According to Bryan Cantrill, you don't need to be open minded about Oracle. It's a waste of the openness of your mind. He says what you think of Oracle is even truer than you think it is. He believes there has been no entity in human history with less complexity and nuance to it than Oracle.
Bryan warns, "Do not fall into the trap of anthropomorphising Larry Ellison. You need to think of Larry Ellison the way you think of a lawnmower. You don't anthropomorphize your lawnmower, the lawnmower just mows the lawn, you stick your hand in there and it'll chop it off, the end. You don't think 'oh, the lawnmower hates me' -- lawnmower doesn't give a shit about you, lawnmower can't hate you. Don't anthropomorphize the lawnmower. Don't fall into that trap about Oracle."
Their stock is 50% higher than it was a year ago.
Not quite sure this is doing them damage.
Tesla stock is 63% higher than it was a year ago, does this prove that each and every decision their leadership made was helpful to the bottom line?
And anyone who is sympathetic to the request, knows that campaigning for the protocol break would require disrupting two or three levels of management above them, forcing powerful people to deal with something they don't care about. And that would be interpreted as wasting important people's time.
So the organization, as a decision making entity, is incapable of recognizing, much less considering, requests for an exception to default behavior.
I worked with a business that operated this way for many years. Even when there were overwhelming reasons to break process, the spark and tinder never got anywhere near each other.
Everyone between the spark and tinder empathized, talked to "somebody" to demonstrate they "tried", and to create an alibi for the inevitable "no" response that came next, while quietly doing everything they could to smother that spark, before it burned them.
> simply updating CDDL to allow integrating ZFS with GPL
That can't be done at this point. Owing to a decision that arose right here from a discussion on HN, the ZFS maintainers adopted a policy in 2016 to opt out of the CDDL's built-in "any subsequent version" clause for new source files:
~/scratch/zfs$ grep --exclude-dir=.git -Ire "Common Development and Distribution License" -A 2 | grep -ie "\(Version 1\.0 only\|\<only\>.*\<version\>\)" | wc -l
821
How about a simpler solution, just relicense everything to BSD / MIT.
ZFS can be run under Linux - combining the Linux kernel with ZFS is a collective work (collection) of two independent works.
(This doesn’t necessarily stop people, but it is read by Debian as “illegal enough” to warrant a splash screen on installing OpenZFS that you’re losing the right to redistribute.)
Oracle is one of the leading researchers in JIT compilers, garbage collectors, and language interpreters.
Interpretation on the fact and metric and the need to tell I leave up to you
My favorite other example of this is when I see a UI redesign that didn't actually benefit anyone and was more a style change than anything, and sometimes actively makes usability worse (cough Liquid Glass cough) In those situations I always think "well, some designers on staff needed to justify their paychecks".
https://www.investopedia.com/terms/p/principal-agent-problem...
E.g. you often imagine cases like a manager making a decision that causes a short term pop in stock price (and bonuses to the manager) to the detriment of the long term health of the company when thinking about the principal agent problem. In the cases I'm thinking about, though, it's more that people rarely can do nothing, even if that's sometimes the best thing to do. E.g. large companies need to have lawyers and designers on staff for lots of reasons. But sometimes there just isn't enough work for these folks to do (even if they need to be "warm" and ready when important work comes along). And if there isn't enough work to do, these people will find work to do.
This is another reason why I think that, even though layoffs are painful, having people "milling about" without clear direction and purpose is the worst for everyone involved. These people will just schedule meetings, insert themselves where it isn't helpful, etc., just to make it seem like they have a purpose.
This could be thought of as a "variant" of the principal agent problem I guess, but this instance of "idle hands are the devil's playthings" is different enough from the "standard" principal agent problem that I don't think it's helpful to conflate these two things.
It is possible to find work in a different area at the company for such in-between times.
For example at the company where I work, a (very capable) secretary whose original role was not needed anymore, but for who there existed a very role in the future was for the in-between time assigned to assist some other department in their reporting duties to regulating authorities.
They gain absolutely nothing by handing over the name and brand - in fact they lose valuable brand recognition.
Obviously most people in the industry hate them with a passion (see this thread as evidence), but many see the association as evidence that they might at least have some expertise with that product set. I certainly don’t agree with their position, but it makes sense commercially.
