Pixel to Em conversion...

[Jukka K. Korpela]

Scripsit Bernhard Sturm:

- -
Jukka: this is true in it's application, but Ben was talking about
it's 'technical' origin.

No, as you can see from your own quotation, Ben claimed that the em _is_
"technically" a measure of width. That's plain wrong. A common
misconception, not a mortal sin, except perhaps if you actually base your
web site design on. Whatever the meaning of "technically" was, the statement
was false.

Read a bit about typography (I am a typographer myself)

Would you believe that I have read a bit about typography and written about
it, too?

The 'em' unit goes back to the Romans using the width of
the letter 'M' to refer to the size of their letters...

That's what I explained in another message in this thread, emphasizing that
the connection was broken in ancient times. To be exact, the inscription
design used the em as the width of _some_ characters, including "M", but
surely excluding e.g. "I".

And it's just a little story, unless people misunderstand it as having some
impact on the em unit in CSS.

Don't only look at how the 'em' unit is implemented in the CSS-specs,
it's sometimes good to know the historical origin of things.

In this case, a little knowledge is just harmful. Historically, the em unit
was never a unit of width alone - only in conjunction with the height.
What's more important, knowing the history, or some distorted version
thereof, seems to make people think that the em unit _is_ the width of "M".

It's not a matter of "implementing" the em unit in CSS specifications. The
CSS unit is simply a reflection of a typographic tradition, in which the em
unit means the size of the font, not the width of anything.

 

[Ben C]

Scripsit Bernhard Sturm:


No, as you can see from your own quotation, Ben claimed that the em _is_
"technically" a measure of width. That's plain wrong.

I accept that. What did it for me was the idea of an "M square".
Although I suppose that makes saying it's a measurement of width moot
rather than wrong. But since I was correcting someone for saying it was
a measurement of height I can't really pretend that's what I meant.

Do we know whether Roman M-squares were actually equilateral?

A common misconception, not a mortal sin,

Thank goodness for that.

Various websites do say it's a measure of width, e.g.
http://www.fontfactory.com/glossary.php. So I reserve some judgment on
the meaning in its wider context. But Mr Korpela makes a good point
about what it means in CSS.

I find this in CSS 2:

http://www.w3.org/TR/REC-CSS2/fonts.html#emsq

15.4.3

Certain values, such as width metrics, are expressed in units that
are relative to an abstract square whose height is the intended
distance between lines of type in the same type size. This square is
called the em square and it is the design grid on which the glyph
outlines are defined.

I missed that because I usually only work with CSS 2.1, from which all
that has been removed and replaced with just a vague reference to "a
concept used in typography".

 

[Jukka K. Korpela]

Scripsit Ben C:

Various websites do say [that the em unit is] a measure of width, e.g.
http://www.fontfactory.com/glossary.php.

There's a lot of bogus information around, and such statements are useful
bogosity indicators. On that particular page, you can see other bogosity
indicators as well, such as "Extended ASCII", "90º" (using masculine ordinal
indicator instead of degree sign - and they claim to be "Type
Professionals"!), "Circero" (for "Cicero"), etc. - I stopped scanning when I
saw a claim that the circumflex is used in Italian.

 

[Ben C]

Scripsit Ben C:

Various websites do say [that the em unit is] a measure of width, e.g.
http://www.fontfactory.com/glossary.php.

There's a lot of bogus information around, and such statements are useful
bogosity indicators. On that particular page, you can see other bogosity
indicators as well, such as "Extended ASCII", "90º" (using masculine ordinal
indicator instead of degree sign - and they claim to be "Type
Professionals"!), "Circero" (for "Cicero"), etc.

Those are quite inexcuscable typos.

- I stopped scanning when I saw a claim that the circumflex is used in
Italian.

According to some more bogus websites it is.

"In Italian it is used in plurals of singulars ending with -io, thus
ending them with a longer i. In modern Italian this is accomplished with
a double or just a single i as in varî, varj, varii, vari ("various",
plural of vario)."

 

[Jukka K. Korpela]

Scripsit Ben C:

Those are quite inexcuscable typos.

