Hello all,
I am dealing with a paper about standards and Free Software and I've found out that the term 'open standard' is a very common term in literature (and commercial advertising). According to most of the definitions I found, an 'open standard' can be patented and, at best, subject to RAND licensing policies. This translate into 'open standard _can_ be impossible to implement in Free Software'.
Now, since I am summarizing in this paper what defines a standard that is implementable in Free Software, it would be nice to propose also a term that is non controversial like 'open standard'.
Free standard is not good: I don't think it's savvy to replicate the fight between 'open' and 'free'. A friend whispered 'non discriminatory standard' but he agrees that the negation at the beginning is less than optimal. I couldn't think of any more solutions, so I ask here two questions:
1) does it make sense to introduce in the Free Software community a new term that is non-controversial and more precise than the generic 'open standard'?
2) if yes, what would that term be?
Any suggestion appreciated. bye stef
On Wed, 2006-07-19 at 10:02 +0200, Stefano Maffulli wrote:
- does it make sense to introduce in the Free Software community a new
term that is non-controversial and more precise than the generic 'open standard'?
There is an ongoing debate over what constitutes an "open standard", see for example the definitions that ETSI have proposed to Global Standards Collaboration (GSC):
"An open standard is developed, approved, and maintained, by a collaborative, transparent and consensus-based process, open to all materially affected and interested parties.
"The standard is subject to FRAND Intellectual Property Right (IPR) policies which do not mandate, but may permit, at the option of the IPR holder, licensing essential intellectual property without compensation.
"The standard is published and made available to the general public under reasonable terms (including for reasonable fee or for free)."
This matches closely the definition of "open standard" used in telecommunications and electronics. It doesn't particularly match what you're talking about. It seems to make sense to me, then, that "open standard" is not a sufficient term.
- if yes, what would that term be?
Probably "royalty free standard". While it doesn't encapsulate all the problems in a single name, it's already a known phrase, and there will always be new problems in the future. If you can use the standard without having to pay a fee, it seems to me that it would be difficult to make the standard incompatible with free software.
Cheers,
Alex.
On 19-Jul-2006, Stefano Maffulli wrote:
Now, since I am summarizing in this paper what defines a standard that is implementable in Free Software, it would be nice to propose also a term that is non controversial like 'open standard'.
First advice: don't have "non-controversial" as a goal, or you will fail. The very idea of freedom in software is controversial, so you must enter any effort acknowledging this.
As for a suggested term: I find it descriptive to discuss "open, freely-implementable standard". That at least is more precise, and leaves the option open to talk about free software implementations of the standard.
Free standard is not good: I don't think it's savvy to replicate the fight between 'open' and 'free'.
That fight exists only to the extent that the *concept* of freedom in software has been corrupted by other forces. You can't avoid that conflict by choosing some other term, except by avoiding the concept of freedom altogether. You can prepare your responses in advance, though, which helps.
- does it make sense to introduce in the Free Software community a
new term that is non-controversial and more precise than the generic 'open standard'?
Precision is good. Low ambiguoty is good. Non-controversial is only possible to the extent that the discussion of freedom in software is non-controversial.
Hi,
Stefano Maffulli ha scritto:
Now, since I am summarizing in this paper what defines a standard that is implementable in Free Software, it would be nice to propose also a term that is non controversial like 'open standard'.
[...]
- does it make sense to introduce in the Free Software community a new
term that is non-controversial and more precise than the generic 'open standard'?
Since the paper will try to define another idea of 'standard', that is: that standard which can be implemented within the Free Software and therefore compatible with certain kind of licences (such as GPL), yes i believe that not only it makes sense to introduce another term, but it is *necessary*.
'Open standard' *already* refers to a set of standards that, in some cases, would not be compatible with FS. Using that same name to point to another concept (i.e: only those standard which have full compatibility with FS), will necessarily lead to confusion and misunderstanding.
- if yes, what would that term be?
Considering that the adjective 'open' is already used in 'open standard', i would avoid to restrict that adjective adding more ones to it: we would end up having not an *appropriate name* but a *descriptive* term, and since descriptive names are not real names they are very prone to be shortened by common and daily usage.
To conclude, i would suggest to take some distances from the 'open'/'free' terminology and use a new term, something like: fair standard community standard transparent standard bright standard or something like that, i don't have lot of imagination, but i hope i explained what i meant :P
Greetings, Giacomo
On Wed, 2006-07-19 at 10:02 +0200, Stefano Maffulli wrote:
- if yes, what would that term be?
well, thanks everybody for the comments.
