Open standard but not open access: Schematron author complains about ISO paywall

The original inventor of a popular XML standard, Rick Jelliffe, who created Schematron, has protested that the widely used text is now behind a paywall at standards body ISO. Schematron is a language for validating XML, designed for processing XML documents and reporting on errors. Version 1.0 was developed in 1999, since when …

[Peter Galbavy]

"The main reason for coming to this conclusion is that there is no longer a technically equivalent specification in the public domain, as it was the case originally,"

Uh huh. Embrace, extend, extinguish. "We invited the public standard in for dinner. We had the public standard FOR dinner. Chianti, anyone?"

Standards bodies are orders of magnitude more experienced at weasel words than humans and so what did we expect?

[Snake]

Even better

"This standard is an ISO/IEC standard and it is unfair to say that ISO and IEC, since this is a joint management and decision, has reneged of the freely available status. We are not reneging the freely available status). It’s simply not requested by the committee," a spokesperson told us."

This has to rank as one of the biggest admitted gaslighting episodes in history.

'We are governed by a committee, of which the committee members consists of our management. The committee [of which holds our own managers!] has made a decision...not to include retaining open source status.

But don't hold it against us ("we didn't make the decision").

Are they REALLY this stupid!

...I guess the answer is: Yes, yes they are.

[Anonymous Coward]

The Final Word

ISO 3103:2019 - The tea brewing standard - costs 58 Swiss Francs for 8 pages or about $8 per page.

[Anonymous Coward]

ISO standards should be free

The sad thing is that ISO standards must be self-supporting. But for the amount of money that they spend on other things, they could offer free PDF versions and only charge for printed versions. Especially troubling as the work is all done by volunteers (before being passed up to the bureaucracy.

[Mishak]

the work is all done by volunteers

Yes, this is the worst bit. They get others to produce the IP and then make money from publishing. They are not even willing to cover the travel costs for the members; it is quite normal for meetings to move round the globe, so someone from Europe may have to travel to Hawaii to attend a 5 day meeting at some expensive hotel (I'm not saying they should pay to stay at that hotel, but...).

[Anonymous Coward]

Re: the work is all done by volunteers

They're as bad or worse than Elsevier

[SGJ]

Free the standards

The true measure of a standard’s value should be the extent to which it has been adopted not the amount ISO can charge for it. For large organisations the cost of purchasing a standard isn’t a deterrent but for small organisations it definitely is.

For example, the ideas in the dozen or so standards in the ISO/IEC 27000 information security management system family would surely be of more benefit out in the real world than behind a paywall for up to CHF 178 each.

Academic papers are slowly moving to an open access model (faster in some disciplines than others) and standards should do the same.

[Yet Another Anonymous coward]

Re: Free the standards

Of course you don't have to read them, but you do have to have a paid for copy of each standard filed in your quality management system somewhere - the ISO standard says so.

And every couple of years they change a comma, update the year suffix and you have to buy it again.

[John Brown (no body)]

Re: Free the standards

"And every couple of years they change a comma, update the year suffix and you have to buy it again."

Sounds like academic text books required for university courses. Especially those written by the lectures. They get updated every year and you can be sure there will be required reading in the current one that's not in the cheap used one from last year.

[eldakka]

Re: Free the standards

> Of course you don't have to read them, but you do have to have a paid for copy of each standard filed in your quality management system somewhere - the ISO standard says so.

I couldn't afford that one, so wasn't aware of that.

[Pascal Monett]

"is a process that requires validation for each edition"

Of course it does - now .

Interesting to note that this Swiss organization decides this year to change the historically-free download status of a very minor document.

Personally, I would believe that a standards body should have the obligation to publish its standards for free.

You put your standards behind a paywall and people will use another standard.

[FuzzyTheBear]

Re: "is a process that requires validation for each edition"

And pretty soon .. https://xkcd.com/927/

[Anonymous Coward]

Difficult to agree with this more

I'm involved with some ISO standards, one level removed from the core working groups because jesus, life is just too short for that.

Looking in from the outside is like looking back in time, before open source and open standards, when "bazaar" was somewhere you brought a carpet, not how you designed a spec. The sheer volume of boilerplate and bureaucracy is staggering, with endless rounds of approvals from national standards bodies who - as best as I can tell - have only limited knowledge of the standards involved, and whose comments revolve around tweaking words, punctuation and the like. Which is ironic because I've purchased and worked from a number of specs, some of which are missing enough technical detail to make implementing them involve a large degree of guesswork. And naturally any revisions to those errors require another round of boilerplate documents and national-standards approvals.