No one thinks of Oracle when they see JavaScript.
It literally wouldn't surprise me if when asked, the legal team simply responded "it's standing policy".
At this time, but their ownership and past behavior indicates that if Deno or anyone else tries to have a paid offering, there’s a non-zero chance Oracle will come sniffing for low effort money.
I’m not sure if that’s even possible under US law though.
Whoever thinks it's a good idea to bet on the altruism of a giant faceless corporation is dumb.
So, is "X abuses IP law" hatred is out of principle or because folks seem to be in love with Sun and Google and hate Oracle and Microsoft.
Oracle's contributions are less clear-cut, particularly if you don't count all the acquired "achievements."
I’d be surprised if Oracle released the trademark without a fight to the end. They have a special way of decimating open source projects.
Evaluated by how useful they are to society at large, many businesses should not exist forever--or even for very long. Xerox PARC, Kodak, and Netscape are examples of companies (or, in PARC's case, a division of a company) that contributed significantly to their fields before becoming defunct. Those contributions aren't worsened or inferior, somehow, because the companies that engineered them are gone.
Whether or not a company is still in business only tells you whether a company is good at keeping itself alive. Over time, that quality is increasingly disconnected from whether a company produces valuable goods or services.
At a couple points my org had hiring crunches and leadership’s short term solution was to find employees from other orgs that could be “loaned” to us. The quality was universally jaw-droppingly low. I had to do code reviews and they would do the craziest junior-developer no-standards stuff that would cause their PRs to get rejected repeatedly, because not only did they make dumb decisions, they didn’t even understand the explanations of why they were dumb decisions. It was infuriating and a horrible waste of time, and the second time around, we tried to say we don’t want that kind of help, but leadership insisted that the free manpower was not optional.
Both Java the language and OpenJDK the main runtime & development kit have had much more money and manpower poured into them under Oracle than they ever had previously. Both continue to advance rapidly after almost dying pre-Oracle acquisition.
MySQL 8 (released in 2018) was a massive release that brought many long awaited features (like CTEs) to the database, although MySQL's development have stalled during the past few years.
Oracle employs several Linux kernel developers and is one of major contributors (especially to XFS and btrfs): https://lwn.net/Articles/1022414
Not top 3 or even top 10, but better than most companies out there.
That's all I can remember.
edit: after thinking about it for a couple more minutes, they're also the main developer of GraalVM — the only high quality FOSS AOT compiler for Java (also mentioned by a sibling comment), and are writing one of the major relatively lightweight modern alternatives to Spring (the other two being Micronaut and Quarkus): https://helidon.io
9.0 is finally released and we are now at 9.3. While nothing big or exciting with every release but development is steady. MySQL 8.0 will reach EOL in April 2026 so every should move to 8.4 LTS soon(ish) and 9.7x should also be LTS by then. I know most on HN is about Postgres but modern MySQL is decent. I think a lot of people still have MySQL from 5.0 era. Which is also somewhat true with Java as well.
I think Oracle do contribute lot of open source code, they just dont get the credit or brag about it.
Many already moved to MariaDB, because development stalled after Oracle bought Sun (which bought MySQL).
People actually used to like the products that made these companies explode. Windows 2000 was cool; Facebook was cool; Google was cool. Whenever they stop being pure unadulterated evil for a few minutes, very quickly a lot of people are willing to forgive them and welcome them back into polite society.
But as long as most geeks can remember, Oracle has never been cool. At its peak, the company enabled a world of snooty BOFH DBAs, selling unreasonably expensive products to well-oiled middle-managers. And then they started "acqui-squeezing" adjacent products, blackmailing their own customers, and suing everyone in sight. They could cure cancer tomorrow, gifting all the related IPs to the world, and most geeks would still see them as scum trying to whitewash their image - and they would probably be right. There are some great engineers in Oracle, but their management is all that is wrong with capitalism.
SQL is kind of shitty, but the relational database model is so great that it makes up for it. And until the rise of entity component systems, SQLite, PostgreSQL, and MariaDB, all the decent implementations were proprietary software. Separately, although SQL's implementation of transactions is even more broken than its implementation of the relational model, transactional concurrency is also great enough to make up for SQL, and, again, all the usable implementations used to be proprietary.
