redis lawyer that insisted we called it:
Redis™*
*Redis is a trademark of Redis Labs
The dumb thing is they insisted the footnote be on the first webpage Redis™ was displayed, as-if I had control over the order of the pages people viewed on the website.
Since then I’ve started removing Redis as a dependency from as many projects as I can.
Yup, technologically illiterate and developer hostile behaviour like that is your canary in the coalmine going belly up. If you stick around after that, you have only yourself to blame for what follows.
I'm obviously not a lawyer, but from what I can see, a project like redis_py is way older than any trademark registration from Redis lab.
If I'm not wrong looking at the trademark, database, they have registered them recently (2018,2019, and a lot very recently 2023) in a very predatory move.
In my opinion, this usage should be considered legitimate based on prior art or something like that.
Anyway, in all case I hope that projects will not give up to this asshole move from redis lab.
Maybe it might be possible to rename the project, but also forbid redis lab to use the name of the projects it had targeted. Like a poison pill. I think that it is legitimate for them to consider that an usage of name s like redis_py by Redis would be counterfeiting of well known brand of these projects. In the end, it might be assumed that it is the fault of Redis to have not enforced any copyright or branding on these projects for more than 10 years despite the fact that they couldn't have ignored their existence and success.
I'm even quite sure that we can find a publication of Redis lab with how-to use any of these open source projects...
> Anyway, in all case I hope that projects will not give up to this asshole move from redis lab.
How much are you willing to pay to finance their attorneys?
Legal bullying like this does not necessarily indicate a willingness to actually sue.
Whether somebody's comfortable finding out is, of course, a different matter.
Probably something up to 50$. Redis lab is not that big as a company so they might not be able to hire top notch lawyers.
But regarding the history of free software that is made of individuals fighting the overreach of predatory corporations, I will be very saddened if everyone of all these projects would give up...
Maybe the fsf or software conservancy can assist some of the projects if they were to ask it?
> Probably something up to 50$.
$50 might get you 5 minutes of an attorneys time. For someone to fight this in court will in the hundreds of hours of billable time. None of these open source client library maintainers is likely going to be able to fund that.
> Redis lab is not that big as a company so they might not be able to hire top notch lawyers.
Redis is a company with a billion dollar valuation that has raised hundreds of millions of dollars, they can afford good lawyers.
Prior art for redis_py clearly belongs to Redis, regardless of when the trademark is actually filed. This is because Redis had to have existed before redis_py.
There are organisations like the Software Freedom Conservancy who, even if they themselves can't help, can probably help you try and find somebody who can.
I'm not suggesting any particular maintainer *should* decide to fight, or even consider whether they want to try - that's very much up to them - but assuming that fighting is untenable in advance without exploring options is something I'd consider an unnecessarily unfortunate outcome.
Is this going to become a trend? Project starts making trademark claims years after a downstream library has been using it without issue. Bit rich, given how accessible libraries can bootstrap an initial product ecosystem.
Is it ever safe to use a trademark name? Will all packages now have to be generic enough to imply the product for which they connect? It”s not a Java/AWS/Redis/Wordpress/Shopify extension, but a thingbob, which just so happens to connect to That-Which-Shall-Not-Be-Named.
You can mention a trademark. You can't use the trademark in the name of your own product. Even in cases where you're legally OK (e.g. mentioning your product is better than <trademark>, or is made for <trademark>) they can still sue you, though it'll be much easier to win the case.
This is called nominative fair use. But Redis's own guidelines for nominative fair use acknowledge that there are cases where you can "use the trademark in the name of your own product": https://redis-doc-test.readthedocs.io/en/latest/topics/trade...
>you may only name it "XYZ for Redis™"
However, I'm not sure where courts would land on a name like "redis-rs". There are arguments on both sides, and I'm not sure if there's enough case law to make the legal outcome clear.
I'd certainly think "redis-rs" is liable to cause confusion as being an official Redis product.
I suspect Redis is going to pull in the various popular Redis libraries under their umbrella and make them incompatible with Valkey. It seems like a defensive move against Valkey.
If I was in charge of any of these libraries, I would replace any mention of Redis with Valkey and move on.
Sure. You can do that. But the community needs a single standard. Open standard that all in-memory data stores like Redis, ValKey, DragonflyDB, KeyDB, Kvrocks will support. Otherwise you will end up in a vendor lock.
> Otherwise you will end up in a vendor lock.
