Theory: "information density"

0 views
Skip to first unread message

Jorn Barger

unread,
Jan 21, 2001, 5:19:01 AM1/21/01
to
My main web-design page-- http://www.robotwisdom.com/web/ --currently
uses the expression 'content-centered' as the umbrella-concept under
which all my design ideas try to find shelter...

But "content-centered" _sounds_ completely tautological, so I'd like to
find a replacement that's a little more self-explanatory. And one idea
might be to build on the idea of 'information density'...

Nothing is more frustrating for me as a web-surfer than being forced to
step thru a 'hierarchical' table of contents: several pages in a row,
each of which offers just its own very-short table of contents for the
next layer down. I've called this the 'stripped grapes' model because
it offers just a skeleton with all food-value removed, when it could
(and should) be redesigned to enrich the information density as much as
possible-- by collapsing all the layers into a single page, including a
summary of the content of all the info-pages, and 'promoting' samples of
the best content from each page to the top layer.

This is what I mean by content-centered. Any page that offers content
(ie, all pages, hopefully) should try to _optimise_ that content, which
normally means raising the information density (and info quality, too, I
guess). Hi-density/hi-quality information includes:

- clear summaries
- simple lists and tables and timelines
- images that really illustrate
- vivid pullquotes
- links that are carefully chosen and clearly described
- forms you can use to get quick answers

A hi-density page should include all of these, arranged clearly and
compactly. Instead, if you analysed the info-density of the average
page, you'd probably get something like:

10k of info graphics
30k of navigation graphics
40k of decorative graphics

1k of content-related links
10k of navigation links
40k of 'filler' links (eg motivated by e-commerce, not info-content)

5k of content text

So you're waiting for 136k to load to get 16k of info.

(And rendering-time also has to be factored into the info-density
calculation: every added layer of tables (especially) further dilutes
your score. Also breaking up text over multiple pages.)


more: http://www.robotwisdom.com/web/

--
http://www.robotwisdom.com/ "Relentlessly intelligent
yet playful, polymathic in scope of interests, minimalist
but user-friendly design." --Milwaukee Journal-Sentinel

Peter Boersma

unread,
Jan 22, 2001, 7:54:02 AM1/22/01
to
On Sun, 21 Jan 2001 04:19:01 -0600, jo...@mcs.com (Jorn Barger) wrote:

>My main web-design page-- http://www.robotwisdom.com/web/ --currently
>uses the expression 'content-centered' as the umbrella-concept under
>which all my design ideas try to find shelter...
>
>But "content-centered" _sounds_ completely tautological, so I'd like to
>find a replacement that's a little more self-explanatory. And one idea
>might be to build on the idea of 'information density'...

[snip onion-model description]

>This is what I mean by content-centered. Any page that offers content
>(ie, all pages, hopefully) should try to _optimise_ that content, which
>normally means raising the information density (and info quality, too, I
>guess). Hi-density/hi-quality information includes:
>

[snip list]

>A hi-density page should include all of these, arranged clearly and
>compactly. Instead, if you analysed the info-density of the average
>page, you'd probably get something like:

[snip percentages]

A series of similar indexes, a tool to measure them, and a large
number of examples can be found at:

The Rating Game
http://stein.cshl.org/~lstein/rater/

with an explanantion (from 1997!) at:
Sifting The Wheat From The Chaff
http://www.webtechniques.com/archives/1997/05/webm/

Peter
--------------------------------------------
Peter Boersma Satama Amsterdam
http://www.satama.nl/~peter/
computer-human interaction? http://sigchi.nl

Jorn Barger

unread,
Jan 22, 2001, 8:58:40 AM1/22/01
to
Peter Boersma <peter.boersma@_REMOVE_THIS_satama.com> wrote:
> [snip onion-model description]
> [snip list]

> [snip percentages]
>
> A series of similar indexes, a tool to measure them, and a large
> number of examples can be found at:
>
> The Rating Game
> http://stein.cshl.org/~lstein/rater/
>
> with an explanantion (from 1997!) at:
> Sifting The Wheat From The Chaff
> http://www.webtechniques.com/archives/1997/05/webm/

Yeah, I recognise these, so I should cite them in the future.

(I'd enumerate the insights he's missing, but you're obviously not
interested in learning new ideas!)

