EU Advocate General: Technical Standards must be freely available [pdf]
curia.europa.eu
curia.europa.eu
Prices are generally single to low double-digit Euros (even for standards that are hundreds or 1000+ dollars).
If you are a solo contractor like me (and hence only need one copy), DON'T get the single licence copies. You need to use some BS DRM software that binds the file to your computer and is a PITA.
Get the organisation one, pay for two licenses, and you'll be given a regular PDF instead (and still save hundreds of dollars).
You obviously will only get the Estonian specific annexes, but those are normally optional anyway, and generally available for free when downloading the "free sample" of a standard for a specific country.
The PDF-selling companies argue that they need that money to organize committee meetings for developing further versions of standards etc., but that leaves me wondering - we have so many other standards-setting institutions which don't rely on shady PDF sales, so why not take that model instead?
[1] https://openregulatory.com/accessing-standards/ (Disclaimer: My company)
No. Because it costs money to run these organizations. They are either funded by the government (which means taxes, which means us), membership fees or other activities. Last I checked everyone needs to feed themselves or their family. This "it's because it's capitalism" thing gets old pretty quickly.
What should be objectionable is precisely when organizations are government (taxes, us) funded and they sell us standards you are supposed to work and test against for, well, government-mandated certification requirements. So...we pay for their work through taxation...for them to sell us the products they create...and then we pay for the testing services against documents we cannot freely access even though we paid for their development. That's just messed-up.
This is where it gets complicated. If USA or European taxpayers fund the development of these standards, should others outside of those regions have to pay or contribute in any way? China is no-longer a poor agrarian society. And yet they benefit from work done in the EU and USA for decades at great cost to taxpayers.
Not sure what the answer to this is other than to point out these things are not simple. There's the "well, it benefits everyone" angle, which brings us back to "Why are these organizations charging for PDF's at all?".
What's the difference? What you described is capitalism.
If we lived in an economic system that didn't pay those organizations, they simply wouldn't exist.
Nitpick: if we lived in an economic system with universal basic income it's not unlikely people would decide to work on standards anyway.
> People aren't going to do that work for free.
Counterexample: all open source software.
Reality example: Open source only exists because people earn a living doing something else. It’s cost is very far from zero. In fact, if we accounted for it I would bet FOSS is, as a category, the most expensive software on the planet.
Nothing is free. A sad yet necessary truth that must be understood.
Do also note that most people spending countless hours on standards work on them either on their own time, or more usually, their affiliated organisation time (university, company...). Standard bodies rarely pay for actual standard development.
They also do nothing. Or go on vacations. There's a very small group of people (on a percentage basis) who develop enough drive and passion to work on FOSS at the level of domain expertise and dedication necessary to get anywhere.
It is important to make a distinction between developers who devote a non-trivial amount of time to FOSS and those who might scratch an itch once or twice in a codebase, never to be heard from again. While all contributions are valuable, GitHub is full of stagnating projects where occasional contributions from random developers simply isn't enough to keep them going.
Look at real active projects and you'll discover that the number of dedicated developers devoting the kind of time and effort necessary to sustain and drive the project forward can often be counted with one or two hands. That is evidence enough of what I am saying.
Why?
Well, there are millions of qualified software developers around the world who cover every domain in software development. I think we can agree that most of them have their basic needs covered. And yet, you don't see millions of developers flocking to work on FOSS.
Why is that?
Because the scenario you paint, for the most part, does not align with reality at scale.
The number of people who, as you say, "work on stuff that they care about personally regardless of how much or how little it earns them" is very, very small, a rounding error. You can't get very far on a FOSS project --particularly if measured across years-- without a very small core group that does all al heavy lifting.
BTW, the fact that FOSS can thrive with just a handful of developers driving a project and little random contributions from others here and there is fantastic. The ecosystem work very well and there's plenty of evidence to show this to be true.
My only point is that we should not pretend that millions of people will flock to FOSS if their basic needs are met. This sounds like one of those universal basic income arguments. And it simply isn't true. People don't function that way. If their basic needs are met, the last thing most people would do is sit in front of a computer for ten hours a day to write code for free.
You seem to be debating some other claim (about how FOSS can be maintained well) than people wanting to do things they care about when they are "settled". What I meant there is that they don't have to do their day jobs, which most of those millions you mention have to.
