Best accessibility practice? - HTML markup to represent dates

[Andy Dingley]

I'm using this at present:
<p title="Publication date" >[28/09/2005 15:00]</p>

Works fine on screen, but Fangs/Jaws just reads it as "left bracket
twenty-eight slash zero slash two thousand five fifteen colon zero right
bracket" Really it needs something more to indicate that it _is_ a
date.

The [...] brackets would be better done with CSS, :before and content: ,
but the boss wants it to work under IE too. The site is firmly UK
based, so d/m/y is necessary and won't be getting internationalised.

So, what's the best markup for representing a date, with good
accessibility to screen readers?

[Jukka K. Korpela]

Andy Dingley wrote:

I'm using this at present:
<p title="Publication date" >[28/09/2005 15:00]</p>
Such a date notation is alienating to more than half of mankind, and it
is guaranteed to cause serious problems at least when the day number is
less than 13, since then a considerable minority (including most people
in the U.S.) will understand in a manner different from the intended
meaning.

Ideally, an author would enter a date in a format he finds suitable,
have it automatically transferred to an internationalized format (such
as ISO 8601) internally, and user agents would display the date to each
user in a format chosen by the user. This format might have to be
different for different content languages, if the user so chooses.

We are very far from such a situation in HTML authoring, partly because
HTML lacks any markup for dates (as well as times, monetary amounts,
decimal numbers, and other data that should be internationalized
internally, localized externally).

The conclusion is that in practice, there are just two sensible ways of
presenting dates in HTML documents:
- use a language-dependent notation where the month is expressed using a
word, or perhaps a conventional alphabetic abbreviation, e.g. "September
28, 2005" or, if conciseness really matters, "Sep 28, 2005"
- use the ISO 8601 date notation, such as "2005-09-28", usually with an
explanation or pointer to an explanation of the notation; this is often
the best approach in tables, chronological lists, and multilingual material.

What about the title? Yes, what about it? It is an optional sweetener,
nothing to be relied upon. If the user does not know and cannot deduce
from other information that [28/09/2005 15:00] is publication date and
time for something (which?), title="Publication date" will not help
much. People who are puzzled by some information won't usually move the
mouse to see if there's a tooltip, especially when the text looks no
different from normal text and makes no suggestion "mouse over me!".

(As an aside, for some decades I though that "date" only means time
specified with a granularity of a day, i.e. identification of day,
month, and year. Only later did I realize that my understanding of
English was imperfect in this respect: "date" may mean that, or it may
mean "date and time", i.e. time specified with a granularity of a
second, or even more exactly. I would still refrain from assuming that
everyone knows the extended meaning of "date".)

Regarding the use of markup, I don't think [28/09/2005 15:00]
constitutes a paragraph even under a liberal interpretation. It's better
to use <div> or <span> markup than semantically wrong markup, like <p>
for a non-paragraph.
Works fine on screen, but Fangs/Jaws just reads it as "left bracket
twenty-eight slash zero slash two thousand five fifteen colon zero right
bracket" Really it needs something more to indicate that it _is_ a
date.
We would need such markup, but we have to do without.
The [...] brackets would be better done with CSS, :before and content: ,
but the boss wants it to work under IE too.
Quite a clever boss you have.

What makes you think you need brackets in the first place? Will all
readers understand them the intended way?
The site is firmly UK
based, so d/m/y is necessary and won't be getting internationalised.
Wrong idea. When you're on the World Wide Web, you cannot know the
language, race, sex, or species of your visitors. You can just assume,
and you might mostly assume right, which means your assumptions will be
false at times.

Surely people in the United Kingdom understand the notation
September 28, 2005, and understand it unambiguously?
Whether they understand, say, 11/09/2005 the way you want depends on
their background; if someone just moved from the U.S., well,...
In fact, I'm afraid the United Kingdom has been Americanized in many
ways so much that users can be uncertain about the interpretetation of
notations - how could they know that your page doesn't just use a date
notation that software written in the U.S. spits out?

(Besides, [28/09/2005 15:00] is far from traditional British style, too,
with the leading zero and with the 24-hour clock. That is, it isn't even
stylish. Anyway, make clarity and unambiguity the first priority, at
least for dates.)
So, what's the best markup for representing a date, with good
accessibility to screen readers?

<div>Published in September 28, 2005.</div>
(It's better to use September than Sep, since a screen reader will
hardly know or guess that here "Sep" should be read as an abbreviation.
Using <abbr title="September">Sep</abbr> would mostly be pointless,
though perhaps theoretically correct by some book. It would distract
many people using a graphic browser that shows the abbreviation with a
dotted underline.)

[Mark Parnell]

In our last episode, Andy Dingley <di*****@codesmiths.com> pronounced to
comp.infosystems.www.authoring.html,alt.comp.accessibility:

<p title="Publication date" >[28/09/2005 15:00]</p> <snip> The [...] brackets would be better done with CSS, :before and content: ,
They would be better dropped altogether IMHO.
but the boss wants it to work under IE too.
Understandably.
So, what's the best markup for representing a date, with good
accessibility to screen readers?

Write it out in full, something along these lines:

28th September, 2005 at 3pm.

--
Mark Parnell
http://clarkecomputers.com.au

[Thomas Jespersen]

On Thu, 29 Sep 2005 00:18:57 +0100, Andy Dingley wrote:

I'm using this at present:
<p title="Publication date" >[28/09/2005 15:00]</p>

That is not only a problem for screenreaders but also for other browsers
because some countries use MM/DD/YYYY and others use DD/MM/YYYY format
which can make it pretty hard to guess what date you are talking about.

So my suggestion is to write it out fully:

28 September 2005 at 3 PM

--
Now Playing:
Rod Modell - Grand Bend

[Jake]

In message <st********************************@4ax.com>, Andy Dingley
<di*****@codesmiths.com> writes

I'm using this at present:
<p title="Publication date" >[28/09/2005 15:00]</p>

Works fine on screen, but Fangs/Jaws just reads it as "left bracket
twenty-eight slash zero slash two thousand five fifteen colon zero right
bracket" Really it needs something more to indicate that it _is_ a
date.

