Proper headings

[Eustace]

From time to time I am reworking webpages that I have made in the past,
trying to improve them both technically and otherwise. At this point,
I've been considering the use of headings. The webpage in question

https://files.nyu.edu/emf202/public/fr/limericks.html

is in Greek. It's structure is like this (I have changed here some
numbers to what I currently think they should be):

<h1>Limericks</h1>
<h6>by author</h6>

<h4>Introduction</h4>
....

<h5>Disclaimer</h5>
....

<h2>poems' section 1</h2>

<h3>POEM 1 TITLE</h3>
....
<h3>POEM 2 TITLE</h3>
....

<h2>poems' section 2</h2>

<h3>POEM 1 TITLE</h3>
....
<h3>POEM 2 TITLE</h3>
....
<h3>POEM 3 TITLE</h3>
....

Some of the titles are within tables. h1, h2, h3 (title of the page,
section of poems, title of poem) represent the basic structure of the
webpage. What I am not sure is what heading numbers the introduction,
the disclaimer, and the author's name should have. Above I have made
them h4, h5, h6, but I am not sure it's the correct way. Any ideas?

emf

 

[dorayme]

From time to time I am reworking webpages that I have made in the past,
trying to improve them both technically and otherwise. At this point,
I've been considering the use of headings. The webpage in question

https://files.nyu.edu/emf202/public/fr/limericks.html

is in Greek. It's structure is like this (I have changed here some
numbers to what I currently think they should be):

<h1>Limericks</h1>
<h6>by author</h6>

No, a heading must describe the topic *directly* under it. You
might be using h6 because its default style is small. Best not to
do this. Just think of what a heading *really means*. Style is a
separate issue and can b handles by CSS.

And read though some of

<http://www.w3.org/TR/html401/struct/global.html#h-7.5.5>

"A heading element briefly describes the topic of the section it
introduces. Heading information may be used by user agents, for
example, to construct a table of contents for a document
automatically.

There are six levels of headings in HTML with H1 as the most
important and H6 as the least. Visual browsers usually render
more important headings in larger fonts than less important ones."

 

[Jukka K. Korpela]

No, a heading must describe the topic *directly* under it.

No, a heading is a heading for part of the document implicitly defined by
the overall use of headings. The HTML recommendations are implicit and vague
in semantic issues like this, though they make feeble attempts at describing
proper use of headings. The little known HTML standard tries to make it
clearer, though only with extra syntax, and you need to read the User's
Guide to the standard to things described clearly:
http://www.scss.tcd.ie/misc/15445/UG.HTML#H1

To put it simply, h1 is a heading for anything that follows it, up to the
next h1 or the end of document, whichever occurs first. And h2 is a heading
for anything that follows it, up to the next h2 or h1 or end of document.

You might be using h6 because its default style is small.

Which is an odd default setting, strangely described and perhaps even
recommended in CSS recommendations. The foolish defaults generated foolish
usage, where h6 is used just to get "fine print", and this in turn implies
that browser vendors don't change the defaults, as they think that would
"break existing documents".

This, and a few other things like the huge default size (with bolding) for
h1, imply that one should always use CSS to suggest reasonable heading
formatting, as appropriate for the document's nature and style.

Regarding the original issue, it's a good rule of thumb that it's waste of
time or worse to try to "clean up" existing pages, unless they are a)
commercially or otherwise very important, b) frequently changed or c) being
renewed regarding purpose, structure, and content, not just code "clean up".
(Item "a" could be omitted from the list, since it almost certainly implies
the need for "b" or "c".)

Assuming that "a", "b", or "c" applies to the OP's page
https://files.nyu.edu/emf202/public/fr/limericks.html
it is evident that it should have only one h1 element. The authors name
should either be made part of that element or presented outside any heading
elements, since as such, it's not a heading. (It could be made a heading if
it preceded some "about the author" section.) You could write e.g.
<h1>Heading of the page<br><small>by A. Author</small></h1>
and style the small element separately if desired.

If the page's substance content is poems, and the rest is explanations, then
you could have one section with h2 heading for the explanations and another
section with h2 for the poems, making the poem headings h3, of course.
Alternatively, and probably better (think about a table of contents
generated from headings), you could make each poem a section at the 2nd
level, with h2 heading.

