| TITLE: | Disposition of comments report - ISO/IEC CD 15445 ISO-HTML |
| SOURCE: | Roger Price, David Abrahamson |
| PROJECT: | JTC1.18.43 ISO-HTML |
| PROJECT EDITOR: | Roger Price, David Abrahamson |
| STATUS: | Out'put of Editors' Meeting |
| ACTION: | For approval by WG4 |
| DATE: | October 14th 1997 |
| DISTRIBUTION: | WG4 and Liaisons |
| REFER TO: | SC18 N 5794 dated 1997.07.23, WG4 N1931, WG4 N1933, WG4 N1935 |
| REPLY TO: | Dr. James David Mason
(ISO/IEC JTC1/SC18/WG8 Convenor) Lockheed Martin Energy Systems Information Management Services 1060 Commerce Park, M.S. 6480 Oak Ridge, TN 37831-6480 U.S.A. Telephone: +1 423 574-6973 Facsimile: +1 423 574-0004 Network: masonjd@ornl.gov http://www.ornl.gov/sgml/wg8/wg8home.htm ftp://ftp.ornl.gov/pub/sgml/wg8/ |
This document provides the disposition of comments following the successful balloting of ISO/IEC CD 15445 ISO-HTML. The editors would like to thank the national bodies who have commented for their work which has contributed considerably to improving the text.
Your editors are now working as authorised in close collaboration with the World Wide Web Consortium (W3C) HTML working group. A summary of the ongoing harmonization work is now available. As a result of this harmonization work we have grouped the comments into the following classes:
The comments are reviewed in the following text in the order in which they appear in SC18 N 5794 dated 1997.07.23. Some comments raise several points and are treated as if they were multiple comments.
A number of comments occur repeatedly or are of such importance that they merit particular discussion:
These will be discussed here in the Introduction, and reference made to these remarks in the disposition.
Several NB's have asked for the document element to be <HTML> rather than <ISO-HTML>. This distinction was introduced on a temporary basis to avoid confusion during discussions involving IETF, W3C and ISO/IEC texts. Throughout this document we will continue to call ISO/IEC 15445 "ISO-HTML" and use the element name <ISO-HTML> to avoid any possible confusion with other HTML specifications, but we will return to the use of <HTML> element name when preparing the final specification for publication.
After the balloting period began, the World Wide Web Consortium proposed HTML 4.0 for review. Since this proposition has been used as a basis for the harmonization discussions, and since it's introduction was specifically allowed for in the Requirements we shall refer to HTML 4.0 instead of HTML 3.2 throughout this document.
As mentioned above, your editors are actively working in close cooperation with the W3C to harmonize the HTML 4.0 and ISO-HTML specifications. Please see the summary of the ongoing work. As a result, some minor changes have been made to the text. For example, some style attributes such as ALIGN and VALIGN have been removed following expert advice received from the W3C. This means that comments on the specification of those attributes are no longer applicable.
The specification ISO/IEC 15445 is aimed at readers who have a working knowledge of HTML but who are not SGML experts. The editors recognise that such readers find ISO 8879 very difficult to read and are discouraged. This has led in the past to the creation of a bad image in influential Internet newsgroups. In order to make the specification more useful to the intended readers, the editors have included explanatory material and specific references to the SGML Handbook which would not be needed by SGML experts.
The CD text required the correct nesting of <H1>-<H6>. It is
recognised that some HTML documents are generated from other structured sources
such as databases or richer SGML DTDs and further processing is not intended.
To support authors wishing to simplify the production of ISO-HTML in such a
case, the corrected text introduces the NoStructure facility
which eliminates the need to respect the correct structuring of sections.
SC18/WG8 have expressed the wish that ISO-HTML should not add to the structures of "HTML" as generally understood from the IETF and W3C specifications. See recommendation 8 of WG8 1927 "Recommendations of the Barcelona Meeting". The corrected text clarifies this, and ensures that all conforming ISO-HTML documents are HTML documents as defined by W3C HTML 4.0.
The CD text gave the impression that conforming systems were required to support the internal subset of the document type declaration [Goldfarb p.403]. The corrected text makes it clear that this is not so by:
Disposition: Accepted
Architectural forms. Disposition: Referred to the W3C HTML group
Remove the constraint that HTML documents must be consistent with the hierarchical organization of the headings. Disposition: Accepted. See Introduction.
Disposition: Moot. However the editors feel that the translation should be done by a native French speaker.
ICADD must form part of ISO-HTML, even if in an informative annex. Disabled needs are just as important as internationalization. Disposition: Accepted in harmonization with W3C.
The doctype element must be <HTML>, not <ISO-HTML>.... Disposition: Accepted. The document element was temporarily changed to ISO-HTML to prevent confusion during the many discussions which involved the IETF, W3C and ISO/IEC specifications. See above.
... The DTD should be gleaned from the PUBLIC identifier. Disposition: Accepted. Text clarified. See Introduction.
The <B1>..<B6> element types should be named <div1>..<div6>. \ldblquote b\rdblquote connotes a ranked version of bold. Disposition: Accepted. The <B1>...<B6> have been removed from the text.
The DTD must prove itself by being able to markup the ISO-HTML standard. This will test whether the goals have been met. Disposition: Accepted. Already done on an informal basis.
The public entity set should be brought into the DTD. These at least should include ISOlat1, ISOgrk1, ISOcyrl, and the other simple non-symbol public entity sets. Fonts for these are almost universally deployed on computers and browsers. Disposition: Harmonize with the W3C HTML group.
The width and height attributes should be reintroduced, e.g. for the <img> tag. Disposition: Rejected. Removed following expert advice from W3C. See Introduction.
There is concern that the current content model for the element type BODY may
not allow footnotes or appended matter inline. If this is so, it should be
explicitly noted in the text of the standard.
Disposition: Moot. The introduction of <DEL> and <INS> as a
result of the harmonisation work shows that it is possible to introduce
inclusions in body of the document, but that it would require an ammendment to
the specification rather than use of
%body.extend;. Since ISO-HTML is a "subset"
of HTML 4.0, such a facility would first have to be
introduced into the W3C specification.
Page 9, Section 1, Paragraph 29, Change to read: \lquote The fully structured DTD allows for the exact specification of a longer document's logical structure.\rquote Disposition: Not applicable. (Clause 1 is not a part of the normative text submitted for review.)
Disposition: Rejected. SC18/WG8 is actively working in close cooperation with the W3C to harmonize the HTML 4.0 and ISO-HTML specifications. Please see the summary of the ongoing work.
The title of this standard should be HTML rather than ISO-HTML. Disposition: Accepted. See Introduction. The document element was temporarily changed to ISO-HTML to prevent confusion during the many discussions which involved the IETF, W3C and ISO/IEC specifications.
This standard should include no extension to the W3C's HTML 3.2. Disposition: Accepted. Every ISO-HTML document will automatically be W3C HTML 4.0 conformant. See Introduction.
There should be no use of SHORTREF and USEMAP to avoid complicated implementation. Disposition: Accepted. We have redrafted the clause on conformance to clearly separate "Conforming" and "Validating" systems. A typical browser is Conforming but not Validating. Conforming systems will not be expected to parse the DTD and thus will not see the SHORTREF and USEMAP. See Introduction.
There is no need for supporting the <NOP> element. Disposition: Accepted. The <NOP> element is needed only to solve a syntax problem in the DTD. Since Conforming systems do not need to parse the DTD, they never see the <NOP> element
The <APPLET> element, which appears in the W3C's HTML 3.2 recommendation, should still be supported. Disposition: Harmonize with W3C. The <APPLET> element was ommitted following expert advice from the W3C. At the recent harmonization meeting, this decision to omit <APPLET> was confirmed by the W3C representative. See Introduction.
First paragraph should be removed. Disposition Accepted.
In second paragraph change "works" to "has worked", Disposition Accepted.
...remove "edited by David Raggett"... Disposition Accepted.
...and change URL to absolute URL of 3.2 specification (http://www.w3.org/pub/WWW/TR/REC-html32) rather than the URL currently shown, which always points to the latest development version of the HTML spec within W3C. Disposition Accepted. Normative reference when available.
In last list item, replace "Internet Draft" by "RFC 2070" Disposition Accepted.
After last item add reference to the 4/1/97 version of 3.2 of HTML specification (http://www.w3.org/pub/WWW/TR/REC-html32). Disposition Accepted. Normative reference when available. See Introduction.
Re item 11, changes in a TC to 8879 currently being prepared should make it unnecessary to introduce the NOP element. Disposition Moot
Re item 14, forcing bodies to begin with compulsory section content before the first heading is unacceptable. ... Disposition Accepted. Authors will be able to choose whether or not they wish an exact hierachy of headings. See Introduction.
Re item 24, the dropping of the type attribute needs more justification than is
given here, ... It becomes essential to explain how the class attribute can be
used to differentiate between types of lists (e.g. <ol class=alpha>
and
<ul class=menu>). Disposition Harmonized. (Note:
Item was 23.)
However, the use of NAMES rather than CDATA for classes is going to make it difficult to distinguish between capital and lowercase versions of lists numbered alphabetically or using roman numerals.... Disposition Accepted. CLASS is now CDATA.
In the first paragraph, add "(SGML)" after "Standard Generalized Markup Language". (It is part of the formal title of ISO 8879, and reinforces the preceding use of the undefined acronym.) Disposition Accepted.
The use of <ISO-HTML> in place of
<HTML> is an unnecessary imposition as it makes it
impossible for any existing document to conform to the standardized DTD.
Disposition See above.
The document type declaration is unnecessarily complex. ... Disposition Accepted. Relevant clauses have been reworked to make it clear that for normal use, no internal subset is used. See Introduction.
This declaration .... Its compulsory inclusion at the start of the document instance is all that is required to 'standardize' the use of HTML. Disposition Accepted. Relevant clauses have been reworked.
Change "Committee Draft" to "standard". Disposition Rejected. The phrase "Committee Draft" is the value of a general entity which reflects the current status of the text. The editors do not wish any text to appear which calls itself a "standard" when it is not.
In Notes 2, 3 and 4 remove reference to Gol90 within square brackets. (ISO standards should only reference non-ISO documents when they are declared as part of the normative references. There is no reason why The SGML Handbook should be included as a normative reference.) Disposition Rejected. This specification is aimed at people who know HTML, but who may not know SGML. For such readers, ISO 8879 is unreadable, and Gol90 which includes the normative text is much to be preferred. See Introduction.
Item 2 is misinterperable. ...
The wording of the ISO-HTML system
identifier is not valid. ...
What happens if a browser can display both
ISO and other variants of HTML. ... Disposition Accepted. The text on
documentation of Conforming and Validating systems has been revised.
Note 7 should added (in brackets) to Note 6, HTML. ... Disposition Not applicable, note 7 removed.
In the paragraph preceding Note 8, replace "SGML International Standard [Gol90, p487]" by "ISO 8879:1986". HTML. ... Disposition Accepted in part. Reference to Gol90 will be kept since ISO 8879 is unreadable by non-SGML specialists. See Introduction.
Delete Notes 8 and 9. Disposition Not applicable. Notes marked "Editors to NB experts:" are not a part of the specification.
In 1. delete "by the control functions and". (Only those ISO 10646 control characters whose function has been declared in the SGML declaration are valid in ISO-HTML documents.) Disposition Accepted.
In 2. how can systems determine the extent of "that part of the coded character set they have in common"? Disposition Moot, but its not an objective of this text to explain this.
Add references to ISO 639, ISO 3166, ISO 8879, RFC 1630, RFC 1738, RFC 1808, RFC 1866, RFC 2070 and the 14th January 1997 HTML 3.2 specification. Disposition Accepted, with HTML 4.0 (work in progress) in place of HTML 3.2.
In Definition 4 change "Table 1 on page 11" to "Figure 1". Disposition Accepted.
Why does HyTime need to be defined? Disposition Accepted. Removed.
In definition of W3C, place final URL in brackets. Disposition Accepted. Angle brackets added.
In first paragraph delete "as documented in .... 1996" and text after the list. Disposition Accepted.
Delete "[YNAD95]". Disposition Not applicable. This reference was generated automatically in the LaTeX rendering of the SGML source code. The source is marked up <bibref refid=rfc2070 page=yes> which is rendered by the ISO system using the approved style.
Shouldn't alignment also be definable on B1-B6 (or DIV1 to DIV6)? Disposition Rejected. <B1>-<B6> are internal to the DTD and are not avalable to the author. Their function is only to "remember" the current depth of nesting, so they cannot carry any user attributes.
As these attribute values are normally defined using the
%justify and %valign parameter entities these two
entities should be defined/explained as part of this sub-clause.
Disposition Not applicable.
In example 2 it seems illogical to place the alignment attribute on the list item element rather than on the ordered list element. (Why should this item be treated differently from any other in the same list?) Disposition Not applicable. Following expert advice from the W3C, the alignment attribute has been removed.