This would be in the interests of the current market leader though - eg. Redis.
And what prevents valkey devs from maintaining their compatibility layer to accommodate those changes?
that's just asking to waste your life forever catching up.
Eternal vigilance is the price of liberty or something.
I switched to Valkey, and not looking back. It's actively maintained and compatible with Redis so far.
Kvrocks (under Apache foundation umbrella) is another interesting Redis-compatible alternative, this one is targeting memory and SSD/NVME storage since it is implemented on top of RocksDB. So depending on your needs it may be a more efficient choice compared to Valkey.
That's a cool one, thanks, I find RockDB-like solutions a very nice niche for performance computing (e.g. ClickHouse, erigon), but I think with all the backing by huge corporations, Valkey will be the replacement for Redis.
This serves as a lesson when naming projects -- call it something different to the core product. If you must have a package called ${name}-rs make it a wrapper around your repo.
When I started using Redis, Redis was a true open source project. There was no company yet.
suffix the language name plus name would give us redis-py.
Which in turn would create the filename redis_py.py
Which would immediately raise the question why not a dot.
redis.py
> There are only two hard things in computer science: cache invalidation and naming things.
Lawyers won't allow any of those. It has to be something like Client for Redis™ in Python™.
Trademarks don’t (usually) cover file names and function names. See for example the classic GNU Emacs tetris.el debacle…
And 3. Off by one errors.
You can’t do that and expect it to be used. When someone needs a redis client library they search for ‘redis’ not ‘thingamabob’.
Lots of projects do it. I look at starts and update frequencies. My toml parser is called taplo, for example.
Since search isn't limited to names, but includes description, they will still find it
All true, but lets be honest: For the technical users searching a library, nothing beats having The Keyword being part of the name.
There is a new comment by antirez in the past few minutes: https://github.com/redis-rs/redis-rs/issues/1419#issuecommen...
Trademark claims are also defended by the otherwise useless PSF. Here GvR objects to someone continuing Python-2 uńder the name "python-2.8" as well as "py28". Informally, bringing in lawyers was threatened at one point:
https://github.com/naftaliharris/tauthon/issues/47
The project was renamed to "tauthon".
Oh dear.
I see Redis Inc. have decided to go full Nagios.
Never go full Nagios.
Though admittedly Nagios' attempt to pull similar assholery wrt CPAN did end up being a source of some amusement to me: http://p3rl.org/Nagios::Plugin
I hope the http://crates.io team react similarly.
So, after reading the Github[1] comments, Redis needs a rust client library. It's hilarious to watch the legal and PM figure this out. No strategy. Just trying to solve "enterprise" needs because I'm sure someone asked for a rust client they'd pay over the moon for.
[1] https://github.com/redis-rs/redis-rs/issues/1419
I would love for someone to fight out these bullshit trademark claims. Are you seriously going to win against a crate author for using your business name after letting them do it for 10 years and even featuring and recommending them on your own website?
(Referencing https://github.com/redis-rs/redis-rs/issues/1419 )
Who is going to pay to defend? The rust author is making it clear they want no part of this legal mess. Which is going to be true for nearly any open source contributor. The only people with legal representation are going to be the VCs looking for value extraction.
Yes, that’s the problem. I understand the perspective of the maintainers but it leads to a situation where the large corporation can do illegal stuff and bully people around because it’s too much effort to fight back.
That's basically our entire society...
Hey welcome to the legal system!
FSF?
I've sometimes wondered if it would be possible to have a voluntary organisation funded by developers and other tech folks whose sole purpose is to fight legal battles against companies and other organisations looking to bully contributors on dodgy legal grounds.
I'm sure a lot of these legal threats would go away very quickly if the possibility of a large scale legal battle was on the horizon every single time. Or at least, they'd be a lot more willing to settle quickly if their opponent was fairly evenly matched resources wise.
https://sfconservancy.org/ ?
Let them fail. Does anyone today remember SCO chasing virtually everyone on Earth after unix licensing violation and copyright infringement claims? There is no shortage in KV stores with similar properties so if redisinc. wants to continue their path without delivering actual value to the customers especially to enterprise customers willing to spend money they know exactly where they will land in several years from now.
I have to thank Redis Inc. for bringing on this shitshow. I had no idea about Valkey and now I do. Corporate hats should consider the loss of general trust and negative marketing their actions produce.
And yeah, I wish FSF or some other, strong and well-funded org would be more "broken glass" and take issue with even the tiniest of legal fights over OSS.