The decorative dividers before and between the poems look confusing (as they
suggest a fundamental division of the page, not consistent with its
structure) and a bit pompous, but especially the texts between them, now
marked as h1, confuse the casual reader. If they relate to the division of
the page into sections, then the heading issue needs to be reconsidered; and
then it's odd that a "heading" is both preceded and followed by a major
visual divider - what is it a heading for, then?

BY THE WAY, THERE IS NO REASON TO "SHOUT" I.E. WRITE TEXT IN ALL UPPERCASE,
IN HEADINGS OR OTHERWISE. IT MAKES TEXT LESS READABLE AND LOOKS CHILDISH OR
OUTDATED. EVEN IF YOU REALLY WANT YOUR HEADINGS TO APPEAR IN ALL LOWERCASE,
WRITE THEM IN NORMAL, "MIXED" CASE IN THE ELEMENT'S CONTENT AND USE THE CSS
SETTING TEXT-TRANSFORM: UPPERCASE.

The page seems to contain W3C recommended icon links for making absurd
noises claiming the document to be "valid HTML" and "valid CSS" (see
http://www.cs.tut.fi/~jkorpela/html/validation.html#icon ), but apparently
the author never bothered testing the links. Clicking on them results (on
normal modern browsers) in

"Sorry! This document can not be checked.
No Referer header found!"

The reason here is that the page has been transferred with https, for no
good reason. Worse still, if I try to access
http://files.nyu.edu/emf202/public/fr/limericks.html
the server redirects to the https address! Of course, this is probably the
server administrators' fault, but it still has an impact on the page. (If
you don't use such foolish icon links, probably the only downside of forced
https is unnecessary overhead in traffic, i.e. the page loads unnecessarily
slow, though usually the effect is not serious. However, one day the server
admin forgets to renew their certificate for the server or mess up with it,
and then you have serious problems.)

(I'm not surprised at seeing that the "W3C CSS Validator" actually reports a
syntax error, if the page is checked by URL.)

 

[dorayme]

No, a heading is a heading for part of the document implicitly defined by
the overall use of headings.

As I understand it, a top level heading, in a paradigm case, is
something that informs about the rest of the document, including
sub headings - "All about birds". An h2 is about all the things
under it including its sub headings - "All about Finches". An h3
is about all the things under it, including its sub headings -
"All about Blue Finches". And so on.

I had not meant by "direct" to suggest that an h1 is about only
any bits before the next sub heading. If the top general walks
by, everyone from the slightly lesser to the lowliest of potato
peelers in the kitchen must salute him and be directly smart
about it.

 

[Rob W.]

Op 28-3-2010 10:57, dorayme schreef:

[snipped]
If the top general walks by,
everyone from the slightly lesser to the lowliest of potato
peelers in the kitchen must salute him and be directly smart
about it.

That may be how they do it on Mars,
but where I live generals don't enter the kitchen
(and nobody slautes either).

 

[dorayme]

Op 28-3-2010 10:57, dorayme schreef:

[snipped]
If the top general walks by,
everyone from the slightly lesser to the lowliest of potato
peelers in the kitchen must salute him and be directly smart
about it.

That may be how they do it on Mars,
but where I live generals don't enter the kitchen
(and nobody slautes either).

Then you better go and see the henry King classic Twelve O'Clock
High where everyone saluted Gregory Peck if he valued his life.
And further than this, you cheeky young thing - <g> - cooks are
not always in the kitchen even on earth. In fact it would be a
scandal if they were and had to piss in there.

 

[Eustace]

No, a heading must describe the topic *directly* under it. You
might be using h6 because its default style is small. Best not to
do this. Just think of what a heading *really means*. Style is a
separate issue and can b handles by CSS.

And read though some of

<http://www.w3.org/TR/html401/struct/global.html#h-7.5.5>

"A heading element briefly describes the topic of the section it
introduces. Heading information may be used by user agents, for
example, to construct a table of contents for a document
automatically.

There are six levels of headings in HTML with H1 as the most
important and H6 as the least. Visual browsers usually render
more important headings in larger fonts than less important ones."

I had added the <h6> last while wondering what I should do about it, and
knowing that I had to do something about it.

(1) An obvious alternative is:

<h1>Limericks<br>
<span>by author</span></h1>

<h2>Introduction</h2>
....

<h3>Disclaimer</h3>
....