I have my own thoughts on FOSS (most of it is "finished" in that it served a purpose and there is no need for maintenance, even if it's imperfect and buggy), but we are not at a state where there are large groups of people who have the means not to care about regular income at all to know what they'd do.
FWIW, in the worst of times, it was the aristrocacy that pushed science and arts forward, because they were the only ones who had the means to do it: it wasn't as fast paced as today, bit it didn't stop either. We've since "democraticised" science and arts (gamifying it a bit) by also making jobs out of it and increasing access to education.
Sure, we don't need everyone to care about everything, but there will always be a critical mass of people caring about critical stuff.
In not-capitalism, there's a possibility this might not be true. Open source developers MIGHT NOT need to earn a living doing something else.
That's all well and fine. Except this is a fantasy. It has not existed ever in human history, does not exist today and the laws of physics say it cannot exist in the future.
If we limit the discussion to reality, well, FOSS isn't even close to free. Before a person can devote effort to working on FOSS, they have to sustain themselves.
Look, there's nothing whatsoever wrong with this reality. FOSS has gone incredibly far and deep under this model. There's nothing wrong with recognizing that FOSS exists due to what one might call charity. It's people working at whatever they do who then donate some of their free time to whatever they care about in the FOSS domain. And that's fantastic. Let's not pretend it's free though. It costs every single contributor a non trivial amount of time to participate in the effort. And that isn't free for anyone on this planet.
What I am saying is that this isn't a bad thing, yet it is a reality that should not be swept under the rug.
FALSE, FALSE and FALSE.
>there's a possibility this might not be true
Zero, zilch, nada. People will not do grunt work for free because it's not fun.
>Will people create wealth if they can’t get paid for it? Only if it’s fun. People will write operating systems for free. But they won’t install them, or take support calls, or train customers to use them. And at least 90% of the work that even the highest tech companies do is of this second, unedifying kind.
I think there are very few people on the planet, if any, who would consider writing a standards document for some medical equipment to be fun enough to do free of cost. In fact I doubt it's even possible to make standards documents for physical things like medical equipment free of cost. It's an entirely different ball game from software standards.
Weird comment to make on the literal internet, where all the standards (Whether IETF, W3C, or WHATWG) are freely available.
What is weird about giving a counter-example?
> IETF and those standards organizations are funded by other people BTW, they don't consist entirely of individuals putting their own money in.
Sure, nobody is claiming that no money is involved at all. The point is: they do not charge for standards, we, the people who read standards, do not fund them. And yet they exist. Not only do they exist, but they maintain some of the most widely implemented standards in the world. Clearly this proves it is possible to not charge for standards and still have standards.
Nope.
Capitalism: An economic and political system in which a country's trade and industry are controlled by private owners for profit.
Maybe the problem is they are not teaching shit in our schools. No, actually, I am wrong. They are teaching shit. Literally.
Funny guy. Comedian in training?
Here, let me help you:
https://www.rd.com/article/what-does-literally-mean/
It's a good read, you will, literally learn something that might literally make your head explode.
:)
I could just as easily claim that “capitalism” has some separate colloquial definition.
What is funny how you were so interested in dictionary definitions before. You’re a pedant when it serves you and when it doesn’t, you’re not. Typical HN. Hope you smug sense of self satisfaction is compensating for whatever you’re missing :)
Thanks for the insult and personal attack. Always a sign of someone with whom one can have enlightened conversations.
> Most standards organizations have SOME way of making money, because we live in capitalism.
which could be shortened as:
> Most things that keep existing have SOME way of making money, because we live in capitalism.
And you rebuttal that it cost money to do anything is a confirmation that it is due to capitalism. I would argue that it is the definition of capitalism.
I do not mean this to be a value judgment.
Sorry my friend. It isn't. It costs money to do anything on this planet, even if it isn't what most would understand to be currency. Barter is also a form of currency.
Altruism aside, nobody is going to give someone a dozen eggs without getting something in return. And that something has to be worth at least what it took to get the 12 eggs. Because, well, doing something at a loss is not sustainable. Eventually you hit zero.
And, of course, you have to protect yourself from unknowns --five egg-laying chickens are eaten by a wolf. Which means you need to get something more than what it takes to make those eggs or you will, again, hit zero when you can't sustain things because of various events.
At this point this isn't even profit. You are working for exactly what you put into making those eggs and a small insurance hedge because shit happens.