Jerry Muelver

unread,
Jan 22, 2001, 9:40:29 AM1/22/01
to
On Mon, 22 Jan 2001 12:54:02 GMT,
peter.boersma@_REMOVE_THIS_satama.com (Peter Boersma) wrote:

<snip>


>A series of similar indexes, a tool to measure them, and a large
>number of examples can be found at:
>
> The Rating Game
> http://stein.cshl.org/~lstein/rater/
>
>with an explanantion (from 1997!) at:
> Sifting The Wheat From The Chaff
> http://www.webtechniques.com/archives/1997/05/webm/
>

Hmmm.... hotanimebabes.com got an Information Rating of 100
on Linc's Rater, and robotwisdom only scored 94.6. Looks
like Jorn's got his work cut out for him to get that
information density up to snuff.

---- jerry
--

Jerry Muelver

unread,
Jan 22, 2001, 9:44:42 AM1/22/01
to
On Mon, 22 Jan 2001 07:58:40 -0600, jo...@mcs.com (Jorn
Barger) wrote:

>Peter Boersma <peter.boersma@_REMOVE_THIS_satama.com> wrote:
>> [snip onion-model description]
>> [snip list]
>> [snip percentages]
>>
>> A series of similar indexes, a tool to measure them, and a large
>> number of examples can be found at:
>>
>> The Rating Game
>> http://stein.cshl.org/~lstein/rater/
>>
>> with an explanantion (from 1997!) at:
>> Sifting The Wheat From The Chaff
>> http://www.webtechniques.com/archives/1997/05/webm/
>
>Yeah, I recognise these, so I should cite them in the future.
>
>(I'd enumerate the insights he's missing, but you're obviously not
>interested in learning new ideas!)

For someone advertising "Relentlessly intelligent


yet playful, polymathic in scope of interests, minimalist
but user-friendly design." --Milwaukee Journal-Sentinel

you make leaps over some remarkable presumptive gaps. I
could enumerate the weak assumptions you rely on, but your'e
obviously not interested in anyone's analysis but your own.

---- jerry
--

Jorn Barger

unread,
Jan 22, 2001, 10:16:00 AM1/22/01
to
Jerry Muelver <je...@hytext.com> wrote:
> robotwisdom only scored 94.6. Looks
> like Jorn's got his work cut out for him to get that
> information density up to snuff.

the page i rated is my joyce overview: http://www.robotwisdom.com/jaj/

and it's my showpiece of applied info-density theory, so i recommend you
look at it before sneering.


(yeesh)

Jerry Muelver

unread,
Jan 22, 2001, 12:22:58 PM1/22/01
to
On Mon, 22 Jan 2001 09:16:00 -0600, jo...@mcs.com (Jorn
Barger) wrote:

>Jerry Muelver <je...@hytext.com> wrote:
>> robotwisdom only scored 94.6. Looks
>> like Jorn's got his work cut out for him to get that
>> information density up to snuff.
>
>the page i rated is my joyce overview: http://www.robotwisdom.com/jaj/
>
>and it's my showpiece of applied info-density theory, so i recommend you
>look at it before sneering.
>

Okay. Done. 94.8 for Information on Linc's scale. That's
ever so much better.

You realize, don't you, that Joyce just plain made up a lot
of that "Information"? Except, maybe, the Periodic Table
stuff.

BTW I don't sneer. I may snirk on occasion, and sometimes
I'll snile, but sneering exhibits too much hostility for my
gentle nature.

---- jerry
--


Jerry Muelver

unread,
Jan 22, 2001, 12:26:43 PM1/22/01
to
On Mon, 22 Jan 2001 09:16:00 -0600, jo...@mcs.com (Jorn
Barger) wrote:

>Jerry Muelver <je...@hytext.com> wrote:
>> robotwisdom only scored 94.6. Looks
>> like Jorn's got his work cut out for him to get that
>> information density up to snuff.
>
>the page i rated is my joyce overview: http://www.robotwisdom.com/jaj/
>
>and it's my showpiece of applied info-density theory, so i recommend you
>look at it before sneering.
>
>
>(yeesh)

I love this quote from your showpiece --

"Because Joyce is so obscure, it's easy for anyone who's
intellectually ambitious to become the local Joyce expert by
reading Ellmann and a few other sources. When these local
experts come together in groups, there's a tendency to
battle for prominence, using Ellmann et al as the standard
of measure.

"Original research is mostly discouraged by this process,
though, so you're left with a lot of belligerence defending
the status quo against new ideas."

That's been my own opinion for a while, now.

----- jerry
--


Jorn Barger

unread,
Jan 22, 2001, 12:33:46 PM1/22/01
to
Jerry Muelver <je...@hytext.com> wrote:
> Okay. Done. 94.8 for Information on Linc's scale. That's
> ever so much better.
>
> You realize, don't you, that Joyce just plain made up a lot
> of that "Information"? Except, maybe, the Periodic Table
> stuff.
>
> BTW I don't sneer. I may snirk on occasion, and sometimes
> I'll snile, but sneering exhibits too much hostility for my
> gentle nature.


Negative-infinity for information, on the Robot Wisdom Scale.

Eric Jarvis

unread,
Jan 22, 2001, 12:39:24 PM1/22/01
to
on Mon, 22 Jan 2001 11:26:43 -0600, je...@hytext.com wrote...

speaking as a Bloomsday boy (I still haven't come up with a convincing
theory to explain the choice of date by it being my birthday, but give
me time)...am I the only person in the entire world who has read
Ulysses from cover to cover and yet hasn't become obsessed by the
book?

there doesn't seem to be any middle ground (apart from me)

but I digress

Ulysses is actually an interesting parallel to hypertext...Joyce was
attempting something similar, trying to write with a greater "density"
of information than a standard narrative...80 years on it still scares
the hell out of most people...so how long will it be before we can
experiment on the web with a similar "density" of information and
still hold an audience

or will a new generation of web educated kids read Ulysses and just
think it's a kind of quaint but unchallenging read?...will they
understand Finnegan's Wake?...now that is a scary thought

--
eric
"the alternative to seeing things in black and white is
to see them in full colour"

Jerry Muelver

unread,
Jan 22, 2001, 2:10:05 PM1/22/01
to
On Mon, 22 Jan 2001 11:33:46 -0600, jo...@mcs.com (Jorn
Barger) wrote:

>Jerry Muelver <je...@hytext.com> wrote:
>> Okay. Done. 94.8 for Information on Linc's scale. That's
>> ever so much better.
>>
>> You realize, don't you, that Joyce just plain made up a lot
>> of that "Information"? Except, maybe, the Periodic Table
>> stuff.
>>
>> BTW I don't sneer. I may snirk on occasion, and sometimes
>> I'll snile, but sneering exhibits too much hostility for my
>> gentle nature.
>
>
>Negative-infinity for information, on the Robot Wisdom Scale.

Information is the resolution of uncertainty, which puts
zero at the very bottom of the scale. For instance, if you
flip a coin and hide it under your hand, I am 50% uncertain
of the result. So I ask you, "What did you get?" You reply,
"Heads." That resolved my uncertainty about the coin toss,
so that reply was information.

But if you flip the coin onto the tabletop where I can see
it, and I look at it and see that it is heads, and I am 0%
uncertain about the result, and then you say, "Heads," that
utterance resolves no uncertainty, and therefore contains no
information.

All of which means that any measure of information density
has to be made from the viewpoint of the receiver/observer,
not the formulator/sender. So you can see that if someone
has once read your page, then goes back for another look
(stranger things have happened), the ratio of useful
information to useless noise plummets. If you would grant
that in a global community not everyone visiting your page
understands English, then for Chinese visitors your page has
very close to zero information regardless of its format,
vocabulary, and navigational components.

It's a cruel world, and we can't always have things the way
we want them.

---- jerry
--

Stylewriter

unread,
Jan 22, 2001, 6:49:12 PM1/22/01
to
I don't follow any of the meandering, off topic, replies to
date, John. So I'll just barge straight in and be as clear
as your goodself.

In article <1enkfx1.1sv1pjr1eg6ksgN%jo...@mcs.com>,
jo...@mcs.com says...
____> My main web-design page-- http://www.robotwisdom.com/web/ --currently
____> uses the expression 'content-centered' as the umbrella-concept under
____> which all my design ideas try to find shelter...
____>
____> But "content-centered" _sounds_ completely tautological, so I'd like to
____> find a replacement that's a little more self-explanatory. And one idea
____> might be to build on the idea of 'information density'...