The [...] brackets would be better done with CSS, :before and content: ,
but the boss wants it to work under IE too. The site is firmly UK
based, so d/m/y is necessary and won't be getting internationalised.

So, what's the best markup for representing a date, with good
accessibility to screen readers?

Or in HPR (3.04)

"twenty-eight divided by nine divided by two-thousand-and-five fifteen
colon zero zero"

No commonality amongst readers, then ;-)

So. The only 'safe' way to write it would be a variation on:

<p title="Publication date" >28th September, 2005 at 3pm.</p>

-------------------------------------------------------------------------
----------------

In an ideal world we'd probably write it as:
<p title="Publication date" ><date form="dmyt">[28/09/2005
15:00]</date></p> or something similar.

UAs would then speak it/display it according to local usage.

Regards.

--
Jake (ja**@gododdin.demon.co.uk -- just a 'spam trap' mail address)

[Andy Dingley]

Andy Dingley wrote:

so d/m/y is necessary and won't be getting internationalised.

Thanks for your comments, even though (this being Usenet) they were
less than useful 8-(

The site _will_ present visible content in the form "[21/09/2005
15:00]"
This is not my choice, it's what the dezyner called for. Yes, there's
an internationalisation issue. Yes the brackets are eye-candy. Yes, the
leading zero is somewhat odd for UK tradition. However none of these
issues are even up for debate - this is the design, this is what it
_must_ look like. RFC2119 applies.

[Reinder Verlinde]

In article <11**********************@z14g2000cwz.googlegroups .com>,
"Andy Dingley" <di*****@codesmiths.com> wrote:

Thanks for your comments, even though (this being Usenet) they were
less than useful 8-(
You should be more grateful. You did get some feedback on how other
screenreaders read this. You also did get good feedback on why
'[21/09/2005 15:00]' is a bad idea.
The site _will_ present visible content in the form "[21/09/2005
15:00]"
This is not my choice, it's what the dezyner called for.

Reread your question (article
<st********************************@4ax.com>), and answer us how we
should know from that post that changing the looks of the date is
unacceptable.

With this extra bit of info, I think the problem is unsolvable with
current technology.

We could try to work around the problem, though. Would it be acceptable
to have a user preference (stylesheet change?) that changes the visible
form of the date?

Would it be acceptable to present the date twice (for example once in
the required form, and once as white-on-white, really small, or
whatever, so that only screen-readers would see it?

Neither solution would be good, but they could be better than doing
nothing.

Reinder

[Lachlan Hunt]

Andy Dingley wrote:

I'm using this at present:
<p title="Publication date" >[28/09/2005 15:00]</p>
...
So, what's the best markup for representing a date, with good
accessibility to screen readers?

Take a look at the hCalendar microformat, it may be suitable for your needs.
http://microformats.org/wiki/hcalendar

--
Lachlan Hunt
http://lachy.id.au/
http://GetFirefox.com/ Rediscover the Web
http://GetThunderbird.com/ Reclaim your Inbox

[Alan J. Flavell]

On Thu, 29 Sep 2005, Andy Dingley wrote:

Andy Dingley wrote:

so d/m/y is necessary and won't be getting internationalised.
Thanks for your comments, even though (this being Usenet) they were
less than useful 8-(

On the contrary, they will be useful for all those readers who care to
learn from them.
The site _will_ present visible content in the form "[21/09/2005
15:00]"
This is not my choice, it's what the dezyner called for.
It doesn't change the facts of the WWW, though.
Yes, there's an internationalisation issue. Yes the brackets are
eye-candy. Yes, the leading zero is somewhat odd for UK tradition.
However none of these issues are even up for debate

Welcome to usenet. I'm surprised at your reaction to this - it's not
as if you haven't been around for quite a while already.

It was an interesting question, and usenet discussed it, with some
informative results. I *can* say that from a fairly impartial
position, since I didn't really get involved on this particular
thread.

The fact that you now say you can't really apply anything that you may
have learned from it, is your problem, not ours; but usenet never was
a mere help desk, as you well know.

have fun

[Andy Dingley]

Lachlan Hunt wrote:

Take a look at the hCalendar microformat,

Thanks! Not sure yet if I can use it (although I'm hopeful) but that's
certainly an interesting thing to bear in mind.

[Jukka K. Korpela]

Andy Dingley wrote:

The site _will_ present visible content in the form "[21/09/2005
15:00]"
This is not my choice, it's what the dezyner called for.
You can and should explain the implications to anyone who is responsible
for the decision, instead of trying to alleviate the consequences
(especially since any achievable alleviation would be in a grotesque
disproportion to the damage caused by the decision).

Don't try to hide their mistakes. It won't help anyone.

The best you could achieve is probably title=
"The publication date is the twenty-first of September in
the year two thousand five at three o'clock PM"
and it would help only on browsers that actually make that text
available to the user, and mainly in aural presentation.

You don't want to do that. Especially since it would also disturb many
people, in addition to helping some.
However none of these
issues are even up for debate - this is the design, this is what it
_must_ look like. RFC2119 applies.

Don't be ridiculous. RFC 2119 applies to RFCs, and you know that.

[Andy Dingley]

Jukka K. Korpela wrote:

You can and should explain the implications to anyone who is responsible
for the decision, instead of trying to alleviate the consequences

They're a paper-based magazine publisher. I'm still fighting the
"enhancement" to switch to pixel-sized fonts and browser-snigging to
decide on what size to use.

My "number one ultra important nothing else must be done until it's
fixed" bug at present is the fact that text sizes are slightly
different on IE and Firefox. The fact that they're also going to vary
between screens, versions, user settings etc. is still a long way from
registering. And don't mention the "right click here to save video
file" alt texts on the thumbnails. So although I might _like_ to
question the date format, it's not a practical proposition for any
earth-bound non avian-pig dependent meaning of "practical".