Barbra Streisands's Effect?
This looks a lot like the wordpress issue… i think the oss wars of the 2020s will be fought around trademarks?
One more reason to move to valkey i guess.
That issue looks reasonable, there isnt anything about trademarks there, why bring it up?
It looks to be about providing support to paying users of Redis in a library that doesn't have much goals.
And they offered to pay for it to do it.
Do ctrl+f for „trademark“, there are currently 12 occurrences. Here’s what the Redis representative wrote:
> These are not my words nor the message I want to convey. In a private call, I said that it feels strange to use the brand name for a library that keeps supporting all the forks indefinitely and does not grant compatibility for the product itself. I also said that it does not make a lot of sense to pretend to rename a crate and replace it with the same software or an alternative library, thus creating confusion and frustration. However, companies do consider protecting their trademarks where their reputation is challenged. This does not need any explanation. In my private call with Armin, I highlighted that. Of course, there is no plan to take action regarding the trademark, but to find a reasoned and agreed-upon way forward to avoid introducing another Redis Rust client library where changes, improvements, and innovation can be introduced.
So basically the point is to make thinly veiled threats of copyright enforcement to „agree to a solution“.
The Mafia also agrees with you on a payment plan after highlighting that a fire would be a shame in your establishment.
He published the emails where it says they want to own the library so they can set the road map for it, they're offering to pay for it.That quote even says they're not looking to take action over the trademark.
Your response is alittle over emotional.
I just quoted the part where the Redis representative describes a private call. How does the content of the emails negate that?
> That quote even says they're not looking to take action over the trademark.
Yeah, he’s basically saying „also, we think you’re infringing but we won’t have to enforce that, right? I’m sure we can come to an agreement, wink wink“.
Maybe I’m reading this completely wrong but this reminder is basically a threat to sue you to death if you don’t comply.
> Maybe I’m reading this completely wrong
IMO you are because your attempting to "read between the lines" and add things that aren't there while ignoring things that contradictory to what this hypothetical lawsuit would include.
He says they don't want to cause a mess of renaming, or forking because it would be worse off and confusing for everyone
For users of Node.js, Java, Python and very soon Go - Valkey-Glide: https://github.com/valkey-io/valkey-glide It will stay out of the hand of Redis, can promise that, and if you use another key-val db and want to contribute to its compatibility with it, we will be happy, come talk to us.
While it's difficult to compare the pettiness, this is definitely sillier than the WordPress thing as far as trademark enforcement goes.
Trademark-as-a-Weapon is becoming too common now. Are we going to be forced to use legally distinct (and confusingly different) names for all our libraries now to defend against this nonsense? Pick projects based on the brand protection posture of the licensors? What a waste of time.
At least they offered to buy it.
Is that any different from someone offering to buy a browser extension to add malware?
the library would most likely remain OSS even after the take over
What makes you think that?
Yeah, if they don’t add malware. I doubt redis will add malware to specifically their rust client.
OSS authors can add malware too.
The risk here is they intentionally make it incompatible with Valkey, similar to how Elasticsearch client libraries complain when a newly added slogan header is missing
That's almost guaranteed. Valkey is going to have to fork and maintain their own client libraries.
I don’t think they could do that in a way that would prevent a fork from undoing it.
What are they hoping to get out of it? Wouldn't this just inconvenience their own users?
Consolidation of control and brand recognition. As soon as Redis revealed their hostile intentions, forks split off. Valkey seems to be the favored one, and in my direct experience is what projects are using moving forward. If we don't intend to use Redis' commercial services we have no reason to use them given the license fuckery.
Like IBM with Redhat, they tried to overstep their position and exert undue control, so the community rejected their status as BDFL. Now IBM has Rocky and Liberty to deal with, and Redis has Valkey rapidly gaining credibility as the go-to solution. The social contract of FOSS goes both ways, and greed is righteously punished with the loss of power.
This was the obvious play when the original Garantia Data-now-Redis Labs went on their acquisition spree and subsequent buttoning up of the ecosystem via their cloud/hosted enterprise offering even before 2013 and then taking the Redis name for themselves. I believe it was only slowed down because of the community aspect pushing back against the stupid licensing changes of it, otherwise they would have gone full steam into it to get even more control. It's so sad to see a good product like that be hung out on the whims of some scummy leeches.
I’m sure most of it is trademark preservation. Trademarks are enforce it or lose it.