<h2>poems' section 1</h2>

<h3>POEM 1 TITLE</h3>
....
<h3>POEM 2 TITLE</h3>
....

<h2>poems' section 2</h2>

<h3>POEM 1 TITLE</h3>
....
<h3>POEM 2 TITLE</h3>
....
<h3>POEM 3 TITLE</h3>
....


My problem with this is that the first <h2> subsection is not equivalent
to the other 2, nor is it's <h3> equivalent to the other <h3>s.

(2) The way to avoid this would be to assign:

<span>Introduction</span>
....

<span>Disclaimer</span>
....

omitting the <h2> and <h3>. The problem with this is the intro and the
disclaimer are not included in the webpage's structure.

(3) But maybe a better idea would be to enclose them within

<div class="into">
....
</div>

and *define their <h2> and <h3> separately with CSS in parallel to the
<h2> and <h3> of the main part of the webpage*, which is already
enclosed within

<div class="poems">
....
</div>

I am not sure whether the idea to have 2 sets of parallel headings
within different (and unequal) sections is OK, but it seems logical enough.

What do you think of these ideas?

Thanks,

emf

 

[Eustace]

No, a heading is a heading for part of the document implicitly defined
by the overall use of headings. The HTML recommendations are implicit
and vague in semantic issues like this, though they make feeble attempts
at describing proper use of headings. The little known HTML standard
tries to make it clearer, though only with extra syntax, and you need to
read the User's Guide to the standard to things described clearly:
http://www.scss.tcd.ie/misc/15445/UG.HTML#H1

To put it simply, h1 is a heading for anything that follows it, up to
the next h1 or the end of document, whichever occurs first. And h2 is a
heading for anything that follows it, up to the next h2 or h1 or end of
document.


Which is an odd default setting, strangely described and perhaps even
recommended in CSS recommendations. The foolish defaults generated
foolish usage, where h6 is used just to get "fine print", and this in
turn implies that browser vendors don't change the defaults, as they
think that would "break existing documents".

This, and a few other things like the huge default size (with bolding)
for h1, imply that one should always use CSS to suggest reasonable
heading formatting, as appropriate for the document's nature and style.

Regarding the original issue, it's a good rule of thumb that it's waste
of time or worse to try to "clean up" existing pages, unless they are a)
commercially or otherwise very important, b) frequently changed or c)
being renewed regarding purpose, structure, and content, not just code
"clean up". (Item "a" could be omitted from the list, since it almost
certainly implies the need for "b" or "c".)

The rules that apply to professional webmasters do not necessarily apply
to amateur ones. Since the number of my webpages is yet small, from time
to time, I review them, considering this a way to upgrade my
web-authoring skills.

Assuming that "a", "b", or "c" applies to the OP's page
https://files.nyu.edu/emf202/public/fr/limericks.html
it is evident that it should have only one h1 element. The authors name
should either be made part of that element or presented outside any
heading elements, since as such, it's not a heading. (It could be made a
heading if it preceded some "about the author" section.) You could write
e.g.
<h1>Heading of the page<br><small>by A. Author</small></h1>
and style the small element separately if desired.

If the page's substance content is poems, and the rest is explanations,
then you could have one section with h2 heading for the explanations and
another section with h2 for the poems, making the poem headings h3, of
course.

Yes, this seems a good idea. See my reply to Dorayme.

Alternatively, and probably better (think about a table of

contents generated from headings), you could make each poem a section at
the 2nd level, with h2 heading.

The decorative dividers before and between the poems look confusing (as
they suggest a fundamental division of the page, not consistent with its
structure) and a bit pompous, but especially the texts between them, now
marked as h1, confuse the casual reader. If they relate to the division
of the page into sections, then the heading issue needs to be
reconsidered; and then it's odd that a "heading" is both preceded and
followed by a major visual divider - what is it a heading for, then?

Most of the poems are satirical (e.g. The voyage of Ulysses - Ulysses
from Atlantic City / set sail for Crete / in a small boat with oars /
but fortunately the Coast Guard / returned him safely home.) and part of
the satire is implicitly comparing modern Greek-Americans, who not
infrequently have ancient Greek names, to their glorious ancient
ancestors, which is of course absurd. I have spent time and effort on
the decorative dividers, that form an integral part of the webpage and
contribute to its overall satirical effect on the reader, and I don't
think I have finished with them yet.