[Jukka K. Korpela]

"Andy Dingley" <di*****@codesmiths.com> wrote:

My "number one ultra important nothing else must be done until it's
fixed" bug at present is the fact that text sizes are slightly
different on IE and Firefox.
Have you already told them how to achieve global totalitarian dictatorship?
(It's the only way to come even close to guaranteeing that everyone sees
their text in exactly the same size. But you know this, don't you.)
And don't mention the "right click here to save video
file" alt texts on the thumbnails.
I won't, since you did.
So although I might _like_ to
question the date format, it's not a practical proposition for any
earth-bound non avian-pig dependent meaning of "practical".

Au contraire. It seems that the only practical proposition is to tell them
that what they ask for is impossible to achieve and would be completely
wrong if it were possible. The wrongly fixed data format is apparently just
a small factor in the catastrophe, but it's still a part of it. I'm sorry
if you cannot afford to tell them all that (no need to say that they'd hire
someone else - that much is obvious).

But don't expect us to help you in wasting time in doing some pointless
cosmetic or theoretizing changes that would have at most a minuscule impact
on the (in)accessibility, and an impact in an unknown direction.

--
Yucca, http://www.cs.tut.fi/~jkorpela/
Pages about Web authoring: http://www.cs.tut.fi/~jkorpela/www.html

[Matt Silberstein]

On Thu, 29 Sep 2005 08:02:44 +0300, in
comp.infosystems.www.authoring.html , "Jukka K. Korpela"
<jk******@cs.tut.fi> in <dh**********@phys-news1.kolumbus.fi> wrote:

[snip]

The conclusion is that in practice, there are just two sensible ways of
presenting dates in HTML documents:
- use a language-dependent notation where the month is expressed using a
word, or perhaps a conventional alphabetic abbreviation, e.g. "September
28, 2005" or, if conciseness really matters, "Sep 28, 2005"
Strunk and While argue for 28 Sept 2005. Such a date does not need any
punctuation since the numbers are separated by letters. I write the
S&K version.
- use the ISO 8601 date notation, such as "2005-09-28", usually with an
explanation or pointer to an explanation of the notation; this is often
the best approach in tables, chronological lists, and multilingual material.

I like 2005-09-28 because it self sorts. But I am lazy.

[snip]

--
Matt Silberstein

Do something today about the Darfur Genocide

Genocide is news | Be A Witness
http://www.beawitness.org

"Darfur: A Genocide We can Stop"
www.darfurgenocide.org

Save Darfur.org :: Violence and Suffering in Sudan's Darfur Region
http://www.savedarfur.org/

[Dr John Stockton]

JRS: In article <dh**********@phys-news1.kolumbus.fi>, dated Thu, 29
Sep 2005 08:02:44, seen in news:comp.infosystems.www.authoring.html,
Jukka K. Korpela <jk******@cs.tut.fi> posted :

Andy Dingley wrote:

I'm using this at present:
<p title="Publication date" >[28/09/2005 15:00]</p>
Such a date notation is alienating to more than half of mankind,

If you mean the half that includes the Chinese, I doubt whether they are
that intolerant. AIUI, unlike others, they no longer consider all
outsiders to be *unspeakable* barbarians; I think they'd just sigh at
the backwardness of the West.
Surely people in the United Kingdom understand the notation
September 28, 2005, and understand it unambiguously?
Yes, but there's no need to use a notation tainted by Americanism on a
British site, where 28th September 2005 / 28 Sept 2005 would be natural
and comprehensible even by visitors from elsewhere.

<div>Published in September 28, 2005.</div>

<div>Published on 28th September 2005.</div> // "on" is not needed

Putting I & O adjacent on the keyboard was NOT a good idea.

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 IE 4 ©
<URL:http://www.jibbering.com/faq/> JL/RC: FAQ of news:comp.lang.javascript
<URL:http://www.merlyn.demon.co.uk/js-index.htm> jscr maths, dates, sources.
<URL:http://www.merlyn.demon.co.uk/> TP/BP/Delphi/jscr/&c, FAQ items, links.

[Dr John Stockton]

JRS: In article <dh**********@phys-news1.kolumbus.fi>, dated Thu, 29
Sep 2005 15:39:19, seen in news:comp.infosystems.www.authoring.html,
Jukka K. Korpela <jk******@cs.tut.fi> posted :

The best you could achieve is probably title=
"The publication date is the twenty-first of September in
the year two thousand five at three o'clock PM"

FYI, in proper English we don't generally use the abbreviations AM & PM
in conjunction with the time in words; and we don't ever write the year
like that either; and using o'clock is uncommon unless necessary to
indicate time.

"Published on the twenty-first of September two thousand and five at
three in the afternoon"

"Published on the twenty-first of September two thousand and five at 3
PM"

"Published on 21st September 2005 at 3 PM"
You can no doubt read English perfectly; but you sometimes write it,
albeit grammatically, with a slight foreign accent!

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 MIME. ©
Web <URL:http://www.merlyn.demon.co.uk/> - w. FAQish topics, links, acronyms
PAS EXE etc : <URL:http://www.merlyn.demon.co.uk/programs/> - see 00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm etc.

[mbstevens]

Dr John Stockton wrote:

You can no doubt read English perfectly; but you sometimes write it,
albeit grammatically, with a slight foreign accent!

We should also be succenct, as in:

<Though you seem to read perfectly and write grammatically, you
sometimes write with a foreign accent!>

For the modern reader, words like 'albeit' instead of 'though' add two
syllables and the white noise of a less familiar word. Here it leads
to the nested structure of an adverbial clause. You get information
across better when you don't ask the reader to push clauses onto their
mental stack for later retrieval.

The clause is short -- no big problem here. But it does sound old
fashioned and too formal for a newsgroup.

(Oim sorry, Gov., just coun't 'elp meself.)
;)
--
mbstevens
http://www.mbstevens.com

[James Pickering]

mbstevens wrote:

Dr John Stockton wrote:

You can no doubt read English perfectly; but you sometimes write it,
albeit grammatically, with a slight foreign accent!
We should also be succenct ..........