You can enforce by requiring an agreement, and making the terms reasonable enough that an OSS project can agree to them (this isn't entirely easy, but it's not that hard either).
Also, there's the thing where these other projects have been using the name since forever, so there's already a history of non-enforcement.
The best time to start enforcing their trademark (from a legal perspective) was ten years ago. The second best time is now. Negotiating many agreements with many OSS projects is a lot of legal work which they may not want to pay for. And none of this undermines my original point that their primary motivation is probably just trademark preservation. That they did a poor job previously doesn’t mean they aren’t trying to do better now, and that they could use less aggressive methods doesn’t mean they aren’t primarily focused on trademark preservation. It’s the simple explanation. Occam’s razor.
You mean the trademark that wasn't filed until 2018[0]?
That is part of it....these projects are older than the trademark, by quite a bit.
[0]: https://trademarks.justia.com/878/05/redis-87805452.html
I don’t know why you think you are dunking on me here, I’m not saying they are right, just what their likely motive is.
It wasn't intended as a dunk. Was just trying to add a bit more context to the situation.
What does Redis do that you can’t do trivially with something else?
Postgres or SQLite for example.
It's akin to asking the following:
What does a car do that you can't trivially do with something else ? Like a boat or an airplane ?
You simply cannot compare an in-memory Key-Value store, to a RDBS. And that's on a conceptual level. If you then consider performance, network latency and the guarantees, we're absolutely comparing airplanes to donkeys: You wouldn't use an airplane to climb up a rough mountain, and you certainly wouldn't prefer a donkey to travel to the other side of the world.
SQLite has the option to run on a pure in-memory database.
SQLite is not a server, you can't use it as a shared cache or job queue or pubsub mechanism.
On the other hand Redis is a bad database so I would say the use cases don't overlap.
You can use SQLite as a queue it is better suited to the task than redis.
SQLite supports atomic operations - the key requirement of a message queue.
That is a very bold claim. A message queue can have many different requirements: Some message queues tolerate lost messages, others don't. Some require a notification mechanism (i.e. a push model instead of a polling model). Other require precise management of dead letters.
All this to say that, yes, you theoretically CAN use SQLite as a message queue. But atomic operations are usually WAY lower on the "must have" list for a message queue than other parameters.
The easiest way you could verify this is by looking at the guarantees provided by some of the most known MsgQueues.
Atomic operations are the baseline requirement for implementing a queue.
This is what ensures only one reader gets a message. SQLite has it, redis does not.
> is better suited to the task than redis
That is a claim not an argument, and again it can't be a shared queue because no networking.
Or do you believe Redis doesn't have atomic operations?
Sic transit gloria mundi
Looks like Mysqlization tragedy, the best would be when FOSS maintainers join their corporate agile sprints to meet roadmap demands, with naming rights gun aimed under table lol. First time I hear about Valkey, it acts as MariaDB in this drama. /s
This open source mess must be changed.
Provide a payment infrastructure to OS developers, otherwise you are left without options.
Anyone can fork your hard work,
anyone with more resources than you can push you out of your own work and expropriate it,
noone pays for your hard work.
When someone realizes this and changes the license the hardhat OS advocates turn against you and fork your hard work, essentially trying to deprive you of any income.
Who on fucking earth thinks this is a good model? And who benefits, anyone dared to ask the question?
For a starter: Force github to implement proper monetization options, that is visible and allows the creators to ask a sum for their work.
Your work will remain open source, buy you must pay a low sum. If anyone is against this, then go, create your own project and don't use publicly available code.
Plenty of OSS developers/companies have survived and make a living out of their work. You just have to provide enough value on top of your project with your closed-source services that people are inclined to pay for them, instead of hacking up their own solutions. Or consultation.
No one forces you to make your project OSS. It's a trade-off you should consider carefully before you do. But because people keep doing it, it's clear there are massive benefits to it.
And many open source projects are just modules extracted from the company code base to gather developer mindset and to reduce costs by harnessing outside developers. It's, as such, a fine model as it is. Rug-pulling and exertion of power through stupid lawyer BS is what kills those projects if anything. If people are not paying for your OSS addons, perhaps they aren't that good after all.
My theory is that open source was never intended to be based on free work.
Rather, it is an open collaboration between companies, which realize that it will lower their cost, while not being a competitive advantage for either party. It improves the profit margin for everyone.
What you're proposing is to convert Github into a small-business platform that sells "open source" software products, which is entirely different.