BY THE WAY, THERE IS NO REASON TO "SHOUT" I.E. WRITE TEXT IN ALL
UPPERCASE, IN HEADINGS OR OTHERWISE. IT MAKES TEXT LESS READABLE AND
LOOKS CHILDISH OR OUTDATED. EVEN IF YOU REALLY WANT YOUR HEADINGS TO
APPEAR IN ALL LOWERCASE, WRITE THEM IN NORMAL, "MIXED" CASE IN THE
ELEMENT'S CONTENT AND USE THE CSS SETTING TEXT-TRANSFORM: UPPERCASE.

I mainly use uppercase for titles of poems... Anyway using
text-transform: uppercase seems a little too complicated and is not a
very good option for Greek because it transforms lowercase accented
letters to uppercase accented letters, while normally Greek uppercase
letters should not be accented. In another webpage:

https://files.nyu.edu/emf202/public/mt/CantoGeneralEL.html

where I used font-variant: small-caps; I had to be careful to write the
letters without accents in the code.

The page seems to contain W3C recommended icon links for making absurd
noises claiming the document to be "valid HTML" and "valid CSS" (see
http://www.cs.tut.fi/~jkorpela/html/validation.html#icon ), but
apparently the author never bothered testing the links. Clicking on them
results (on normal modern browsers) in

"Sorry! This document can not be checked.
No Referer header found!"

I've considered using just the W3C icons without the links to avoid
these problems, communicating at least to the visitor the fact that I
took time to test the webpage.

The reason here is that the page has been transferred with https, for no
good reason. Worse still, if I try to access
http://files.nyu.edu/emf202/public/fr/limericks.html
the server redirects to the https address! Of course, this is probably
the server administrators' fault, but it still has an impact on the
page. (If you don't use such foolish icon links, probably the only
downside of forced https is unnecessary overhead in traffic, i.e. the
page loads unnecessarily slow, though usually the effect is not serious.
However, one day the server admin forgets to renew their certificate for
the server or mess up with it, and then you have serious problems.)

I have complained to the server administrator about using https for the
_public_ folder of the account but to no avail. When they forget to
renew the certificate would be an excellent time to complain again even
more strongly.

(I'm not surprised at seeing that the "W3C CSS Validator" actually
reports a syntax error, if the page is checked by URL.)

emf

 

[Jukka K. Korpela]

In future, please quote or paraphrase just the specific statements or
questions that you are commenting on. Hang around on Usenet for a while and
you'll see the point (and the high cluelessness factor of fullquoters).

Yes, this seems a good idea. See my reply to Dorayme.

I saw it and I did not see your point. You describe a logical approach, then
discard it on strange grounds:
"first <h2> subsection is not equivalent to the other 2, nor is it's <h3>
equivalent to the other <h3>s."
Why should they be "equivalent"? It suffices that they are at the same
structural level.

Most of the poems are satirical

You explain this in detail, but you don't say how it relates to the issue.

I have spent time and effort on
the decorative dividers, that form an integral part of the webpage and
contribute to its overall satirical effect on the reader, and I don't
think I have finished with them yet.

I don't see how they would contribute to a supposed satirical effect. But if
you wish to use ancient-looking decoration, why don't you put it into the
left or right margin, where it would look like a horizontal divider?

I mainly use uppercase for titles of poems...

AS I EXPLAINED, THAT'S A BAD IDEA.

Anyway using
text-transform: uppercase seems a little too complicated

There's nothing complicated in it. The complication is the desire to use
uppercase only, for a writing system that has used mixed case for a very
long time.

and is not a
very good option for Greek because it transforms lowercase accented
letters to uppercase accented letters, while normally Greek uppercase
letters should not be accented.

"Normally" as by which norm? If you think that accents should be lost in
uppercasing, you can take your case to the Unicode Consortium, which defines
the case mappings that software vendors are supposed to apply. But you would
lose that battle. Dropping accents is just bad practice. It is not part of a
cultural heritage. On the contrary, it was caused by real or assumed
technical limitations in using accents on uppercase letters. The academies
for French and Spanish languages have declared that uppercase letters shall
preserve accents, and I don't see any reason why authorities on Greek
language would not come, or would not have come, to the same conclusion.

In another webpage:

https://files.nyu.edu/emf202/public/mt/CantoGeneralEL.html

where I used font-variant: small-caps; I had to be careful to write
the letters without accents in the code.