Succinct, maybe?
(Oim sorry, Gov., just coun't 'elp meself.)

Me neither. :-)

--
James Pickering
http://jp29.org/

[mbstevens]

James Pickering wrote:

mbstevens wrote:

Dr John Stockton wrote:

You can no doubt read English perfectly; but you sometimes write it,
albeit grammatically, with a slight foreign accent!

We should also be succenct ..........

Succinct, maybe?

(Oim sorry, Gov., just coun't 'elp meself.)

Me neither. :-)

Got me, you silver tongued dev... dev.... How DO you spell that.
:)

[Dr John Stockton]

JRS: In article <vi********************************@4ax.com>, dated
Thu, 29 Sep 2005 23:38:54, seen in news:comp.infosystems.www.authoring.h
tml, Matt Silberstein <Re**************************@ix.netcom.com>
posted :

Strunk and While argue for 28 Sept 2005. Such a date does not need any
punctuation since the numbers are separated by letters. I write the
S&K version.

They are alien writers, devoid of authority in the questioner's
locality. There's something unreasonable about indicating the ninth
month by a word which, unlike Sep & September, actually means seven in
at least one well-known language.

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 MIME. ©
Web <URL:http://www.merlyn.demon.co.uk/> - w. FAQish topics, links, acronyms
PAS EXE etc : <URL:http://www.merlyn.demon.co.uk/programs/> - see 00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm etc.

[boclair]

Dr John Stockton wrote:

JRS: In article <vi********************************@4ax.com>, dated
Thu, 29 Sep 2005 23:38:54, seen in news:comp.infosystems.www.authoring.h
tml, Matt Silberstein <Re**************************@ix.netcom.com>
posted :
There's something unreasonable about indicating the ninth month by a word which, unlike Sep & September, actually means seven in
at least one well-known language.

Concise OED: Sept.abbr.1September. 2Septuagint

Concise OED has no listing for Sep

Tim Morris

[James Pickering]

boclair wrote:

Dr John Stockton wrote:

JRS: In article <vi********************************@4ax.com>, dated
Thu, 29 Sep 2005 23:38:54, seen in news:comp.infosystems.www.authoring.h
tml, Matt Silberstein <Re**************************@ix.netcom.com>
posted :

There's something unreasonable about indicating the ninth

month by a word which, unlike Sep & September, actually means seven in
at least one well-known language.

September -- seventh (Latin: septum) month in the Roman calendar --
ninth month in the Gregorian calendar.

--
James Pickering
http://jp29.org/

[Matt Silberstein]

On Fri, 30 Sep 2005 23:52:15 +0100, in
comp.infosystems.www.authoring.html , Dr John Stockton
<jr*@merlyn.demon.co.uk> in <rV**************@merlyn.demon.co.uk>
wrote:

JRS: In article <vi********************************@4ax.com>, dated
Thu, 29 Sep 2005 23:38:54, seen in news:comp.infosystems.www.authoring.h
tml, Matt Silberstein <Re**************************@ix.netcom.com>
posted :

Strunk and While argue for 28 Sept 2005. Such a date does not need any
punctuation since the numbers are separated by letters. I write the
S&K version.

They are alien writers, devoid of authority in the questioner's
locality. There's something unreasonable about indicating the ninth
month by a word which, unlike Sep & September, actually means seven in
at least one well-known language.

And they probably wrote September actually, I shortened it by habit.

--
Matt Silberstein

Do something today about the Darfur Genocide

Genocide is news | Be A Witness
http://www.beawitness.org

"Darfur: A Genocide We can Stop"
www.darfurgenocide.org

Save Darfur.org :: Violence and Suffering in Sudan's Darfur Region
http://www.savedarfur.org/

[Stan Brown]

On Fri, 30 Sep 2005 23:52:15 +0100 in
comp.infosystems.www.authoring.html, Dr John Stockton favored us
with...

JRS: In article <vi********************************@4ax.com>, dated
Thu, 29 Sep 2005 23:38:54, seen in news:comp.infosystems.www.authoring.h
tml, Matt Silberstein <Re**************************@ix.netcom.com>
posted :

Strunk and While argue for 28 Sept 2005. Such a date does not need any
punctuation since the numbers are separated by letters. I write the
S&K version.

They are alien writers, devoid of authority in the questioner's
locality. There's something unreasonable about indicating the ninth
month by a word which, unlike Sep & September, actually means seven in
at least one well-known language.

As Fowler pointed out in other contexts, English grammar is not a
branch of logic.

If asked, he would probably call the abbreviation "Sept." for
September a STURDY INDEFENSIBLE -- illogical but nonetheless
standard.

--
Stan Brown, Oak Road Systems, Tompkins County, New York, USA
http://OakRoadSystems.com/
HTML 4.01 spec: http://www.w3.org/TR/html401/
validator: http://validator.w3.org/
CSS 2.1 spec: http://www.w3.org/TR/CSS21/
validator: http://jigsaw.w3.org/css-validator/
Why We Won't Help You:
http://diveintomark.org/archives/200..._wont_help_you

[Stan Brown]

On Fri, 30 Sep 2005 20:47:17 GMT in
comp.infosystems.www.authoring.html, mbstevens favored us with...

Dr John Stockton wrote:

You can no doubt read English perfectly; but you sometimes write it,
albeit grammatically, with a slight foreign accent!
We should also be succenct, as in:

<Though you seem to read perfectly and write grammatically, you
sometimes write with a foreign accent!>

For the modern reader, words like 'albeit' instead of 'though' add two
syllables and the white noise of a less familiar word.

Is the modern reader bothered by the misspelling of "succinct"?

I get so tired of people presuming to instruct others in English
usage and then making glaring mistakes themselves.
Here it leads
to the nested structure of an adverbial clause.

There is no adverbial clause in the sentence you quote: there are two
independent clauses joined by the conjunction "but". "albeit
grammatically" is an adverbial PHRASE.