And yes, it does seem to suck standards into it's orbit, but never release them. Once an ISO prefix is applied, copyright seems to be transferred to ISO (I may be wrong on this, haven't checked the details). And your ability to distribute it via any process other than ISO is lost.

I thoroughly approve of the need for standards, but if I had to choose a standards body I would go W3C or ETSI. ISO feels more like an academic publisher, focussed on the bottom line and nothing else.

[Yet Another Anonymous coward]

Re: Difficult to agree with this more

That's the idea.

ISO 9001 - quality standards in manufacturing

ISO 9002 - quality standard sin services

Now adding

ISO 9003 - quality standards in monopolies

ISO 9004 - quality standards in ripping off customers

ISO 9005 - quality standards in Ponzi schemes

[Dante Alighieri]

Negotiable Virtue

I like your ISO 9002 and wish to partake.

Which establishment should I approach. Purely for research processes.

Icon...

[Yet Another Anonymous coward]

Re: Negotiable Virtue

You really don't want ISO standardised sin !

[Dave559]

Re: Negotiable Virtue

But should you deviate from the standard, you must be corrected! And I understand some people are very willing to be quite thoroughly corrected.

[emag]

Re: Negotiable Virtue

I can only imagine the ASiN.1 parsing needed.

[Giles C]

Re: Difficult to agree with this more

I was once told this by an ISO 9001 auditor.

Before you were making junk - now you are making well documented junk.

I.e.

If you decide the final test before shipping test a piece of fragile electronics by dropping it off the roof of the office, then as long as you follow the set procedure it will pass the quality test (but not necessarily be any good to the customer afterwords.

Fortunately my work these days is way away from quality control processes.

[Anonymous Coward]

Re: Difficult to agree with this more

You're probably also making junk that costs twice as much as it did before. My organisation looked into an ISO type quality assurance process, and fairly soon came to the conclusion that our customers were happy to pay for work, not so much for paperwork. So now we expect them to put on their own quality "wrapper" if they want it.

[J. Cook]

Re: Difficult to agree with this more

Yep. that particular ISO standard only means that you have a well documented process on place for the manufacture of a product; says nothing about if the product is actually functional or not.

[Anonymous Coward]

Re: Difficult to agree with this more

", copyright seems to be transferred to ISO (I may be wrong on this, haven't checked the details)"

Yeah - you really haven't checked the details have you?

[John Brown (no body)]

Re: Difficult to agree with this more

"and whose comments revolve around tweaking words, punctuation and the like."

I can see where this might actually be quite important when defining an international standard such that it means precisely the same thing to all speakers of the primary published language, let alone making sure the translations also define exactly the same thing.

[John Sager]

Re: Difficult to agree with this more

I've purchased and worked from a number of specs, some of which are missing enough technical detail to make implementing them involve a large degree of guesswork

Standards ought to have an accompanying Rationale document that explains why certain design choices were made. When the process of defining the ADA language was running, the team I was in was writing a compiler (not for ADA) and we were interested enough to get the docs for the four candidates. All had a Rationale doc to describe how the choices were made for various language features. This must have aided immensely the various ADA compiler writers at the time.

[steelpillow]

"there is no longer a technically equivalent specification in the public domain"

"ISO standards should not be cited normatively by national laws of any country."

Indeed.

Let Rick, or the w3c, or whomever it may concern, fork the few changes they have made.

Then start on the next ISO paywall along...

[Mishak]

WG21 (C++) manage this very well

The C++ standard is published as ISO/IEC 14882:2020 (current).

The final committee drafts that are sent to ISO for standardization are available on the web at http://www.open-std.org/jtc1/sc22/wg21/docs/standards. These are not ISO standards, but they are what was sent to ISO.

Full LaTeX source for each is also available as tags in github - https://github.com/cplusplus/draft.

[Anonymous Coward]

"contain intellectual property of demonstrable economic value"

Do they get royalties on them? Or just ISO swallow the money to pay for the Louis XIV lifestyle this kind of organizations usually like?

[steamnut]

They are not alone

As someone who has to submit electronic products for CE (ok, now UKCA here) approval it annoys me that all of the EU standards, that they themselves create, are charged for. And they change them quite often. As the standards are downloaded there is no actual cost to the EU. The only way to see if a standard applies to your product you have to buy it.

[Doctor Syntax]

The best solution would be to fork it from the previous free version, perhaps with an announcement that the ISO version is deprecated. Ultimately the standard that matters is the one everyone follows. If the free version is regarded as definitive then ISO can charge whatever they want for their text and it won't matter.

