Book People Archive

Re: feedback to umichigan on "books and culture", part 1

john said:
>    it may be worth stepping back a bit
>    and seeing what issues are in play here.

thanks, john, but i'm not sure you "stepped back" far enough.


>    Past commenters have agreed that it is useful to keep track of 
>    the page sequencing and the page numbering of a book

"keep track of" is a relatively ambiguous expression.

both of these variables need to be _determinable_...


>    In most cases, sequencing can be determined from numbering 
>    in a straightforward way, but there are various exceptions 
>    (odd numbering schemes, unnumbered pages, restarting
>    sequences, numbering mistakes., etc.)

if you want to talk about _whether_ there are "exceptions",
we can discuss that.   but to assert it as some kind of "fact"
is incorrect.   i will repeat that there are no "exceptions" that
cannot be handled with filenames that are _wisely_ chosen,
and will prove it with any counter-cases you can bring up.


>    So, some relevant questions include:   Should both 
>    sequencing and numbering be explicitly noted somewhere?

it might surprise you to hear me say that the answer is yes,
because we haven't yet progressed the thread to the point
where making the sequencing explicit becomes a good idea.
that's due to the "discussion", not with my treatment of it...


>    (I haven't seen anyone say outright that explicit sequencing 
>    *anywhere* is inherently a bad idea, but it has been argued
>    that it's relatively unimportant

it might also surprise you to hear me say that sequencing
_should_ be done explicitly, and not just be "determinable";
both systems should give the same result, for redundancy.
but again, nobody moved the discussion to that point yet.


>    and may confuse readers if they have to deal with page 
>    *sequence* numbers instead of or alongside "regular" page numbers)

two sets of numbers _will_ confuse readers, some of them,
and most of them at one time or another.   it's a _certainty_,
from murphy's law, so the word "may" is really inaccurate...


>    Which of those two things, if any, should be noted in page filenames?
>    Possible answers include:
>    a) Numbering only (Bowerbird)
>    b) Sequencing only (akin to what happens by default in many PDFs)
>    c) Neither; they should just be in the metadata (UMich)
>    d) Both (Jon Noring)
>    e) It doesn't matter much as long as the files are well managed

if you really want to get deeper into this discussion, you will need to
specify exactly whether you're talking about filenames and/or url's,
and -- on a broader level -- what you make the end-user deal with...


>    The position you take will probably be more convincing to readers 
>    if you can show a real-world example of where your recommended 
>    choice has helped you or your users, or where a different choice 
>    has led to problems.

i've pointed out any number of examples where "a different choice"
leads to problems.   every pagenumber i've given from the mabie book
has forced everyone who wanted to look at it to do the "add 6" exercise,
a totally unnecessary and confusing hassle.   and that's a case where
_at_least_ the offset is consistent throughout the book, so a person
can compute it without actually having to go verify each composed url,
which is what one is required to do when the offset in a book changes.
(and since a person usually will not know whether the offset changes
in a specific book, they have to verify each url just as a matter of course.
this borders on ridiculous.)

and thus it should be obvious that avoiding this ludicrous situation
is indeed something that will "help your users", wouldn't you agree?

-bowerbird