Which RDBMS software has/have become the best in the world after that?
I think you could make reasonable arguments for SQLite, Postgres, MariaDB, Impala, Hive, HSQLDB, SPARK, Drill, or even Numpy, TensorFlow, or Unity's ECS, though those last few lack the "internal representation independence" ("data independence") so central to Codd's conception.
What's your opinion?
Also, though, horizontal scaling is a lot less important now than it was 20 or 30 years ago. https://www.servethehome.com/2025-server-starting-point-inte... says AMD has 192 cores per socket, and you can get two-socket motherboards, so 384 cores total. And you can stick 12 128GiB DDR5-6000 DIMMs in it, so 1.5 tebibytes of RAM, and a single SSD is 30 terabytes, and SSDs can commit a transaction group durably in typically 0.1 milliseconds. And those 384 cores (EPYC 9005, so Zen 5c, https://www.servethehome.com/amd-epyc-9005-turin-turns-trans...) are 2.25GHz and typically about 2.2 instructions per clock (https://chipsandcheese.com/p/zen-5-variants-and-more-clock-f...), and they support AVX512.
As one rough estimate, 2.25GHz with AVX512 (at 1 IPC) means you can do 36 billion column-oriented 32-bit integer operations per core per second, which with 384 cores means 13 trillion 32-bit integer operations per core per second. On one server. So if you have a query that needs to do a linear scan of a column in a 13-million-row table, the query might take 300μs, but you should be able to do a million such queries per second. But normally you index your tables so that most queries don't need to do such inefficient things, so you should be able to handle many more queries per second than that!
(Each socket has 12 DDR5 channels, totaling 576 gigabytes per second to DRAM per socket or 1.13 terabytes per second across the two of them, so you'll get worse performance if you're out of cache. And apparently you can use 512GiB DIMMs and get 6 tebibytes of RAM!)
So, if you need more than one server for your database, it's probably because it's tens or hundreds of terabytes, or because you're handling tens of millions of queries per second, or because your database software is designed for spinning rust. Spinning rust is still the best way to store large databases, but now the cutoff for "large" is approaching the petabyte scale.
I think the space of databases that are under ten terabytes and under ten million queries per second is large enough to cover almost everything that most people think of when they think of "databases".
https://newsletter.pragmaticengineer.com/p/code-review-on-pr...
Oracle is a parasite.
You're probably thinking of Genericisation. This isn't a law in the sense you probably mean, there is no statute about it, no legislature wrote it, nobody signed anything. Instead Genericisation is a legal doctrine related to the core idea in trademark law that we can't have exclusive use of descriptive marks.
Suppose you make a Big Car and you try to trademark "Big Car" as your exclusive mark for this new product. That's just describing the car, it's generic so you can't do that, it's OK to trademark "Giganticar" or "Waterluvian Car" or something because people can describe what their similar product is with the words "big car" but if you were allowed to own "Big Car" they can't do that.
Genericisation says well, if your product is so successful that now everybody knows what a "Waterluvian" is, and most people shown a new big car from say, Ford, say "Waterluvian" so that even Ford's sales people struggle to teach the guys on the forecourt not to call this a "Waterluvian" - that's now a generic term, you can't stop Ford just saying they're making a Waterluvian.
Genericisation only applies for crazy famous stuff. Kleenex is an example because your mom knows what a kleenex is, the guy who mows your lawn knows, Elon Musk knows, everybody knows, that's actually famous. Javascript probably wouldn't meet that requirement. My mother does not know what Javascript is, my boss does, because he's a software engineer, and maybe the average numerate graduate knows, but I wouldn't bet a lot of money on it.
Dilution is a related idea, also for very famous things. Dilution says for these famous things it's not OK to use the famous mark for any other purpose even though it's not related. So Disney toilet paper isn't OK, Coca-Cola brand vibrators, not OK, and so on. Nobody thinks the vibrator is a beverage, but Coke is so famous that doesn't matter. That doesn't impact here either.
To google something has for decades for millenials meant search online in any way.
Trademark law is dumb and inconvenient. Only people owning trademarks disagree.