Now we get into capitalism: Your eggs are better than those someone else offers. You deliver. You have better packaging. If an egg arrives cracked, you replace it. You created a system to be able to package eggs safely with less work.
You recognize this delivers value. You determine you can charge more for your eggs than the raw cost of production. You also want to expand so you can service more people in town and maybe even neighboring towns.
You need money to expand...because the construction isn't going to be free. So, you tack on as much profit onto your price for a dozen eggs as the market is willing to pay and hope for the best. That last part is important. Someone else could come out with equivalent eggs and tack-on less profit. Unless you have a monopoly, you do not control profit, the market does.
And that, in a super-simple nutshell, is how capitalism happens.
Absolutely every person on the planet is a capitalist.
Why?
Everyone tries to earn more than the raw costs to exist. That's your profit. It allows you to save to get things you want, navigate tough times, support a family, go to school, etc.
In other words, capitalism is part of our very DNA. It isn't some weird evil thing that's been imposed on society.
We have family from the old USSR. You wouldn't believe the things they had to do to survive. Capitalists to the core...because the shit they lived in was just that, shit.
> Altruism aside ...
We need resources to keep existing and we need people to work to make resources, in almost every case people need other's work to survive.
The concept that these transactions can and should be conceptualized via their monetary value (rather than civic duty, status, noblesse oblige, etc.) is capitalism.
Feudal nobles needed food to eat the same as us, but they did not live in a capitalist society.
We could convert their military power, their family links to nearby nobles, and the willingness of their subjects to obey them to some kind of monetary value, but all of these things did not behave as capital.
> Everyone tries to earn more than the raw costs to exist. That's your profit. It allows you to save to get things you want, navigate tough times, support a family, go to school, etc.
This is scarcity not capitalism, if you only have subsistence level of resources you have 0 capital.
The main characteristic of a capitalist system if that resource ownership and production happen in a market.
for example Mexican avocado cartels compete in a globalized capitalist market, but the farmer they extort live in very different system.
---
I am not passing any kind of judgment on capitalism here, not saying whether we can/should move away from it, but there have been societies not built on capitalism.
The EU or US could just decide to pay for it. I would guess that this is politically easier to swallow in Europe even though it makes as much if not more sense economically in the US.
That sounds logistically very difficult. The government would need to decide _which_ standards are worth paying for. You certainly couldn't have everyone who wants to just bill the government for however much they decide their standard "costs" to develop.
Unless it's a standard that you need to follow by law, it doesn't make sense for the government to be in charge of making it public.
We'd want to pay for that too!
Or we could (And I have always thought this would be helpful) have a FOSS standards forum. Want a standard, make a Git issue, people discuss it for a bit, someone makes a markdown pull request.
It would be pretty sensible for the EU to pay the USB consortium enough money to make the standard free for all EU citizens - I think that one is already free but you get the idea. Especially because the EU wants USB ports to be mandatory on cellphones.
ISO SQL standard - 187 CHF https://www.iso.org/standard/76583.html
ANSI SQL standard - 237 USD for non members https://webstore.ansi.org/standards/iso/isoiec90752016
It's readable - though obviously you're going to need experience in RDBMS theory (Codd's paper, etc) to grok it.
It is very verbose, though.
Fortunately, the only people who need the ISO SQL spec are the vendors: the people writing SQL engines and tooling: it's next-to-useless for people who are designing databases or writing queries against them, chiefly because no RDBMS implementation, ever, has come close to implementing the full specification, and everyone has their own extensions to SQL - so all the vendors put out their own documentation and the world's largely been happy with that for the past ~40 years that SQL's been relevant.
It's not too hard to find the standards in PDF form if you know where to look - and as someone who does a fair bit of SQL, I'll say I've only ever referred to the standards to back-up my more controversial posts on StackOverflow.
it would be useful to have that standard. Thanks, will have a hunt.
.
6. Aggregate Queries Can Contain Non-Aggregate Result Columns That Are Not In The GROUP BY Clause
For example to find the highest paid employee:
SELECT max(salary), first_name, last_name FROM employee;
.
That's straightforward, no?
max(salary) over(partition by emp, dept order by salary desc) as top_emp
...
where top_emp = 1
I don’t know exactly where in the spec it’s defined, though - but I’ve seen people say it was introduced in SQL99.
