Re: List structures

Discussion in 'HTML' started by dorayme, Mar 17, 2009.

  1. dorayme

    dorayme Guest

    In article
    <1iwp1pb.1qpnyo8o3ccqvN%>,
    (D.M. Procida) wrote:

    > dorayme <> wrote:
    >
    > > > Which do you prefer, the nested list version:

    >
    > > > or the headered version:

    >
    > > > I think the former is to be preferred, but I can't see a tremendous
    > > > reason for preferring one over the other.

    > >
    > > Well, you should not prefer the first to the second because the first is
    > > all wrong and would make a doc invalid.
    > >
    > > Fix it up and then I will tell you which is to be preferred and why.
    > > This offer is open for 21 days. But do not delay too much, my shop is
    > > very busy and the stock goes in and out quickly. <g>

    >
    > All right then, apart from the typo, which do you prefer?
    >


    OK, I am suitably chastized. Did not look closely! <g>

    First markup: <http://dorayme.netweaver.com.au/markup1.html>

    Second markup: <http://dorayme.netweaver.com.au/markup2.html>

    Well, one thing we can say is that the first markup is simpler than the
    second in that it uses one actual element to make do for what the second
    uses four. If you are impressed by Ockam's Razor, you will surely use
    the first. It elegantly and simply and less disjointedly represents the
    structure inherent in what you are saying.

    Let's suppose that more than a monk's thoughts from about 800 years ago
    is needed before a final decision. Here is more.

    It is almost always a good idea in these matters to compare the choices
    you face in the event of your author style sheet being not available and
    a default used by the browser.

    If the browser does use its styles then both choices have their
    strengths and weaknesses on mere presentational grounds in tests that I
    have conducted. The all one list is neat and understandable, the meaning
    is represented by indentation and differential bulleting. Personally I
    feel the lack of font-weight in the primary list items does not help out
    enough and I am irritated by the primary list item bulleting. But the
    meaning is terribly clear!

    In the more profligate second version, this lack of font-weight is made
    up for. But that is the trouble, it is made up with too much of a
    vengeance and second order headings are awfully big. Not only that, the
    vertical spacings between headings and lists are too big and ugly. On
    the other hand, the lack of bullets for the headings is nicer than the
    sight of the first order bullets in the first mark up imo.

    I would say on aesthetic grounds alone, the first markup wins by half a
    length.

    But presentation is *everything* when it comes to human observers'
    appreciation of meaning and there is more than aesthetics involved. If
    your list is longer than the example, and let us suppose it is so we can
    answer a more general query, then the first would win hands down on
    communication in visual browsers because it better uses the screen area.
    If a user can get your content with less scrolling and effort then they
    get the meaning better. Simple as that. On this ground, again, the first
    markup wins.

    I will discuss the issue under the assumption that maintenance and
    author styling efforts are important factors if you really want. But
    perhaps this will do for now.

    I desist for the moment telling you about a wonderful and apt third way
    to organise your lists in a table, but we should keep it in mind as a
    player when complex lists are involved.

    --
    dorayme
    dorayme, Mar 17, 2009
    #1
    1. Advertising

  2. dorayme

    dorayme Guest

    In article <>,
    dorayme <> wrote:

    > In article
    > <1iwp1pb.1qpnyo8o3ccqvN%>,
    > (D.M. Procida) wrote:
    >
    > > > > I think the former is to be preferred, but I can't see a tremendous
    > > > > reason for preferring one over the other.


    >
    > First markup: <http://dorayme.netweaver.com.au/markup1.html>
    >
    > Second markup: <http://dorayme.netweaver.com.au/markup2.html>
    >
    >
    > I desist for the moment telling you about a wonderful and apt third way
    > to organise your lists in a table, but we should keep it in mind as a
    > player when complex lists are involved.


    Here is a third type of way, it still does not beat the sheer clarity of
    the first markup but only because the default styles are so severe! I
    have no real idea why default style sheets do not put a modicum of
    padding and border and border-collapse in. So I put this in to show what
    the bare minimum of style would make it look like.

    Third markup: <http://dorayme.netweaver.com.au/markup3.html>

    Looks pretty clear to me and if default styles were better thought out
    along the lines you can see, I would say this is best of all to
    communicate the event schedule at least in visual browsers. I welcome
    comments about the effects for other modalities.

    Note that if you remove the author styles in this table markup, it
    simply looks shithouse and markup1 wins yet again! That markup 1 is
    quite a thoroughbred, no?

    I think it is a very telling mistake of browser makers to adopt a
    default style sheet that does not include the bare minimum of styles I
    include above. To me, it says loud and clear: we do not really
    understand the function or semanticity of tables!

    Or is it, oh gee... no... a sheer yellow streak, a miserable cowardly
    giving into the table layout brigade! This perfectly fits their crime.
    Folk who use tables for sheer page layout and not for organizing lists
    would not want borders by default! Poor lambs, they would have to
    actually remove default borders. I am disgusted! <g>

    The meaning of a table is best brought out by a few simple borders and a
    modicum of padding. This presentation is part and parcel of user
    understanding. Here is a case where open mindedness is simple mindedness
    or sheer wickedness as per above.

    The border in a default situation is quite important for human
    recognition of what is going on, meaning-wise in a table properly used.
    For so called tabular data, or, as I prefer, as a way to organise lists
    and bring their relationships out to human observers.

    This does not mean I advocate borders full stop. Authors can use their
    judgement. With suitable cell spacing and/or padding and sometimes
    elegant backgrounding, no-border is a *perfectly* good option. But in
    the default situation, it is an almost perverse thing to leave out. If
    you don't believe me, remove my styles completely and see how awful and
    mildly confusing it looks in FF, to take an example.