Change "writing directional" to "writing direction". Disposition Accepted.
Add description of %dirn parameter entity here, and then delete
the %dirn parameter entity definition from subsequent definitions
(nearly every one!)
Disposition Accepted. Relevant clauses will be re-worked.
Add statement to the effect that all ISO-HTML IDs must be unique within a document. ... Disposition Accepted. New subclause will be drafted to clarify this point.
Replace "This Committee Draft does not specify three-letter primary tags, however their description may be found in the "Ethnologue" [Gri92]" by a reference to Part 2 of ISO 3166. Disposition Refer to W3C. We need a harmonized treatment of language tags.
Is the reference to RFC 1766 in the next paragraph correct, or should it be a reference to RFC 1866? Disposition Yes - RFC 1766 is "Tags for the Identification of Languages"
If it is correct add RFC 1766 to the list of normative references. Disposition Accepted.
If point 1 is followed surely point 2 serves no purpose. Are both required? Disposition Both are needed - see RFC 2070.
Note 19 is not true - ISO 9573 has not been updated to define "names for all the characters specified by ISO 10646". The note should be deleted. Disposition Accepted. Editors were misinformed by NB expert. Note removed.
In first text paragraph, delete "of gossamer". Disposition Accepted.
The next sentence is incorrect. HTML links do not have two anchors which are "both identified by their address". Only the target anchor is identified by an address. The address of the source anchor is implied from its position in the content (unless it has been assigned a value for the name attribute). If these two terms must be retained then copy the definition of them from 15.2 to this clause. Disposition Harmonized. See RFC 1866. The NB expert may be thinking of the common interpretation rather than the IETF description we use.
Delete "[BL94]" Disposition Not applicable. This reference was generated automatically in the LaTeX rendering of the SGML source code.
In the paragraph following the list change "source anchors" to "target anchors". Disposition Accepted. (Good reading!)
In the last paragraph remove both references to "[Fie95]". Delete "[BL94]" Disposition Not applicable. These references were generated automatically in the LaTeX rendering of the SGML source code.
Make last sentence a separate paragraph. Disposition Accepted.
In example change backslash for normal (forward) slash. Disposition Accepted. Backslash was introduced in error by the SGML->LaTeX conversion.
Delete reference from end of Note 24. Disposition Accepted. All notes marked "Editors to NB experts:" will be deleted.
Check whether it would be better to replace the <NOP>
element with a reference to the new facility to be provided to allow empty
strings in replacement parameter entities in models that will be defined as
part of the SGML on the Web Technical Corrigendum to SGML later this year.
Disposition Moot.
In 2 delete reference to Gol90. Disposition Rejected. Not only is the ISO 8879 text unreadable, but the audience for ISO/IEC 15445 is also expected to be unskilled in SGML. The editors will give as much help as possible in providing useful references to readable texts. See Introduction.
Delete 4 and 5. (The parameter entities are automatically invoked as part of the definitions of the appropriate elements in the internal subset.) Disposition Accepted. The text should say that step 4 comes before step 5.
Delete Note 20. (REL and REV must be CDATA so that they can contain subset.) Disposition Accepted.
Change "Please see" to "See". (Apply same rule to all subsequent clauses.) Disposition Accept.
For REL add "Defines relationship of source anchor to target anchor." Disposition Accepted. Description of REL has been considerably improved as part of the HTML 4.0 harmonization. See Introduction.
For REV add "Defines relationship of target anchor to source anchor." Disposition Not applicable. REV removed following expert advice from W3C. (For further review and harmonization.)
Add example of REL and REV, e.g.: Disposition See above.
Providing %justify ...
Disposition Not applicable. %justify; removed following
expert advice from W3C.
... and %dirn have been defined previously as suggested above,
they can be removed from 16.1,... Disposition Accepted.
In the first paragraph of 16.2, how can an address contain "such information as ... signature and authorship"? Typically the only thing that goes in with the address are phone and fax numbers, e-mail addresses and web-site addresses. It would seem more suitable to cite these rather than signatures. Disposition Accepted.
It might be an idea if the last sentence of the description of the ALT attribute in 17.3 was placed in an emphasized note to ensure that this strong recommendation is clearly seen. (It cannot be stressed highly enough!) Disposition Accepted. Change "strongly" to "very strongly", and markup as <hp4>, the maximum highlighting allowed by the ISO DTD.