> The Advocate General’s Opinion is not binding on the Court of Justice. It is the role of the Advocates General to propose to the Court, in complete independence, a legal solution to the cases for which they are responsible. The Judges of the Court are now beginning their deliberations in this case. Judgment will be given at a later date.
It's going to be for something dumb like the exact dimensions of a fire safety tag, but if you don't tell me the law then it's a secret law.
Charging for standards is a way to prevent competition.
By the way, there should be a special place in hell for the people who designed this EU standards website:
https://standards.cencenelec.eu
You can search for standards, then you can pay for them in a webshop, but you can't link to the main page for each standard as there is some kind of URL-cookie mechanism that works hand-in-hand with state on the server that will break down as soon as you start using the URLs elsewhere.
If a government wants to write compliance with a proprietary standard into a law, then they should be required to buy or license that standard
You still have to shell out for at least a day per standard per year for auditing but it would be nice if the standards themselves were free. Every little helps as Mr Tesco is fond of saying.
(EDIT - speling)
Or maybe the law could create its own official standard that simply paraphrases what the closed standard says, effectively making it open.
Say you own a house. The government needs to take your house to demolish it to build a road. The government is within its rights to just take it (after due process and fair payment).
Now say you own a document that you make people pay to read. The government needs to take your document to make it part of the law to build a building code. Why doesn't the government exercise same kind of rights to just take it (after due process and fair payment)?
That's not true. If 3000 people a year would have paid you $100 for access over the next 20 years, that's $300,000, which as a future income stream has some PV X. You'll be indifferent [1] between collecting the trickle of revenue from customers paying the paywall, and a one-time lump sum from the government of X [2].
[1] You'll be indifferent by definition. Because you'd say "Yes please" to a lump sum of $1 trillion, "No thanks" to a lump sum of $1, and PV is defined to be the crossover point where your answer changes.
[2] If you have a discount rate of 5%, then X ~= $300,000 / 1.05^10 or so, depending on how front-loaded the payments are and how much you care about payments more than 20 years in the future.
> quite unfair in fact
If the government pays you less than X, in theory it's unconstitutional, plain and simple. If the legal system doesn't accept your projected revenue, or discount rate, or you have to pay the lawyers and expert witnesses too much to legally establish the basis for those calculations...well then, the problem isn't the government taking your document to make it part of the law. The problem is the legal system is too inefficient or unreasonable, which has a different set of fixes (e.g. make laws about what calculations the government has to use in this situation, and make the government pay the plaintiff's attorney fees if the case is sufficiently clear-cut instance of "the government didn't follow its own official rules and this should never have gotten to court in the first place")
This part doesn't follow.
Why doesn't the government make their own document? Eminent domain exists for land because it's not possible to make more land in the same place.
(Sorry this is a US-centric explanation, I don't really understand how the EU system works)
Sometimes legislators say "We don't feel like having government employees write this particular set of regulations from scratch. Instead we'll make a law that says such-and-such a document written by a private organization is effectively promoted to a law."
There are three separate questions here: (1) Should legislators promote documents to laws? (2) Should legislators be allowed to promote documents to laws? (3) If legislators do promote a document to a law, how should the legal system interpret that action?
If you say "(1): No," then you should vote for a Congress / President that doesn't think it's okay to promote documents to laws. But sometimes a Congress / President might be elected that does think it's okay, and they might promote some documents to laws.
US law currently says "(2): Yes." If you think it should say "(2): No" you probably need to change the system. I.e. pass a Constitutional amendment, which is extremely difficult, much harder than merely voting in a Congress / President that agrees with you. Good luck on that.
That means it comes down to (3). Which is where I make the analogy with eminent domain. You had a document that people were paying you money for. The government made that document part of the law, and the law must be freely available, so now because of the government's actions, people aren't paying you money because a copy of your thing is now free.
Sure, the government could have not promoted the document to a law. Or the system could be set up so the government isn't allowed to promote documents to laws.
But those things that could have happened didn't happen. Instead, what actually happened is the government promoted your document to a law.
Why didn't they make their own document instead of promoting yours? It doesn't matter at this point. They did promote your document, and they are allowed to do it under the current system.
The obvious ways to resolve the conundrum are: (a) Make people pay to see the law, or (b) Make the law free and have the taxpayers compensate you for the lost revenue of the document, or (c) Make the law free and you just have to suck it up, too bad, so sad.