Alex Hudson suggested:
Probably "royalty free standard"
It is a fine term, but probably it carries some confusion: what is royalty free? The patent license? or the access to the specification? I think it is acceptable to pay for the specifications of a standard, provided that implementing the standard and distributing software in source form is allowed. So probably royalty free is not the best solution.
Ben Finney suggested:
"open, freely-implementable standard"
not bad, but long (and even longer in Italian: implementabile liberamente)
Sean Daly suggested:
"open unencumbered standard"
and Sam Liddicott added an 'and' to it. That is a fine term, too. But like non-discriminatory it carries a negation in front. In any case I couldn't find a simple translation in Italian and gave up on this too.
Giacomo Poderi suggested:
to take some distances from the 'open'/'free' terminology and use a new term, like: fair standard
and I stop here. I like this: fair is a good term. Like in "fair trade" or "fair play" it carries a positive meaning, non discrimination is included. IMHO we have a clear winner here. What do you think?
cheers stef
On Wed, 2006-07-19 at 19:22 +0200, Stefano Maffulli wrote:
Alex Hudson suggested:
Probably "royalty free standard"
It is a fine term, but probably it carries some confusion: what is royalty free? The patent license? or the access to the specification?
I think generally it's well-accepted to mean implementing the standard - it's not about how much it costs to get a standard document, or not.
The issue with it is that while the terms may be royalty-free, people could dream up other restrictions.
Giacomo Poderi suggested:
to take some distances from the 'open'/'free' terminology and use a new term, like: fair standard
and I stop here. I like this: fair is a good term. Like in "fair trade" or "fair play" it carries a positive meaning, non discrimination is included. IMHO we have a clear winner here. What do you think?
I think it's very much in the eye of the beholder and too descriptive. If you start asking for fair standards, people will say their standards are fair.
"Reasonable" has already been used to mean "you can pay a reasonable fee" for our licence. I don't see why fair couldn't mean "you can pay a fair fee" for our licence.
Cheers,
Alex.
On Wed, 2006-07-19 at 18:34 +0100, Alex Hudson wrote:
I think it's very much in the eye of the beholder and too descriptive. If you start asking for fair standards, people will say their standards are fair.
"Reasonable" has already been used to mean "you can pay a reasonable fee" for our licence. I don't see why fair couldn't mean "you can pay a fair fee" for our licence.
I see your point and I agree with it: fair may be not a good term in English after all. In Italian fair is translated into 'equo' (for example, fair trade is commercio equo e solidale). 'Equo' (latin aequus) carries a more evident meaning of equity. The message to send is that there is truly no discrimination.
Looking at my English dictionary the only term with the same latin root in English is 'equitable' but that is not an easy word, I guess. Is there some other term that can summarize Royalty-free and Non-discriminatory?
Jeroen said:
I've seen the EC using this definition too,
It's not the EC that uses that definition but IDABC (http://europa.eu.int/idabc/) in its European Interoperability Framework (http://ec.europa.eu/idabc/en/document/2319/5644)
I think that definition of Open Standard is too restrictive also for free software, not only for BSA, but this is another topic and I'll save it for the next discussion :)
cheers stef
On Thu, 2006-07-20 at 07:40 +0200, Stefano Maffulli wrote:
Looking at my English dictionary the only term with the same latin root in English is 'equitable' but that is not an easy word, I guess. Is there some other term that can summarize Royalty-free and Non-discriminatory?
Not really. "Equitable" is common English, but tends to be more applied to contracts, agreements or settlements - meaning both sides come out equal.
I guess you can say "RF-AND" (Royalty-free and non-discriminatory), as opposed to RAND, and people would figure out what you were saying.
Cheersm
Alex.
At Thu, 20 Jul 2006 11:25:02 +0100, Alex Hudson wrote:
I guess you can say "RF-AND" (Royalty-free and non-discriminatory), as opposed to RAND, and people would figure out what you were saying.
Royalty-free is always non-discriminatory.
Jeroen Dekkers
On Fri, 2006-07-21 at 13:52 +0200, Jeroen Dekkers wrote:
At Thu, 20 Jul 2006 11:25:02 +0100, Alex Hudson wrote:
I guess you can say "RF-AND" (Royalty-free and non-discriminatory), as opposed to RAND, and people would figure out what you were saying.
Royalty-free is always non-discriminatory.
That really rather depends on your point of view; if RF was always non-discriminatory we probably wouldn't be even having this discussion.