There's no reason to omit diacritic marks from small caps.

By the way, you don't get real small caps on the web. What you get is
uppercase letters in reduced size. Real small caps are different. (For
example, the height of a real small caps letter is at least approximately
the same as the height of lowercase letters. The "fake" small-caps that
browsers produce and considerably taller.)

I've considered using just the W3C icons without the links to avoid
these problems, communicating at least to the visitor the fact that I
took time to test the webpage.

Why would that interest the visitor, even if it were true? (As I wrote, your
page did _not_ pass the "W3C CSS Validator" test, and you apparently ignored
the warning that the markup validator issued.) Besides, validation is not
testing.

I have complained to the server administrator about using https for
the _public_ folder of the account but to no avail. When they forget
to renew the certificate would be an excellent time to complain again
even more strongly.

It won't happen very soon. (Their certificate was recently renewed for 2
years.)

 

[dorayme]

....

I had added the <h6> last while wondering what I should do about it, and
knowing that I had to do something about it.

OK...

(1) An obvious alternative is:

<h1>Limericks<br>
<span>by author</span></h1>

I have used this myself ... as in

<http://members.optushome.com.au/droovies/opinion/drugLaws.html>

To make it useful, style the span, perhaps to be a smaller
percentage font-size than the main size for the h1.

<h2>Introduction</h2>
...

<h3>Disclaimer</h3>
...

The introduction h2 is one reasonable way, I guess. You could
also just introduce things under the the h1 in a paragraph (as I
have done at the above URL. An actual "Introduction" looked a bit
fussy so I just set it not to display visually).

<h2>poems' section 1</h2>

<h3>POEM 1 TITLE</h3>
...
<h3>POEM 2 TITLE</h3>
...

<h2>poems' section 2</h2>

<h3>POEM 1 TITLE</h3>
...
<h3>POEM 2 TITLE</h3>
...
<h3>POEM 3 TITLE</h3>
...


My problem with this is that the first <h2> subsection is not equivalent
to the other 2, nor is it's <h3> equivalent to the other <h3>s.

Equivalence is a tricky notion. Be careful. What it means for two
things to be the same or similar depends entirely on the pattern
being employed. And there are *no rules* to say what the correct
patterns must always be.

(2) The way to avoid this would be to assign:

<span>Introduction</span>
...

<span>Disclaimer</span>
...

omitting the <h2> and <h3>. The problem with this is the intro and the
disclaimer are not included in the webpage's structure.

Are you sure? See my cryptic remarks above about equivalence. How
firm a grip does anyone have on what is and what is not in
general part of a web page structure?

(3) But maybe a better idea would be to enclose them within

<div class="into">
...
</div>

and *define their <h2> and <h3> separately with CSS in parallel to the
<h2> and <h3> of the main part of the webpage*, which is already
enclosed within

<div class="poems">
...
</div>

I am not sure whether the idea to have 2 sets of parallel headings
within different (and unequal) sections is OK, but it seems logical enough.

What do you think of these ideas?

I think you should keep it simple and think of the h2 headings as
anything directly under the main heading and topic. What you did
first up...

I have misgivings about the disclaimer being entitled to a sub
heading under the introduction heading. Why not just have it as a
last paragraph in the introduction. Or, depending on its nature,
stick a link to it in a footer.

 

[Eustace]

Sorry for the off-topic question, but does modern Greek have accents
and, if so, how are they used?

It had inherited several diacritical signs from ancient Greek, but since
1982 it only has 1, to indicate, like in Spanish, which syllable should
be stressed. Unlike Spanish, it's used in all polysyllabic words. Also,
the accent goes _before_ an accented capital vowel at the beginning of a
word, not on top of it, and it's never used with all-uppercase words in
the titles of books or newspapers. See for example a sample front page at

http://www.theonion.com/articles/mr-special-foreign-man-wont-read-anything-not-writ,1958/

ΕΛΕΥΘΕΡΟΤΥΠΙΑ, ΠΡΩΤΟΦΑÎΗΣ, ΕΙΜΑΣΤΕ

or

http://www2.rizospastis.gr/firstpage.do?size=small&publDate=24/3/2010

emf

 

[Jukka K. Korpela]

Sorry for the off-topic question, but does modern Greek have accents
and, if so, how are they used?

Well it's not _completely_ off-topic...

