From willemw@ace.ace.nl Fri Jan 15 15:22:08 1999 Received: from ace.ace.nl (root@ace.ace.nl [193.78.104.92]) by dkuug.dk (8.8.7/8.8.7) with ESMTP id PAA23632; Fri, 15 Jan 1999 15:22:04 +0100 (CET) (envelope-from willemw@ace.ace.nl) Received: from ace.ace.nl (willemw@localhost.ace.nl [127.0.0.1]) by ace.ace.nl (8.8.7/8.8.7) with ESMTP id PAA28304; Fri, 15 Jan 1999 15:21:59 +0100 To: iso14766@dkuug.dk, sc22wg15@dkuug.dk Cc: nni38122locales@dutiwv.twi.tudelft.nl Subject: Some comments on ISO 14766, WD3 From: Willem Wakker Reply-To: Willem Wakker Date: Fri, 15 Jan 1999 15:21:59 +0100 Message-ID: <28301.916410119@ace.ace.nl> Sender: willemw@ace.nl Here are some comments on ISO 14766 WD3, produced during a few hours spare time. Maybe some more comment will follow (but don't wait for it!). 1- before a `real' review of the document can be made, the document should proofread by an native english standardswriter and be spell checked. As an example: 1st sentence 2nd paragraph of the scope: what are "requirements for making POSIX suitable in the culture"?? "options needed of the POSIX standards"?? This really puts the reader off. 2- the title of the document promises the reader Guidelines. However, there is no Guideline to be found in the document ... In the Scope section it is mention that "This TR provides a guideline .."; is the TR to be seen as one (1) guideline? In that case the title should be changed. 3- In TR 10000 a careful distinction is made between a Profile ("a set of basestandards with options etc ..") and the document in which the profile is specified (the ISP). TR 10000 provide requirements on both. 14766 should make the same distinction: there should be a clear separation between the requirements on the Nationale Profiles (and Nationale Locales) (i.e. the requirements on the standardsselection, options etc) and the requirements on the documents in which these are NPs (NLs) are specified (required chapters, layout etc). 4- Update the references, include TR 10000-3 (all parts of TR 10000 are dated 1998), and bring in the standard format. 5- I would suggest to include in section 4 the Profile definition from TR10000-1 (with a reference to that document). 6- TR 10000-3 defines the concept of "OSE Profiles" and identifies POSIX OSE Profiles (like .13 and .10, now ISP 15287-1 and -2). Strictly speaking: those are POSIX Profiles. Are they intended to fall under the current definition of 4.1? 7- In 4.2, a POSIX National Profile is defined as a subset of a POSIX Profile. What is in the rest of the POSIX Profile of which the POSIX National Profile is a subset? Or are we talking here about sets of profiles (all the POSIX Profiles, including the POSIX OSE Profiles of which the POSIX National Profiles are a subset)? Please clarify what is intended. 8- Is a POSIX National Locale a Profile? Can it exist without a POSIX National Profile? Or is it always a (mandatory, optional) part of a POSIX National Profile? 9- section 4.4 is another horrible example of what is ment under 1. Please first think about what you want to say, then read a textbook on how to say it (or copy it from a standard on conformance) and then write it down. 10- In my view only applications are Internationalized: the access and/or manipulation of national/cultural dependent information is done through standardized interfaces; localization is then the act of selecting one specific locale. I would never call the act of making an application platform capable of supporting this functionality "Internationalization" of the application platform as this is something completely different from Internationalizing applications. 11- Portability (4.8): sight .... There is binary portability, source code portability (even people protability!). What do you mean? Why is it needed here? What is "(KS)" here and in 4.10, 4.11. 12- ISP (4.11): see definition in TR 10000-1. Please note that an ISP is a document, not a profile! 13- A distinction is made (for instance in the 1st sentence of 5.1) between "POSIX [based] [I,i]international Standards" and POSIX. What is the difference. Please make the terminology consistent. 14- The five `dash-items' in 5.1 are no purposes. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa Willem Wakker email: cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc ACE Consulting bv tel: +31 20 6646416 van Eeghenstraat 100 fax: +31 20 6750389 1071 GL Amsterdam, The Netherlands www: http://www.ace.nl eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee