Theory: 'action hypertext'
Jorn Barger
and Willem de Kooning: [1]
"At a certain moment the canvas began to appear to
one American painter after another as an arena in which to
act-- rather than as a space in which to reproduce, redesign,
analyze, or 'express' an object, actual or imagined. What was
to go on the canvas, was not a picture but an event."
I'd like to propose an analogous theory called 'action hypertext' in which
the webpage captures, not a static image of html-tags being typed in, but
a dynamic, evolving image of the web-researcher ***sifting thru search-
results*** on a particular topic and gradually identifying and linking the
best resources on each subtopic.
Authoring tools, then, need to evolve to support this:
- each time you do a search, the results are cached on your own machine, in
the form of a to-do list that you continually prune and annotate
- this pruned-and-annotated list can be published immediately as a webpage,
and becomes the basis of an ongoing exploration of the topic [2]
- links might be categorised by 'action-stage': looks-promising-but-not-yet-
skimmed, skimmed-and-tentatively-approved, skimmed-but-some-doubts, not-
ideal-but-the-best-so-far, read-fully-and-approved, etc
- the authoring tool records-- on the page itself-- the date the page was
begun, and the last date it was modified
- hundreds of separate pages on different topics can be juggled, and any
time you find a new resource you can easily insert it into the appropriate
page
- the authoring-tool recognises that others' pages can vanish, and monitors
for this. It could even remember the original search-- and cache the page
itself!-- so that it can propose likely substitutes.
[1] http://google.com/search?q=reproduce.redesign.analyze+1952
[2] see also: http://www.robotwisdom.com/web/academia.html
Jerry Muelver
Barger) wrote:
>In 1952, Harold Rosenberg explained the 'action painting' of Jackson Pollock
>and Willem de Kooning: [1]
>
> "At a certain moment the canvas began to appear to
> one American painter after another as an arena in which to
> act-- rather than as a space in which to reproduce, redesign,
> analyze, or 'express' an object, actual or imagined. What was
> to go on the canvas, was not a picture but an event."
>
>I'd like to propose an analogous theory called 'action hypertext' in which
>the webpage captures, not a static image of html-tags being typed in, but
>a dynamic, evolving image of the web-researcher ***sifting thru search-
>results*** on a particular topic and gradually identifying and linking the
>best resources on each subtopic.
>
To what useful end, Jorn? It looks like you are proposing an
automated meta-blog. Stream-of-conscious blogging may be
entertaining to Joyce fanatics, but it only gets in the way
of someone actually trying to retrieve information. The raw
logging of a web-researcher's fumbling efforts to access
information is only anti-information. It would be like
trying to develop a biography by installing a traffic-cam on
a corner sometimes traversed by the subject.
Note that it is Pollock's paintings ("events" if you will),
not home movies of him painting, that comprise the art.
Would you rather read a novel finished to the author's
satisfaction, or slog through a log of the keystrokes used
to do the writing?
---- jerry
--
Nick Kew
at the keyboard of jo...@enteract.com (Jorn Barger) wrote:
> In 1952, Harold Rosenberg explained the 'action painting' of Jackson Pollock
> and Willem de Kooning: [1]
>
> "At a certain moment the canvas began to appear to
> one American painter after another as an arena in which to
> act-- rather than as a space in which to reproduce, redesign,
> analyze, or 'express' an object, actual or imagined. What was
> to go on the canvas, was not a picture but an event."
What a terribly, terribly provincial and blinkered viewpoint[1].
The same could clearly be said of a great deal of art from around the
world, going back for centuries. Take a look for example at Giotto's
highly interconnected (or perhaps, in the language of hypertext, linked)
frescoes in the Basilica di San Francesco at Assisi.
> I'd like to propose an analogous theory called 'action hypertext' in which
> the webpage captures, not a static image of html-tags being typed in, but
> a dynamic, evolving image of the web-researcher ***sifting thru search-
> results*** on a particular topic and gradually identifying and linking the
> best resources on each subtopic.
Since you've posted this to some WWW groups, you're presumably thinking
in terms of RDF here? Or if not, perhaps you'd like to reference your
argument to current RDF-based work in progress?
> Authoring tools, then, need to evolve to support this:
Now that depends very much on consensus-building, and once again brings
us back to RDF - which is gradually being supported by tools.
> - each time you do a search, the results are cached on your own machine, in
> the form of a to-do list that you continually prune and annotate
>
> - this pruned-and-annotated list can be published immediately as a webpage,
> and becomes the basis of an ongoing exploration of the topic [2]
>
> - links might be categorised by 'action-stage': looks-promising-but-not-yet-
> skimmed, skimmed-and-tentatively-approved, skimmed-but-some-doubts, not-
> ideal-but-the-best-so-far, read-fully-and-approved, etc
>
> - the authoring tool records-- on the page itself-- the date the page was
> begun, and the last date it was modified
>
> - hundreds of separate pages on different topics can be juggled, and any
> time you find a new resource you can easily insert it into the appropriate
> page
>
> - the authoring-tool recognises that others' pages can vanish, and monitors
> for this. It could even remember the original search-- and cache the page
> itself!-- so that it can propose likely substitutes.
I see a lot of the ideas that were floating around in the mid-90s, as the
web gained popularity but those with any kind of vision also sought to
look beyond its limitations (particularly the crippled HTML linking model,
and more recently XML's more complex but still limiting one). Indeed,
I seem to recollect you as someone with an idea or two.
I also see ideas that *are* being explored by W3C - and thus with the
potential for industry support - today.
I was there not only expressing ideas, but implementing them (a working
Holistic Hypertext[2] implementation was demoed in the original WebThing
in 1995). The lesson I learned back then is that it's not sufficient
to do (let alone just to think) interesting things, you need to work
in a context where you will be seen. In the absence of a spare billion
to spend on marketing, W3C may be the best available forum today.
[1] Of course, taken in context, the original author may well
have been saying something perfectly reasonable.
[2] Better known to non-Web-based theorists as Open Hypertext; I used
the term Holistic Hypertext for a wholly web-based implementation.
--
Nick Kew
Site Valet - the mark of Quality on the Web.
<URL:http://valet.webthing.com/>
Jukka K. Korpela
> To what useful end, Jorn?
Before taking Jorn Barger seriously, you might wish to check his posting
history. Naturally, the pointless crossposting without setting followups
(now set by me) is symptomatic enough, but you could check
<http://groups.google.com/groups?
selm=Xns917181BDF023jkorpelacstutfi%40193.64.137.17>
where I, as response to Jorn Barger's claim that he never made a personal
attack on anyone, proved that he actually has used words like "idiot" and
"crypto-fascist" about an individual (about Jakob Nielsen, incidentally).
--
Yucca, http://www.cs.tut.fi/~jkorpela/
Jorn Barger
> [...] It looks like you are proposing an automated meta-blog.
If you generalise 'blog' to include any page with offsite links,
then sure.
> Would you rather read a novel finished to the author's
> satisfaction, or slog through a log of the keystrokes used
> to do the writing?