But so are many other laws. We all just have to follow them all anyway.
I find that convenient, despite owning no trademarks.
Just like how we take for granted that you can count on your food label not lying to you and your food not having dangerous ingredients, with proper inspections and such.
I guess I don't feel bad not knowing this though, as the language really does have nothing to do with the company it's insane that they even hold a trademark for it.
https://www.engadget.com/2014-02-26-when-carl-sagan-sued-app...
I'd donate.
Python was invented earlier, but didn’t see wide use until later.
And that they were both massively accelerated by the level of interest in the early WWW is undeniable. No other general purpose languages can say that except perhaps Perl, and it slowly burned out.
So the JVM has a runtime-enforced nominal type system (and object model) with classes.
But JS, to my knowledge, only has primitive types enforced at runtime, and no nomimal class system, unless you basically implement it yourself?
Uh, edit: maybe I get you now, it does have that in a way. But prototype identity and "instanceof" are rarely used in practise.
Maybe I'm missing your point here. Answering at late local time.
It would be so great to have a nominal type system in the browser though.
So many JS librarlies have their own version of it, and it causes insufferable headaches when combined with TypeScript.
Like, they use complicated hacks to make sure that their library objects are not structurall/duck typed.
Yes, the typing and semantic models are wildly different. The point is that they’re primitive in a way that the other widespread alternative, C++, did not inherit from its Cfront heritage.
that's what I was aiming at, maybe poorly.
There's tons of libraries that use some kind of runtime-observable instance property as a tag, to mimic nominal typing in JS.
The same thing is also possible using prototype identity, if you either use the class keyword syntax sugar introduced with ES5 (?), or if you manually do OOP using prototypes. But the latter is very uncommon.
It seems to be more common to add a property like
$_$_$____superlib__$-inst_WALRUS
> To plead a claim of fraud, petitioner must plead that: (1) respondent made a false representation to the USPTO; (2) respondent had knowledge of the falsity of the representation; (3) the false representation was material to the continued registration of the mark, and (4) respondent made the representation with the intent to deceive the USPTO.
> A claim of fraud must set forth all elements of the claim with a heightened degree of particularity [...] Indeed, “the pleadings [must] contain explicit rather than implied expressions of the circumstances constituting the fraud.” In addition, intent to deceive the USPTO is a specific element of a fraud claim, and must be sufficiently pleaded
> Essentially, Petitioner’s theory of fraud is based on allegations that the specimen of use submitted with Respondent’s maintenance documents do not show use by the proper party. It is well-settled that the proper ground for cancellation is the underlying question of whether the mark was in use in commerce, not the adequacy of the specimens [...] the insufficiency of the specimens, per se, does not constitute grounds for cancellation; the proper ground for cancellation is that the term has not been used as a mark
To file for a mark or renewal of a mark and claim ownership of something you do not own is insane. It's not like this is a 5 second process or that there isn't a lot of money riding on this-- this sort of thing is super serious and incredibly important! You're telling me no one at Oracle or their counsel was able to catch this in review before filing? As far as I can tell, in the renewal for the mark[1], Node.js was the sole specimen provided as an example of mark use! Come on...
EDIT: Sorry, correction, they have three specimens attached to the renewal, two of which seem to be the same. Clearly an insurmountable amount of work and too complicated to validate.
[0]: https://ttabvue.uspto.gov/ttabvue/v?pno=92086835&pty=CAN&eno...
[1]: https://tsdr.uspto.gov/documentviewer?caseId=sn75026640&docI...
However, I do believe the word has been diluted and genericized and hope the USPTO chooses to release it.
A good argument to avoid losing a trademark to genericization is to show that there is an actual generic term that overlaps with the trademark, but then the trademark is not the generic term itself.
Examples:
Nintendo → Video Game Console
Post-it → Sticky Note
Xerox → Photocopy
etc ...
In the case of JavaScript, there's no generic term to allude to; JavaScript is the generic term, which might weigh towards the argument of genericization.
Err, that's not a given by any stretch. This is exactly what the suit is trying to prove. They are not a rightful holder of the trademark. They've failed to show use in commerce, and one of their examples of use was someone else's.
And here's one (trivial, but valid) use of it [2].