No they aren’t. â€Extended ASCII†is a factual error. They got the facts
wrong, instead of just pressing the wrong key by accident. Using wrong
character instead of the degree sign is a character-level error – not an
accidental typing mistake. Any typographer should know better. Mistyping â€Ciceroâ€
can be classified as an accidental typo, and it is indeed inexcusable (i.e.,
not excusable) in a purported reference material when it appears _twice_.

Please read a reputable book on typography if you seriously doubt that I am
not right here. Surfing around bogus web pages and citing them takes you
nowhere.

According to some more bogus websites it is.

So why do you keep citing bogosities? Besides, what you quote from an
unspecified source says that the circumflex is _not_ used in Italian; it
claims that it _was_. There’s a difference, especially if one pretends to
present _reference_ information on typography.

 

[Bernhard Sturm]

According to some more bogus websites it is.

which is true... the use is not common, but I wouldn't call the
reference to it 'bogus'. I know that Mr. Korpela is going to argue on
this as well
Maybe this url will help him a bit to identify the bogus sites on the web:

http://en.wikipedia.org/wiki/Circumflex
http://www.scudit.net/mdaccento5.htm

HTH
bernhard

 

[Bernhard Sturm]

So why do you keep citing bogosities? Besides, what you quote from an
unspecified source says that the circumflex is _not_ used in Italian; it
claims that it _was_. There’s a difference, especially if one pretends
to present _reference_ information on typography.

please, don't be silly: if a modern italian website is going to quote
the title of an (not so old) italian book (e.g. Pisacane, Carlo
[1818-1857] "Scritti varî, inediti o rari"), would you then also claim,
that the circumflex is of no typographical use in the Italian language?
The italian source of Wikipedia indicates, that the circonflesso is
still in use in modern Italian language according to the rule Ben quoted:

salario(sing) -> salarî (plural)
simposio(sing) -> simposî (plural)

http://it.wikipedia.org/wiki/Accento_circonflesso

cheers
bernhard

 

[Bernhard Sturm]

Scripsit Ben C:

Various websites do say [that the em unit is] a measure of width, e.g.
http://www.fontfactory.com/glossary.php.

There's a lot of bogus information around, and such statements are
useful bogosity indicators. On that particular page, you can see other
bogosity indicators as well, such as "Extended ASCII",

As I recall there is in fact an 'extended ASCII' existing. It's more
precicely described as 'high ASCII' because 'extended' could imply that
the ASCII was 'extended' from 7-bit to 8-bit at a later stage (which is
not the case).
however I used to work on old IBM machines, and on IBM PC DOS you can
use the ECS (Extended Character Set) which uses 8-bit values to refer to
the higher part of the ASCII-chart in order to map also European
characters. Again: the information about the 'extended ASCII' is not
that 'bogus' as you are calling it (otherwise you should blame IBM for
labelling it ECS, but then we would also have to blame Microsoft for
labelling their OS as 'Windows' which has nothing to do with a real
window .

bernhard

 

[Jukka K. Korpela]

Scripsit Bernhard Sturm:

As I recall there is in fact an 'extended ASCII' existing.

That's because you have read bogus information. That was understandable - as
was the wrong idea about the em unit - until correct information was given.

Again: the information about the 'extended ASCII' is not
that 'bogus' as you are calling it

It definitely is. Common people are excused for their ignorance on this
issue, and on the em unit issue, but people who write references are not.
And ignorance stops being excusable at the point where you start and keep
arguing, without any factual grounds, in public against correct information.

The ASCII code is defined by an American National Standard. There is only
one ASCII standard in force. I have actually acquired and read it. Have you?

What people call "extended ASCII" is just _different_ 8-bit codes, and none
of them carries "ASCII" in their name. This is where the wrong information
becomes harmful and not just ignorance: people will act as if their
references to "extended ASCII" referred to one specific code.

(Similarly, thinking that the em unit is the width of "M" - without ever
bothering to check how wide "M" is in em units - is just a piece of common
ignorance _until_ you start setting things up in a manner that is based on
the misconception. For example, it would be logical to set width: 30em for a
text input field if you have set its font to monospace and want it to be 30
characters wide. Quite logical, since all characters have equal width there.
Just the premise about em being the width of "M" is completely wrong.)