Since the Web allows offsite links, and since there are always new
resources becoming available-- offsite-- to link, and also since old
resources may 404 at any time, your analogy of a 'finished' webpage
is inapplicable (which is exactly what I was trying to say).
.
Jorn Barger
> > I'd like to propose an analogous theory called 'action hypertext' in which
> > the webpage captures, not a static image of html-tags being typed in, but
> > a dynamic, evolving image of the web-researcher ***sifting thru search-
> > results*** on a particular topic and gradually identifying and linking the
> > best resources on each subtopic.
>
> Since you've posted this to some WWW groups, you're presumably thinking
> in terms of RDF here?
No, just simple old HTML, for the output. (How the authoring app stores its
database is irrelevant.)
> Or if not, perhaps you'd like to reference your
> argument to current RDF-based work in progress?
You got the wrong dude, sorry.
.
Jerry Muelver
Barger) wrote:
>Jerry Muelver <je...@allmyfaqs.com> wrote:
>> [...] It looks like you are proposing an automated meta-blog.
>
>If you generalise 'blog' to include any page with offsite links,
>then sure.
>
Instead of erecting a straw man argument, how about dealing
with mine? What does including any page with offsite links
have to do with my point?
You said, "I'd like to propose an analogous theory called
'action hypertext' in which the webpage captures, not a
static image of html-tags being typed in, but a dynamic,
evolving image of the web-researcher ***sifting thru
search-results*** on a particular topic and gradually
identifying and linking the best resources on each subtopic.
Your own blog is a listing of such search results. Updating
it frequently makes it "dynamic" in the the sense you are
using the term. Now it looks like you are proposing an
automated meta-blog (familiar words...) by logging the
"image of the web-researcher *** sifting thru
search-results***." If that's not what you meant, you
shouldn't have said it. If it =is= what you meant, how would
the blogger's "sifting thru search-results" be of any
interest or value to anyone?
>> Would you rather read a novel finished to the author's
>> satisfaction, or slog through a log of the keystrokes used
>> to do the writing?
>
>Since the Web allows offsite links, and since there are always new
>resources becoming available-- offsite-- to link, and also since old
>resources may 404 at any time, your analogy of a 'finished' webpage
>is inapplicable (which is exactly what I was trying to say).
That begs the question, Jorn. Your "action hypertext" is no
more immune to link rot than any other page on the Web. Your
"action hypertext" is nothing more than the "Semantic Web"
in Jorn's clothing. See http://www.we.org/sw/ for the
original concept, along with subsequent action hypertext
updates "linking the best resources on each subtopic".
---- jerry ---- "Many of the cairns constructed along the
Path to Web Mastery are built with familiar stones."
--
Jorn Barger
> [...] What does including any page with offsite links
> have to do with my point?
Apparently you're unaware that my weblog is a very minor facet
of my website/research. You could try here for a better sample:
http://www.robotwisdom.com/science/thera.html
> Your own blog is a listing of such search results. Updating
> it frequently makes it "dynamic" in the the sense you are
> using the term. Now it looks like you are proposing an
> automated meta-blog (familiar words...) by logging the
> "image of the web-researcher *** sifting thru
> search-results***." If that's not what you meant, you
> shouldn't have said it. If it =is= what you meant, how would
> the blogger's "sifting thru search-results" be of any
> interest or value to anyone?
Elementary logic error: my weblog is one example of web-research,
but (especially since I never mentioned it!) it's not a good,
representative example... because it has no topic. (I did
mention "on a particular topic", which you chose to trim.)
> [...] Your "action hypertext" is no
> more immune to link rot than any other page on the Web.
I was proposing an authoring tool that recognised the 'action
hypertext' paradigm, including but not limited to the linkrot
problem, and assisted in reducing that problem.
> Your "action hypertext" is nothing more than the "Semantic Web"
Well, if you take away the idea of author-embedded semantic tags,
and add the idea of an authoring tool that understands the
process of topical search, then sure. ;^/
--
Robot Wisdom Weblog: http://www.robotwisdom.com/ "If you worry that
reading the news online will rob you of the serendipity factor you get
with the newspaper, Jorn Barger solves the problem." --Dan Gillmor
Jerry Muelver
Barger) wrote:
>Jerry Muelver <je...@allmyfaqs.com> wrote:
>> [...] What does including any page with offsite links
>> have to do with my point?
>
>Apparently you're unaware that my weblog is a very minor facet
>of my website/research. You could try here for a better sample:
> http://www.robotwisdom.com/science/thera.html
>
I guess I was mislead by this advisory at the bottom of
"thera" -- "Before you leave this site: Be sure you've
checked out Jorn's weblog which offers daily updates on the
best of the Web-- news etc, plus new pages on this site. See
also the overview of the hundreds of pages of original
content offered here, and the offer for a printed version of
the site."
>> Your own blog is a listing of such search results. Updating
>> it frequently makes it "dynamic" in the the sense you are
>> using the term. Now it looks like you are proposing an
>> automated meta-blog (familiar words...) by logging the
>> "image of the web-researcher *** sifting thru
>> search-results***." If that's not what you meant, you
>> shouldn't have said it. If it =is= what you meant, how would
>> the blogger's "sifting thru search-results" be of any
>> interest or value to anyone?
>
>Elementary logic error: my weblog is one example of web-research,
>but (especially since I never mentioned it!) it's not a good,
>representative example... because it has no topic. (I did
>mention "on a particular topic", which you chose to trim.)
>
"Chose to trim"? I think not. Elementary reading error....
Read again. My quote was --
<quote>
You said, "I'd like to propose an analogous theory called
'action hypertext' in which the webpage captures, not a
static image of html-tags being typed in, but a dynamic,
evolving image of the web-researcher ***sifting thru
search-results*** on a particular topic and gradually
identifying and linking the best resources on each subtopic.
</quote>
If you don't like my selection of examples, offer some of
your own. After all, it's =your= theory... after a fashion.
I suppose I could have suggested
http://www.robotwisdom.com/jaj/ since the same argument
applies -- how would the blogger's "sifting thru
search-results" be of any interest or value to anyone?
>> [...] Your "action hypertext" is no
>> more immune to link rot than any other page on the Web.
>
>I was proposing an authoring tool that recognised the 'action
>hypertext' paradigm, including but not limited to the linkrot
>problem, and assisted in reducing that problem.
>
>> Your "action hypertext" is nothing more than the "Semantic Web"
>
>Well, if you take away the idea of author-embedded semantic tags,
>and add the idea of an authoring tool that understands the
>process of topical search, then sure. ;^/
Or if you substitute for the understanding authoring tool
any of the readily-available search engines capable of
reading semantic tags, sure again.
Consider how the inescapable bias and tunnel-vision of a
single-author web-research project might be mitigated by
multi-author contributions to the process, with the added
breadth of expertise and experience such additional authors
would bring, as well as their wealth of treasured and tested
links to core articles on sub-topics and annotations to
same, and you might have something worth considering. In
fact you would have a wiki.