It seesm clear to me that the 'umbrella-concept ' under
which your site operates is minimalistic and structureless
information'.

____>
____> Nothing is more frustrating for me as a web-surfer than being forced to
____> step thru a 'hierarchical' table of contents: several pages in a row,
____> each of which offers just its own very-short table of contents for the
____> next layer down. I've called this the 'stripped grapes' model because
____> it offers just a skeleton with all food-value removed, when it could
____> (and should) be redesigned to enrich the information density ....

Absolutely agree. Three licks shold get any visitor to any
element of content they seek.

____> .....as much as
____> possible--

This is where you start to loose it ....

____> by collapsing all the layers into a single page, including a
____> summary of the content of all the info-pages, and 'promoting' samples of
____> the best content from each page to the top layer.
____>
____> This is what I mean by content-centered. Any page that offers content
____> (ie, all pages, hopefully) should try to _optimise_ that content, which
____> normally means raising the information density (and info quality, too, I
____> guess).

The problem with your style is that you are not optimising
anything. What you are doing to stripping the content of
all structure and focus. Human communication is not the same
as computer or robot communication - as much as you may wish
that it were. Human communication depends on shape, form,
focus, hierarchy and digestibility. Your theory and site
lack of these elements. It may be efficient in
communciations information quality down a telephone line -
but it is not efficient as communicating information to a
human being.

____> Hi-density/hi-quality information includes:
____>
____> - clear summaries
____> - simple lists and tables and timelines
____> - images that really illustrate
____> - vivid pullquotes
____> - links that are carefully chosen and clearly described
____> - forms you can use to get quick answers
____>
____> A hi-density page should include all of these, arranged clearly and
____> compactly.

Your site lacks summaries, simplicity. There are NO images
despite your definition. A link such as 'service like
Atomz.com ' cannot be describedas clearly described.

Theproblem with your approach is that it is as inappropriate
and pointless as the graphic artist gone mad approach. One
end of the spectrum is obsessed with everything but the
content - while the other end is obsesse with everything but
the presentation.

Both approaches are doomed to dismal failure in the face of
human communication dynamics which demand a balanced,
stuctures and focussed approach for any kind of success to
result.

Few visitors to a site are so single minded on obtaining the
information on 'that' site that they want to do nothing
other than read 5,000 information-packed words on the
relevant topic. The vast majority of site visitors are
seeking bite sized chunks of information that may or may not
be present in the site being visited. They do not wish to
read through 5,000 words to come to the conclusion that the
chunk they seek is not after all there.
And most of the rest of the visitors are not seeking any
specific piece of information, per se. They simply wish to
get a 'flavour' of the content.
In the realms of the commercial site, your approach is a
suicidal 'trip' for any prospective business. In the realms
of the reference site, your approach is guarenteed, imho, to
bore the pants of any inquirer.

Regards,

Stylewriter

joshs...@my-deja.com

unread,
Jan 22, 2001, 7:42:10 PM1/22/01
to
In article <1enkfx1.1sv1pjr1eg6ksgN%jo...@mcs.com>,

jo...@mcs.com (Jorn Barger) wrote:
> My main web-design page--
http://www.robotwisdom.com/web/ --currently
> uses the expression 'content-centered' as the
umbrella-concept under
> which all my design ideas try to find shelter...
>
> But "content-centered" _sounds_ completely
tautological, so I'd like to
> find a replacement that's a little more self-
explanatory. And one idea
> might be to build on the idea of 'information
density'...
>

[snip snip]

[another snip]

All of this smack's of Jakob Nielsen's ideas on
usability on the 'net (and his book "Designing
Web Usability). I have some disagreements with
his theories as well, but he does balance
"information density" with usability.

I mean, what if I removed all the paragraph
breaks, graphics, and white space from a
document? The whole page would be one clutter of
text: tons of information, but hardly usable.

As opposed to reducing the file size (which
Nielsen also does, across the board), there's
also the "screen" size. Perhaps no more than 10%
of the screen should be devoted to navigation,
for instance. Of course, a 640x120 navigation
bar is only 10% of a 1024x768 screen, but is over
25% of 640x480. So good luck there...