I'm sure Ellison lawyer's can come up with thousands of examples of JavaScript being used within the context of Oracle's business activities.
The way to go is fight for genericization (or start calling it ECMAScript, lmao).
1: https://tsdr.uspto.gov/#caseNumber=75026640&caseSearchType=U...
2: https://docs.oracle.com/en/database/oracle/oracle-database/2...
Id argue the opposite. The wording makes no reference to oracles ownership of the product or name that is JavaScript. And ECMA is reffered to as the "maker" of the standard.
If anything, this is an example by Oracle themselves using the trademark in a generic context.
Its like cocacola calling themselves "a producer of fanta" and referring to a the food and drug administration to define what that means.
I doubt the writer of that doc was aware that Oracle owns the JavaScript trademark.
But for "JavaScript"? What else is there? "JS"?
Edit: I guess there's "ECMAScript", but who actually says that (aside when they legally need to)?
It can be both.
>Everyone knows JavaScript isn’t an Oracle product
But older people should know that it was a Sun product and Oracle bought Sun.
Edit: Sun actually only licensed the name. But in the renewal it points to an Oracle product called Oracle JavaScript Extention Toolkit.
https://tsdr.uspto.gov/documentviewer?caseId=sn75026640&docI...
Javascript was written at Netscape.
> With JavaScript, an HTML page might contain an intelligent form that performs loan payment or currency exchange calculations right on the client in response to user input. A multimedia weather forecast applet written in Java can be scripted by JavaScript to display appropriate images and sounds based on the current weather readings in a region. A server-side JavaScript script might pull data out of a relational database and format it in HTML on the fly. A page might contain JavaScript scripts that run on both the client and the server. On the server, the scripts might dynamically compose and format HTML content based on user preferences stored in a relational database, and on the client, the scripts would glue together an assortment of Java applets and HTML form elements into a live interactive user interface for specifying a net-wide search for information.
> Java programs and JavaScript scripts are designed to run on both clients and servers, with JavaScript scripts used to modify the properties and behavior of Java objects, so the range of live online applications that dynamically present information to and interact with users over enterprise networks or the Internet is virtually unlimited. Netscape will support Java and JavaScript in client and server products as well as programming tools and applications to make this vision a reality.
> "Programmers have been overwhelmingly enthusiastic about Java because it was designed from the ground up for the Internet. JavaScript is a natural fit, since it's also designed for the Internet and Unicode-based worldwide use," said Bill Joy, co-founder and vice president of research at Sun. "JavaScript will be the most effective method to connect HTML-based content to Java applets."
This was all implemented, and Java applets had full interoperability with JavaScript. Applets could call JavaScript functions, and JavaScript functions could call applet methods. Of course over time people gave up on Java applets and JavaScript became a good enough language to write real application logic directly in it. It's true that JavaScript now has virtually nothing to do with Java, but that wasn't the case initially, and the name has at least some logic behind it.
One more tidbit I just learned: there was a Netscape/Sun deal around the name, so the registered trademark has some legal history. It's not that Sun (and then Oracle) just claimed rights to something Netscape made.
There is some irony in that Ryan isn’t acknowledging Node.js own trademark in his post, given that he was the person who announced the Node.js trademark.
https://nodejs.org/en/blog/uncategorized/trademark
So he wants Node.js trademark to be acknowledged, but doesn’t acknowledge it himself.
Oracle wants the JavaScript trademark acknowledged, and he doesn’t want to acknowledge that either.
This all seems very silly to me.
Of all places to put trademark acknowledgement, it’d be there - and it’s missing.
Javascript has become such a ubiquitous term that its copyright status is increasingly tenuous. Node.js by contrast has no such problem, yet. Most of the industry supports this initiative, and dumping on the people willing to invest the time and money to fix it once and for all, over seemingly irrelevant things feels petty.
It looks like JScript is still trademarked by Microsoft, why not ask them to do whatever the community thinks is right for ECMAScript names and then we can all refer to the language a little faster?
In TypeScript 7, the compiler will be written in Go instead of TS. But the compiler will still produce JS code as its output and so Node.js is still relevant for running that JS code.
Or is there something else about TypeScript 7 that will make Node.js irrelevant?