---- jerry
--
Jorn Barger
> >Apparently you're unaware that my weblog is a very minor facet
> >of my website/research. You could try here for a better sample:
> > http://www.robotwisdom.com/science/thera.html
>
> "thera" -- "Before you leave this site: Be sure you've
> checked out Jorn's weblog which offers daily updates on the
> best of the Web-- news etc, plus new pages on this site. See
> also the overview of the hundreds of pages of original
> content offered here, and the offer for a printed version of
> the site."
So, I made a theoretical post about hypertext authoring, and you
jumped to the conclusion it referred to my most popular page,
ignoring the fact that I specified "on a particular topic"?
> >Elementary logic error: my weblog is one example of web-research,
> >but (especially since I never mentioned it!) it's not a good,
> >representative example... because it has no topic. (I did
> >mention "on a particular topic", which you chose to trim.)
>
> "Chose to trim"? I think not. Elementary reading error....
> Read again. My quote was --
>
> <quote>
> You said, "I'd like to propose an analogous theory called
> 'action hypertext' in which the webpage captures, not a
> static image of html-tags being typed in, but a dynamic,
> evolving image of the web-researcher ***sifting thru
> search-results*** on a particular topic and gradually
> identifying and linking the best resources on each subtopic.
> </quote>
That was just cut-and-paste-- immediately after, you trimmed the
relevant part when you made your argument.
> If you don't like my selection of examples, offer some of
> your own. After all, it's =your= theory... after a fashion.
It's a question of your intellectual honesty-- if you select
examples that don't match my "particular topic" specification,
then _you're_ erecting the straw man. And I did suggest the
Thera page as an example, which you again ignored.
> I suppose I could have suggested
> http://www.robotwisdom.com/jaj/ since the same argument
> applies -- how would the blogger's "sifting thru
> search-results" be of any interest or value to anyone?
I'd submit quite confidently that sifting thru search-results
is the highest calling of a Web researcher.
That page got over 200 hits yesterday (a quiet Sunday). My weblog-
'sifting' got thousands. More than 200 other pages on my site got
ten or more, and this number grows daily as my site gets better known.
I often get thank-yous that single out my pages as the most useful
treatment people have found on the given topic.
I'm a little puzzled why you don't recognise the value in that Joyce
page-- is it that you're overwhelmed by the density of information
it offers? (It tries to give a 'what-you-should-know' for every
major Joycean subtopic-- would you be happier if it were broken
down into many subpages-- after the conventional hypertext style--
with a top-page that consisted of simple topic-headers?)
I'm probably guilty of demanding too much from some readers...
> Or if you substitute for the understanding authoring tool
> any of the readily-available search engines capable of
> reading semantic tags, sure again.
Sure what? Are you suggesting semantic tags are already useful
for sifting thru search results? (I find that very hard to believe.)
> Consider how the inescapable bias and tunnel-vision of a
> single-author web-research project might be mitigated by
> multi-author contributions to the process, with the added
> breadth of expertise and experience such additional authors
> would bring, as well as their wealth of treasured and tested
> links to core articles on sub-topics and annotations to
> same, and you might have something worth considering. In
> fact you would have a wiki.
And then all you have to do is search those splendid pages...
and sift the search-results! Gosh, it'd be nice if there were
an authoring tool that understood that sifting process...
.
Jerry Muelver
Barger) wrote:
>Jerry Muelver <je...@allmyfaqs.com> wrote:
>> >Apparently you're unaware that my weblog is a very minor facet
>> >of my website/research. You could try here for a better sample:
>> > http://www.robotwisdom.com/science/thera.html
>>
>> I guess I was misled by this advisory at the bottom of
>> "thera" -- "Before you leave this site: Be sure you've
>> checked out Jorn's weblog which offers daily updates on the
>> best of the Web-- news etc, plus new pages on this site. See
>> also the overview of the hundreds of pages of original
>> content offered here, and the offer for a printed version of
>> the site."
>
>So, I made a theoretical post about hypertext authoring, and you
>jumped to the conclusion it referred to my most popular page,
>ignoring the fact that I specified "on a particular topic"?
>
Read, Jorn, read. "...advisory at the bottom of 'thera'..."
There's a clue there that I went to your recommended page.
It's a mass of links focusing on the alleged date of a
specific event, just as you promised. It is a non-example of
"active hypertext", just as I promised, as susceptible to
link rot as any other page on the web. It is the log of one
man's sifting through search results, a meta-blog log.
>> >Elementary logic error: my weblog is one example of web-research,
>> >but (especially since I never mentioned it!) it's not a good,
>> >representative example... because it has no topic. (I did
>> >mention "on a particular topic", which you chose to trim.)
>>
>> "Chose to trim"? I think not. Elementary reading error....
>> Read again. My quote was --
>>
>> <quote>
>> You said, "I'd like to propose an analogous theory called
>> 'action hypertext' in which the webpage captures, not a
>> static image of html-tags being typed in, but a dynamic,
>> evolving image of the web-researcher ***sifting thru
>> search-results*** on a particular topic and gradually
>> identifying and linking the best resources on each subtopic.
>> </quote>
>
>That was just cut-and-paste-- immediately after, you trimmed the
>relevant part when you made your argument.
>
The fact is that my cut-and-paste was complete, whereas you
accused me of eliding important information when quoting
you. If you make charges of intellectual dishonesty, you own
the responsibility for proving your case by evidence rather
than emulation.
>> If you don't like my selection of examples, offer some of
>> your own. After all, it's =your= theory... after a fashion.
>
>It's a question of your intellectual honesty-- if you select
>examples that don't match my "particular topic" specification,
>then _you're_ erecting the straw man. And I did suggest the
>Thera page as an example, which you again ignored.
>
Web to Jorn -- I =did= do the Thera page, and found it
lacking. As I've explained several times before, it is a
meta-blog log, plain and simple.
>> I suppose I could have suggested
>> http://www.robotwisdom.com/jaj/ since the same argument
>> applies -- how would the blogger's "sifting thru
>> search-results" be of any interest or value to anyone?
>
>I'd submit quite confidently that sifting thru search-results
>is the highest calling of a Web researcher.
>
Well, yeah -- if you define Web researcher as a guy who
sifts through search-results, of course it's his highest
calling. If I define a Web researcher as a guy of obfuscates
arguments, we have modestly different but largely similar
highest calling.
>That page got over 200 hits yesterday (a quiet Sunday). My weblog-
>'sifting' got thousands. More than 200 other pages on my site got
>ten or more, and this number grows daily as my site gets better known.
>I often get thank-yous that single out my pages as the most useful
>treatment people have found on the given topic.
>
Proving what? If there was any validity to this "evidence",
it would be to show that the process of sifting and
collating works without requiring a software tool like an
"understanding editor".