I highly recommend reading up on usability as a
whole, as opposed to narrowly defining success in
terms of "information density."

-Josh Segall


Sent via Deja.com
http://www.deja.com/

Jerry Muelver

unread,
Jan 23, 2001, 12:14:44 AM1/23/01
to
On Mon, 22 Jan 2001 23:49:12 GMT, Stylewriter
<scam...@eggs.net> wrote:

>I don't follow any of the meandering, off topic, replies to
>date, John. So I'll just barge straight in and be as clear
>as your goodself.
>

<snip>


>
>Few visitors to a site are so single minded on obtaining the
>information on 'that' site that they want to do nothing
>other than read 5,000 information-packed words on the
>relevant topic. The vast majority of site visitors are
>seeking bite sized chunks of information that may or may not
>be present in the site being visited. They do not wish to
>read through 5,000 words to come to the conclusion that the
>chunk they seek is not after all there.
>And most of the rest of the visitors are not seeking any
>specific piece of information, per se. They simply wish to
>get a 'flavour' of the content.
>In the realms of the commercial site, your approach is a
>suicidal 'trip' for any prospective business. In the realms
>of the reference site, your approach is guarenteed, imho, to
>bore the pants of any inquirer.

Jorn misses the point that communication is the eliciting of
a desired response. Like his hero James Joyce, he assumes
that vomiting profundity is enough to do the job. Both
neglect to take into account the reaction of the reader.
Notice what noise and thunder are generated around
explanations of what The Author really meant. At some point,
someone is bound to point out, "If a piece of writing needs
a bunch of apologists explaining what the author really
meant, the author apparently did not do a very good job in
the first place."

---- jerry
--


Jorn Barger

unread,
Jan 23, 2001, 1:57:20 AM1/23/01
to
Stylewriter <scam...@eggs.net> wrote:
> Your site lacks summaries, simplicity. There are NO images
> despite your definition.

The page http://www.robotwisdom.com/web/ which you're looking at is
_about_ the theory, it's not at all an example of applying the theory.

As I said in a previous post, the best example is:

http://www.robotwisdom.com/jaj/

Eric Jarvis

unread,
Jan 23, 2001, 5:39:28 AM1/23/01
to
on Mon, 22 Jan 2001 13:10:05 -0600, je...@hytext.com wrote...

> If you would grant
> that in a global community not everyone visiting your page
> understands English, then for Chinese visitors your page has
> very close to zero information regardless of its format,
> vocabulary, and navigational components.
>

so how do you compute it if the page itself is in English but has a
link to a Chinese translation?...what if it also has links to
translation into Arabic, Bengali, Hindi, Japanese, Swahili, Russian,
Dutch, Spanish, Portuguese, German, French, Danish, Finnish and
Hungarian? (this is what I'm aiming at for the end of next year)

does each link to a new language add to the level of
information?...and how does the number of people who speak the
language affect this?

--
eric
"live fast, die only if strictly necessary"

Jerry Muelver

unread,
Jan 23, 2001, 6:52:16 AM1/23/01
to

Computing "information density" is a parlor game -- nothing
more. It's a metric with no utility, like computing the
number of angels that can dance on a pin. In web design, as
in all communication endeavors, what matters is the result.
Communication is the eliciting of a desired response. If you
are not getting a response you can use to assess the effect
of your communicative action, you were not communicating in
the first place.

Why do people insist on just putting out messages without
regard to the effect of the message? Why do they act as if
any uttering or writing or posting is communication? First,
because they don't understand the communication process. And
second, because it makes them feel good. Such one-way
messaging is only consumptive communication -- the effect is
on the sender, who is getting the response he desires
internally. That's not very social behavior, is it?

---- jerry
--

Eric Jarvis

unread,
Jan 23, 2001, 7:41:36 AM1/23/01
to
on Tue, 23 Jan 2001 05:52:16 -0600, je...@hytext.com wrote...