Even the most modern form of Greek, after the writing reform in the 1980s,
uses stress accent, though just one: the acute accent (called tonos in
Greek), as opposite to three different accents (which had lost their
difference in pronunciation in ancient times). It also uses another
diacritic mark, the dieresis (called dialytika in Greek).

The stress accent is used in all words except unstressed monosyllabics, to
indicate the position of stress, which is essential and distinctive (i.e.
may be the only difference between words).

A common Greek keyboard has a key for adding a stress mark or a dieresis or
both on the following vowel, so typing should not be a problem. The keyboard
driver converts the combination to a single character (Unicode code point)
that consist of a letter and a diacritic mark (or two of them). This works
for uppercase letters, too. The glyphs may differ from what you expect,
since e.g. while the lowercase alpha has the stress mark above it, the
uppercase alpha has it on the left of it.

You can enter such characters on a web page as such, provided that you use a
character encoding that allows it. Even the old 8-bit encodings for Greek
(such as "Windos Greek") contain such "precomposed" characters for capital
letters, too. They lack some rare characters such as capital iota with
stress mark and dieresis, but such characters would only appear in
all-uppercase text and could, if desired, be written there using combining
diacritic marks.

In principle, you could always enter a letter with a diacritic mark as a
sequence of a base letter and a combining diacritic mark. E.g., if you used
character references in HTML for some reason, you could use Αĭ
(capital alpha followed by combining acute accent) instead of Ά
(capital alpha with stress mark). It does not pay off, though. On moder
browsers you probably get the same rendering, but old browsers may do odd
things with combining diacritics, e.g. putting them over the base letter
crudely in a fixed position independently of the base letter shape.

Besides, there are still people who write modern Greek in the old,
"polytonic" system. And if you would like to quote a pre-1980s book in
Greek, it would be appropriate to use its orthography. That's quite possible
in HTML, either using UTF-8 or using character references, though you might
want to find a suitably advanced editor that lets you type the characters
conveniently.

 

[Eustace]

I saw it and I did not see your point. You describe a logical approach,
then discard it on strange grounds:

I did not exactly discard it. I was just not sure, and was asking for a
second opinion. Eventually I accepted the option.

I don't see how they would contribute to a supposed satirical effect.
But if you wish to use ancient-looking decoration, why don't you put it
into the left or right margin, where it would look like a horizontal
divider?

I suspect that the effect of the meander divider in the context of my
webpage is clearly different for Greeks than it is for non-Greeks. There
was also a great difference when I presented orally some of the
limericks to Americans and to Greek-Americans. The Greek-Americans would
start laughing even before I had said anything funny yet.

AS I EXPLAINED, THAT'S A BAD IDEA.

I disagree on this. But it's possible that the issue is written
languages conventions. In any case, I would consider the issue, as the
former one, as in final analysis mainly an issue of personal aesthetic
judgment, and hence differences of opinion are to be expected.

There's nothing complicated in it. The complication is the desire to use
uppercase only, for a writing system that has used mixed case for a very
long time.

The complication I meant was extra keystrokes. I find it simpler to type
the particular text directly in uppercase when it has to appear in
uppercase on the webpage.

"Normally" as by which norm?

It was a mistake to write "normally". I should have written that Greek
uppercase letters in all-uppercase words are _never_ accented. Period.

If you think that accents should be lost in

uppercasing, you can take your case to the Unicode Consortium, which
defines the case mappings that software vendors are supposed to apply.

This might be a good idea. Now, how do I go about doing this? I have
also made a (small) intervention on the Wikipedia page on Greek
punctuation. It's not the first time the computer higher authorities
make mistakes with modern Greek. The Greek Stardards Organisation had
forgotten to include the greek semicolor ('ano teleia') in ELOT928 /
ISO-8859-7(!)... Similar oversights should not be infrequent with many
other minor languages.

There's no reason to omit diacritic marks from small caps.

As I said, the acute is never used with all-uppercase words in Greek.
The fact that they are smaller size does not make any difference whatsoever.

Why would that interest the visitor, even if it were true?

I consider it a matter of principle, but I am well aware that others,
especially in this newsgroup, have entirely different opinions.

(As I wrote,

your page did _not_ pass the "W3C CSS Validator" test, and you
apparently ignored the warning that the markup validator issued.)