>I'm a little puzzled why you don't recognise the value in that Joyce
>page-- is it that you're overwhelmed by the density of information
>it offers? (It tries to give a 'what-you-should-know' for every
>major Joycean subtopic-- would you be happier if it were broken
>down into many subpages-- after the conventional hypertext style--
>with a top-page that consisted of simple topic-headers?)
>
Let me divest you of your puzzlement. I am bemusedly whelmed
by the density of non-information it offers. That a man's
writings inspire such such a huge conflagration of
pseudo-intellectual energy (the "Joyce Industry"?) is surely
an indication that he failed the most basic test of a writer
-- to communicate. Far more has been written in attempts to
help him do his job than the effort deserves. Why not expend
the energy on content analysis of Pentacostal glossalia?
Then we could have a "Speaking in Tongues Industry".
>I'm probably guilty of demanding too much from some readers...
>
More likely, readers demand too little of some writers.
>> Or if you substitute for the understanding authoring tool
>> any of the readily-available search engines capable of
>> reading semantic tags, sure again.
>
>Sure what? Are you suggesting semantic tags are already useful
>for sifting thru search results? (I find that very hard to believe.)
>
Good one, Jorn -- complain about the taste of the apple pie
while the blossoms have yet to open on the tree.
>> Consider how the inescapable bias and tunnel-vision of a
>> single-author web-research project might be mitigated by
>> multi-author contributions to the process, with the added
>> breadth of expertise and experience such additional authors
>> would bring, as well as their wealth of treasured and tested
>> links to core articles on sub-topics and annotations to
>> same, and you might have something worth considering. In
>> fact you would have a wiki.
>
>And then all you have to do is search those splendid pages...
>and sift the search-results! Gosh, it'd be nice if there were
>an authoring tool that understood that sifting process...
>
I agree. Such a tool would obsolete the scads of
search-bloggers still mired in the fascination of digging
through their diapers -- "Oh wow! Look what I found! And I
probably did it all myself!"
---- jerry ---- "Discoveries on the Path to Web Mastery were
hidden in plain view all along."
--
Jorn Barger
> >So, I made a theoretical post about hypertext authoring, and you
> >jumped to the conclusion it referred to my most popular page,
> >ignoring the fact that I specified "on a particular topic"?
>
> There's a clue there that I went to your recommended page.
I recommended that page in my _followup_ to your first, jumping-to-
conclusions post.
> It's a mass of links focusing on the alleged date of a
> specific event, just as you promised.
'Promised'... on the page itself? Did you fail to notice that these
links are sorted by subtopic, and that I've tried to extract
generalisations from the pages linked?
(I get the impression that you lack a certain element of open
intellectual curiosity.)
> It is a non-example of
> "active hypertext", just as I promised, as susceptible to
> link rot as any other page on the web.
Now you've jumped to the conclusion that you can recognise active
hypertext by the absence or presence of linkrot?!? Odd, given
that my original post proposed an authoring tool for preventing
linkrot to _support_ active hypertext, thus confirming that linkrot
is an inevitable and urgent aspect of the genre.
My definition was that active hypertext uses an evolving page to
'depict' a web-researcher sifting search-results-- which is very much
how the Thera page is evolving.
> It is the log of one
> man's sifting through search results, a meta-blog log.
'Log' would normally imply the ordering is chronological by search-date,
but my topical pages are arranged by subtopic. ('Meta' isn't really
the appropriate term either, is it? I'm not sifting others' weblogs.)
> The fact is that my cut-and-paste was complete, whereas you
> accused me of eliding important information when quoting
> you.
You pasted the full quote, and then requoted just the parts that fit
your preconceptions. You have yet to acknowledge that your example
was poorly chosen because you ignored my specs.
> [..] I =did= do the Thera page, and found it
> lacking. As I've explained several times before, it is a
> meta-blog log, plain and simple.
Your current message is your first that addresses the _content_ of
that Thera page (aside from the standard footer).
And please define 'meta-blog log'. How is it a narrower term than
my original attempt at paraphrase: "any page with offsite links"?
> Well, yeah -- if you define Web researcher as a guy who
> sifts through search-results, of course it's his highest
> calling.
You're welcome to propose a better definition. I can't imagine what
you're objecting to.
> If there was any validity to this "evidence",
> it would be to show that the process of sifting and
> collating works without requiring a software tool like an
> "understanding editor".
Who said 'requiring'? Isn't one purpose of newsgroups like these to
field suggestions about improvements to existing tools? (If not,
whyever not???)
> I am bemusedly whelmed
> by the density of non-information it offers. That a man's
> writings inspire such such a huge conflagration of
> pseudo-intellectual energy (the "Joyce Industry"?) is surely
> an indication that he failed the most basic test of a writer
> -- to communicate.
Apparently you didn't notice, a couple of years back, when Ulysses
was widely hailed as the greatest novel of the 20th century. It really
isn't appropriate for you to dismiss my hypertext design because you
don't appreciate the _content_ my pages treat!
> > Are you suggesting semantic tags are already useful
> >for sifting thru search results? (I find that very hard to believe.)
>
> Good one, Jorn -- complain about the taste of the apple pie
> while the blossoms have yet to open on the tree.
You offered XML search-engines as an adequate substitute for my proposed
authoring tool. Since we can't know how well they'll work until the
XML tags are designed and implemented, and since the task of design
is overwhelming, and the task of implementation is just wishful thinking...
I don't see that as much of a substitute!
> Such a tool would obsolete the scads of
> search-bloggers still mired in the fascination of digging
> through their diapers -- "Oh wow! Look what I found! And I
> probably did it all myself!"
Didn't I see a blog on your homepage? Hmmm?
Jerry Muelver
Barger) wrote:
>. . .
>(I get the impression that you lack a certain element of open
>intellectual curiosity.)
>
Shall I interpret that as an ad hominem, with "open
intellectual curiosity" a Jocyean code phrase for "ignorant
uneducated lout"? Would that be an intellectually honest
interpretation?
. . .
>My definition was that active hypertext uses an evolving page to
>'depict' a web-researcher sifting search-results-- which is very much
>how the Thera page is evolving.
>
And my objection was that such a depiction offers nothing
nothing of value to the accumulation of knowledge. Who cares
what the sifting process was? Who designates the sifters?
Who shall sift the sifters?
. . .
>'Log' would normally imply the ordering is chronological by search-date,
>but my topical pages are arranged by subtopic. ('Meta' isn't really
>the appropriate term either, is it? I'm not sifting others' weblogs.)
>
I agree. "Log" is not completely accurate. We should stick
with your phrase - "...an analogous theory called 'action
hypertext' in which the webpage captures, not a static image
of html-tags being typed in, but a dynamic, evolving image
of the web-researcher ***sifting thru search-results*** on a
particular topic and gradually identifying and linking the
best resources on each subtopic...." with the added provisio
that the sequence of the sifting is not significant, and
since the page does not log changes, neither are edits and
revisions. What's important is the final static page, the
web-searcher's undated, unsequenced blog.... Whoops....
. . .
