From Julia_Flanders@Brown.edu Wed Oct 24 11:12:19 2001
Date: Tue, 23 Oct 2001 18:58:03 -0400
From: Julia Flanders <Julia_Flanders@Brown.edu>
To: Lou.Burnard@oucs.ox.ac.uk, Syd Bauman <Syd_Bauman@Brown.edu>
Subject: frontmatter comments and fixes

At 9:18 PM +0100 10/23/01, Lou Burnard wrote:
>Thanks! very much appreciated... I was planning to get Sebastian to redo
>the pdf in separate fascicles and put them on the website, but that
>wouldnt get done as fast. As to the comments, I think if Julia doesnt
>mind, it would be most helpful to pass any substantitive
>changes/redrafting by the whole board rather than just apply them, so I'd
>prefer her to email them to the list or to me. Obviously straightforward
>typos can be fixed without discussion, but I think we may need some
>additional prose as well.

Well, I just had a handy hour or so, so here are the things I've found:

A general thought/question on the composition of the front matter: at 
the moment, each piece of the front matter carries a date (either 
explicitly in the case of the two Introductory Notes, or implicitly 
in the original Preface which is clearly not from the present day), 
but there's nothing from the current editors describing the current 
document. Are you planning to write such a thing? I assume you are, 
but in case you're not let me say that I think it would be very 
useful and would put the other bits into a proper perspective; at the 
moment it's a little odd reading about the various prior layers of 
revision (and particularly odd if one knows what further revisions 
are supposed to be taking place). An even more radical approach would 
be to write a Whole New Introduction which would summarize the 
factual information in the existing material but present it in a 
voice from the present day. This would have the advantage of clarity 
and brevity, but the disadvantage of losing all the lovely phrases of 
thanks (which I think have a great deal of merit). I'm not sure how I 
feel about this approach, but it might be worth considering.

Global:

--all instances of phrases taking the form "X-compliant" or 
"X-conformant" (where X could be "TEI", "XML", "SGML", etc.) should 
be hyphenated. Current usage is inconsistent.

--all list items should be terminated with the same character (at 
least, there should be consistent usage within a given list), 
preferably a semicolon in cases where each list item is a clause, and 
a comma or nothing at all where each list item is just a phrase; the 
last item gets a period or nothing. Current usage is inconsistent.

--should we/could we use en-dashes for number ranges? (fussy fussy)

Introductory Note  (June 2001):

--first list, second bulleted item, "generates a set...": "dtd" 
should be capitalized

--first list, third bulleted item, "can be processed...": "special 
purpose" should be hyphenated

--fourth paragraph, "As noted elsewhere...": the phrase "corrigible 
errors" sounds redundant; I realize this phrase has a history within 
the TEI documents, but here (read for the first time) it sounds odd 
(especially in the context: a number of corrigible errors were 
corrected...).

--fifth paragraph, "A major design goal...": "...to ensure that the 
DTD fragments generated by it..." sounds awkward; maybe better as "to 
ensure that the DTD fragments it generates..."

--in general, the fifth paragraph feels a little confusing in its 
treatment of backwards compatibility. I don't know if my level of 
comprehension is representative; when we say "any document conforming 
to the original TEI P3 SGML DTD would also conform to the new XML 
version of it, though backwards compatibility is not guaranteed", 
where does the non-guarantee come in? that is, if we feel we've 
achieved the goal of not breaking existing documents, how are we not 
backwards compatible? if it's just that we *think* we've achieved it 
but we don't want to *guarantee* it, then it might be  better to 
rephrase: "...new XML version of it. Although backwards compatibility 
is not absolutely guaranteed, we believe..."

--second list, fourth bulleted item, "All the examples...": add a 
period at the end of the parenthetical sentence.

--last paragraph, "Much work remains...": delete comma following "the 
Gentle Introduction to SGML"


Introductory Note  (May 1999):

--second para after the heading "Specific changes in the DTD", "Where 
possible...": add space after italicized "not"

--bulleted list: several problems here:
	--characters after each item are inconsistent, particularly 
the first few items
	--in the item "added <dateLine>" the bullet is insufficiently indented
	--in the prepenultimate item, "change content" should be 
"changed content"
	--what is currently the last item, "Finally..." should 
probably not  be part of the list at all, or if it is, it shouldn't 
start with "Finally". But since it contains a list, better to just 
make this a new paragraph.


Preface:

--the first line is an example of the oddness of the layered front 
matter; "the result of over five years' effort" makes one do a double 
take. At the very least, a date should be added to the Preface 
indicating that it's the original preface, written in 1994 (?).

--bulleted list: again, make the terminal punctuation consistent (I'd 
just remove it since the items are so short)

Acknowledgements:

--Committee on Metalanguage and Syntax: unlike the other committees, 
this one includes the chair as part of the list of members instead of 
keeping it on a separate line; should be consistent; need to insert 
the phrase "Members 1990-1992" and remove the semicolon following 
"(Queen's University)".

--para beginning "In addition, many members...": the designation 
"Myrdal Norway" without punctuation looks odd; add a comma?

--list of work groups: the work group names are inconsistently 
capitalized; I'd go for the low style and just capitalize the first 
word of the name.

--TR12 Literary prose: remove semicolon after Liege

--AI5 Print dictionaries: "Department of Defense" looks weird there 
all by itself without a country...

--para following the heading "Advisory Board": "life time" should be 
one word, no hyphen.

--list of organizations, Association for History and Computing: add 
an "r" to "Max-Planck-Institut fu..."

--list of organizations, Dictionary Society of America: elsewhere in 
this document, "independent consultant" has been capitalized; looks 
more dignified that way but I don't care which we choose.

--Steering Committee Membership: is the steering committee still in 
operation? if so, the date ranges with no end date are OK, but 
otherwise should we provide end dates?

Best, Julia
