Claims of Conformance (was: Re: CLDR and ICU)

[Ken Whistler]

On 7/26/2012 4:20 PM, Richard Wordingham wrote:

Perhaps I've read too much into
http://www.unicode.org/policies/logo_policy.html  .  The implication is
that untrue or misleading claims using the word 'Unicode' are
contravening the trademark.


That's more on the level of making sure that when you use the Unicode Mark
mark, you are actually referring to the Unicode Standard, the Unicode 
Consortium,

and so on.

You cannot slap the Unicode Mark on a self-publication of UTF-37 with your
own idiosyncratic code tables and call that Unicode. *That* would be a 
violation

of the trademark.

It is a whole nother kettle of fish when somebody says of their product
This product conforms to the Unicode Standard, Version 6.2.0. There
would be nothing misleading about their use of the Unicode Mark in
such a case -- they are actually referring to the actual standard which
claims the trademark. The reference is not misleading.

But the *claim* of conformance could be false, if their product is examined
in detail. (Or tested, or reverse engineered, or whatever.) And *that*
is the part that the Unicode Consortium has neither the personnel nor
the inclination to be chasing after. The Consortium cannot police such
claims, especially for a standard as widely implemented as this one.

The same would apply to claims of conformance to the other standards,
such as UCA, LDML, Unicode Regex, and so on.

--Ken

Re: Claims of Conformance (was: Re: CLDR and ICU)

[Andrew West]

On 27 July 2012 00:42, Ken Whistler k...@sybase.com wrote:

 It is a whole nother kettle of fish when somebody says of their product
 This product conforms to the Unicode Standard, Version 6.2.0. There
 would be nothing misleading about their use of the Unicode Mark in
 such a case -- they are actually referring to the actual standard which
 claims the trademark. The reference is not misleading.

Yet such a claim would be wrong according to the Trademark Policy
page, because they omitted the ® symbol and used the word conforms
(they should have stated This product is compliant with the Unicode®
Standard, Version 6.2.0.).  The page clearly states that any claim of
conformance is not allowed to be made if the Unicode Word Mark
guidelines are not followed (e.g. omitting the ® after Unicode, or
using a verb other than use, implement, support, or are
compliant with), which implies that any wrongly formulated or
formatted claims of conformance are null and void, and should not be
accepted by potential users of the product.

I am sure we have discussed how stupid this page is on this list
before, and I for one refuse in principal to add the ® symbol to
Unicode when, for example, I claim conformance to the Unicode 6.1
normalization algorithm for BabelPad.  Perhaps people should be wary
of using my software because the Unicode Word Mark is misused, but
more likely they will think that Unicode's (oops!) trademark policy is
a little bit silly.

Andrew

Re: Claims of Conformance (was: Re: CLDR and ICU)

[Philippe Verdy]

You may still use the terms Universal Character Set. However, claims
of conformance is declarative. It may have unexpected bugs soemwhere
in which cas the claim should come with either a disclaimer of
warranty, or with a reliable contact address for the support to have
these conformance bugs corrected or more reliably detected and avoided
possibly with discussed workarounds.

But as long as the conformance is not proven false, such a use of the
expression conforms to the Unicode standard along with the
disclaimer of warranty or with the support offer with limted liability
of the author or distributor should be OK. However for references to
the character encodings, it will still be better to use the registered
encoding abbreviations of the supported UTF's. Conformane to a
standard UTF does not mean full conformance with everything in TUS.

For example I can still see many fonts that claim to be made for
Unicode, or in Unicode version. They are not more conformant than many
fonts that do not claim it, only because they support only a smaller
subset of the UCS. There does not exist any font that fully supports
every characters assigned in the Unicode/ISO/IEC 10646 standards, but
stil there are claims in the used font names (without clarifications).
In my opinoon all those claims are false, simply because there's no
conformance rules defined in TUS for fonts (almost everything in those
claims is not found in fonts thelselves, but in rendering engines,
wich are parsing the text and trying to map glyph using the separate
specifications of the supported font formats).

2012/7/27 Andrew West andrewcw...@gmail.com:
 On 27 July 2012 00:42, Ken Whistler k...@sybase.com wrote:

 It is a whole nother kettle of fish when somebody says of their product
 This product conforms to the Unicode Standard, Version 6.2.0. There
 would be nothing misleading about their use of the Unicode Mark in
 such a case -- they are actually referring to the actual standard which
 claims the trademark. The reference is not misleading.

 Yet such a claim would be wrong according to the Trademark Policy
 page, because they omitted the ® symbol and used the word conforms
 (they should have stated This product is compliant with the Unicode®
 Standard, Version 6.2.0.).  The page clearly states that any claim of
 conformance is not allowed to be made if the Unicode Word Mark
 guidelines are not followed (e.g. omitting the ® after Unicode, or
 using a verb other than use, implement, support, or are
 compliant with), which implies that any wrongly formulated or
 formatted claims of conformance are null and void, and should not be
 accepted by potential users of the product.

 I am sure we have discussed how stupid this page is on this list
 before, and I for one refuse in principal to add the ® symbol to
 Unicode when, for example, I claim conformance to the Unicode 6.1
 normalization algorithm for BabelPad.  Perhaps people should be wary
 of using my software because the Unicode Word Mark is misused, but
 more likely they will think that Unicode's (oops!) trademark policy is
 a little bit silly.

 Andrew