(a) just feels icky because it effectively means the law is a secret. Penalizing people for violating laws they can't know about (without paying) seems pretty unfair and maybe violates some legal principles (due process maybe), I dunno the specifics, not a lawyer. (c) definitely seems wrong too because it seems like "the government took my stuff and didn't pay me for it" falls (or should fall) under eminent domain even if the thing the government took isn't land. That leaves (b), I can't really think of any other options
Renumeration.
They would simply pay "fair market value" for the rights, though likely there'd be a push for a cooperative between nations which spread the cost widely, making the actual cost to any given nation relatively tiny.
The second option would mean they would still get sued by the copyright holders of the closed standards, paraphrase or not.
The better option is probably for the government to re-publish the standards as-is, with some compensation for the copyright holders.
Purpose is that the government doesn't have to rule all details, but can leave that to domain experts and ongoing research.
This means that they become de facto law, even if they do not adhere to fundamental principles of the law: transparency, and fairness.
Standard paywalls thus should be seen as a barrier for the rule of law, not just barrier for innovation.
The prices for access to standards made sense in the era, when every document had to be printed instead of downloading as PDF.
What about prohibiting standarization for easily outdatable things ? Like for example not using RSA anymore ? Some big government could finally propose something better for their peoples without backdoring it ? Look it is like with carring gun - you really want to sabbotage ammo your people are using to defend ??
Then, the IBC allows you to claim fire ratings if you match an Underwriter's Laboratory tested assembly (for example, a fire-rated wall) which buying PDFs from UL is another additional expense.
Point being, there's a rabbit hole of paywalled technical standards in construction.
But then again with how wrong it is usually...
If you want "free" standards, look what happened to so-called web standards dominated financially by FAANG, and the catastrophic results to the extant web. Might as well say the web as such doesn't exist anymore.
The fact that standards are often difficult to parse is more a side-effect of compromise wording, non-English authors, and technical terseness, and that they effectively act more as a contract between the member companies, which already have the institutional knowledge to interpret the wording.
What are some good examples of Open Standards and Open Web Standards for EU and other world regions?
Open Standards: https://en.wikipedia.org/wiki/Open_standard
Open Standards > Specific definitions of an open standard: https://en.wikipedia.org/wiki/Open_standard#Specific_definit...
Web Standards: https://en.wikipedia.org/wiki/Web_standards
E.g. W3C, IETF, WHATWG, and ECMA develop open standards.
> The Web Monetization API allows websites to automatically and passively receive payments from Web Monetization-enabled visitors.
From https://interledger.org/faq/ :
> Web Monetization is being proposed as a W3C standard. Using the Interledger Protocol, the Web Monetization proposed standard aims to make it easier for web creators to generate income from their work without relying on advertising, site-by-site subscriptions or tracking models.
IIUC, ILP messages would help financial regulators - such as FTC CAT Consolidated Audit Trail - audit traditional ledgers and cryptoassets?
Interledger was contributed to W3C and has undergone significant major revision; including packetization of liquidity into many small transactions. FWIU, W3C Interledger Protocol is a W3C spec but by producing IETF-style numbered RFCs, their process slightly differs from the W3C WG Working Group model (with a page, a mailing list; and one or more git Repositories with Issues: github.com/orgname, github.com/orgname/readme, github.com/orgname/orgname.github.io, ).
Independent, state, academia, and industry are all invited to contribute time to open [web] standards that solve for many users and use cases.
W3C Process Document > 2. Members and the Team: https://www.w3.org/Consortium/Process/#Organization
"The IETF process: an informal guide" > 2.6. Bodies involved in the process: https://www.ietf.org/standards/process/informal/
Ecma By-laws > Art. 3: Membership: https://www.ecma-international.org/policies/by-laws/
Ecma TC39 (ECMAscript (JS)) Process matrix: https://tc39.es/process-document/
After all, isn’t the entire value of a spec in the verification that it is correctly implemented?
To put my money where my mouth is, I only buy certified “made for iPhone”/“certified Magsafe compatible” accessories now and have had a much better experience than I did before.
The purpose of the current system is to defray the costs of the printing press, since there was no other way to disseminate information 100 years ago. Now that we have the Internet, paying for individual standards is unjustifiable and should be ended. This is made worse because of the increasing number of standards that effect anything useful. Countless problems have happened because people cannot affordably access the hundreds of standards they are supposed to be using.