If I have patents which cover standard X, and my licence is "You may implement my patents so long as you do not provide the source to your software", I think most here would consider that to be discriminatory against free software.
The fact that I don't charge for that license doesn't really come into it.
I agree that any requirement for royalty is probably the largest stumbling block, though.
Cheers,
Alex.
At Fri, 21 Jul 2006 13:05:44 +0100, Alex Hudson wrote:
On Fri, 2006-07-21 at 13:52 +0200, Jeroen Dekkers wrote:
At Thu, 20 Jul 2006 11:25:02 +0100, Alex Hudson wrote:
I guess you can say "RF-AND" (Royalty-free and non-discriminatory), as opposed to RAND, and people would figure out what you were saying.
Royalty-free is always non-discriminatory.
That really rather depends on your point of view; if RF was always non-discriminatory we probably wouldn't be even having this discussion.
If I have patents which cover standard X, and my licence is "You may implement my patents so long as you do not provide the source to your software", I think most here would consider that to be discriminatory against free software.
The fact that I don't charge for that license doesn't really come into it.
I agree that any requirement for royalty is probably the largest stumbling block, though.
If something is made available royalty-free, everybody is free to use it. That's the only sensible definition of royalty-free. If only a small or big group is able to use it without paying, it simply isn't available royalty-free.
Jeroen Dekkers
On Fri, 2006-07-21 at 14:35 +0200, Jeroen Dekkers wrote:
If something is made available royalty-free, everybody is free to use it.
Or that everyone is able to use it for free.
My previous example of a licence was available to all: the discrimination comes on the part of the developer willing or not to accept its terms. It's still available royalty-free to all, though.
While you might not accept the definition, it's well-accepted in general that "royalty free" doesn't imply any other non-financial meanings.
Cheers,
Alex.
|| On Wed, 19 Jul 2006 19:22:01 +0200 || Stefano Maffulli stef@zoomata.com wrote:
sm> I like this: fair is a good term. Like in "fair trade" or "fair sm> play" it carries a positive meaning, non discrimination is sm> included. IMHO we have a clear winner here. What do you think?
I see more problems with fair than with free, to be honest.
While with free we always have to explain the difference between freedom and price, freedom is comparatively well-defined when looking at other terms, in particular open.
On the other hand: ask someone what is fair and the answers you will get will deviate much, much more. Microsoft certainly considers it fair to pay patent royalties -- so will some other companies. This is a term that can backfire badly on us.
I think you should consider taking reference to "Open Standard" and "Free Standard" and then use simply "Standard" consistently, defining it along the lines that work for Free Software and everyone else.
Because in the end, and "Open Standard" is something of an oxymoron: if only one or a few players are using it, it is merely a proprietary format or interface, but not a standard.
Regards, Georg
Am Mittwoch, den 19.07.2006, 19:41 +0200 schrieb Georg C. F. Greve:
I see more problems with fair than with free, to be honest.
I concord with this.
Here a vision about a sort of categorizing the standards into 3 levels:
1) Free Standard
Fully documented with at least one Free Software reference implementation.
2) Open Standard
Fully documented.
3) Proprietary Standards
Not fully documented and/or covered by patents or other freedom restrictions.
Today everybody is talking about "Open Standard", IBM, Microsoft, etc. This results in an awareness about Open Standards in decision makers of the PA and the economy.
It will be difficult to propagate the term of "Free Standard" against the PR departments of IBM and Co. pushing the term of "Open Standard".
Even being it difficult it is worst trying, but maybe it would be wiser to try to create an official and widely accepted definition of "Open Standard", which matches our views of "Free Standard". This would mean, that PR departments of IBM and Co. will work for the propagation of "Free Standard".
Maybe it could be done creating a definition together with the ODF Alliance, OASIS and other well known entities.
I don't know if this is possible, I just hope their is a chance. Can someone with more insight give a comment?
Happy hacking! Patrick
At Wed, 19 Jul 2006 20:25:52 +0200, Patrick Ohnewein wrote:
Today everybody is talking about "Open Standard", IBM, Microsoft, etc. This results in an awareness about Open Standards in decision makers of the PA and the economy.
It will be difficult to propagate the term of "Free Standard" against the PR departments of IBM and Co. pushing the term of "Open Standard".
Even being it difficult it is worst trying, but maybe it would be wiser to try to create an official and widely accepted definition of "Open Standard", which matches our views of "Free Standard". This would mean, that PR departments of IBM and Co. will work for the propagation of "Free Standard".