> On Tue, 23 Jan 2001 10:39:28 -0000, Eric Jarvis
> <webm...@befrienders.org> wrote:
>
> >on Mon, 22 Jan 2001 13:10:05 -0600, je...@hytext.com wrote...
> >> If you would grant
> >> that in a global community not everyone visiting your page
> >> understands English, then for Chinese visitors your page has
> >> very close to zero information regardless of its format,
> >> vocabulary, and navigational components.
> >
> >so how do you compute it if the page itself is in English but has a
> >link to a Chinese translation?
> >
> >does each link to a new language add to the level of
> >information?...and how does the number of people who speak the
> >language affect this?
>
> Computing "information density" is a parlor game -- nothing
> more. It's a metric with no utility, like computing the
> number of angels that can dance on a pin. In web design, as
> in all communication endeavors, what matters is the result.
> Communication is the eliciting of a desired response. If you
> are not getting a response you can use to assess the effect
> of your communicative action, you were not communicating in
> the first place.
>
> Why do people insist on just putting out messages without
> regard to the effect of the message? Why do they act as if
> any uttering or writing or posting is communication? First,
> because they don't understand the communication process. And
> second, because it makes them feel good. Such one-way
> messaging is only consumptive communication -- the effect is
> on the sender, who is getting the response he desires
> internally. That's not very social behavior, is it?
>

fizzackly

any theoretical analysis of web page design means nothing if divorced
from theoretical study of web usage...it's an old hobby horse of
mine...as far as I am concerned a web site isn't complete until it's
being read...content and design are all very important factors, but
there is only one measure of it's success or failure...the affect it
has on the user

the reason people like to concentrate on ideas related to the
production of a web site is that there is some control there...if it
was simply a matter of meeting a set of fixed criteria, then design
would be a process entirely under the designer's control...good for
the ego, but impossible in reality

it's scary to think of the web as a huge and complex thing that we
barely understand the first thing about, yet are becoming totally
reliant on...but it's the truth

colin

unread,
Jan 23, 2001, 4:12:43 AM1/23/01
to

> This is where you start to loose it ....

must remeber that one. classic!


Matthias Gutfeldt

unread,
Jan 23, 2001, 8:58:26 AM1/23/01
to
Jerry Muelver wrote:
> Communication is the eliciting of a desired response. If you
> are not getting a response you can use to assess the effect
> of your communicative action, you were not communicating in
> the first place.

This is a description of a highly manipulative communication sequence.
According to your model, B's response would in turn try to elicit a
desired response in A, and A would have no really reliable way of
knowing whether he was successful or not. Basically, A and B are trying
to cheat each other into saying what they want to hear. I know this is
an accurate description of many situations, but I wouldn't consider it a
general definition of how communication works.


> Why do people insist on just putting out messages without
> regard to the effect of the message? Why do they act as if
> any uttering or writing or posting is communication? First,
> because they don't understand the communication process.

Your understanding of "the communication process" might not be their
understanding of "the communication process".


> And second, because it makes them feel good.

What's wrong about feeling good? In the thread on why we help newbies,
feeling good was one of the reasons. It's as valid as any other reason
for communication.


Matthias

Jerry Muelver

unread,
Jan 23, 2001, 9:32:56 AM1/23/01
to
On Tue, 23 Jan 2001 14:58:26 +0100, Matthias Gutfeldt
<wo...@gmx.at> wrote:

>Jerry Muelver wrote:
>> Communication is the eliciting of a desired response. If you
>> are not getting a response you can use to assess the effect
>> of your communicative action, you were not communicating in
>> the first place.
>
>This is a description of a highly manipulative communication sequence.
>According to your model, B's response would in turn try to elicit a
>desired response in A, and A would have no really reliable way of
>knowing whether he was successful or not. Basically, A and B are trying
>to cheat each other into saying what they want to hear. I know this is
>an accurate description of many situations, but I wouldn't consider it a
>general definition of how communication works.
>

What's the alternative definition? How can you call anything
"communication" if you don't work the result of the action
into the equation? Is the "desired" aspect troubling? Let's
notch it up a level of abstraction from the specific to the
general case, so folks won't clutch up about the presumed
manipulative aspect of the model:

* Communication is the eliciting of a response.

* Effective communication is the eliciting of a desired
response.

Now we can include consumptive communication and obnoxious
structured noise (tobacco ads, for instance) in the larger
case, and productive discourse in the more specific case.

