A disagreement over the PostgreSQL trademark [LWN.net]

This release on PostgreSQL.org describes an ongoing disagreement over the PostgreSQL trademark:

In 2020, the PostgreSQL Core Team was made aware that an organization had filed applications to register the 'PostgreSQL' and 'PostgreSQL Community' trademarks in the European Union and the United States, and had already registered trademarks in Spain. The organization, a 3rd party not-for-profit corporation in Spain called 'Fundación PostgreSQL,' did not give any indication to the PostgreSQL Core Team or PGCAC that they would file these applications.

to post comments

A disagreement over the PostgreSQL trademark

Posted Sep 13, 2021 21:11 UTC (Mon) by flussence (guest, #85566) [Link] (1 responses)

This is a career-ending grift for the guy who started it. It's going to be a huge waste of resources for everyone involved, and even if PostgreSQL lets him off as lightly as they say they will, nobody will trust him within a mile of other FOSS projects afterwards.

A disagreement over the PostgreSQL trademark

Posted Sep 13, 2021 22:57 UTC (Mon) by chris_se (subscriber, #99706) [Link]

Yeah, he's actually based his professional career around PostgreSQL (giving talks at conferences, etc.). The blog post by the PostgreSQL core team is actually the first search result if you type in his last name and PostgreSQL. Talk about biting the hand that feeds you...

And it's not like he didn't have a lot of chances to remedy this before it got to this point, but apparently he doubled and tripled down.

I wish the PostgreSQL team the best of luck in resolving this as painlessly and quickly as possible.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 0:54 UTC (Tue) by NYKevin (subscriber, #129325) [Link] (23 responses)

Here's the view from the other side: https://postgresql.fund/blog/postgres-core-team-attacks-p...

I know very little about any of this, so I'm reserving judgment for now, but I find it hard to believe that this guy is some kind of corporate false flag. OTOH, registering a competing trademark puts PGCAC in a really uncomfortable position, as failing to object now could harm the defensibility of their trademark later. His surprise at their opposition is itself surprising, frankly, and suggests to me that he doesn't have an adequate understanding of trademark law to be doing this in the first place.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 2:31 UTC (Tue) by Paf (subscriber, #91811) [Link]

This doesn’t seem like much of a both sides situation - it’s extremely clear that this person is trying to squat on Postgres trademarks which are clearly not theirs. The best case scenario is this person is naive/foolish, etc.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 6:11 UTC (Tue) by gfernandes (subscriber, #119910) [Link] (14 responses)

That is definitely from someone either legally naive, or intentionally making a grab for trademarks they have no right over.

There is no "distributed" trade mark.

I do wonder if that person is willing to put up his house for distributed ownership. And I wonder if he understands the dilution of control such an ownership might result in.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 7:11 UTC (Tue) by chris_se (subscriber, #99706) [Link] (13 responses)

> There is no "distributed" trade mark.

In fact, the PostgreSQL core team have to do something here, or they might even lose their trademark. Trademark law is something that is inherently opposed to the "distributed" concept - there's something called "trademark dilution".

A disagreement over the PostgreSQL trademark

Posted Sep 17, 2021 10:21 UTC (Fri) by tialaramex (subscriber, #21167) [Link] (12 responses)

No. "We have to do something" is a myth, a myth which conveniently excuses stuff like the Mouse going after artists and children because of some reasoning like "Well they had to, if they let one little girl dress as Cinderella without an official Disney-branded costume then magically all Pixar movies are Public Domain" when in reality Disney is simply litigious because they have the money and you don't.

Trademark owners can sit back and do _nothing_. Infringement does not, in fact, get rid of your trademark. You probably _should_ sit back and do nothing in most cases. This is probably not one of those cases because it's intended to challenge the existing PostgreSQL core team, but in general you do not risk "losing your trademark" simply because you don't insist on chasing everybody who uses the mark without permission.

Lack of *use* eventually gets rid of your trademark, but that's fine. If PostgreSQL doesn't release any software for 50 years, presumably we're all OK with somebody else using that name for new software even if it's a database.

Dilution only affects *very* famous brands, things you could pull ten random people off the street and at least nine would recognise. Think Disney, Nike, Walmart, Coca-cola - not PostgreSQL, Docker, or even Linux. Dilution is the argument that any mention of or even allusion to this very famous mark affects the "real" user of the mark. Coca-cola T-shirts seems like it's obviously from the Coca-cola corporation, it would be _weird_ if it's actually some outfit from South Dakota with four employees - even though the famous brand is most associated with a beverage, not clothing. Likewise you'd assume Nike webcams were from the company behind the famous clothing/ footwear brand, and not some garage startup in San Francisco. But again, they don't have to sue, they just have the option and in most cases being so famous comes with the money to afford to sue early if they want to.