I don't think that this is a valid argument. It is fairly difficult to propagate the term "Free Software" as well. Everything is plagued by "Open Source", including the companies you refer to. It doesn't mean that we should give up because PR departments, large/small companies, journalists, and unfortunately, Free Software developers, use the wrong term.
So, even if it would be difficult to propagate a new term, I think it's worth the effort.
Am Freitag, den 21.07.2006, 20:27 +0300 schrieb Yavor Doganov:
I don't think that this is a valid argument. It is fairly difficult to propagate the term "Free Software" as well. Everything is plagued by "Open Source", including the companies you refer to. It doesn't mean that we should give up because PR departments, large/small companies, journalists, and unfortunately, Free Software developers, use the wrong term.
You are surely right about the terms "Open Source" and "Free Software".
I am asking myself if the term "Open Standard" would be fine for us. I followed long discussions trying to create some definition of Open Standard or Free Standard. Today we have no definition and just an undefined term "Open Standard" to which people gets more and more familiar.
It's very difficult to propagate a term, when another is already known for a sort of concept. It is much easier to introduce a term with the concept than renaming the already familiar term of a concept.
So, even if it would be difficult to propagate a new term, I think it's worth the effort.
You are right: If the the term "Open Standard" is not acceptable for the Free Software community, it will be worth to fight for the introduction of a new term.
My question is: Is "Open Standard" an acceptable term for the Free Software community and can we create our definition under the label "Open Standard"?
Happy hacking! Patrick
At Fri, 21 Jul 2006 20:18:08 +0200, Patrick Ohnewein wrote:
I am asking myself if the term "Open Standard" would be fine for us. [...] My question is: Is "Open Standard" an acceptable term for the Free Software community and can we create our definition under the label "Open Standard"?
To be honest, until this discussion I always thought that "Open Standard" = "Free Standard". I just used "free" because "open" is at least ambiguous, needless to mention the association with "Open Source".
To answer your question: I think that nowadays, when the Free Software Movement is constantly threatened to be buried by the "Open Source Campaign", it is our duty to point out the differences and establish a new term. I think that the new term should be called "free standard", because namely freedom is what distinguishes it from the already spread out term "open standard".
At Wed, 19 Jul 2006 19:41:03 +0200, Georg C. F. Greve wrote:
Because in the end, and "Open Standard" is something of an oxymoron: if only one or a few players are using it, it is merely a proprietary format or interface, but not a standard.
An Open Standard isn't an oxymoron IHMO. You've got proprietary standards like MP3 and de facto standards like MS Word.
The Dutch government actually requires patents to be licensed royalty-free in their definition of Open Standard (see http://www.ososs.nl/index.jsp?alias=watisos, in the middle there is the English definition). I've seen the EC using this definition too, but got critised for that by BSA/CompTIA/etc, so I'm not sure they are still doing that.
I think using this definition is the way to go: nobody is against Open Standards, the only thing you've convince people of is that a standard isn't open when patents aren't licensed royalt-free.
Jeroen Dekkers
Jeroen Dekkers jeroen@vrijschrift.org claimed:
An Open Standard isn't an oxymoron IHMO. You've got proprietary standards like MP3 and de facto standards like MS Word.
"De facto standards" are often not standardised in any meaningful way, so that phrase is an absurdity. Just look at all the small variations in so-called-standard Microsoft Word files.
[...] http://www.ososs.nl/index.jsp?alias=watisos [...] I think using this definition is the way to go: nobody is against Open Standards, the only thing you've convince people of is that a standard isn't open when patents aren't licensed royalt-free.
It may be beneficial to support someone else's definition, too.
Regards,
|| On Wed, 19 Jul 2006 22:31:15 +0200 || Jeroen Dekkers jeroen@vrijschrift.org wrote:
jd> An Open Standard isn't an oxymoron IHMO. You've got proprietary jd> standards like MP3 and de facto standards like MS Word.
You're missing my point, I think.
I know that these are referred to as standards by their vendors, but ultimately what makes a standard a standard? To me it is the availability for everyone to work and interoperate on the basis of that standard independent of their competitors.
So even though people like to speak about "proprietary standards" or "de-facto standards" for marketing purposes, I don't consider them standards, at all.
In fact I would encourage everyone to speak about "proprietary formats" and/or "proprietary interfaces" in such cases.
