- "hypertext" describes a set of UI conventions. linking to things is not new
- rendering static documents is distinct from interactive dynamic ones
- forms are static documents
- rendering documents is not remotely related to their mode of transmission.
Post
the lines and arrows—they link the appropriate text range to the comment, so it can be read as intended, or expanded into thought bubbles. the arrow in fact serves as a decoupling. i'm confident that joyce's arrow was a metatextual marker
None of Samuel Johnson's more ornate hand as he wrote and scratched out lines for the preface to his celebrated Dictionary or improved the style of other authors' prose to illustrate proper usage.
linguistic prescriptivist? writing a dictionary? this is my surprised face
Now what kind of museum display would a disk make? ''On this disk are the last three revisions of Moby Dick.'' Would anyone believe it? Would there have to be a computer continually at work so that visitors could see the disk's contents on the screen?
the author of this piece rocks!!! now he seems to get closer: the computer must be at work, because it must display multiple distinct views (in this case, mutually exclusive states of some region). the manuscript of joyce uses the arrow construction to toggle comment visibility (using the brain and eye).
but what is this museum paradox? the museum, i now see, was a very clever analogy: by necessity, the static artifact serves as synechdoche for historical process
the case of moby-dick in the museum (cleverly) conflates two artifacts: that of moby-dick, and that of the computer and its software. these versions of moby-dick—they have names, and audiences. that is to say: the versions all together are coherent only in the context of a stateful editing process—i.e., in the mind of a specific user!
the user may be herman melville, or perhaps his editor. in either case (i give the author grace), a named version might be synthesized from selecting the named alternative of a span (or default, or choose).
WE JUST MADE KCONFIG FOR MOBY-DICK!
now he begins to romanticize what amounts to posthumous privacy invasion:
What if Emily Dickinson had had a word processor, leaving all her unpublished poems in it or sending an occasional printout to a friend?
computer surveillance excites him!
Would there have been any equivalent of those penciled underlinings that have been seized as clues to her choice among variations of the same poem?
you don't know her! does the dash speak like a tally to you? i see it as, sometimes, a bridge, a hope—perhaps the underlines were beats, cycling through variations, polyrhythms
What would take the place of that torn flap of an envelope on which she wrote, ''Was never Frigate like''
the torn flap is an act of intimacy, of ecstasy, of secrecy—the envelope, torn, has just experienced rebirth (or took an unexpected path). could it be like a dog ear, a leaf?
- a trial beginning for what became the famed ''There is no Frigate like a Book''?
the mail moves! it takes flight! in a book perhaps she might set sail between ports of chapter?
Would her computer ever tell us how in ''I'm Nobody! Who are You?'' she toyed with the familiar phrase ''the livelong day'' as well as the striking ''livelong June''?
but she didn't mean to share that with you!
perhaps she worried terribly about the implication of "familiar" day? perhaps she was entranced by the prospect of a "livelong june day", and its rhythmic time? "june" and "you" are so close—a portmanteau?
How would we identify her three typical stages of composition: worksheet draft, semifinal draft, fair copy?
she obviously never wrote about it, yet you speak of "typical" and impose classifications. would emily call her final draft "fair"? did she ever guess the "copy" might be copied?
the straitjacket of the word processor and its error messages seems little different to the gall of describing her "typical system" in terminology that implies she ever consented to the reproduction of her "final copy". both of these are straitjackets designed by men to contain her
Maybe exhibition-goers will be interested in other things anyway, by the time computers become literary memorabilia.
the author tears himself away from his extended voyeurist excursion, skillfully covering his tracks as a consideration of the tastes of exhibition-goers.
i will allow him the grace of assuming his "typical system" line was intentionally brusque and reductive. perhaps it was a warning! perhaps he expounds at length because it delights him that she remains mysterious!
Perhaps they will concoct theories on why an author plunged into a computer known for stern discipline instead of ''user-friendliness.''
oh i know this one!!!! https://www.chronicle.com/article/how-iowa-flattened-literature/
i have concocted many a theory around bazel at twitter, but the one that fits best is simply that pants is user-friendly, while bazel exerts hierarchical control. and the smell of musk was in the water
Whether an author tried the more difficult word-processing programs, with many rote commands to remember, or stayed with the ones designed with easy mnemonic aids.
actually literally vim vs emacs. in 1983 he could see the flame wars building.....
Whether an author caught on quickly and sped ahead, a counterpart to the touch typists of yesteryear, or piled up the ''error messages'' caused by ignorant keyboard commands.
ok we've had our ups and downs but i now understand his humor and think the museum analogy was so good. he satirically invoked a teacher arrogantly seeking to typicize emily dickinson, while in fact seeing her as eternal. or something.
and of course he would see error messages as an insult! they are an insult! particularly the ones that "pile up"......c++????
one of my students showed me clang's error: too many errors error message and:
(1) that's not a fucking compile error! that's obviously an info log
(2) this isn't explained in the help message. it was completely fucking unclear whether clang was experiencing an ICE or something. nope! it was a "helpful" change to the default behavior.
(3) it's a mean thing to say to a student! it reads like "you're hopeless, i'm not even grading this"
why would you ever limit errors by default?
(a) improved the oops! all errors! benchmark by 50% [demonstrating why benchmarks are fascist]
(b) now you have to add a new flag to get the same behavior again, which breaks gcc compat
(c) if you want to restrict access to build error logs to people with access to modify the clang cli [google internal corporate politics]
@hipsterelectron wellll yeah but if you assume that the source input is usually more signal than noise, then as the number of errors goes up so do the odds that the compiler is just fully off in the weeds. At some point you have to assume the compiler is the problem, not the program.