Again, if you're going to offer usage advice, you might trouble to
get your terms right.

Moving beyond the details of how you phrased your criticism, I don't
see that your suggested version is superior to the original.

--
Stan Brown, Oak Road Systems, Tompkins County, New York, USA
http://OakRoadSystems.com/
HTML 4.01 spec: http://www.w3.org/TR/html401/
validator: http://validator.w3.org/
CSS 2.1 spec: http://www.w3.org/TR/CSS21/
validator: http://jigsaw.w3.org/css-validator/
Why We Won't Help You:
http://diveintomark.org/archives/200..._wont_help_you

[Andy Dingley]

On Thu, 29 Sep 2005 23:38:54 GMT, Matt Silberstein
<Re**************************@ix.netcom.com> wrote:

Strunk and While argue for

Strunk and White are not seen as particularly authoritative in the
non-US English-speaking world.

[Dr John Stockton]

JRS: In article <g4********************************@4ax.com>, dated
Sat, 1 Oct 2005 12:36:38, seen in news:comp.infosystems.www.authoring.ht
ml, Andy Dingley <di*****@codesmiths.com> posted :

On Thu, 29 Sep 2005 23:38:54 GMT, Matt Silberstein
<Re**************************@ix.netcom.com> wrote:

Strunk and While argue for

Strunk and White are not seen as particularly authoritative in the
non-US English-speaking world.

That seems unreasonably tactful.

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk DOS 3.3, 6.20; Win98. ©
Web <URL:http://www.merlyn.demon.co.uk/> - FAQqish topics, acronyms & links.
PAS EXE TXT ZIP via <URL:http://www.merlyn.demon.co.uk/programs/00index.htm>
My DOS <URL:http://www.merlyn.demon.co.uk/batfiles.htm> - also batprogs.htm.

[Dr John Stockton]

JRS: In article <6O*****************@news-server.bigpond.net.au>, dated
Fri, 30 Sep 2005 23:23:46, seen in news:comp.infosystems.www.authoring.h
tml, boclair <bo*****@bigpond.net.au> posted :

Dr John Stockton wrote:

JRS: In article <vi********************************@4ax.com>, dated
Thu, 29 Sep 2005 23:38:54, seen in news:comp.infosystems.www.authoring.h
tml, Matt Silberstein <Re**************************@ix.netcom.com>
posted :

There's something unreasonable about indicating the ninth

month by a word which, unlike Sep & September, actually means seven in
at least one well-known language.

Concise OED: Sept.abbr.1September. 2Septuagint

Concise OED has no listing for Sep

Dictionaries represent usage, and do not judge reasonability. The COD
is rather small. One should not expect too much from Universities in
the Western Hemisphere. Chambers Dictionary has Sep.

Sept is a word meaning seven in at least one language; AFAIK, Sep is
not.

I know well enough *why* it's called September-for-7; the names of
Sep..Dec should have been changed at the appropriate time.

IMHO, the preferred mark-up should use YYYY-MM-DD, and systems not
treating that well should be upgraded. A speech used should be able to
choose to have the month read as a word in his local language and
accent.

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 MIME. ©
Web <URL:http://www.merlyn.demon.co.uk/> - w. FAQish topics, links, acronyms
PAS EXE etc : <URL:http://www.merlyn.demon.co.uk/programs/> - see 00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm etc.

[Andy Dingley]

Dr John Stockton wrote:

Andy Dingley <di*****@codesmiths.com> posted :

Strunk and White are not seen as particularly authoritative in the
non-US English-speaking world.

That seems unreasonably tactful.

It's a forged post 8-)

[Andreas Prilop]

On Fri, 30 Sep 2005, boclair wrote:

Date: Fri, 30 Sep 2005 23:23:46 GMT

Concise OED has no listing for Sep

Look into your own header.

[Johnny Winther Ronnenberg]

Andy Dingley wrote:

I'm using this at present:
<p title="Publication date" >[28/09/2005 15:00]</p>

Works fine on screen, but Fangs/Jaws just reads it as "left bracket
twenty-eight slash zero slash two thousand five fifteen colon zero
right bracket" Really it needs something more to indicate that
it _is_ a date.

The [...] brackets would be better done with CSS, :before and
content: , but the boss wants it to work under IE too. The site is
firmly UK based, so d/m/y is necessary and won't be getting
internationalised.

So, what's the best markup for representing a date, with good
accessibility to screen readers?

Although this tread was fun to read, it seems that no one had the thought,
why don't we have at a set of
<date><day></day><month></month><year></year><time></time></date> tags?

We have a language attribute and the possibility to select a specific
codepage, which will make most modern browsers use language specific
settings. What would be more natural, than a set of tags, that adapts to
national standards?

Then we might be able to pass beyond "Me Tarzan, you Jane" dialogs and users
do not have to guess whether 2005-5-1 means fifth of January or first of
May. Both are valid in different languages just as the same goes for date
with the year in the end.

To put it mildly, dates are a pain in the ass, you never know what they
actually means. It would be nice to know was this written 2 days ago, 3
months ago or 5 years ago. But due to the different date systems one rarely
have a chance to find out, unless one is lucky and both day and month
exceeds 12 ;-)

Best regards

Johnny Winther Ronnenberg

[Tim]

On Fri, 25 Nov 2005 20:54:15 +0100, Johnny Winther Ronnenberg sent:

To put it mildly, dates are a pain in the ass, you never know what they
actually means. It would be nice to know was this written 2 days ago, 3
months ago or 5 years ago. But due to the different date systems one rarely
have a chance to find out, unless one is lucky and both day and month
exceeds 12 ;-)

What's the name of the thirteen month? ;-)

--
If you insist on e-mailing me, use the reply-to address (it's real but
temporary). But please reply to the group, like you're supposed to.

This message was sent without a virus, please destroy some files yourself.