A disagreement over the PostgreSQL trademark

Posted Sep 17, 2021 12:11 UTC (Fri) by anselm (subscriber, #2796) [Link] (4 responses)

because of some reasoning like "Well they had to, if they let one little girl dress as Cinderella without an official Disney-branded costume then magically all Pixar movies are Public Domain"

That would be highly unlikely given that trademark law and copyright law are entirely different ball games.

What companies try to avoid are situations where words like “to google” mean “search something on the Internet” rather than “use specifically the Google search engine to search something on the Internet”. It is often a consequence of having been too successful and become what is affectionately called “a household name”. The Hoover, Kleenex, and Xerox companies, among others, also have sad tales to tell about this. By coming down like a ton of bricks on presumptive diluters of its trademarks, the Disney corporation has – in spite of its early and sweeping success in the area – so far managed to avoid a world where the word “disney” is synonymous with “animated feature film”, at the cost of making themselves unpopular with third-party purveyors of Cinderella costumes and their customers. We may not like them for it but it's how their legal departments roll.

Disneyfication

Posted Sep 22, 2021 21:55 UTC (Wed) by tialaramex (subscriber, #21167) [Link] (3 responses)

Nah, Disneyfication still has exactly the sort of unpleasant meaning the Disney corporation would prefer was unavailable, but while they could sue people for using that word (you can sue people for anything whatsoever) they would clearly lose, because the most plausible targets are big media empires and (aside from the ones actually owned by Disney) those are no strangers to frivolous lawsuits and would defend themselves. If you go after say The Atlantic for writing about Disneyfication, they are going to hire people to defend the suit and win. e.g. "How the Olympics got Disneyfied" is still right there for you to read.

Likewise, US corporations once upon a time threatened many fledgling UK computer magazines demanding they write ® everywhere because the names of all their computers, software etc. were legally registered trademarks. The magazines, and their publishers, and the publishers' lawyers, told them to fuck off. Actually this wasn't so long after 1971, so perhaps they told them “We refer you to the reply given in the case of Arkell v. Pressdram” which has the same meaning.

You're correct that trademark and copyright are separate, and yet this type of muddled thinking is nevertheless all too common.

Disneyfication

Posted Sep 23, 2021 8:58 UTC (Thu) by anselm (subscriber, #2796) [Link] (2 responses)

Disneyfication still has exactly the sort of unpleasant meaning the Disney corporation would prefer was unavailable,

If you hold a trademark on a term that doesn't mean that term becomes “unavailable” to anyone but you. It just means that nobody but you gets to sell something that – by virtue of what it's called or looks like – could be mistaken (by reasonable customers) to be your product.

“Disneyfication” is not something that the Disney corporation is marketing. Hence there is no risk of someone else's product to be mistaken for “Disneyfication”. The Disney corporation legal team is probably not enthusiastic about the existence of the term given that it is not exactly complimentary, but trademark law gives them no leverage to outlaw it for everybody else (or, as you say, they could try but it would be a waste of time, money, and effort), and they know it. Having said that, even if the Disney corporation had a trademark on the term “Disneyfication”, that wouldn't prevent the likes of The Atlantic from writing about it – it would just prevent other companies from marketing a similar product under the same name.

The problem Google, Hoover, and friends have is not that purveyors of other search engines or vacuum cleaners are trying to pass off their own offerings as Google/Hoover/… imitations, it's that they're so successful in the market that the general public is using “hoover” to say “vacuum the carpets” rather than “use specifically a Hoover brand vacuum cleaner (and not, say, a Samsung, Roomba, or any of the other vacuum cleaner brands commonly available to customers) to vacuum the carpets”, which if the word has to exist at all is the usage that the Hoover corporation would vastly prefer. Unfortunately for these companies, that sort of thing is very hard to police. Marketing departments everywhere are trying to navigate a treacherous path between the highly desirable goal of making their product's name very well-known among the public and the very undesirable effect of making it a synonym for the entire market segment; this is why they earn the big bucks.

Disneyfication

Posted Sep 23, 2021 11:47 UTC (Thu) by tialaramex (subscriber, #21167) [Link] (1 responses)

But this completely undercuts your earlier point. All the corporations would dearly like to control the narrative, and indeed they all publish "official" trademark rules saying you mustn't do any of the things they don't want you doing. But those aren't the law.

