Management
Technology
- Applications
- Business Intelligence
- Development
- Hardware
- Mobile & Wireless
- Networking
- Internet
- Operating Systems
- Security Products
- Servers & Datacentre
- Storage
Toolbox
Training
Books
White Papers
Webcast
Resource Centre
A Jesuit's Guide to Open Standards
August 03, 2009
Posted by: Glyn Moody
One of the hottest topics in computing today is that of open standards. That's largely being driven by the interest in them from national governments – particularly in Europe. As a result, what was once a dry-as-dust subject has suddenly become crucially important in terms of winning large government contracts.
The idea behind truly open standards is to create a level playing field so that everyone can compete on an equal and fair basis. The benefits are obvious: it ensures a true Darwinian selection process is possible, leading to lower costs and more rapid innovation. Moreover, it's attractive to those handing out major government contracts since it avoids lock-in: if a standard is truly open, anyone can support it without needing to ask permission or pay licensing fees.
Since the move to open standards is pretty hard to argue against – not least because it would be impolitic to argue for less competition and more uneven playing fields – those who stand to lose from this trend have had to resort to Jesuitical casuistry – as here, which introduces the concept of “balance” to open standards:
The balance issue I speak of stems from the idea that many in the “open standards” discussion have placed an over-emphasis on implementers of standards – particularly those who want to implement under licensing that conflicts with patent terms. Essentially the simplified argument is that because those licensing models are not compatible with patent terms, then there should be no IP protections in the standards process so those implementers can implement. That is not balance – that is imbalance.
The logic here seems to be that there would be an “imbalance” in open standards if it were insisted that patents were excluded – because balance obviously means having standards with and without patents. While it's true that creates a “balance”, it's a purely linguistic one; the fact is that patent-encumbered standards requiring licensing fees cannot, by definition, be open. That's because they do not create level playing fields: there is always one or more players who occupy a privileged position. So the balance is entirely specious.
What's particularly curious is that the same posting – from Microsoft's Jason Matusow – actually includes an extremely good articulation of why it is in the interests of companies to take part in the creation of truly open standards – that is, without trying to assert patents:
The vast majority of ICT standards are created based upon some original contribution from a commercial entity. These submissions are done in order to lay the foundation for a standard, and that decision to contribute is based on the idea of getting some return on the contribution. The return is NOT necessarily about royalties. The return may be that a product that includes the standard as part of it will do better in the marketplace because of the broader adoption of the baseline technology as a standard. The return might be in improved interoperability of a given product or service due to the adoption of that standard. There are competitive reasons for contributions – the hope may be to displace a competitor who is using a non-standardized solution.
This is spot-on, and completely undermines the later statement that:
As a society, we want contributions to standards to continue. If you make the standards environment hostile to contributors…they will contribute less.
They won't, for the reasons Matusow made before.
Two final points...about his “two final points”:
There is no evidence to date that a GPL product can’t implement a standard with IP restrictions. There are many, many RAND standards implemented in GPL-licensed products today.
If Matusow means products that include both GPL'd code and RAND (reasonable and non-discriminatory) standards, that's certainly true, but beside the point, which is that you can't include RAND standards *in* GPL'd code. As the guide to the GNU GPLv3 explains:
Whenever someone conveys software covered by GPLv3 that they've written or modified, they must provide every recipient with any patent licenses necessary to exercise the rights that the GPL gives them.
That means *without* payment, not just for a reasonable payment, which may be small, but is still impossible for software that can be freely distributed.
And as for the second point:
I am all for the idea of “open standards,” but to me it is all about process and participation.
That is rather rich in the light of the OOXML standardisation “process” at the ISO. As I wrote elsewhere recently, after listing a number of well-documented examples of Microsoft's “participation” there:
Taken individually, these dubious actions might be dismissed by Microsoft as “minor lapses”, “misunderstandings” or actions of an atypical “rogue” manager – as was done in Sweden, where an offer was made to support partners financially if they attended the key meeting of the national body and voted in favour of OOXML. But taken together they suggest a consistent philosophy of being prepared to use whatever means necessary in order to gain the required number of votes.
Against that background of a standardisation process being bent to breaking point, complaints about the *balance* of open standards ring rather hollow.
Follow highlights from ComputerworldUK on Twitter
Sign up for our Daily Newsletter
The UK IT News widget Get it for your site!