>
>And please define 'meta-blog log'. How is it a narrower term than
>my original attempt at paraphrase: "any page with offsite links"?
>
Perhaps "auto-blog" would be more precise. The "web
searcher" is sifting through his own search results,
dynamically posting the results (not the sequence) of his
search through his results. Do I have that right? At least,
that's what "Thera" shows.
...
You said, "I'd submit quite confidently that sifting thru
search-results is the highest calling of a Web researcher."
I said --
>> Well, yeah -- if you define Web researcher as a guy who
>> sifts through search-results, of course it's his highest
>> calling.
>
You now add --
>You're welcome to propose a better definition. I can't imagine what
>you're objecting to.
>
I'm objecting to the use of tautological "definitions". I
feel like Alice talking to the Caterpillar.
>> If there was any validity to this "evidence",
>> it would be to show that the process of sifting and
>> collating works without requiring a software tool like an
>> "understanding editor".
>
>Who said 'requiring'? Isn't one purpose of newsgroups like these to
>field suggestions about improvements to existing tools? (If not,
>whyever not???)
>
You said "Authoring tools, then, need to evolve to support
this:...." followed by a list of specified functions. I took
that to mean you were suggesting that the process of
producing "action hypertext" =required= the use of a
special-function tool. I confidently submit that any
reasonable reader would make the same conclusion.
...
I said (re: Joyce page) --
>> I am bemusedly whelmed
>> by the density of non-information it offers. That a man's
>> writings inspire such such a huge conflagration of
>> pseudo-intellectual energy (the "Joyce Industry"?) is surely
>> an indication that he failed the most basic test of a writer
>> -- to communicate.
>
>Apparently you didn't notice, a couple of years back, when Ulysses
>was widely hailed as the greatest novel of the 20th century.
That must be some of that sarcasm-on-the-web I hear so much
about....
>It really
>isn't appropriate for you to dismiss my hypertext design because you
>don't appreciate the _content_ my pages treat!
>
It isn't? Then was would be an appropriate dismissal?
First, I wasn't in on the voting. Second, your hypertext
design is really only a catalog of criticism and
cross-criticism -- critics hailing or dissing each other. It
is "web research" to show that you can do "web research",
not for any discrenable productive end. It's a blog locked
on a theme.
. . .
>You offered XML search-engines as an adequate substitute for my proposed
>authoring tool. Since we can't know how well they'll work until the
>XML tags are designed and implemented, and since the task of design
>is overwhelming, and the task of implementation is just wishful thinking...
>I don't see that as much of a substitute!
>
Really? You offer a theory to be supported by an imaginary
"understanding" authoring tool with magical properties and
functions, and you object to my suggestion that you are
merely describing the Semantic Web proposals in terms more
acceptable to non-geek literary critics? You offer a
description of an apple blossom diet, and object to my apple
pie recipe rebuttal? Why not get concrete with your
proposal, so objections and suggestions have some substance
to affect? For instance, why not consider the infrastructure
needed for your "active hypertext". How about a Universal
Text Data format, as proposed on
http://www.cs.tut.fi/~jkorpela/data/utd.html ? Wouldn't that
be more productive than claiming that my wishful thinking is
difficult to implement, whereas yours is magically superior?
...
I suggested --
>> Such a tool would obsolete the scads of
>> search-bloggers still mired in the fascination of digging
>> through their diapers -- "Oh wow! Look what I found! And I
>> probably did it all myself!"
>
You now counter --
>Didn't I see a blog on your homepage? Hmmm?
What? A diversionary straw man argument from the Champion of
Intellectual Honesty?
First, the "blog" you may be referring to could be any of
dozen or more of my home pages (or "name pages") on wikis I
have established or joined, so we are dealing with an
ambiguous referrent here. Second, I have never in this
discussion objected to blogs or their implemention, or even
to the potentially trivial nature of their content -- I have
only offered the thus-far unrefuted contention that your
"active hypertext" fantasy is only a Semtantic Web
implementation disguised as a blog. Third, my so-called
"blogs" exist in open-access wiki communities where anyone
is free to participate, even edit(!), ideas, proposals, and
discussions.
What would a Joyce-analysis page look like if it were a
wiki, not under the complete dominant control of a sole
philosopher-king "web researcher", hmmmm?
---- jerry
--
Jorn Barger
> [...] your hypertext
> design is really only a catalog of criticism and
> cross-criticism -- critics hailing or dissing each other.
Well, no. That's not even close as a description of the content,
much less the design. (Academic infighting _is_ a useful dramatic
'hook' when available, though.)
> You offer a theory to be supported by an imaginary
> "understanding" authoring tool with magical properties and
> functions,
Quoting my original post:
> - each time you do a search, the results are cached on your own machine, in
> the form of a to-do list that you continually prune and annotate
Magical, or self-evident?
> - this pruned-and-annotated list can be published immediately as a webpage,
> and becomes the basis of an ongoing exploration of the topic
Magical, or self-evident?
> - links might be categorised by 'action-stage': looks-promising-but-not-yet-
> skimmed, skimmed-and-tentatively-approved, skimmed-but-some-doubts, not-
> ideal-but-the-best-so-far, read-fully-and-approved, etc
Magical, or self-evident?
> - the authoring tool records-- on the page itself-- the date the page was
> begun, and the last date it was modified
Magical, or self-evident?
> - hundreds of separate pages on different topics can be juggled, and any
> time you find a new resource you can easily insert it into the appropriate
> page
Magical, or self-evident?
> - the authoring-tool recognises that others' pages can vanish, and monitors
> for this. It could even remember the original search-- and cache the page
> itself!-- so that it can propose likely substitutes.
Magical, or self-evident?
> why not consider the infrastructure
> needed for your "active hypertext".
The infrastructure is plain old HTML, thanks, as I said early on. (You
may have a shiny new hammer, but I'm not talking about nails!)
> How about a Universal Text Data format, as proposed on
> http://www.cs.tut.fi/~jkorpela/data/utd.html ?
I see:
-- lots of hard-to-read underlined blue text
-- author-embedded semantic tags
-- an etext-format that can't be posted straight to netnews, unlike:
http://www.robotwisdom.com/web/parsing.html
> What would a Joyce-analysis page look like if it were a
> wiki, not under the complete dominant control of a sole
> philosopher-king "web researcher", hmmmm?
Um, less-consistent navigation, less quality-control?
Jerry Muelver
Barger) wrote:
. . .
>
>Quoting my original post:
>
>> - each time you do a search, the results are cached on your own machine, in
>> the form of a to-do list that you continually prune and annotate
>
>Magical, or self-evident?
>
Wiki
>> - this pruned-and-annotated list can be published immediately as a webpage,
>> and becomes the basis of an ongoing exploration of the topic
>
>Magical, or self-evident?
>
Wiki
>> - links might be categorised by 'action-stage': looks-promising-but-not-yet-
>> skimmed, skimmed-and-tentatively-approved, skimmed-but-some-doubts, not-
>> ideal-but-the-best-so-far, read-fully-and-approved, etc
>
>Magical, or self-evident?