Disney can't stop me writing about Disneyfication of the education system, Google can't stop me writing that people should stop Googling my name. Both could sue me (but that might be a bad idea) of course, but the law is clear in both cases. Far from being somehow a magic ingredient that has prevented words like "Disneyfication", or because of an imaginary obligation to do so, Disney sues trivial infringers because it can, because it is rich.

Disneyfication

Posted Sep 24, 2021 10:06 UTC (Fri) by anselm (subscriber, #2796) [Link]

Far from being somehow a magic ingredient that has prevented words like "Disneyfication", or because of an imaginary obligation to do so, Disney sues trivial infringers because it can, because it is rich.

Yes, and because they apparently enjoy coming across as assholes.

The word “Disneyfication” isn't a trademark issue at all, because nobody has a trademark on “Disneyfication” and even if they did, it wouldn't prevent anyone from talking about the phenomenon. I've never claimed that corporations have an “obligation” to sue. What corporations do want (apart from preventing competitors from selling products that pretend to be the original) is to avoid their product's name becoming an equivalent of the word “hoover” standing in for “any vacuum cleaner regardless of actual brand”. Lawsuits (or the threat of such) are one tool that is available to them in the endeavour to achieve this; whether these will get them anywhere is generally up in the air, and there is also a PR angle to be considered.

This is not entirely theoretical. For example, here in Germany Google successfully leant on the editors of Duden (the leading German-language dictionary, comparable to the OED or Merriam-Webster in the USA) to change the entry for googeln from “use a search engine such as Google to search something on the Internet” to “use Google® to search something on the Internet”. Whether this will get Google anywhere in their attempt to rescue “googling” remains to be seen, but it looks like they're not going down without a fight.

A disagreement over the PostgreSQL trademark

Posted Sep 17, 2021 12:27 UTC (Fri) by chris_se (subscriber, #99706) [Link]

I want to make clear that I said they have to do something *here*, I didn't say any trademark holder has to do something in all cases.

While you are completely correct that enforcement is not required around edge cases or people doing minor things, you can actually lose your trademark if you don't enforce it at all. I agree with you that "having to enforce it" is often used as an excuse to be horrible to other people, and I despise that kind of behavior. But in this specific case the other side registered a trademark of the same name (in a different category, but clearly related to the same thing, unlike e.g. the detergent when it comes to the Linux trademark), and that makes a huge difference. If Pepsi started releasing their soda under the Coca-Cola brand without permission, and Coca-Cola did nothing about it for years, they would probably lose their trademark at some point. That's different from a 10 year old kid selling their home-made soda under the name Coca-Cola during a public event, where Coca-Cola would definitely not lose their trademark if they didn't enforce it.

A disagreement over the PostgreSQL trademark

Posted Sep 17, 2021 19:13 UTC (Fri) by calumapplepie (guest, #143655) [Link] (5 responses)

No.

There are multiple types of trademark, and there are multiple ways to infringe. Trademark dilution is only one of the ways that infringment can occur. It is, as you said, when the trademark is used in a completely different industry than the original, but when the original is so big that it is in the public mind. This isn't even stuff like Coke t-shirts: since Coke does, in fact, sell t-shirts, using their logo in your own would be traditonal infrigment. Dilution is just any appearance of the mark in a way that harms the brand (other than nominative usage: a news article about Coke being made with the blood of innocents can use the logo without trademark infringment, though libel is possible if the article didn't actually uncover any innocents). For instance, if an actor in an adult film is wearing Nike shoes, that is dilution: unlike with traditional infringment, you don't need to show confusion.

Any (and I mean ANY) use of your trademark on a product similar to that provided by you MUST be sued as infringment. That is because the core purpose of a trademark is to uniquely identify a product as coming from a specific location. If a customer looks at a product, and incorrectly decides that it is yours, then you are required to sue the creator of that product. If you tolerate such products, then you lose the trademark: products with the mark can no longer be associated with you anyways, so there is no point in giving you the ability to control who has the mark.

This isn't limited to the exact list of offerings by your company (though it is more limited than the dilution): for instance, if I open a store that repairs New Balance shoes, I can't advertise myself as a "New Balance Repair Store", even though New Balance is a brand far smaller than the giants that dilution would apply to. "New Balance Health and Wellness" shop would not infringe, however: "new balance" is an english phrase, and so it doesn't get the same protections as other, more unique phrases (such as Kodak).