[Jukka K. Korpela]

"Johnny Winther Ronnenberg" <jo************@XYZtdcadsl.dk> wrote:

Although this tread was fun to read,
About two months ago, yes. Anyone who wishes to check what was discussed has
to check Google Groups or maybe some other archive.
it seems that no one had the
thought, why don't we have at a set of
<date><day></day><month></month><year></year><time></time></date> tags?
Are you sure? Maybe it was too obvious to be mentioned that we would not
discuss the notation of dates that much if there were markup for them.
We have a language attribute and the possibility to select a specific
codepage, which will make most modern browsers use language specific
settings.
Sorry, but you have apparently no idea of what the lang attribute means and
causes. It is not meant to select a codepage, and no browser is crazy enough
to treat it that way. Hardly any browser uses any language specific settings
when you use the lang attribute, though many of them mess around with fonts,
making the default font depend on the declared language, if you have not
specified any font.
What would be more natural, than a set of tags, that adapts to
national standards?
I could list down a few thousand more natural things, like the sunshine and
landscape covered with snow. Anyway, national standards are something
completely different, and there are typically thousands of them in each
country, and few if any of them relate to HTML authoring for the WWW.
Then we might be able to pass beyond "Me Tarzan, you Jane" dialogs and
users do not have to guess whether 2005-5-1 means fifth of January or
first of May. Both are valid in different languages just as the same goes
for date with the year in the end.
You apparently picked up a date notation that does not correspond to _any_
standard or authoritative recommendation. Better luck next time.
To put it mildly, dates are a pain in the ass, you never know what they
actually means. It would be nice to know was this written 2 days ago, 3
months ago or 5 years ago. But due to the different date systems one
rarely have a chance to find out, unless one is lucky and both day and
month exceeds 12 ;-)

Exactly how many plausible interpretations can you find for the expression
"November 26, 2005" in an English-language document, or "26. marraskuuta
2005" in a Finnish-language document?

--
Yucca, http://www.cs.tut.fi/~jkorpela/
Pages about Web authoring: http://www.cs.tut.fi/~jkorpela/www.html

[Norman L. DeForest]

On Sat, 26 Nov 2005, Tim wrote:

On Fri, 25 Nov 2005 20:54:15 +0100, Johnny Winther Ronnenberg sent:

To put it mildly, dates are a pain in the ass, you never know what they
actually means. It would be nice to know was this written 2 days ago, 3
months ago or 5 years ago. But due to the different date systems one rarely
have a chance to find out, unless one is lucky and both day and month
exceeds 12 ;-)

What's the name of the thirteen month? ;-)

"Qudrat" in the Bahá'í calendar which has 20 months in a year.

"Adar II" in the Hebrew calendar for leap years that have 13 months.

"Mac" in the Mayan calendar (Yucatan Mayan language; the Guatemalan Mayans
used slightly different month names but I don't have a listing of them)
which has 18 months in a year.

--
Norman De Forest http://www.chebucto.ns.ca/~af380/Profile.html
"> Is there anything Spamazon DOESN'T sell?
Clues. The market's too small to justify the effort."
-- Stuart Lamble in the scary devil monastery, Fri, 13 May 2005

[Safalra]

On Fri, 25 Nov 2005 20:54:15 +0100, Johnny Winther Ronnenberg wrote:

Andy Dingley wrote:

[snip acccessibility of dates]

Although this tread was fun to read, it seems that no one had the thought,
why don't we have at a set of
<date><day></day><month></month><year></year><time></time></date> tags?

You've chosen an interesting combination of specificity here - you're
requiring a division into days, months, and years, even though some
calendars don't break up dates in this fashion, but you don't require a
simiilar division of times into hours, minutes, and seconds.

In line with the general ideas of mark-up something like this would be
better, if it existed:

<p>We'll meet up <time time="20051129">next Tuesday</time> to discuss the
future of HTML.</p>

Here the time element (which could be used to represent a more specific
time than the date given in the example) would have an attribute specifying
an ISO-8601 formatted time, while allowing the page author to write the
time in whatever way they felt suitable.

These are all just ideas for the future, of course.

--
Safalra (Stephen Morley)
http://www.safalra.com/hypertext/

[Dr John Stockton]

JRS: In article <43*********************@dread11.news.tele.dk>, dated
Fri, 25 Nov 2005 20:54:15, seen in news:comp.infosystems.www.authoring.h
tml, Johnny Winther Ronnenberg <jo************@XYZtdcadsl.dk> posted :

Then we might be able to pass beyond "Me Tarzan, you Jane" dialogs and users
do not have to guess whether 2005-5-1 means fifth of January or first of
May. Both are valid in different languages just as the same goes for date
with the year in the end.

Who do you think uses Y D M as the date field order?

The correct form is 2005-05-01 for 2005 May 1st -- ISO 8601.

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 MIME. ©
Web <URL:http://www.merlyn.demon.co.uk/> - w. FAQish topics, links, acronyms
PAS EXE etc : <URL:http://www.merlyn.demon.co.uk/programs/> - see 00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm etc.

[Johnny Winther Ronnenberg]

Jukka K. Korpela wrote:

"Johnny Winther Ronnenberg" <jo************@XYZtdcadsl.dk> wrote:

it seems that no one had the
thought, why don't we have at a set of
<date><day></day><month></month><year></year><time></time></date>
tags?
Are you sure? Maybe it was too obvious to be mentioned that we would
not discuss the notation of dates that much if there were markup for
them.

There is nothing obvious in that HTML lacks a needed tag. Databases,
spreadsheets and word processors uses a special tag for dates and are
therefore able to show you a date in another language format.

Given that a set of tags existed and where used correct, the user agent
could very easily convert a date according to the user's preference or the
conventions of a given language. This would make it possible to request all
dates in a form one where accustomed to. Just as you can choose to use a
personal style sheet that show pages different from what the designer
intended.

Beyond that it would help in the conversion to other document types,-)

We have a language attribute and the possibility to select a specific
codepage, which will make most modern browsers use language specific
settings.