>
>> Why do people insist on just putting out messages without
>> regard to the effect of the message? Why do they act as if
>> any uttering or writing or posting is communication? First,
>> because they don't understand the communication process.
>
>Your understanding of "the communication process" might not be their
>understanding of "the communication process".
>

Certainly. That works much along the same lines as an
understanding of good design, acceptable social conduct, and
the whole panapoly of human values.

>> And second, because it makes them feel good.
>
>What's wrong about feeling good? In the thread on why we help newbies,
>feeling good was one of the reasons. It's as valid as any other reason
>for communication.
>

I didn't say there was anything wrong with feeling good. In
fact, I offered "feeling good" as a legitimate motivator for
personal expression. The real point is that one who mounts a
communication activity solely because it makes him feel good
to do so, really has little to complain about if the
response is not quite what he desired. If the goal is
effective communication, the "feel good" aspect has to be
shifted from "look how clever I am" to "that really worked
well".

---- jerry
--

Peter Boersma

unread,
Jan 23, 2001, 11:39:17 AM1/23/01
to
On Mon, 22 Jan 2001 07:58:40 -0600, jo...@mcs.com (Jorn Barger) wrote:

>Peter Boersma <peter.boersma@_REMOVE_THIS_satama.com> wrote:

>> A series of similar indexes, a tool to measure them, and a large
>> number of examples can be found at:

[snip reference]


>
>Yeah, I recognise these, so I should cite them in the future.

Yeah, I showed them to you before, but you're obvi... no, forget that.

Look, I mean no harm, but there are other approaches, and you should
read them. Have a look at the work of Jared Spool (since you hate
Jakob Nielsen] and see what he comes up with after testing with real
users.

Jared's company is called User Interface Engineering, and can be found
online at:
http://www.uie.com (redirects)

Michael Stutz

unread,
Jan 23, 2001, 11:50:54 AM1/23/01
to
In article <94ik11$kil$1...@nnrp1.deja.com>, <joshs...@my-deja.com> wrote:

>As opposed to reducing the file size (which
>Nielsen also does, across the board), there's
>also the "screen" size. Perhaps no more than 10%
>of the screen should be devoted to navigation,
>for instance. Of course, a 640x120 navigation
>bar is only 10% of a 1024x768 screen, but is over
>25% of 640x480. So good luck there...

I've always favored a one-line text menubar. But I've been wondering
whether there ought to be *0%* of the screen devoted to navigation --
no navigation controls in the document at all. Like the Tufte
quotation from the old pre-corporatized photo.net site, about how the
navigation interface should be part of the document itself -- the less
extra controls, the easier reading becomes. The idea of navigation
controls seems almost counter-hypertext, not futuristic -- the more
controls, the more a document is locked into one specific context.

Philip Stripling

unread,
Jan 23, 2001, 5:32:40 PM1/23/01
to
m...@dsl.org (Michael Stutz) writes:

>
> I've always favored a one-line text menubar. But I've been wondering
> whether there ought to be *0%* of the screen devoted to navigation --
> no navigation controls in the document at all. Like the Tufte
> quotation from the old pre-corporatized photo.net site, about how the
> navigation interface should be part of the document itself -- the less
> extra controls, the easier reading becomes. The idea of navigation
> controls seems almost counter-hypertext, not futuristic -- the more
> controls, the more a document is locked into one specific context.

Well, there's hypertext and then there's hypertext. I favored the browse
and graze technique of Web design myself, but people were emailing me
asking me where stuff was. To me, it was obvious, but then I was writing
the pages with the links in the text instead of in navigation areas, so I
knew where all those links were. Until it got too big for me to remember.

So I put in a table of contents and linked to that on every page. People
_loved_ having a TOC because then they didn't have to browse through and
pay attention to what they were reading to find stuff -- they could get to
it directly.

Controls absolutely _are_ counter-hypertext. And for many people, that's a
plus. But with HTML, I can try to have both. I have to say, though, that
for my users, the navigation links and TOC are heavily used to avoid having
to read a whole Web file to get where they want to go. Some people do want
to browse (although whether reading a page with a couple of dozen
hyperlinks per paragraph is a pleasure seems to be in some dispute), but
some don't, so I try to have it both ways.