>
Wiki
>> - the authoring tool records-- on the page itself-- the date the page was
>> begun, and the last date it was modified
>
>Magical, or self-evident?
>
Wiki
>> - hundreds of separate pages on different topics can be juggled, and any
>> time you find a new resource you can easily insert it into the appropriate
>> page
>
>Magical, or self-evident?
>
Wiki
>> - the authoring-tool recognises that others' pages can vanish, and monitors
>> for this. It could even remember the original search-- and cache the page
>> itself!-- so that it can propose likely substitutes.
>
>Magical, or self-evident?
>
Magical.
>> why not consider the infrastructure
>> needed for your "active hypertext".
>
>The infrastructure is plain old HTML, thanks, as I said early on. (You
>may have a shiny new hammer, but I'm not talking about nails!)
>
Looks like my hammer hits your nails right on the head,
except for the magical one.
I suggested --
>> How about a Universal Text Data format, as proposed on
>> http://www.cs.tut.fi/~jkorpela/data/utd.html ?
>
You reply --
>I see:
>-- lots of hard-to-read underlined blue text
Those would be "links".
>-- author-embedded semantic tags
Those would be "categorizations" like your magic authoring
tool would do.
>-- an etext-format that can't be posted straight to netnews, unlike:
> http://www.robotwisdom.com/web/parsing.html
>
That would be -- irrelevant. Why are we talking about
posting straight to netnews all of a sudden?
>> What would a Joyce-analysis page look like if it were a
>> wiki, not under the complete dominant control of a sole
>> philosopher-king "web researcher", hmmmm?
>
>Um, less-consistent navigation, less quality-control?
Also less personal-bias-filtered, broader in range of
analysis, deeper in scope of analysis, more openly
contentious, more contempuous of sloppy research and
analysis, subject to realptime peer evaluation, and a more
accurate depiction of the "dynamic, evolving image of the
web-researcher ***sifting thru search-results*** on a
particular topic" since page revisions are stored and
retrievable.
Sounds like the perfect answer to your dilemma of wanting
"active hypertext" while lacking an "understanding authoring
tool".
---- jerry ---- "It helps to look down once in a while to
make sure you're stepping, not stumbling, on the paving
stones on the Path to Web Mastery."
--
Jorn Barger
> >> - each time you do a search, the results are cached on your own machine, in
> >> the form of a to-do list that you continually prune and annotate
>
> Wiki
Actually, that wasn't the question, but if Wiki really already does
this,
can you point me to a webpage that documents how it's implemented? (I
tried
to find some content at Wiki.org, but just went around in circles--
Does
'The Book' link to an etext or an ad? Just an ad, I discover by trial
and
error. Does 'The Software' link to docs? Beats me:
PerlConvertEolScript, WikiWaySources, TipsForSiteOperators are pretty
opaque as anchortext...)
> >> - links might be categorised by 'action-stage': looks-promising-but-not-yet-
> >> skimmed, skimmed-and-tentatively-approved, skimmed-but-some-doubts, not-
> >> ideal-but-the-best-so-far, read-fully-and-approved, etc
> >Magical, or self-evident?
>
> Wiki
Actually, that wasn't the question, but if Wiki really already does
this,
can you point me to a webpage that documents how it's implemented?
> >> - hundreds of separate pages on different topics can be juggled, and any
> >> time you find a new resource you can easily insert it into the appropriate
> >> page
> >Magical, or self-evident?
>
> Wiki
Actually, that wasn't the question, [etc]
> >> - the authoring-tool recognises that others' pages can vanish, and monitors
> >> for this. It could even remember the original search-- and cache the page
> >> itself!-- so that it can propose likely substitutes.
> >Magical, or self-evident?
>
> Magical.
Which part? Surely not caching the original search, and/or the
original
page. But if it has these, it would be pretty easy to repeat the
search
and then look for the greatest statistical overlap with the vocabulary
of the original page.
> Looks like my hammer hits your nails right on the head,
> except for the magical one.
First off, you switched hammers-- you were pounding on the Semantic
Web,
most of the way. (Second off, please post those urls I requested
above,
if you want to convince me you finally understand the issues.)
> >-- lots of hard-to-read underlined blue text
> Those would be "links".
Has ciwa* theory really not advanced to a stage where underlined blue
text is agreed to be less-readable than non-underlined text of a
maximally-contrasting color? (For tables of contents, I usually
make just the bullet be the anchor, and if they're #-links I use '#'
for the bullet.)
> >-- author-embedded semantic tags
>
> Those would be "categorizations" like your magic authoring
> tool would do.
See above-- you've jumped to conclusions yet again.
> >-- an etext-format that can't be posted straight to netnews, unlike:
> > http://www.robotwisdom.com/web/parsing.html
>
> That would be -- irrelevant. Why are we talking about
> posting straight to netnews all of a sudden?
Because it's the great unsolved annoyance of text markup-- how to
translate marked-up pages into something convenient to post.
The conventional solution is to demand everyone replace their favorite
tools with more-complex (ie, buggy, slow, expensive) tools that
support
the markup-flavor-of-the-month. But a more net-savvy approach is
the use setext-like netnews-style markup, with footnotes for the
messy stuff. (See that url you quoted but obviously didn't visit.)
> >> What would a Joyce-analysis page look like if it were a
> >> wiki, not under the complete dominant control of a sole
> >> philosopher-king "web researcher", hmmmm?
> >
> >Um, less-consistent navigation, less quality-control?
>
> Also less personal-bias-filtered,
The disadvantage of multiple biases (Wiki, Slashdot, Memepool,
Plastic)
is that you can't easily learn who you trust. One author per website
is my philosophy.
> broader in range of analysis,
My active-hypertexts certainly aim for a broad range, but they have
the advantage that I can annotate that range with my own reactions,
highlighting the ones I think deserve the most trust. (If readers
don't trust me, why would they want to spend any time at my site at
all, anyway?)
> deeper in scope of analysis,
It's usually only as deep as the deepest contributor, although
occasionally miracle partnerships happen.
> more openly contentious,
You were just dissing me for using academic infighting as a 'hook'!?
> more contempuous of sloppy research and analysis,
Again, it's only as good as its best individual.
> subject to realtime peer evaluation,
I have suggestion boxes for that (or I'd rather use newsgroups).
> and a more
> accurate depiction of the "dynamic, evolving image of the
> web-researcher ***sifting thru search-results*** on a
> particular topic" since page revisions are stored and
> retrievable.
This sounds like it would be useful about once a year...
Jerry Muelver
Barger) wrote:
Wiki info at http://www.usemod.com/cgi-bin/wiki.pl
A wiki on information architecture (that would be the topic
under discussion here) is at http://iawiki.net (click on
Recent Changes link near the top to see what's been going on
lately). Wiki source code is readily available, so specific
enhancements can be readily implemented. For instance,
MeatBall and AllMyFAQs use the same engine, but AllMyFAQs
has a more verbose, in-context type of search result.