<<newer entry | back to blogs index | older entry>>
Advert
Email this article to a friend or colleague:
PLEASE NOTE: Your name is used only to let the recipient know who sent the story, and in case of transmission error. Both your name and the recipient's name and address will not be used for any other purpose.
- This article is now being printed.
What are your views on this subject? Use the form below to post a comment on this article up to 1000 characters.
Click below to add 'A Jesuit's Guide to Open Standards' to your blog.
If you do not have a ComputerworldUK Account and would like to use this feature, please Register.
If you are a registered, logged-in user, this will post the title and first paragraph of this story to your blog to share with your readers.
Advert
Newsletters
Sign up to our Weekly Round-Up for the past week’s not-to-be-missed news analysis and insight delivered straight to your inbox.
WHITE PAPERS
-
Legal risks: Employee use of the internet and email
Exploring the challenges facing IT Mangers today and vital steps to ensure safe internet an email use by employees. -
Phishing for victims
This White Paper examines the phenomenon of phishing. It explains the potentially catastrophic threat it presents to all kinds of organisation. Exploding some widespread myths, it lights up the murky waters where phishing first emerged and where it continues to evolve. But it also highlights what your business can do to blunt the threat.
About the author
Glyn Moody Glyn Moody's look at all levels of the enterprise open source stack. The blog will look at the organisations that are embracing open source, old and new alike (start-ups welcome), and the communities of users and developers that have formed around them (or not, as the case may be).
Comments received
License Nerd said on Monday, 03 August 2009
GPL is a RAND license, Glyn, so it's hard to see how they are as incompatible as you claim. In fact, almost every open source or free software license is also a RAND license. Basically RAND is a subset of the requirements for a free or open source license.
Some RAND licenses will also be free software licenses, and some RAND licenses will be incompatible with free software licenses. It all comes down to what terms are in a particular license beyond the minimal RAND requirement of being reasonable and non-discriminatory.
al said on Monday, 03 August 2009
Does this mean we can expect MS to start creating GPLv3 standards? Standards need to be _free_ and _open_. This is incompatible with the MS business model. They have always been incapable of competing on a level playing field. That isn't going to change...
Glyn Moody said on Monday, 03 August 2009
Well, GPL is an example of a particular kind of RAND licence - one where the licensing fees are zero. Any other kind of RAND licence with fees would be incompatible.
So, yes, it does come down to the terms, but as far as costs are concerned, there's only *one* class of RAND that would be compatible with the GPL - and an (uncountably) infinite class that are not.
Glyn Moody said on Monday, 03 August 2009
@al: strangely, I believe it will. We've already seen MS move huge distances towards free software; I think it will eventually embrace it fully. Why? Because MS doesn't care about licences, it cares about making a profit: the future lies with open source code, in contrast to the continuing problems with Vista...
Math Nerd said on Monday, 03 August 2009
(Takes off license nerd hat, puts on math nerd hat)
While it is true that there are an infinite number of possible licenses, it is a countable infinity, not an uncountable infinity. Each license can be expressed as a finite sequence of UTF-8 characters. Take the bits from each byte of those representation in sequence and string them together, and put a 1 bit on front, and treat that sequence of bits as a binary integer. Each possible license corresponds to a bit string generated this way. Given any such bit string, the processes is reversible to recover the license text.
This gives you a one to one correspondence between the set of all possible licenses and a subset of the positive integers, which shows that the set of all possible licenses is countable.
Sam Johnston said on Monday, 03 August 2009
Glyn,
Thanks for taking the time to elegantly dissect Matusow's "argument" for redefining "open" as "closed".
I spend most of my time these days trying to make "Open Cloud" a reality so this is an interesting data point to have for my future dealings with Microsoft.
You may be interested in the "Microsoft definition" of "open standard" in Wikipedia[1] courtesy Vijay Kapoor:
"Let's look at what an open standard means: 'open' refers to it being royalty-free, while 'standard' means a technology approved by formalised committees that are open to participation by all interested parties and operate on a consensus basis. An open standard is publicly available, and developed, approved and maintained via a collaborative and consensus driven process."
That's right, "'open' refers to it being royalty-free".
"Open" always trends towards "more open". Resist at your own peril Microsoft.
Sam
1. http://en.wikipedia.org/wiki/Open_standard
Glyn Moody said on Monday, 03 August 2009
@mathnerd: welcome back. It's a neat 1-to-1 mapping, but it omits all those licences where the fee is an amount that cannot be represented by a finite sequence of UTF-8 characters - for example if I charge pi Euros, or other irrational numbers.