Sorry, but you have apparently no idea of what the lang attribute
means and causes. It is not meant to select a codepage, and no
browser is crazy enough to treat it that way. Hardly any browser uses
any language specific settings when you use the lang attribute,
though many of them mess around with fonts, making the default font
depend on the declared language, if you have not specified any font.

Sorry but you are wrong, if you use the lang attribute and set the needed
codepage in the metatags. Modern browsers IE5+ FF and Opera will use
language elements correct. My native language is Danish and Danish has tre
letters that are not used in any other languages, something similar goes for
Norwegian and Swedish.

According to the standard, chapter 8, user agents should be able to use
correct lettering of a defined language. But I know for certain that IE has
to be told which code page to use otherwise it will show Danish letters in
very little meaningful way;-)

Anyway, national standards
are something completely different, and there are typically thousands
of them in each country, and few if any of them relate to HTML
authoring for the WWW.

HTNL authoring is about writing document to an audience, which has very
different concepts of how things should be presented, among them are dates.
No standard in the world will ever influence how ordinary people percept
things, especially not across language barriers.

English is a very good proof of that, its not one language, its several
based on different cultures. As far as I know, there are at least five or
six official versions each with a special grammar, spelling and conventions.
None of them are more correct than the other, but in daily life's both
academics as well as the working man use at different and simpler grammar
and spelling, which also are used in books, newspapers etc. too.

If the language standard does not reflects the actual use of the language it
becomes ready to be deprecated ;-) This goes to for every spoken language in
the world.

HTML is a structural mark-up language, it is not defined for English only,
if it where, it probably would not have support for languages that's are
read from right to left ;-)

You seem to think that I am barking up the wrong tree, but you are deadly
wrong. My wife who are a secretary in City Hall of Copenhagen sees daily
birth, wedding and divorce certificates from every corner of the world and
she can tell you, that it its pure hell, to guess how six digits with spaces
between shall be read.

In Denmark dates are written dd-mm-yy as a standard although yy-mm-dd are
allowed. Two different interpretations are possible 06-01-03 take a guess.
The date is on your divorce papers according to Danish laws you cannot be
married before your divorce is final. The officials therefore have to
request verifications from embassies all over the world.

Both in and out of Europe the following combinations can and do occur,
dd-mm-yy, mm-dd-yy, yy-mm-dd and yy-dd-mm. So even official conventions in
public offices differ and creates headaches for the people who are dependent
on exact information.

So dates can be inaccessible to fully normal and well functioning people;-)
Best regards
Johnny Winther Ronnenberg

[Alan J. Flavell]

On Sun, 27 Nov 2005, Johnny Winther Ronnenberg wrote:

Jukka K. Korpela wrote:

Sorry, but you have apparently no idea of what the lang attribute
means and causes. It is not meant to select a codepage, and no
browser is crazy enough to treat it that way.

[...]
Sorry but you are wrong, if you use the lang attribute and set the needed
codepage in the metatags.
You don't understand what you're talking about.
Modern browsers IE5+ FF and Opera will use language elements
correct.
Define what you mean by "correct". As far as I can see, you haven't
mentioned /anything/ yet which is likely to be usefully influenced by
the HTML lang attribute (as opposed to the HTTP character encoding
scheme).
My native language is Danish and Danish has tre letters that are not
used in any other languages, something similar goes for Norwegian
and Swedish.
Of course. You still don't understand how things work in HTML.
I'd recommend that you find out *before* contradicting those who are
trying to explain it to you.

Your use of the antique term "code page" is already a warning sign.
But I know for certain that IE has to be told which code page to use
otherwise it will show Danish letters in very little meaningful
way;-)

And this has precisely *what* to do with the HTML lang attribute???
(Hint: nothing.)

[Stephen Poley]

On Sun, 27 Nov 2005 16:49:18 +0100, "Johnny Winther Ronnenberg"
<jo************@XYZtdcadsl.dk> wrote:

English is a very good proof of that, its not one language, its several
based on different cultures. As far as I know, there are at least five or
six official versions each with a special grammar, spelling and conventions.

There are no official versions of English (fortunately).

--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/

[Dr John Stockton]

JRS: In article <43*********************@dread11.news.tele.dk>, dated
Sun, 27 Nov 2005 16:49:18, seen in news:comp.infosystems.www.authoring.h
tml, Johnny Winther Ronnenberg <jo************@XYZtdcadsl.dk> posted :

Both in and out of Europe the following combinations can and do occur,
dd-mm-yy, mm-dd-yy, yy-mm-dd and yy-dd-mm.

Where is yy-dd-mm used, except by mistake?

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 MIME. ©
Web <URL:http://www.merlyn.demon.co.uk/> - w. FAQish topics, links, acronyms
PAS EXE etc : <URL:http://www.merlyn.demon.co.uk/programs/> - see 00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm etc.

[Pierre Goiffon]

Dr John Stockton wrote:

Where is yy-dd-mm used, except by mistake?

I was told, Japan ?

[Philip Ronan]

"Pierre Goiffon" wrote:

Dr John Stockton wrote:

Where is yy-dd-mm used, except by mistake?

I was told, Japan ?

yy-mm-dd

--
phil [dot] ronan @ virgin [dot] net
http://vzone.virgin.net/phil.ronan/

[Johnny Winther Ronnenberg]

Dr John Stockton wrote:

JRS: In article <43*********************@dread11.news.tele.dk>, dated
Sun, 27 Nov 2005 16:49:18, seen in
news:comp.infosystems.www.authoring.h tml, Johnny Winther Ronnenberg
<jo************@XYZtdcadsl.dk> posted :

Both in and out of Europe the following combinations can and do
occur, dd-mm-yy, mm-dd-yy, yy-mm-dd and yy-dd-mm.

Where is yy-dd-mm used, except by mistake?

According to my wife its quit common in Sweden, Norway, France, Spain,
Italy, it dos occur in UK and Germany, as well, as in countries on the South
American continent and from some Asian countries and Australia and in other
countries to.