. . .
>The disadvantage of multiple biases (Wiki, Slashdot, Memepool,
>Plastic)
>is that you can't easily learn who you trust. One author per website
>is my philosophy.
>
That could be a problem, but perhaps less of a problem than
you think. A wiki community is based on trust... and it
works. I don't know what an academic community is based on,
but somehow I think "trust" is not a major component.
>> broader in range of analysis,
>
>My active-hypertexts certainly aim for a broad range, but they have
>the advantage that I can annotate that range with my own reactions,
>highlighting the ones I think deserve the most trust. (If readers
>don't trust me, why would they want to spend any time at my site at
>all, anyway?)
>
Easily done on a wiki. Readers could also make annotations,
and would be sure to return to read your subsequent
"clarifications".
>> deeper in scope of analysis,
>
>It's usually only as deep as the deepest contributor, although
>occasionally miracle partnerships happen.
>
They're common on wikis.
>> more openly contentious,
>
>You were just dissing me for using academic infighting as a 'hook'!?
>
I don't think so. I believe I dissed you only for using
tautological definitions.
>> more contempuous of sloppy research and analysis,
>
>Again, it's only as good as its best individual.
>
On a wiki, it's only as good as its best synergy.
>> subject to realtime peer evaluation,
>
>I have suggestion boxes for that (or I'd rather use newsgroups).
>
Argument manipulations, like eliding points you can't
adequately refute, or throwing up a straw man when your
argument falters, that are invisible in "suggestion box" or
newsgroup channels, are exposed to sunlight on a wiki. It's
quite easy to figure out who to trust when the entire
discussion is aviable for single-source scrutiny.
>> and a more
>> accurate depiction of the "dynamic, evolving image of the
>> web-researcher ***sifting thru search-results*** on a
>> particular topic" since page revisions are stored and
>> retrievable.
>
>This sounds like it would be useful about once a year...
Actually, "previous versions" are quite handy, both for
recovery in the event of an "accident" or malicious
modification, and for tracing the development of an emerging
concept like "active hypertext".
I'm not evangelizing wikis here. I'm only suggesting that
many, if not all, of the functions you are seeking might
already be available... or tested and found lacking... by
folks who share your interest in research and analysis even
though they may share your interest in specific topics.
---- jerry
--
Jorn Barger
You made several specific claims about active-hypertext problems it could
solve. Please cite specific urls.
> A wiki on information architecture (that would be the topic
> under discussion here) is at http://iawiki.net (click on
> Recent Changes link near the top to see what's been going on
> lately).
Single-pixel background images cause my slow MacNetscape to freeze,
sorry.
> A wiki community is based on trust... and it works.
[reply suppressed]
> >My active-hypertexts certainly aim for a broad range, but they have
> >the advantage that I can annotate that range with my own reactions,
> >highlighting the ones I think deserve the most trust.
>
> Easily done on a wiki.
Easily done in HTML, too, but there's easy and there's easy. Because
I do this dozens of times a day, I want a custom tool.
> I believe I dissed you only for using tautological definitions.
Let's see: "Stream-of-conscious blogging", "Joyce fanatics", "fumbling
efforts", "anti-information", "Elementary reading error", "inescapable
bias and tunnel-vision", "density of non-information", "pseudo-
intellectual energy", "digging through their diapers", "nothing of
value to the accumulation of knowledge", "magical properties", "the
complete dominant control of a sole philosopher-king"... and, here it
is: "your hypertext design is really only a catalog of criticism and
cross-criticism -- critics hailing or dissing each other. It is "web
research" to show that you can do "web research", not for any
discrenable productive end.""
> Argument manipulations, like eliding points you can't
> adequately refute,
Thanks for reminding me. Are you ready yet to admit you were wrong
in claiming:
-- "you are proposing an automated meta-blog"
-- "Your "action hypertext" is no more immune to link rot than any
other page on the Web."
-- "Your "action hypertext" is nothing more than the "Semantic Web""
-- "any of the readily-available search engines capable of reading
semantic tags" [are an adequate substitute for my authoring tool]
-- "I went to your recommended page" [anachronism]
-- "my cut-and-paste was complete, whereas you accused me of eliding
important information when quoting you"
-- "the Thera page... is a meta-blog log, plain and simple"
-- "your hypertext design is really only a catalog of criticism and
cross-criticism -- critics hailing or dissing each other"
-- "You offer a theory to be supported by an imaginary "understanding"
authoring tool with magical properties and functions"
-- "[author-embedded semantic tags] would be "categorizations" like
your magic authoring tool would do."
> I'm not evangelizing wikis here.
Coulda fooled me.
> I'm only suggesting that
> many, if not all, of the functions you are seeking might
> already be available... or tested and found lacking... by
> folks who share your interest in research and analysis even
> though they may share your interest in specific topics.
Post proof or retract.
Jerry Muelver
Barger) wrote:
>Jerry Muelver <je...@allmyfaqs.com> wrote:
. . .
>> I'm not evangelizing wikis here.
>
>Coulda fooled me.
>
You offered a "theory" (or more precise the proposal of a
theory), and I suggested several methods for its
implementation. Since my messages would actually have helped
you clarify the operation of your magical authoring tool,
and you ducked the opportunity to produce such
clarification, there appears to be little more I can do. I
come, after all, from a wiki milieu, where trust and
cooperation are core cultural values. You've rejected trust,
and insisted on solo production. It looks like I can't help
you.
>> I'm only suggesting that
>> many, if not all, of the functions you are seeking might
>> already be available... or tested and found lacking... by
>> folks who share your interest in research and analysis even
>> though they may share your interest in specific topics.
>
>Post proof or retract.
This is like arguing with a Creationist. You offer magical
functions in a mystery authoring system to accomplish
unproductive goals in the pursuit of trivial
self-indulgence, and I asked you to show how it would (not
even =does=, but =would=) be of value, and you consistently
duck the challenge.
I've shown you wikis, how they work, the kinds of traceable,
trackable "active hypertext" they produce, and you simply
claim "Not what I want."
So, tell us what you =do= want. State a refutable hypothesis
of some sort that can be tested, that does not rely on
tautological definitions, and we can examine your claims of
hypertext superiority.
To restate my case, I see no benefit in your original
description of "active hypertext". Therefor, I see no reason
why anyone would expend any energy on the development of
your magical authoring system. You've offered nothing beyond
a wishlist of user-interface functions to indicate how it
would work. You've rejected suggestions that the specific
functions you seek already exist without your having
invented them.
So, how do you figure to inporve the Web? Produce, or stop
blowing smoke.
---- jerry ---- "It can take all day, sometimes, for the
morning fog to dissipate on the Path to Web Mastery."
--
Michael Stutz
> On 24 Feb 2002 14:01:52 -0800, jo...@enteract.com (Jorn
> Barger) wrote:
>
> >Jerry Muelver <je...@allmyfaqs.com> wrote:
> >> Would you rather read a novel finished to the author's
> >> satisfaction, or slog through a log of the keystrokes used
> >> to do the writing?