In this case especially, there is an obligation to sue in order to maintain the trademark. A reasonable person would conclude that a company entitled "PostgreSQL Services" (as opposed to one that simply advertises such an offering) is either offically part of the people who make PostgreSQL, or that PostgreSQL is a generic term for a general type of product. Thus, unless you want the trademark to become a generic term, you do have an obligation to enforce it or lose it.

(as the other commentator said, this is all completely distinct from copyright: trademark law is to a large degree a consumer protection law, in that it allows consumers to be confident that the products they are buying actually come from the company they are buying from. If the PostgreSQL trademark is lost, it's not a matter of the source code to Postgre becoming public domain, but rather the ability for random people to legally (say) distribute malware under the logo)

A disagreement over the PostgreSQL trademark

Posted Sep 17, 2021 20:02 UTC (Fri) by mpr22 (subscriber, #60784) [Link] (1 responses)

> For instance, if an actor in an adult film is wearing Nike shoes, that is dilution

I would expect that to be considered "de minimis" unless the camera crew was making a great effort to keep the Nike branding visible / refer to it in the script / otherwise draw attention to what brand of shoes the cast are wearing and thereby associate "Nike" with "naughty movies".

A disagreement over the PostgreSQL trademark

Posted Sep 17, 2021 21:07 UTC (Fri) by Wol (subscriber, #4433) [Link]

But if they are genuine real Nike's, how on earth can this be considered a trademark infringement!?

Nike may not wish to be associated with that sort of film, but there's no abuse of trademark. Nike would have to sue for "bringing the brand into disrepute" or somesuch.

It's like copyright has been diluted where anybody can pay a licence fee and use any music, and the artist has to object if they hear of and don't like the use. But they can't sue for breach of copyright because a licence fee has been paid.

Cheers,
Wol

A disagreement over the PostgreSQL trademark

Posted Sep 22, 2021 21:30 UTC (Wed) by tialaramex (subscriber, #21167) [Link] (2 responses)

“Any (and I mean ANY) use of your trademark on a product similar to that provided by you MUST be sued as infringment. [...] If you tolerate such products, then you lose the trademark”

Imagine if calumapplepie was right about this. and thus that every major brand in the world had "lost the trademark" presumably somehow across all jurisdictions -- because of course in practice big brands cannot and don't sue "any use of your trademark" but (even for the litigious ones like Disney) only whichever targets attract their legal staff for whatever reason.

It's absurd once you write it down like that, isn't it? But that's what this myth would have you believe.

But I'm sure calumapplepie isn't just a keyboard warrior here to insist upon their preferred myth but an actual lawyer, and so I look forward to them giving us some credentials and telling us where they're licensed to practice trademark law so that we can go look at local statutes to identify where this "sue or lose it" rule is, if anywhere, in their local law. Wait, actually if they're really a practising trademark lawyer they'd *know* where to look and can just tell us right?

A disagreement over the PostgreSQL trademark

Posted Sep 22, 2021 21:58 UTC (Wed) by Wol (subscriber, #4433) [Link] (1 responses)

Like all legal stuff it's fuzzy.

But the *point* of trademark law is to prevent "passing off". If someone else uses your trademark in a manner which is likely to confuse your customers, and does it without explicit permission, then that is the scenario where you really have to sue.

Otherwise the courts will conclude you don't care. And that WILL cost you the trademark.

(The fuzziness is, of course, deciding what will or won't confuse your customers ... and in this case, I think it seems pretty clear.)

Cheers,
Wol

A disagreement over the PostgreSQL trademark

Posted Sep 23, 2021 10:23 UTC (Thu) by tialaramex (subscriber, #21167) [Link]

You don't have to sue. That's the entire thesis here. You're going to keep going around this loop because you're convinced of a myth.

If you subsequently file suit, the court won't "conclude you don't care". You filed a lawsuit, so obviously you care.

Estoppel means that you can't win if your argument is that behaviour you previously ignored is suddenly now something you cared about and the court should retrospectively fix it. Time's arrow doesn't work that way, and the principle of estoppel reflects that. But Estoppel is deliberately very limited, it's just to prevent the sort of obvious unfairness of e.g. waiting until somebody's use of your mark makes them a nice profit and then swooping in to claim it for your own.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 6:58 UTC (Tue) by GoodMirek (guest, #101902) [Link] (2 responses)

After reading the view of the other side at https://postgresql.fund/blog/postgres-core-team-attacks-p... it seems to me more complicated than just "someone is trying to hijack the Postgres trademark".