In the example in 17.4 remove the backslashes that preceded the % symbols. Disposition Accepted. Backslashes due to error in SGML->LaTeX conversion.
Need to add statement to effect that this element may only be used within the
<HEAD> element.
Disposition Accepted.
In the first paragraph of 19.2 replace "[BLMM94]" with a cross reference to RFC 1738. Disposition Accepted. SGML is "<bibref refid=url>"
Delete Note 32. Disposition Accepted. All notes marked "Editors to NB experts:" will be deleted.
Either in 21.2 or elsewhere in the document, add a description of the 7 level
structure that applies to the use of fonts identified using nested <BIG>
and <SMALL> tags.
Disposition Not applicable. Following expert advice from W3C, <BIG>
has been removed.
Note 33 is invalid because the model for a caption (%text), does not allow paragraphs to be embedded within captions. Disposition Accepted. (Good reading!)
No definition is provided for the %Content-type and %HTTP-Method parameter entities. Disposition Accepted.
Delete "[BLMM94]" from item 1 in 40.3. Disposition Not applicable. These references were generated automatically in the LaTeX rendering of the SGML source code.
Re note 35, the reference should be to ANSI X3.4-1968, but it might make more sense to reference ISO 646 as part of the normative references. Disposition Accepted. We reference ISO 646.
Re note 36, it seems unlikely that you will be able to change the common practice of using & rather than ; as this form of the text is easier to read, as well as being the one most commonly used by HTML browser developers. Disposition Moot.
For the METHOD attribute, define what is meant by "server side effects" Disposition Accepted.
It should be mentioned that many people reserve H1 for the title of the page. Disposition Accepted.
It should be possible to start using headings at a level lower than that defined for H1. Disposition Accepted. See Introduction.
Use of server-side mapping using ISMAP should be discouraged while use of client-side maps using USEMAP should be encouraged within this standard. Disposition Accepted. Added note.
It should be made clear that ISMAP and USEMAP should not be used for the same image. Disposition Accepted. Added sentence.
The example in 50.4 should be based on the use of a client-side map. Disposition Accept. Added basic example and reference to client-side map example.
Is the implied value of MAXLENGTH controlled by SIZE? Disposition Answer: No.
In the defintion of SIZE change "Prefered" to "Preferred". Disposition Accept. Found and changed 3 occurences.
In 51.3.2, the example should illustrate use of the SIZE attribute. (Would be particularly applicable to file name field.) Disposition Moot.
In 51.3.5 add warning that password is not encoded when transmitted. Disposition Accept. Added warning.
In 51.3.7 the wording of the Value attribute for the last element should be
different from the default wording if no value is entered. (Need to mention the
default value somewhere!) Change example to <INPUT TYPE=RESET VALUE="Press
here to reset">.
Disposition Refer to W3C. Any change must be harmonized.
In 51.3.8 mention the default name given to the button if no value is specified. Disposition Refer to W3C. Any change must be harmonized.
Disposition Not applicable. Following expert advice from the W3C, the <ISINDEX> element has been removed. See Introduction.
As mentioned above, there is no obvious reason for replacing the well-known
<HTML> tag by <ISO-HTML>.
Disposition See above.
The description of the role of the CHARSET attribute should be copied from that of the A element rather than being referenced. Disposition Not applicable. Following expert advice from the W3C, the CHARSET attribute has been removed.
In 57.2 the significance of the word "Thus" in the second paragraph is not clear. It should be stated more clearly that the only time that elements with no associated action need to be listed before those that have actions associated with them is if the two areas overlap and the area for which action has been inhibited is to take priority. Disposition Accepted.
In the definition of the NAME attribute change "50 on page 37" to read "(see clause 50 on page 37)" Disposition Accepted. Problem due to SGML->LaTeX conversion.
In the example, change "Flinstones" to "Flintstones" and place two keyword definitions together (after the definition of the Reply-to field). Disposition Accepted.
In the paragraph after the list, the sentence beginning "If a browser chooses to render the %section.content" is invalid. Either the browser supports the relevant behaviour or it displays the section content. As the text is currently worded, if the browser displays the section content then, and only then, it can decide if it also needs to display the contents of the ALT attribute. (Note that there is no ALT attribute defined in the current definition.) Disposition Accept.
Note: This does not, however, conform with the usual way the ALT attribute has been used with objects. Normally the ALT field contains text that is displayed in the status line of the browser while the cursor is placed over the object. Disposition Not applicable.
