Re: Long file names in Dired

[Marcin Borkowski]

Hi, Emanuel!

(Disclaimer: I tried not to sound rude in my email.  If I failed, please
take into consideration that this was not my intent!  I just strongly
disagree with most of what you wrote.)


On 2015-04-25, at 13:20, Emanuel Berg <embe8573@student.uu.se> wrote:

> Marcin Borkowski <mbork@mbork.pl> writes:
>
>> Though I don't agree with you about the publishing
>> industry: authors usually suck at typography, and
>> they usually don't care anyway.
>
> It is just something they never did. I child can learn
> it. I don't know what the publishing houses do all
> day. Typesetting in LaTeX (or even old-school groff

Wow, this seems to cross the boundary of trolling. ;-)

Have you read Bringhurst ("The Elements of Typographic Style")?  Or
Tschichold ("Die neue Typographie")?  (Not that I agree with him, btw.)
Or Hochuli ("Das Detail in der Typografie")?

If by "typography" you mean "putting all the elements where the designer
wants them to be" - then I agree (to some extent - try using LaTeX to
typeset anything containing equations /and/ figures on the grid, or to
wrap a figure with text when the text includes an enumeration - good
luck!) that it's /relatively/ simple (at least most of the time).

If by "typography" you mean "the art and technique of arranging type to
make written language readable and appealing" (this is what Wikipedia
has), then you most probably need /at least/ half a dozen years to get
reasonable experience, and twice to three times that to achive what
could be called mastery.

If you're wondering what the publishing houses do all day, I can tell
you what.  I work for a small journal (two issues per year, each issue
is typically 150-200 pages long), and it's /a lot/ of work.  I work
there with a friend, and our duties are (mainly) to, let's call it,
"make the issue look good".  We receive the files prepared by the
authors, and after several weeks of hard work we produce a pdf file
ready to send to the printing house.  And by hard work I mean hard work
- probably the main reason it's just "hard" and not "impossible" is
several thousand hours of experience of both of us.  (Another reason is
that I have a few hundred lines of Elisp code, automating these parts of
our job that /can/ be automated - it's not even a majority, but it saves
us considerable time of boring, repetitive tasks.)

> which is what I think O'Reilly uses) isn't easy first
> day but after a while you understand how it works.

Well, I don't know groff, but the subtle interactions of the
line-breaking and page-breaking algorithms in TeX, for instance, or
positioning of floats, are not something you would understand after
a "while".  (I have about 20 years of experience with TeX and LaTeX,
both high and low level, and now and again I'm totally astonished and
can't fathom what happened to some page or footnote or figure.)

> Then, after you get the details the way you want it is
> is only a matter of using 99% of the same over and
> over. So it should have been standardized one zillion
> times over by now.

Yeah, it should have been.  Now how do you explain how it's possible
that such a "trivial" issue - putting a bunch of smaller rectangular
things on one big rectangular thing - cannot be solved satisfactorily,
even though it's 2015?  (I'm looking at you, CSS!)

>> The problem is that if LaTeX happens to hyphenate
>> the author's name, or the publisher's name, or the
>> title, or the university name, etc. (which is quite
>> possible), it should do it according to the rules
>> for the language of that particular field.
>
> Is that why you hate BibLaTeX? ... :J

One of the reasons.  I explained the other ones earlier.

> Did you consider disabling hyphenation altogether?

Doesn't make sense - it will lead to disastrous results.  What's more,
it /should not be needed/ - this is a trivial programming problem which
should be /solved/ by standard tools.  (Well, hyphenating itself is not
a /trivial/ problem, but it's more or less /solved/ - though it might be
interesting for some of you to learn that TeX's hyphenating algorithm,
and the process of preparing the patterns for that algorithm, were the
basis in a PhD in CS.)

>> (Another thing - as I mentioned - is that that whole
>> first/von/last/jr stuff is quite English-centered.
>> In non-Germanic languages this is sometimes
>> completely irrelevant - take Russian with its
>> "patronimicum", which is similar to the middle name,
>> but is someting else (it's basically things like
>> "John, son of Jack, Smith).
>
> Yes. In Sweden we instead have that "incorporated" in
> the last name, often: Svensson, Bengtsson, ...
> (literally Sven's son, Bengt's son, etc., but they are
> just family names: the father isn't for this reason
> Sven and Bengt as with the Russian system).