Glyn Moody said on Monday, 03 August 2009
@mathnerd: welcome back. It's a neat 1-to-1 mapping, but it omits all those licences where the fee is an amount that cannot be represented by a finite sequence of UTF-8 characters - for example if I charge pi Euros, or other irrational numbers.
Glyn Moody said on Monday, 03 August 2009
@Sam: well spotted, very interesting. good luck with the open cloud - you'll need it...
Math Nerd said on Monday, 03 August 2009
Pi euros can be represented in a finite number of UTF-8 characters. In hex: 70 69 20 65 75 72 6f 73. That's simply the string "pi euros", which does a pretty good job of representing pi euros. :-)
Harry Split said on Tuesday, 04 August 2009
@Math Nerd: And what about licences where the fee is defined as "the smallest positive number that cannot be expressed with a finite sequence of UTF-8 characters"? :P
Glyn Moody said on Tuesday, 04 August 2009
@Mathnerd: OK... But if the licence fee can be any real number, there are an uncountable number of licence fees, so it's still uncountable...
Glyn Moody said on Tuesday, 04 August 2009
@Harry: you are Kurt Gödel, and I claim my five pounds....
An engineer said on Tuesday, 04 August 2009
Euros are not, however, a continuum of real numbers. No currency is. Currency is always quantized, to the cent, or to the penny. License fees can charge fractional cents, of course, and then ignore the fraction left over after multiplying by the number of licenses. However, it can probably be assumed that the license fees can be quantized to within units of (the minimum quantization limit for the currency) divided by (the population of the Earth). Given that RAND terms imply a maximum possible license fee, the number of possible license fees then becomes finite (not merely countably infinite, but actually finite).
Glyn Moody said on Tuesday, 04 August 2009
@engineer: I'm a *mathematician*, what do I care for your eminently sensible practicalities?
Gerd said on Tuesday, 04 August 2009
It's funny laugh. "Open standard" means with RF patent license or no patents, usable without any restrictions, with no control of a single party. But there is more to it: who invited the Redmond company to comment on European policies with their misguided advice? They have to be told that they are not at home here and need to be more conservative and adapt to our rules. Europe to the Europeans. They should be put in the plane and sent back home to Seattle.
Alex said on Tuesday, 04 August 2009
More stories like this need creating in simpler terms. I'd love to post this to some friends who I'm still struggling to have comprehend Free Software as a better (more humane?) way. But it has too much dependence on knowing the intimate histories referred to.
Maybe I'll manage to give it a go.
Glyn Moody said on Tuesday, 04 August 2009
@Alex: that would be great - please feel free to use it as you find helpful. Unfortunately, my ability to convey this stuff in simpler terms is limited.
Math Nerd said on Tuesday, 04 August 2009
@Glyn; Yes, the real numbers are uncountable, but you can't specify an arbitrary real number in a license. You are limited to the real numbers that can be expressed in the language that the license is written in, and that's merely a countable infinity. You have to be able to identify the number in the license, and any number a human can identify of necessity can be identified in a finite number of symbols.
Jack Dausman said on Tuesday, 04 August 2009
Great stuff. Even the "Jesuit" metaphor fits into the discussion, ala Christopher Kelly's "Two Bits: The Cultural Significance of Free Software." Ironically, I help support an open source initiative at a Children's Residence in Nairobi (Nyumbani), which is happily dependent on Ubuntu/Edubuntu. Nyumbani is Jesuit founded !
Glyn Moody said on Wednesday, 05 August 2009
@Mathnerd: I was afraid you'd say that. This sounds like a well-known problem to me - someone must have sorted this out definitively before; unfortunately I'm an applied mathematician by training, so my number theory isn't good enough to rebut your last point.... Anyone else out there know/care (probably a vanishingly small subset of readers, alas....)?
Glyn Moody said on Wednesday, 05 August 2009
@Jack: I'm actually a big fan of Jesuits. I just hope the next Pope comes from their numbers...
Glen Butterworth, S.J. said on Tuesday, 25 August 2009
"Jesuitical casuistry?" Really. I'm a 21st century Jesuit and you should know that I find this offensive. If you were writing about 17th century theology, then the use of this phrase would be appropriate. In this context it's just unprofessional, anti-Catholic, and inappropriate.
Thanks for receiving my feedback.