Right, at this stage at least I ignore the warnings, and focus and the
mistakes. As long as the Validator says that I passed the test I
consider it satisfactory. The warnings are inconsequential.

Besides, validation is not testing.

I am well aware of that, and this is the reason I use validation,
because I don't care enough to do testing on different browsers and
versions. If a webpage validates and looks OK at my FF and IE I consider
it OK with me. But as I have said, I am an amateur, and I am well aware
that if I were a professional I would see things differently.

BTW, I adjusted the W3C code so that the links now work OK with my
particular (https) webpages.

emf

 

[Athel Cornish-Bowden]

[ ... ]

I guess that's because stress is mostly regular in Spanish, so you only
need to mark the anomalies. But Greek is like Russian with irregular
stress and only one stressed syllable per word. In Russian they only use
a stress accent in textbooks for foreigners though.

and books for children, I think. I have a vague idea that religious
texts have accents, as well, but I'm remembering that from many years
ago, and may be confusing it with something else.

 

[Jukka K. Korpela]

and books for children, I think. I have a vague idea that religious
texts have accents, as well, but I'm remembering that from many years
ago, and may be confusing it with something else.

Even in secular non-textbook books, Russian sometimes uses an accent to
distinguish between otherwise identical words (or forms of a word).

ObHTML: This is a challenge in web authoring, as combinations of a Cyrillic
letter and an acute accent do not exist as precomposed characters. But the
solution is simple: just inser a normal Cyrillic letter following by
combining acute, either as such (if you know how to do that and your
document's character encoding is UTF-8) or as character reference ́.
The quality of the result depends on the font and on the browser.

 

[Athel Cornish-Bowden]

On 2010-03-28 19:28:56 +0200, "Jukka K. Korpela" <[email protected]> said:

[ ... ]

The academies for French and Spanish languages have declared that
uppercase letters shall preserve accents,

Well, that is indeed what the academies declare, but it's a long way
from what most people (and many publishers) do. I'm about the only
person I know who systematically puts accents on capital letters in
French and Spanish texts, and I've often been told that it's not
necessary, and sometimes I've been told that it's wrong.

As with most French rules, the traditional rules are complicated and
fraught with exceptions, like capital É is often printed without the
accent, but À is usually printed with it. It's simpler to do what the
Academy says.

 

[Neredbojias]

It had inherited several diacritical signs from ancient Greek [...]

Just to nitpick you may be interested to know that the ancient Greeks
themselves did not actually use any diacritical marks or indeed any
word divisions. The use of diacritical marks to denote the pitch of
the words and breathings, not the stress mind you, and the use of the
miniscule text came after the ancients themselves were long gone...

I thought so but didn't want to say anything because ancient Greek is
hardly one of my areas of expertise. Still, dabbling in Latin, we (-my
cohorts and I) used to run across the occasional Greek text in often
unlikely places and muse, "Did they really know what they were talking
about?"

 

[Eustace]

It had inherited several diacritical signs from ancient Greek [...]

Just to nitpick you may be interested to know that the ancient Greeks
themselves did not actually use any diacritical marks or indeed any word
divisions. The use of diacritical marks to denote the pitch of the
words and breathings, not the stress mind you, and the use of the
miniscule text came after the ancients themselves were long gone...

Andrew

I was well aware of it, but thanks for pointing it out to the group. The
diacrticals were invented only after Greek had spread outside its
borders, to help those who were learning it as a second language. Or
maybe even later, in the Byzantine period. The net result, however, that
generations of modern Greek pupils had to learn lists of words that were
taking a daseia. I still remember a particular day in school that I
didn't remember the list well enough in the 7th grade

Today, however, some Greek conservatives contend that the intellectual
deterioration of the Greek youth began with the introduction of the
monotonic system...

emf

 

[Athel Cornish-Bowden]

[ ... ]
In another webpage:

https://files.nyu.edu/emf202/public/mt/CantoGeneralEL.html

where I used font-variant: small-caps; I had to be careful to write
the letters without accents in the code.

There's no reason to omit diacritic marks from small caps.

By the way, you don't get real small caps on the web. What you get is
uppercase letters in reduced size. Real small caps are different. (For
example, the height of a real small caps letter is at least
approximately the same as the height of lowercase letters. The "fake"
small-caps that browsers produce and considerably taller.)

