OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

office message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: FYI: It's okay to use UNICODE name as a standard spreadsheet function


FYI: The formula subcommittee has verified that having a function named "UNICODE" is okay with the Unicode Consortium, and won't intefere with the IP mode.  Below is documentation explaining why we believe this, and why we asked.  It's important to record that information at the TC level, so here is that (legal) justification.

==========================================

First, background:

The formula subcommittee has discussed having a function named UNICODE, which takes a character and returns its Unicode/ISO 10646 numeric value.  E.G., UNICODE(" ") is 32, and UNICODE("π") {that's pi} is 960 (U+03C0).  This would be the code point IGNORING any encoding (UTF-8, UTF-7, whatever).  There would be reverse function named UNICHAR, converting number to a character.  These are the obvious extensions to CODE and CHAR, which don't necessarily work for the full Unicode/ISO 10646 set.

I sent an an email to the Unicode Consortium, and got a reply (copied below).  They say there is no problem naming a function "UNICODE".  Since Gnumeric already does this, I'm of a mind to continue the practice (and with a corresponding UNICHAR).  That means there's a corresponding UNICHAR.

The likely workaround would be to name this function "ICODE" (for "International Code"), which would produce a corresponding "ICHAR".    An advantage of this would be to emphasize the international nature of this; really this only involves the standard character encoding agreed upon between ISO 10646 and the Unicode Consortium.  But UNICODE is already used by an application, many developers are familiar with the term Unicode, and the term "international" can be ambiguous (since there are other multi-nation specifications like Latin-1 that aren't the same thing). 


--- David A. Wheeler

(Here's the messages below, with "@" replaced by " at " for anti-spam purposes.)


> ----- Start Forwarded Message -----
> Sent: Mon, 19 Feb 2007 10:10:12 -0800
> From: "Magda Danish (Unicode)" <v-magdad at microsoft.com>
> To: "dwheeler at dwheeler.com" <dwheeler at dwheeler.com>
> Subject: FW: Subj: Use of UNICODE name as a standard spreadsheet function
>  name
> 
>  Dear Mr. Wheeler,
> 
> Thank you for checking with us regarding possible trademark issues. I am happy to inform you that there would be no restriction on naming a function "unicode" or whatever other name you want. Certainly no licensing issues as far as the Unicode Consortium is concerned. I hope this answers your question.
> 
> That said, Unicode technical directors [begin quote] don't see that "UNICODE" is a very good name for a "function...which would return the numeric encoding of a given character." From the way that is stated in the email, it is very unclear what is intended, and the name "UNICODE" doesn't help clarify it. [end quote]
> 
> If you feel there's need for any further discussion with a Unicode technical director, please let me know and I would be happy to put you in touch with one.
> 
> Regards,
> 
> ---------------------------
> Magda Danish
> Sr. Administrative Director
> The Unicode Consortium
> 650-693-3921
> magda at unicode.org
> 
> 
> 
> 
> -----Original Message-----
> Date/Time:    Fri Feb 16 12:46:13 CST 2007
> Contact:      dwheeler at dwheeler.com
> Name:         David A. Wheeler
> Report Type:  Other Question, Problem, or Feedback Opt Subject:  Use of UNICODE name as a standard spreadsheet function name
> 
> Dear Unicode Consortium:
> 
> Hello, I'm chair of the OASIS office formula subcommittee.  We're writing a specification for the exchange of spreadsheet formulas.
> 
> One of the proposed function names is UNICODE, which would return the numeric encoding of a given character.  There will be a few other character-related functions as well.
> 
> Would a function with this name cause problems (trademark, etc.)?
> 
> There must be absolutely NO ROYALTY or other such restriction on ANY implementation.  It must be possible for this specification to be implmented by a program released under the GNU General Public License (GPL), or by Microsoft using its current Office license. We will change the function name, and/or strike all references to Unicode, if that is the alternative.
> 
> We operate under the "RF (Royalty-free) on Limited Terms" Mode of the OASIS IPR Policy, which you can see here:
> http://www.oasis-open.org/who/intellectualproperty.php
> 
> Thanks.
> 
> 
> 
> -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- (End of Report)
> 
> 
> 
> ----- End Forwarded Message -----


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]