The definition of the DECLARE attribute needs clarification. Can you declare the same object more than once? Can you use DECLARE with any other attribute? If not what rules apply to its use? Disposition To be harmonized with W3C.
The example is unclear. What happens if the style calls for alphabetic or roman
numbering? What does it mean to diplay 1998) 12th April? Disposition
Accepted. Example replaced.
The term "Internet Media Type" needs to be formally defined somewhere, or a reference needs to be made to a document in which it is formally defined. Disposition Accepted. Reference RFC 2068.
The example should be changed as it illustrates bad practice. If an element has a language attribute the text inside that element must be in the language specified. Disposition Rejected. The aim of the example is to show clearly that the punctuation is controlled by the language attribute of the quotation element.
In the first para of 68.2, change "a sequence" to "an embedded sequence". Disposition Accepted.
The third paragraph of 68.2 is very dubious. What is the technical difference between a drop-down menu and a list menu? What proportion of browsers make this distinction? Disposition Accept.
In 68.3, and in subsequent clauses, remove references to Gol90 ... Disposition Rejected. Not only is the ISO 8879 text unreadable, but the audience for ISO/IEC 15445 is also expected to be unskilled in SGML. The editors will give as much help as possible in providing useful references to readable texts. See Introduction.
... and place phrase "see ISO 8879:1986 subclause ...." in parentheses. Disposition Reject. See Introduction.
The example in 70.4 is bad. Change it to one that uses Hebrew characters within the SPAN element. (This will serve to reinforce the point that the ISO version of HTML is based on ISO 10646 rather than ISO 8859-1.) Disposition Moot. Example reworked for clarification.
Why is the ALIGN element defined using a full list of permissible terms rather than %justify?
The use of DUMMY as the name for the BORDER attribute is silly. If you cannot retain the BORDER name used by existing browsers at least change it to something useful like SHOW-WITH.
Can a default value be specified for CELLPADDING (e.g. 1 pixel)?
Can percentage values be greater than 100%? Disposition Four comments not applicable. Following expert advice from the W3C, the ALIGN, DUMMY, CELLPADDING and WIDTH attributes have been removed. See Introduction.
Why does the definition of VALIGN have a default value of #IMPLIED when the text says that the default value is middle? Disposition Not applicable. Following expert advice from the W3C, the VALIGN attribute has been removed. See Introduction.
Why does the definition of ALIGN have a default value of #IMPLIED when the text says it is center?
Why does the definition of VALIGN have a default value of #IMPLIED when the text says that the default value is middle? Disposition Two comments not applicable. Following expert advice from the W3C, the ALIGN and VALIGN attributes have been removed. See Introduction.
The second sentence in the first paragraph of 79.2 should not be normative. If the text is to be retained it should be placed as a note after the penultimate paragraph (which explains what a user agent does with the title). Disposition Accepted. Text demoted to note.
Why does the definition of VALIGN have a default value of #IMPLIED when the text says that the default value is middle?
What is the relationship between the default value of align on TR and those specified on TD and TH elements within the row? Disposition Two comments not applicable. Following expert advice from the W3C, the ALIGN and VALIGN attributes have been removed. See Introduction.
The last sentence in Note 68 should be deleted. (Not relevant for international standard.) Disposition Rejected. The editors will give as much help as possible in providing useful references to readable texts. See Introduction.
Change all references to 15xxx to 15445. Disposition Accepted.
Comment on being based on v1.17 of html.decl should be removed. (Not appropriate to international standard.) Disposition Accepted.
Selection of ISO 10646 character set is not dependent on whether a particular product supports it. Replace existing character set definition with one in preceding comment that references ISO 10646, and remove comments re nsgmls. Disposition Accepted for application to final IS text.
Remove comment after ATTCNT (not appropriate for international standard). Disposition Accepted for application to final IS text.
iso-html.dtd should start with references to iso-html.ent and ISOlat1.ent to allow single line reference to the DTD without requiring definition of an internal subset. (See comments on Clause 2.) Disposition Accepted. Internal subset will not be required. See Introduction.
Remove references to RFCs covered by normative references. Disposition Rejected. The editors will give as much help as possible in providing complete references as used in high quality journals. See Introduction.
Remove reference to HyTime. Disposition Accepted.
david@cs.tcd.ierprice@cs.uml.edu