I am wondering whether there are any untrue statements in the presented view of the other side. Doing any fact checking of the claims made there seems hard for a simple end user.

The other side (Spanish non-profit) claims to pursue their actions in order to fix perceived issues with the current state of Postgres trademark handling and lack of registration.
For sure it is pretty unfortunate if they do that without any prior attempt to fix the perceived issues together with PAC and PEU.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 8:15 UTC (Tue) by oldtomas (guest, #72579) [Link]

"For sure it is pretty unfortunate if they do that without any prior attempt to fix the perceived issues together with PAC and PEU"

In my view this is exactly the relevant smell test. If the Fundación hasn't worked towards (and reached) a consensus with the PostgreSQL core team, they should back out *now*.

Going further is bad faith.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 18:33 UTC (Tue) by marcH (subscriber, #57642) [Link]

> After reading the view of the other side at https://postgresql.fund/blog/postgres-core-team-attacks-p... it seems to me more complicated than just "someone is trying to hijack the Postgres trademark".

tl;dr:

- A good question: Who is PostgreSQL?
- An... interesting" answer: me, so I registered the orphaned trademark and offered my "protection" without trying to talk to others about it first. Now the Core Team is launching an "unprecedented attack" on me, I wonder why?

Of course "me" is phrased as "the community" - on every side.

In the final conclusion:
> The Core Team should resign and call for open elections for a new Core Team.

No less. You can tell there's been some bad blood beyond this trademark issue.

The parallel between "distributed copyright" and "distributed trademark" is funny. It's all "Intellectual Property" after all? Can't wait for the "fork trademark" button on Github.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 7:45 UTC (Tue) by chris_se (subscriber, #99706) [Link] (3 responses)

> Here's the view from the other side: https://postgresql.fund/blog/postgres-core-team-attacks-p...

Yikes. Read the negotiation terms he sent to the core team - it appears he's attempting to force changes to the organizational structure of the core team of PostgreSQL. I have no idea how PostgreSQL core team works in detail, and it could very well be that there are things there that I wouldn't like (or maybe I'd be completely fine with it), but the answer to any real or imagined issues there is not to register a trademark and then force something.

What's next, I register a trademark "Gnome" in some other category and respond to requests by the project with the condition that they get rid of client side window decorations, just because I don't like them?

To me this is an awful "view from the other side", because it makes him appear much more unsympathetic to me. My impression after reading both posts is that the PostgreSQL core team remained vague in their statement in some places to protect his reputation at least to some extent, and he completely failed to recognize the olive branch they've extended in that manner.

I do agree that he appears sincere, but that could make the whole proceedings so much more painful should he end up viewing himself as a martyr.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 9:58 UTC (Tue) by intgr (subscriber, #39733) [Link] (1 responses)

> Read the negotiation terms he sent to the core team

Indeed. I dug a bit further, he has in the past also called for a change in the way the Core Team works.
https://postgresql.fund/blog/is-it-time-to-modernize-post...

Trying to force their hand via a trademark dispute seems misguided, almost guaranteed to fail and will just waste resources. Seems like a lose-lose situation. Did he make any attempt to raise these issues privately?

I think he raises some good points, particularly that the Core Team does not have a legal body and PGCAC, the most important organization, actually answers to its own board, not the Core Team.

And that transparency in these orgs seems lacking. Funnily the PostgreSQL Core Team page states "Starting in 2019, the Core Team are aiming to publish regular activity reports", and that was the first and only report ever published. (https://www.postgresql.org/developer/core/)

I'm an outsider so it's guaranteed that I'm missing some information.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 15:12 UTC (Tue) by kleptog (subscriber, #1183) [Link]

I don't disagree that there are possibly points of improvement. The most charitable explanation is that it is a well-meaning person going about it the wrong way.

But as for PGCAC being a problem and possibly going rogue, the answer to that is determined by the articles of incorporation and the by-laws (I think that's the correct term, here we call them statutes). AFAICT they are not on the website (here, they are public, not sure about Canada). Without them you can't really say what the board can and cannot do.

Trademarks & open source are tricky at the best of times. I see they use SPI for donations which is a much more mature organisation.

A disagreement over the PostgreSQL trademark

Posted Sep 14, 2021 22:21 UTC (Tue) by LtWorf (subscriber, #124958) [Link]

> What's next, I register a trademark "Gnome" in some other category and respond to requests by the project with the condition that they get rid of client side window decorations, just because I don't like them?

Oh… brb… got to register a trademark real quick!