 

[Bernhard Sturm]

Scripsit Bernhard Sturm:


That's because you have read bogus information. That was understandable
- as was the wrong idea about the em unit - until correct information
was given.

do I understand you correctly, that you are telling us that mentioning
the term 'extended ASCII' is just plain wrong, and it doesn't exist at all?
If that is the case then could you please ask for the deletion of this
Wikipedia entry:

http://en.wikipedia.org/wiki/Extended_ASCII

I have read it, and I understood it that way, that the 7-bit ASCII forms
the base of all those various extensions. Maybe all the information in
this Wikipedia entry is wrong, as there is no such thing as
'ASCII-extensions'. It should be called 'based-on-ASCII-extensions'?
Enlight me, and it would be a great thing if you really could ask for
the corrections or deletion (if necessary) in the mentioned Wikipedia
entry as many people do rely on such information (including me).

Cheers
Bernhard

 

[Dan]

do I understand you correctly, that you are telling us that mentioning
the term 'extended ASCII' is just plain wrong, and it doesn't exist at all?
If that is the case then could you please ask for the deletion of this
Wikipedia entry:

http://en.wikipedia.org/wiki/Extended_ASCII

There, Wikipedia is being descriptivist rather than prescriptivist,
and noting terminology that can be found in use, without necessarily
implying that it's actually "correct" by any logical standard. It
goes on to note that the usage is criticized, and that it's been used
in a fairly sloppy manner to refer to a whole range of different sets
of characters that go beyond those defined in ASCII, and is thus not a
specific character set / encoding, nor is it endorsed in any way in
conjunction with the ASCII standard.

 

[Jukka K. Korpela]

Scripsit Bernhard Sturm:

do I understand you correctly, that you are telling us that mentioning
the term 'extended ASCII' is just plain wrong, and it doesn't exist
at all?

Of course.

If that is the case then could you please ask for the
deletion of this Wikipedia entry:

Why? I have no interest in raising the level of toilet wall engravings just
to see some other people paint their crap over them. Anyone who wants to
know the facts can consult reliable sources.

 

[Maxx Pollare]

Reality folded in on itself, and somewhere the following words from
myself appeared in history:

"What the hell is an Em, and how big is it in pixels?".

Thanks all...
Got busy ocver the last few days, but I've enjoyed the help.

I'll post back with a weblink of my work later, unless anyone can
sugest a better news group for /that sort of thing/.

 

[John Hosking]

I'll post back with a weblink of my work later, unless anyone can
sugest a better news group for /that sort of thing/.

Depends on why you're posting it. If you want a critique, post in
alt.html.critique. If you need help with the markup, alt.html or
comp.infosystems.www.authoring.html are good. For CSS discussions try
comp.infosystems.www.authoring.stylesheets.

 

[Bernhard Sturm]

Scripsit Bernhard Sturm:


Why? I have no interest in raising the level of toilet wall engravings
just to see some other people paint their crap over them. Anyone who
wants to know the facts can consult reliable sources.

Jukka, I think this is quite an arrogant statement you made.
I always admired your knowledge and profound know-how, but sometimes you
just seem not to be able to accept other facts or point-of-views. If the
above wikipedia source is just 'toilet wall engraving' and not a
'reliable source' to you, then fair enough, but I think it's just not
good enough, if you keep on going criticizing others of not providing
'good' sources to support their statements, but on the other hand you
fail to provide those 'reliable sources' in order to proof your points.
In most cases you just come up with 'I have done that, and I do know it
is that way, do you?'.
Forgive me, but IMHO this is patronising others for their strive of a
(halfway) fair discussion. Although this is not necessary - I believe it
is just good netiquette - but I have never seen you apologising for a
statement you (falsly) made.

bernhard

 

[Bernhard Sturm]

It
goes on to note that the usage is criticized, and that it's been used
in a fairly sloppy manner to refer to a whole range of different sets
of characters that go beyond those defined in ASCII, and is thus not a
specific character set / encoding, nor is it endorsed in any way in
conjunction with the ASCII standard.

that is all true, but as you correctly mentioned the term is in use and
there are various references to it (even from IBM in their manuals for
Lotus Notes, printers or PC DOS
(http://www.redbooks.ibm.com/redbooks/pdfs/gg244459.pdf for example)).
The term is in use, and can be used even if it's not precise (it's
similar to the use of the word 'Homepage' in German in order to refer to
an entire Website; not accurate, but it's in use . Therefore I
wouldn't blame the authors of the website we are discussing for
providing unreliable facts...