Regards, Georg
Georg C. F. Greve ha scritto:
|| On Wed, 19 Jul 2006 19:22:01 +0200 || Stefano Maffulli stef@zoomata.com wrote:
sm> I like this: fair is a good term. Like in "fair trade" or "fair sm> play" it carries a positive meaning, non discrimination is sm> included. IMHO we have a clear winner here. What do you think?
I see more problems with fair than with free, to be honest.
While with free we always have to explain the difference between freedom and price, freedom is comparatively well-defined when looking at other terms, in particular open.
On the other hand: ask someone what is fair and the answers you will get will deviate much, much more. Microsoft certainly considers it fair to pay patent royalties -- so will some other companies. This is a term that can backfire badly on us.
I'm sorry to point out that probably for this there is no real solution: for example Microsoft certainly consider free the software they offer for downloading within their website [0] In the discussion we have been mixed two different aspects 1) What is the definition of a term (therefore what it means) 2) How people use that term in the daily life
We can determine 1), but we cannot determine 2). We can only observe at 2) And this is true for *any* term in any language
Greetings, Giacomo
[0]http://www.microsoft.com/downloads/results.aspx?pocId=&freetext=Free%20s...
|| On Thu, 20 Jul 2006 00:01:59 +0200 || Giacomo Poderi poderi1980@yahoo.it wrote:
gp> I'm sorry to point out that probably for this there is no real gp> solution: for example Microsoft certainly consider free the gp> software they offer for downloading within their website [0]
Nobody said that "free" in English was perfect, but you're missing several points here that make it the best we have:
What Microsoft is referring to on that page is called "freeware" and not Free Software. We obviously know that Microsoft is deliberately trying to ignore the fact that there is a 20 year old scientific definition of "Free Software" that is causing them quite a bit of trouble.
In general it is not the fact that you have enemies that will try to confuse your issue that decides about how good or bad a term is, it is the question of how good you can keep the basis clean.
For Free Software in English there are generally only two ways in which people understand the term: price or freedom. Such a confusion is comparatively easy to clear up. In pretty much any other language you can clearly refer to freedom.
The same is not true for "fair" which means something different for every person on this planet. You might as well use "nice" "good" "lovely" "cozy" "cuddly" or anything else.
Regards, Georg
* Stefano Maffulli wrote, On 19/07/06 18:22:
Sean Daly suggested:
"open unencumbered standard"
and Sam Liddicott added an 'and' to it. That is a fine term, too. But like non-discriminatory it carries a negation in front. In any case I couldn't find a simple translation in Italian and gave up on this too.
Unencumbered has specific meaning, strongly suggesting (in english) that there are no restrictions or controls on use. Other words might be emancipated, unrestricted,harmonious, cordial I agree the "un" is technically a negation but in english it unencumbered and unrestricted read as positive things. Sorry I can't help with the Italian. Maybe you could ask a legal person for the equivalent of "without entailment" in Italian?
and I stop here. I like this: fair is a good term. Like in "fair trade" or "fair play" it carries a positive meaning, non discrimination is included. IMHO we have a clear winner here. What do you think?
"fair standard" sounds like it means the same thing as "RAND", because fair=reasonable. and fair=non-discriminatory. I worry that it therefore fails to even raise controversy because people would think they know what it means when they don't.
Otherwise I like the "open free standard" or "open and free standard."
I think we like "open" because we know what open is supposed to mean, but as acknowledged it often doesn't, the "and-free" works to distinguish it from the wrong sort of openness.
and the "free" could be cash:free or GNU:free and means the same thing in its context here, so that lack of confusion would be a benefit.
Sam
What's wrong with "free standard"? I always use this term and people seem to understand clearly what it means.
On Wed, 2006-07-19 at 21:18 +0300, Yavor Doganov wrote:
What's wrong with "free standard"? I always use this term and people seem to understand clearly what it means.
I strongly agree with this. A standard is free if it is not patent uncumbered. There is no chance of making the mistake between free as in freedom and free as in gratis, as I see no way you can ask money for the implementation of the standard without patents anyway. At most you can use copyright to protect the original specification, but nobody can prevent you to write another description of the standard and distribute it for free or for a fee.
Free standard convey the very notion of being free to implement such standard. I don't see any better wording than Free Standard. Nobody is going to think they cannot charge you money for the book that describes it, there's no chance of misunderstanding that unless you want to do it on purpose..
And no, I don't think it is acceptable that someone can't write it's own version of the specification if he wants to nor I find it acceptable that to implement such standard you are required to buy the official specification if you don't want to.
Simo.