Another issue with 0% navigation is all those people who come into a page
from a search engine. That has happened to me a _lot_, and with no
navigation hints on how to get more info than is on the page, I'm often
left with chopping bits off the end of the URL to backtrack into what seems
to be an interesting site. Often Web authors don't chunk well, so few pages
are stand alone, and navigation is sorely needed.

--
Philip Stripling | email to the replyto address is presumed
Legal Assistance on the Web | spam and read later. email to philip@
http://www.PhilipStripling.com/ | civex.com is read daily.
Resources for freelancers, small businesses, and entrepreneurs.

Stylewriter

unread,
Jan 23, 2001, 7:33:31 PM1/23/01
to
In article <cg4q6toojnfcs9so3...@4ax.com>,
je...@hytext.com says...
____> "If a piece of writing needs
____> a bunch of apologists explaining what the author really
____> meant, the author apparently did not do a very good job in
____> the first place."
____>
An excellent point, J.

:S:

Stylewriter

unread,
Jan 23, 2001, 7:34:28 PM1/23/01
to
In article <1ennx2t.1nfveafr5qvj1N%jo...@mcs.com>,
jo...@mcs.com says...
____> The page http://www.robotwisdom.com/web/ which you're looking at is
____> _about_ the theory, it's not at all an example of applying the theory.
____>

I am left speechless.

:S:

Mark J. Young

unread,
Jan 24, 2001, 1:42:03 PM1/24/01
to
Michael Stutz wrote:
>
> I've always favored a one-line text menubar. But I've been wondering
> whether there ought to be *0%* of the screen devoted to navigation --
> no navigation controls in the document at all. Like the Tufte
> quotation from the old pre-corporatized photo.net site, about how the
> navigation interface should be part of the document itself -- the less
> extra controls, the easier reading becomes. The idea of navigation
> controls seems almost counter-hypertext, not futuristic -- the more
> controls, the more a document is locked into one specific context.

Interesting and compelling point. Navigation bars
seem part of a movement to make webpages look and
feel like desktop software applications. This is
clearly appropriate for some webpages (e.g., web-based
based applications). I question whether the "app" L&F
is appropriate for webpages that are more documentation
oriented.

Mark

Urban Fredriksson

unread,
Jan 25, 2001, 3:48:23 AM1/25/01
to
In article <w3qr91t...@shell.rawbw.com>,
Philip Stripling <phil_st...@cieux.zzn.com> wrote:

>Another issue with 0% navigation is all those people who come into a page
>from a search engine.

I think that in those cases, if they want to read more
it's quite likely they'll go back to the search results.
If I have another document on the same subject, it'd
usually show up there too.

>That has happened to me a _lot_, and with no
>navigation hints on how to get more info than is on the page, I'm often
>left with chopping bits off the end of the URL to backtrack into what seems
>to be an interesting site.

What sort of navigational information do you think I
should have on a document in case it really is the only
one on a certain subject? (Examples are the two last in
the list on my main page and a couple of others I haven't
bothered to have links to at all.)
--
Urban Fredriksson http://www.canit.se/%7Egriffon/
To get rid of an enemy, make him a friend.

Joe Brenner

unread,
Jan 30, 2001, 5:20:52 AM1/30/01
to
Jerry Muelver <je...@hytext.com> writes:

There's a passage in Lilian Hellman's _Pentimento_, where
she describes an encounter with Tallulah Bankhead. Bankhead
has snuck off to have sex with some guy, and Hellman
accidentally walks in on them. Bankhead makes a show of not
being embarrassed, and goes into a routine like "Hey, look
at this guy's penis, have you ever seen one as big as this!"
Lilian Hellman is very dismissive of Bankhead... Bankhead's
schtick reminds her of 20s flapper-types she used to know,
and it just seems stale and pointless. She goes on to say
that she can't imagine that any woman would really care
about penis size, and that the whole obsession actually
strikes her as homosexual at root.

I picked up two things from this:

(a) I agree that it sounds like Tallulah Bankhead was
trying too hard to be outrageous.

(b) I also get a strong impression that there was something
a little repressed about Lilian Hellman. She was a
little too eager to project her standards and reactions
on to others.

Point (b) was presumably not the intended response of the
reader. It would not at first glance make sense to say that
the communication was "ineffective", since I would say that
she was communicating more than she had intended.

Reply all
Reply to author
Forward
0 new messages