An alternative aspect Jelliffe could consider is to whether ISO has the right to refer to the standard as implementing Schematron. The licence is MIT which does allow ISO to change, re-licence and charge for it but does the licence cover the name?

[Anonymous Coward]

Not really anything to do with ISO...

IIRC - ISO only requires that the final published paper with their logo on it is copyrighted. If the original authors want to give the text away, that's their choice.

At the end of the day, most standards exist before they are published by ISO (a process that takes ages, anyway).

[Claptrap314]

Never understood

How some "industry standard" could be anything other than freely available without running afoul anti-cartel laws.

But then, I an inventor, not a lawyer.

[stiine]

Re: Never understood

If any of the ISO standards have been designated as legal requirements (for example, building codes), then in the US, these standards must be available no charge. Doesn't matter if they're ISO, ASE, etc.

[Claptrap314]

Re: Never understood

Huh. Then someone ought to inform the ISO about that...

[J. Cook]

Re: Never understood

... Someone should tell the NEC and the NFPA that, then.

[boblongii]

ISO

ISO exists to make money to pay its staff. So they churn out specifications and get their mates to make them legally binding. I despise them.

[DrSunshine0104]

Similar but Different

Before changing course into IT work, I worked in architecture and building industry in the US. The largely adopted ICC building codes were used by municipalities all throughout the States but were constantly behind a paywall. It finally took a law to achieve open access. Getting to the building codes is still a maze of links, and adverts for paid access, as well as missing some common features to make navigation easy, but you can do it.

I would love a free access to the ISO standards. I hate having to scour university websites to find ISO standards that are at times out of date or incomplete.

If you want it to be a standard, you cannot hide it from the people who will use it.

Hell, I will whitelist their site on my ad blocker for them.

[Yet Another Anonymous coward]

Re: Similar but Different

What's even worse is when you don't need to read the standard, because either the technical work is being done by a supplier, or more often the standard is so vague that it doesn't tell you anything.

but you are required to buy all the standards that refer to your product and keep buying updated version.

[martinusher]

Its why the IETF exists

The problem with ISO and the cost of standards has been around for decades. One of the first places I came across it was in telecommunications where the raft of standards was not only voluminous but also rather expensive. This made implementing networking a time consuming and frustrating experience. Then along came IETF and their DiY approach to standards. A bit chaotic at times but spreading the wealth meant that things were widely implemented and rapidly improved. The older (and usless) standards still exist and are promulgated and incorporated into modern standards work (looked at a WiFi packet closely, anyone?) but as a rule anything that involves the older standards is a) inefficient and b) doesn't work.

[Anonymous Coward]

Sounds like we need ISOhub. I don't know how to set it up if I did I would.

[YetAnotherJoeBlow]

More thieves...

What qualifies these immoral leeches to be gatekeeper of those jejune standards anyways?

Seriously, I have a friend who could be convinced to donate an industrial strength host to make those standards available at no cost as well as publishing their torrents.

These people would sell their mother.

[W.S.Gosset]

Re: More thieves...

No no no no no no!

They would RENT her.

[DrXym]

Put it this way

If someone were to snaffle up all the ISO standards and dump them on some free repository, I for one would not be complaining.

That said, I've read (legitimately obtained) ISO standards that are reminiscent of the "Handbook for the Recently Deceased" in Beetlejuice - impenetrable gobbledegook that says things without explaining those things or putting those things in an order that makes them comprehensible or useful to someone trying to learn them or transfer into an actual product.

[ortunk]

Re: Put it this way

I had to read through 27k and prepare a plan for certification, the amount of reference loops and gobbydygook was impressive.

[adfh]

Same thing with Standards Australia

Standards, required by law to be followed, behind a paywall. Doesn't seem right that something mandated by law is behind a paywall.

[anonanonanonanonanon]

Why is it so troublseome to find specifications?

Although it now seems to now turn up in google search, last year I needed the very specific definition of the EXIF format to try and debug and fix a very annoying problem we had with one implementation we were using. Could I find the damn doc anywhere? No, there were some pages that gave a fairly detailed breakdown, but never quite complete enough. The exif.org website has it, but you cannot find it from the home page. I think I first found it through a chain of links leading somehwere where someone mentioned it in passing in a comment.

[tapanit]

Let's start a campaign for EU regulation to the effect that if anything is called a standard or referred to in any legislation or similar by the EU or any member state, it must be freely available. Might add that any member states are banned from participating in any organization that tries to copyright standards. And a similar federal legislation in the USA and wherever. (I would not go as far as making all ISO employees criminals by definition, though.)

[fidodogbreath]

...We are not reneging the freely available status. It’s simply not requested by the committee...