Cheers
Bernhard

 

[Bernhard Sturm]

Scripsit Bernhard Sturm:


Of course.


Why? I have no interest in raising the level of toilet wall engravings
just to see some other people paint their crap over them. Anyone who
wants to know the facts can consult reliable sources.

such as this RFC from 1975 about the use of "Extended ASCII" in Telnet
sessions?
http://www.ietf.org/rfc/rfc698.txt

 

[Jukka K. Korpela]

Scripsit Bernhard Sturm:

such as this RFC from 1975 about the use of "Extended ASCII" in Telnet
sessions?
http://www.ietf.org/rfc/rfc698.txt

The only thing that RFC 698, a "proposed standard" from 1975, is good for is
checking how a particular protocol was defined. It is surely not, and was
never meant to be, any authority or reliable source of information on any
other matter.

I wonder if you realize how ridiculous it is to cite over 30 years old
technical descriptions that use sloppy terminology. Besides, the "Extended
ASCII" described in that document has nothing to do with the "Extended
ASCII" in other documents cited so far. The more you cite mutually
incompatible uses of "Extended ASCII", the more you give arguments in favor
of my point - but you can stop now, since the point is crystal clear to
anyone who has actually checked some reliable sources.

 

[Jukka K. Korpela]

Scripsit Bernhard Sturm:

Jukka, I think this is quite an arrogant statement you made.

Don't be ridiculous. I don't clean up the shit that other people put around
the Web, and there's nothing arrogant in saying that when the shit has been
cited in public discussion and I was asked to clean it up.

I always admired your knowledge and profound know-how, but sometimes
you just seem not to be able to accept other facts or point-of-views.

There's no "other facts" in this issue, or in the em issue. You can have any
point of view you like, but it does not change the facts.

- - I think it's just
not good enough, if you keep on going criticizing others of not
providing 'good' sources to support their statements, but on the
other hand you fail to provide those 'reliable sources' in order to
proof your points.

If you want to argue about the meaning of the em unit, or the ASCII code,
then you are supposed to check your facts from reliable sources, on your
own. I've already told you to find a book on typography for the first issue
and the ASCII standard for the latter.

In most cases you just come up with 'I have done
that, and I do know it is that way, do you?'.

In matters where it would be pointless to cite references, since the other
side hasn't apparently checked any reliable references. If someone is
willing to argue in public without checking any facts (and reading random
web pages does not count as checking facts), why would he actually check
anything even if I gave an ISBN or ANS number?

Forgive me, but IMHO this is patronising others for their strive of a
(halfway) fair discussion.

This "fair discussion" (about the em unit, and then ASCII) is comparable to
discussing whether 2+2 equals 4 or 7, with some people citing novels that
say 2 + 2 = 7. (Yes, there is an enjoyable novel saying that.)

Although this is not necessary - I believe
it is just good netiquette - but I have never seen you apologising
for a statement you (falsly) made.

Your sentence does not really parse. And I don't really care what you are
trying to say with it. You are just babbling pointlessly, since the factual
issues - em and ASCII (which I raised for comparison, as an example of
matters that people have misconceptions about, and you surely proved that) -
are easily resolved as soon as you check facts, and you are just attacking
me for saying this. If there's anyone who should apologize, it's you.

 

[Ben C]

There's no "other facts" in this issue, or in the em issue. You can have any
point of view you like, but it does not change the facts. [...]
This "fair discussion" (about the em unit, and then ASCII) is comparable to
discussing whether 2+2 equals 4 or 7

Hardly. We're talking about the meaning of two semi-technical terms:
"em" and "extended ASCII". Words mean what people use them to mean and
if documents are found on the web including the Wikipedia that use terms
in a particular way then that is factual evidence about how they're
being used.

Now you say they're used differently in "reliable sources" and "books"
and you are a fairly reliable source yourself. But Bernhard Sturm, who
is a typographer and also a reliable source, has suggested that the "em"
is a unit of width after all. So there are facts on both sides. Enough
to conclude that there is some legitimate variation in the use of these
terms.