It does not surprise me much, my wife has been at Cityhall of Copenhagen for
almost 25 years and they do things too day like they did 100 years ago,
their "official grammar" where forgotten by the rest of the population by
the time my mother considered having me, but its still official grammar
there. I think that, that's the case the rest of the world over to, people
continue their local customs. No matter what a zillion standards says is
correct.

I just a asked my wife again and she says, there is only one exception,
where dates are represented in a meaningful way and that is in passports.
They are in 8 digits and in dd-mm-yyyy or yyyy-mm-dd.

Best regards

Johnny Winther Ronnenberg

[Jan Roland Eriksson]

On Tue, 29 Nov 2005 22:34:37 +0100, "Johnny Winther Ronnenberg"
<jo************@XYZtdcadsl.dk> wrote:

Dr John Stockton wrote:
[...]

Where is yy-dd-mm used, except by mistake?

According to my wife its quite common in Sweden...
No way! it's "totally out of order" here in my home country
(Sweden that is)

[...]
It does not surprise me much, my wife has been at Cityhall
of Copenhagen for almost 25 years...

Well; we Swedes tend to think that the Danes are a bit peculiar :-)

--
Rex

[Pierre Goiffon]

Philip Ronan wrote:

Where is yy-dd-mm used, except by mistake?

I was told, Japan ?

yy-mm-dd

Oops yes
Thanks for the correction !

[Pierre Goiffon]

Johnny Winther Ronnenberg wrote:

Where is yy-dd-mm used, except by mistake?

According to my wife its quit common in Sweden, Norway, France, Spain,
Italy, it dos occur in UK and Germany, as well, as in countries on the South
American continent and from some Asian countries and Australia and in other
countries to.

I can say I've never seen such a date format used in France

[Philip Ronan]

"Pierre Goiffon" wrote:

Johnny Winther Ronnenberg wrote:

Where is yy-dd-mm used, except by mistake?

According to my wife its quit common in Sweden, Norway, France, Spain,
Italy, it dos occur in UK and Germany, as well, as in countries on the South
American continent and from some Asian countries and Australia and in other
countries to.

I can say I've never seen such a date format used in France

I've never seen it in the UK either :-(

--
phil [dot] ronan @ virgin [dot] net
http://vzone.virgin.net/phil.ronan/

[Dr John Stockton]

JRS: In article <43*********************@dread11.news.tele.dk>, dated
Tue, 29 Nov 2005 22:34:37 local, seen in news:comp.infosystems.www.autho
ring.html, Johnny Winther Ronnenberg <jo************@XYZtdcadsl.dk>
posted :

Dr John Stockton wrote:

JRS: In article <43*********************@dread11.news.tele.dk>, dated
Sun, 27 Nov 2005 16:49:18, seen in
news:comp.infosystems.www.authoring.h tml, Johnny Winther Ronnenberg
<jo************@XYZtdcadsl.dk> posted :

Both in and out of Europe the following combinations can and do
occur, dd-mm-yy, mm-dd-yy, yy-mm-dd and yy-dd-mm.

Where is yy-dd-mm used, except by mistake?

According to my wife its quit common in Sweden, Norway, France, Spain,
Italy, it dos occur in UK and Germany, as well, as in countries on the South
American continent and from some Asian countries and Australia and in other
countries to.

I've no recollection of seeing it used in or by in any of those European
countries; or any others in fact. I might have missed seeing it
elsewhere (I saw no Spaniards when I was in Spain); but I'm confident UK
does not use it.

I just a asked my wife again and she says, there is only one exception,
where dates are represented in a meaningful way and that is in passports.
They are in 8 digits and in dd-mm-yyyy or yyyy-mm-dd.

Even American ones?

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 MIME. ©
Web <URL:http://www.merlyn.demon.co.uk/> - w. FAQish topics, links, acronyms
PAS EXE etc : <URL:http://www.merlyn.demon.co.uk/programs/> - see 00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm etc.

[Robi]

Johnny Winther Ronnenberg wrote in message news:43*********************@dread11.news.tele.dk. ..

Dr John Stockton wrote:

JRS: In article <43*********************@dread11.news.tele.dk>, dated
Sun, 27 Nov 2005 16:49:18, seen in
news:comp.infosystems.www.authoring.html, Johnny Winther Ronnenberg posted :

Both in and out of Europe the following combinations can and do
occur, dd-mm-yy, mm-dd-yy, yy-mm-dd and yy-dd-mm.

Where is yy-dd-mm used, except by mistake?

According to my wife its quit common in Sweden, Norway, France, Spain,
Italy, it dos occur in UK and Germany, as well, as in countries on the South
American continent and from some Asian countries and Australia and in other
countries to.

Never seen it like that in Germany, Italy, Spain, France, the UK and Venezuela.
I've never seen it as yy-dd-mm anywhere. The only odd way I had to get accustomed
to is the US way - mm-dd-yy.

I don't know about Asia or Down Under though (sorry Bruce ;-)

[Johnny Winther Ronnenberg]

Dr John Stockton wrote:

Even American ones?

I asked my wife she could not remember whether US passport used one format
or another. But she had a comment, if the year is obvious, we might not
investigate further. But in most cases we call the embassy for confirmation.

To Jan Roland Eriksson, both in Sweden, Norway and Denmark one can find
samples of birth certificates, marriage certificates and divorce papers with
date formats in a "wrong" format. Just as one can in other countries as
well. The important point is to remember that no RFC or ISO standard will
change the mind or culture of public offices, just as regular laws make a
little difference.

The view of the world of the priest rules, when he writes the birth
certificate. If that was not so, my second first name would be illegal in
Denmark. But it is not, if you are a direct descendant of the naval officer
who came to shore when the Danes sunk the English fleet a long time ago. At
the time Winther where not recognizes as a legal surname, but the priest
ruled that as a first name it was ok. So now its legal and protected as if
the priest had written an RFC which where accepted as an standard. The
priest did none of the kind, he just used his own common sense just as
people in public offices do to day. Unfortunately that leads to dates that
are not directly translatable..

Best regards

Johnny Winther Ronnenberg