Funny you picked this example. I should like to do either -- depending
on the nature of my inquiry, and the mood and interest of the moment.
And there's no reason you can't eat your cake and have it, too -- the
'finished' novel and entire log are just two views (of many); you just
might not always want to go there.
We need to be able to view a log of the writing of a novel -- this is
most especially important for the writer himself, who must obviously
be able to see where he is going. This log is the _manuscript_.
Because a manuscript is really just a log of the writing of a book.
Ask any literary critic or aspiring writer how important it is to see
the manuscripts -- there's no better way to study a writer's method
than to _watch it happen_ and see where he goes ...
We see it all plopped out at once with the old paper manuscripts, and
we see it so with computer files, but there's no reason that computers
have to stick with saving only final frozen images (or revisions when
you decide to check them in).
Revision control tools and capture buffers are a start, but in my
ideal writing machine, realtime keystroke record & playback is built
right in (instantly giving the writer robust revision control, since
there's stoppoints at every change in the work).
(How much disk space would you need to keep a lifetime's text log?)
> >Since the Web allows offsite links, and since there are always new
> >resources becoming available-- offsite-- to link, and also since old
> >resources may 404 at any time, your analogy of a 'finished' webpage
> >is inapplicable (which is exactly what I was trying to say).
I think what you're saying, Jorn, and what you've been trying to say,
is that the kind of web-writing that interests you -- and the kind
you've long been doing -- is writing _around_ all the words that are
already out there. You've got an idea or want to write about such and
such a topic, and the first thing you do is scope it out, and see
who's already gone there. Then you paint all these other words and
travels into your page, so that your piece kind of frames what's
already out there. And now the trouble with this kind of writing is of
course what's out there is always slowly moving (the 404s and new
links), which over time kind of shifts the picture in the frame -- so
you're prosposing that new writing tools take these problems (and this
kind of writing) into account, so when a writer makes one of these
"active hypertexts" or framed kind of pieces it knows where to look to
find the picture that you painted ...
Jorn Barger
> [...]
> We need to be able to view a log of the writing of a novel [...]
Especially in cases like Finnegans Wake.
> (How much disk space would you need to keep a lifetime's text log?)
See http://www.robotwisdom.com/jaj/ulysses/clocktime.html
If you wrote 1.5 megs per day (the size of Ulysses), a full lifetime would
be 37 gigs. So if you wrote 150k, then 3.7 gigs.
[...]
> I think what you're saying, Jorn, and what you've been trying to say,
> is that the kind of web-writing that interests you -- and the kind
> you've long been doing -- is writing _around_ all the words that are
> already out there. You've got an idea or want to write about such and
> such a topic, and the first thing you do is scope it out, and see
> who's already gone there.
Yes-- I tend to forget this isn't yet the universal practice.
> Then you paint all these other words and
> travels into your page, so that your piece kind of frames what's
> already out there. And now the trouble with this kind of writing is of
> course what's out there is always slowly moving (the 404s and new
> links), which over time kind of shifts the picture in the frame -- so
> you're proposing that new writing tools take these problems (and this
> kind of writing) into account, so when a writer makes one of these
> "active hypertexts" or framed kind of pieces it knows where to look to
> find the picture that you painted ...
Someday we may be able to just click on a name, and a popup menu will
automatically be generated that offers:
biography
photo
online works
news
fansites
official site
etc
But for now, Google doesn't quite get there, so the author's job is to
pre-sort as much of it as may be useful. Yahoo has a super-dumb
implementation of this in some newswire articles, offering 'websites'
or 'news' for apparently random names within the article. eg:
http://dailynews.yahoo.com/h/cdh/20020323/lo/rallying_the_troops_1.html
.
Michael Stutz
> st...@dsl.org (Michael Stutz) wrote:
> > (How much disk space would you need to keep a lifetime's text log?)
>
> See http://www.robotwisdom.com/jaj/ulysses/clocktime.html
>
> If you wrote 1.5 megs per day (the size of Ulysses), a full lifetime would
> be 37 gigs. So if you wrote 150k, then 3.7 gigs.
Yeah, I read that piece back when you first wrote it. It's nice to
know that a cheap 40GB harddrive is big enough to hold a life-work's
of text, but what I'm wondering is how much more space would you need
to store plaintext writing input that was read & recorded in
_realtime_ ...
> Yahoo has a super-dumb implementation of this in some newswire articles,
> offering 'websites' or 'news' for apparently random names within the article.
> eg: http://dailynews.yahoo.com/h/cdh/20020323/lo/rallying_the_troops_1.html
I wonder if these links ought not to be the job of the author at all,
but should instead be taken care of by the browser or reader software.
So like you could select a region of text in the window you're
reading, then right-click (or whatever) and it'd show a pop-up list
with a set of configurable links: fansites, photos, news, definition,
history, author profile, etc. ...
Jorn Barger
> Someday we may be able to just click on a name, and a popup menu will
> automatically be generated that offers:
>
> biography
> photo
> online works
> news
> fansites
> official site
> etc
>
> But for now, Google doesn't quite get there, so the author's job is to
If you click on a word that's not a name, you might get:
definition
etymology
encyclopedia
image-search
Google-search
GooJa-search
...but the OED points the way to a feature the Web may offer someday,
which might be called the 'biography' of the word-- a _chronological_
inventory of every occasion the word has been used. If the Web ever
gets semanticised enough to tag the dates every document (div, span)
was written, this sorting could be a basic search-engine option.
Again, though, an editor's job will be (as with that OED madman) to
winnow the most interesting uses... and ideally to add some
'biographical' commentary on how the word's use evolved.
.
Michael Stutz
> If you click on a word that's not a name, you might get:
>
> definition
> etymology
> encyclopedia
> image-search
> Google-search
> GooJa-search
'Click' itself might not even be the button of the mouse but good web
readers might offer all sorts of key commands that work on the
selected text ...
> ...but the OED points the way to a feature the Web may offer someday,
> which might be called the 'biography' of the word-- a _chronological_
> inventory of every occasion the word has been used.
> If the Web ever gets semanticised enough to tag the dates every document
> (div, span) was written, this sorting could be a basic search-engine option.
Anyone could always change the timestamps on a file, of course, but
one way to do it right might be for all computers to get their clocks
reasonably synchronized, with tools like chrony[1].
Or no -- way easier than asking everyone to please change their local
systems, the search engine itself should just write a timestamp for
when a file is first spidered (as well as for all subsequent changes).
But if the search engine relies on publisher-provided dates, you
should be able to optionally filter out those files with undesired or
bogus timestamps ...
(Like 01-01-1980 -- c'mon, who did any real work on that day? I
didn't, and I was there!)
Just last night I was thinking about what one would do with a little
black box, whose contents consisted of the entire Internet up to then.
"What's in the box?"
"Uh, a copy of the _Inter_net."
"Well, what's it say? Let's start at the beginning!"