I see.  We have some similar names in Polish, too, but they are also
just family names nowadays.

>> In Polish, we don't have "von"
>
> Why - did the communists wipe out the entire Polish
> nobility? Alright. This is interesting, but what does
> it have to do with BibLaTeX?

Well, not exactly - it's just that the nobility doesn't count anymore.
(And yes, the communists helped that process a lot, as well as Germans
during WW2.)

What does it have to do with BibLaTeX?  Exactly what I wrote about:
different countries (or cultures) have different naming schemes, and
BibLaTeX supports only a small subset of the set of Germanic naming
schemes.  No Slavic ones (granted, the Polish one is trivial, especially
now that it's simplified: even though we don't have any von-like thing,
we did have declension of names depending on sex, and for female names
also on the marital status: the wife of Mr. Nowak used to be called
Nowakowa, and his daughter - Nowakówna.  These forms, however, are very
seldom used now.  AFAIK, in Czech you still have something like this),
no Asian ones, no Icelandic ones.  (I don't know about other languages.)

>> and using "jr" is very rare, though possible, but
>> seems (at least to me) extremely pretentious.
>
> It can also be functional, to separate the two persons
> with the same name. For example, the famous American
> and Mexican boxers - fathers and sons:
>
>     Floyd Mayweather Sr./Jr.
>     Julio César Chávez Sr./Jr.

Of course, but here in Poland we seldom give the son the same name as
his father's.  (And I would argue that doing that might be seen as
pretentious, too.)  Apart from the so-called "celebrities", I know
exactly one (rather eccentric, btw) person who actually did it, and even
though he's Polish, he lives in the US, where it's apparently more
common.

>> Or take Icelandic, where the name structure is
>> totally different than in other languages - many
>> Icelanders /don't even have/ the family name (and
>> they sort their names by the first name!).
>
> What does it matter what exceptional and goofy
> convention some people have somewhere around the
> world? The do have *names* on Iceland. A BibLaTeX
> entry has an author field. Just type the name of the
> the industrious Icelander there!

Not exactly.  A BibLaTeX entry has an author field, which should contain
the last, prefix, first and postfix parts, some separated by commas and
some by spaces, in a strictly defined (and a bit counterintuitive)
order.  Assume that you're an Icelander writing an article in Icelandic,
with a bibliography of works by Icelandic authors.  Assume that you want
them sorted by the first name (apparently, this is what Icelanders do).
How do you make BibLaTeX do it?

>> And there are quite a few people who /do/ care about
>> the references. For instance, I work for a journal,
>> where we (with a friend of mine) are responsible for
>> (among others) typesetting the papers. We sometimes
>> spend/waste quite a lot of time on bibliographies
>> (mainly because authors "don't care" - if they
>> actually used BibTeX and not hand-crafted,
>> inconsistent formatting, things would be a lot
>> easier for us...). I even wrote an Emacs utility
>> which helps transform such inconsistent pile of s##t
>> into proper markup.
>
> Of course references should be correct and consistent
> and software (including Emacs modes) should support
> it. I think BibLaTeX does its job and the problem (?)
> with lack of support for "national hyphenation" isn't
> anything I would ever worry about.

Well, as I mentioned earlier, I do worry about it, working for a journal
which is typeset in Polish, but the articles often quote other works,
mainly in English, but sometimes in German, French or Russian.  Happily,
we don't use BibLaTeX, but amsrefs, which happens to support
multi-language bibliograhic entries.  In this department BibLaTeX does
/not/ do its job, period.  And hyphenation is just the symptom, not the
disease; the disease is that some language-dependent conventions are
hard-wired into the software.  (I guess that this exchange helped /me/
understand the reason for my strong negative feelings for BibLaTeX: as
a (professional) mathematician and a (hobby) programmer, I can't stand
when a theorem/program is not general enough when it should be.  Thank
you, Emanuel!!!)

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University