"Doublethink lies at the very heart of Ingsoc, since the essential act of the Party is to use conscious deception while retaining the firmness of purpose that goes with complete honesty."

[jon honeyball]

Note this from bsi:

https://shop.bsigroup.com/Help/drm/

"You are required to have the FileOpen plug-in installed in order to access all electronic BSI documents that are purchased from the BSI Shop.

FileOpen document rights management solutions are trusted by publishers, corporations, universities and governments worldwide to protect their digital assets from copying, piracy, and unauthorized sharing.

You will need an internet connection to open the documents for the first time. Once your documents have been authenticated, you will be able to access the documents offline."

and

"’I'd like to share the document with my colleague. Can I?

An individual document (in this case, a PDF file) viewed after download from Shop or BSOL is licensed to a sole named user who is permitted to install a single electronic copy of it,for use on a single computer.We understand that users often view their standards from multiple computers however and have built in flexibility for you to open documents on up to three devices to allow for this."

and

"How many times can I print the document?

You are permitted to print a single copy of a document. Printing one or many pages, or the entire document will count as a single print so we recommend that if you need a printed version, that you print the entire document."

[Haudiobe]

Deja vu

https://www.3gpp.org/ftp/tsg_sa/WG4_CODEC/TSGS4_112-e/Inbox/S4-210293.zip

Same for DASH and ISO File Format.

[morges2021]

Deja vu - also for the popular JPEG-1 (ITU-T T.81¦ISO/IEC IS 10918-1) standard

JPEG (Joint Photographic Experts Group) was originally what the abbreviation says. It was neither an ITU, nor ISO/IEC group, but an informal group of experts (with about 15 core experts) from CCITT (ITU) and ISO. It operated from 1986-1993 only in an informal manner with its own rules and documentation - before the formal agreement of joint working between ITU and ISO/IEC came into force in 1993. The differences were in different policies (like patent policy - RF "baseline" and RAND "options", individual participation, not companies, countries, NBs; completely different documentation, meeting rules etc. etc., and approval by the group when the specification was finished. This was the so so-called JPEG-8 R5 Specification in 1990 which was picked up by the Independent JPEG Group (IJG) to create an Open Source Software which came out in 1991 free of charge to anybody.

The Specification then formally was brought into the ITU and ISO/IEC JTC1 respectively, and was independently approved by both parties, in the ITU in 1992 and in ISO/IEC in 1994. The standards bodies where the specification was brought in: CCITT SGVIII "New Image Communication Group" and ISO/IEC JEC1 SC29 WG10. The successor of the ISO/IEC Working Group even today calls itself as "JPEG", but this is more marketing really and not the early group, because that group is just a normal JTC1 Working Group with JTC1 policies.

The ITU and ISO standard which based entirely on the JPEG-8 R5 specification was after parallel approval and publication sold by ISO and ITU separately in their shops. The ISO version was sold both centrally by the ISO in Geneva but also by the national standardization publishers, like DIN Beuth Verlag in Germany.

Today, after almost 30 years the situation of the sell of the submitted JPEG based specification is the following:

Purchase prise for ITU-T T.81¦ISO/IEC 10918-1 standard (according to accesses to the respective online-shops):

ISO Geneva = 198 CHF

DIN Beuth Verlag in Germany = 214,60 EUR

ITU-T Geneva = 61 CHF

Those, who know how ITU publications work will now be surprised how this can be, because from the ITU Server you can normally download the ITU standards in PDF Format free format. But not for this one, because there is this "Committee" in ISO (not the technical guys in JTC1 SC29) who decide that those joint text standards with the ITU which are commercially valuable should not be given away free of charge, Though this is against the announced policy of ISO (namely to provide such joint standards free of charge after 1 year after the first publication)....

Well, the work was done by JPEG on a completely voluntary basis. They submitted the spec free of charge in a format that was already very close to the ITU/ISO joint text publication format both to ITU and ISO. And they sell it for good money. Deja vu....

You may ask, why do they do it. Very simply because this is the business model of ISO and IEC. For many ISO areas (e.g. standardization of I do not know "rubber tyres"...) this business model works well, but for ICT and especially in Open Source ICT space (like the software code of IJG) this does not work. The problem is that to change that is for them very difficult. Both ISO, IEC in Geneva and the National Standardization Organization rely to significant share on the income from sales of publications. Unfortunately, they apparently do not look into details where and how the publications came from and if the business model they follow fits also to the business model in the ICT space of today... But hopefully one day they realize this and correct it.

[Anonymous Coward]

Public libraries should have standards

I remember looking up a British Standard in a public library when I was writing some freeware in the 1990s. I think it was on microfilm.