Just for my interest (as I rarely have occasion to type anything in
Greek, apart from individual letters used as mathematical or chemical
symbols), does the concept of small caps exist in Greek typography?
Some of the things we (anglophone people) often take for granted may
simply be non-existent in other languages, like capitals in Arabic and
Hebrew. Even in Russian, although they have cursive letters they often
seem to prefer l e t t e r s p a c i n g for text that would
probably be in italics if the Roman alphabet were used.

Many years ago (around 1954) I used a Greek textbook that never had any
headings in capitals until one page headed ἊΥΤΟΣ, which caused a lot of
puzzlement, we weren't used to readings things in all capitals. The
teacher wasn't able to explain why this was done, except, perhaps to
underline the importance of the topic, which would probably have been
more effectively achieved by writing αὒτος in a larger size or in bold.
(I may have put the accents and breathings in the wrong places -- it
was a long time ago.)

All of this to say that I share Jukka's distaste for headings in all
caps, though in printed material small caps can look nice (not on the
web, for the reason he gives).

 

[Eustace]

[ ... ]
In another webpage:

https://files.nyu.edu/emf202/public/mt/CantoGeneralEL.html

where I used font-variant: small-caps; I had to be careful to write
the letters without accents in the code.

There's no reason to omit diacritic marks from small caps.

By the way, you don't get real small caps on the web. What you get is
uppercase letters in reduced size. Real small caps are different. (For
example, the height of a real small caps letter is at least
approximately the same as the height of lowercase letters. The "fake"
small-caps that browsers produce and considerably taller.)

Just for my interest (as I rarely have occasion to type anything in
Greek, apart from individual letters used as mathematical or chemical
symbols), does the concept of small caps exist in Greek typography?

I have no idea if the concept of small caps exist in Greek typography!
However, no Greek would be surprised if s/he would come across a title
in small capitals; it would just seem one possible way of writing a
title. Nothing unusual about it. In my Canto General page

https://files.nyu.edu/emf202/public/mt/CantoGeneralEN.html

I just followed the way the titles were written in the book with the
score of the work from which I copied the lyrics (they were in Spanish
as well as in Greek, French, and English). It seemed aesthetically quite
pleasing to me. BTW, to get an idea what Mikis Theodorakis and Pablo
Neruda are, I would highly recommend to listen to the video of "The
United Fruit Co." from my

https://files.nyu.edu/emf202/public/mt/CantoGeneralVI.html

while following the lyrics in the other webpage.

Some

of the things we (anglophone people) often take for granted may simply
be non-existent in other languages, like capitals in Arabic and Hebrew.
Even in Russian, although they have cursive letters they often seem to
prefer l e t t e r s p a c i n g for text that would probably be in
italics if the Roman alphabet were used.

I wouldn't think there is anything unusual if I would see l e t t e r
s p a c i n g in a Greek book. Actually I have used them in some
comments of some of my webpages basically for my own use.

Many years ago (around 1954) I used a Greek textbook that never had any
headings in capitals until one page headed ἊΥΤΟΣ, which caused a lot of
puzzlement, we weren't used to readings things in all capitals.

Now, that's interesting! Wondering why a word was in all capitals would
have never occur to me! So there is a significant difference in language
conventions here. However, I just opened several English (American?) and
Greek books (including of the Loeb Classical Library) that are in my
bookcase, and had no difficulty finding titles in all capitals in each
one of them. I wasn't aware that they are unusual in the UK.

The

teacher wasn't able to explain why this was done, except, perhaps to
underline the importance of the topic, which would probably have been
more effectively achieved by writing αὒτος in a larger size or in bold.
(I may have put the accents and breathings in the wrong places -- it was
a long time ago.)

Yes, capitalized words, bold, italics, letter spacing, all seem
legitimate ways of emphasizing titles to me. I just checked my Masters
Thesis at NYU out of curiosity. And yes, the title at the very beginning
was in all caps, though the subsection titles were not; and my name at
the very end is in "small caps", made (in WordPerfect 5) of letters with
2 different font sizes. It never occurred to me at the time to check the
Chicago Manual of Style on this, though I used it heavily at the time.

All of this to say that I share Jukka's distaste for headings in all
caps, though in printed material small caps can look nice (not on the
web, for the reason he gives).

I have to say that I was totally unaware that this issue existed at all!
But, based on past experience, I am not surprised to find in this
particular newsgroup people with strong convictions on such issues

emf