    --
    dorayme
    dorayme, Mar 17, 2009
    #2
    1. Advertising

  3. dorayme

    D.M. Procida Guest

    dorayme <> wrote:

    > > All right then, apart from the typo, which do you prefer?
    > >

    >
    > OK, I am suitably chastized. Did not look closely! <g>
    >
    > First markup: <http://dorayme.netweaver.com.au/markup1.html>
    >
    > Second markup: <http://dorayme.netweaver.com.au/markup2.html>
    >
    > I would say on aesthetic grounds alone, the first markup wins by half a
    > length.


    I think we should disregard the aesthetics - we can change the
    presentation any way we like.

    If for some reason the user isn't seeing styles, then it could be
    because they've chosen that, or because they can't (maybe they're blind)
    - either way, they're used to seeing content like that, and more
    importantly, the nested list structure gives more clues about the
    relationships within its contents.

    That's why I prefer the first way.

    The downside is that more CSS is required, and the application of a
    suitable class falls to the author, in order to improve the appearance
    in most contexts.

    > I desist for the moment telling you about a wonderful and apt third way
    > to organise your lists in a table, but we should keep it in mind as a
    > player when complex lists are involved.


    Your table would be better achieved by using CSS to style the nested
    list for you - as long as the author keeps the list at two levels.

    Daniele
    D.M. Procida, Mar 17, 2009
    #3
  4. dorayme

    dorayme Guest

    In article
    <1iwpvzi.t602jt13g2dczN%>,
    (D.M. Procida) wrote:

    > dorayme <> wrote:
    >
    > > > All right then, apart from the typo, which do you prefer?
    > > >

    > >
    > > OK, I am suitably chastized. Did not look closely! <g>
    > >
    > > First markup: <http://dorayme.netweaver.com.au/markup1.html>
    > >
    > > Second markup: <http://dorayme.netweaver.com.au/markup2.html>
    > >
    > > I would say on aesthetic grounds alone, the first markup wins by half a
    > > length.

    >
    > I think we should disregard the aesthetics - we can change the
    > presentation any way we like.


    I was speaking in the context of where author styles were not available.
    Aesthetics is not about merely pretty. Aesthetics is good when function
    shines through. And we are talking a particular functional bit of a
    page, not the whole page.

    Preceding my last remark was:

    "... the first markup is simpler than the second in that it uses one
    actual element to make do for what the second uses four. ... It
    elegantly and simply and less disjointedly represents the structure..."

    and

    "... always a good idea ... to compare the choices ... in the event of
    .... a default used by the browser."

    >
    > If for some reason the user isn't seeing styles, then it could be
    > because they've chosen that, or because they can't (maybe they're blind)
    > - either way, they're used to seeing content like that, and more
    > importantly, the nested list structure gives more clues about the
    > relationships within its contents.
    >


    First, the user *always* sees styles. Don't miss this like troops miss
    road-side bombs!

    Where no author styles operate, the vacuum is necessarily immediately
    filled by a browser supplied set of styles. These styles are very
    important communication factors. If the beloved abstract correct element
    to use does a worse job to communicate information than what is regarded
    by the orthodox as an incorrect or even non-semantic element, then maybe
    a different road should be chosen.

    I too regard the relationship within content to be important. But the
    proof is in the pudding and the pudding is not the HTML, it is in the
    results on the countless devices that humans use.

    > The downside is that more CSS is required, and the application of a
    > suitable class falls to the author, in order to improve the appearance
    > in most contexts.
    >


    Certainly it is an important factor how much CSS fiddling is needed to
    bring something up to a fine communication.


    > > I desist for the moment telling you about a wonderful and apt third way
    > > to organise your lists in a table, but we should keep it in mind as a
    > > player when complex lists are involved.

    >
    > Your table would be better achieved by using CSS to style the nested
    > list for you - as long as the author keeps the list at two levels.
    >


    Are you saying that better than

    <http://dorayme.netweaver.com.au/markup3.html>

    would be to make markup 1 look something similar?

    --
    dorayme
    dorayme, Mar 17, 2009
    #4
  5. dorayme

    D.M. Procida Guest

    dorayme <> wrote:

    > > I think we should disregard the aesthetics - we can change the
    > > presentation any way we like.

    >
    > I was speaking in the context of where author styles were not available.
    > Aesthetics is not about merely pretty. Aesthetics is good when function
    > shines through. And we are talking a particular functional bit of a
    > page, not the whole page.


    My point was that we can manage the appearance spearately, so we don't
    need to worry about that here. As for the functional component of the
    aesthetics that can't be dealt with in the styling - well, that's the
    list structure question I was aksing.

    > > If for some reason the user isn't seeing styles, then it could be
    > > because they've chosen that, or because they can't (maybe they're blind)
    > > - either way, they're used to seeing content like that, and more
    > > importantly, the nested list structure gives more clues about the
    > > relationships within its contents.
    > >

    >
    > First, the user *always* sees styles. Don't miss this like troops miss
    > road-side bombs!
    >
    > Where no author styles operate, the vacuum is necessarily immediately
    > filled by a browser supplied set of styles.


    Not necessarily. The browser provides information to the user about the
    content via HTML. Sometimes it does this through styles (bold for
    <strong>, big text for headings) but sometimes it simply says "heading
    level one".

    > I too regard the relationship within content to be important. But the
    > proof is in the pudding and the pudding is not the HTML, it is in the
    > results on the countless devices that humans use.


    No, the proof of the pudding is in the *eating*.

    That aside, since humans use countless devices to read HTML, I shouldn't
    try to second-guess them - I should follow the same rules that the
    devices should be following.

    Where the rules need interpretion that isn't dead simple, then I still
    think that aiming to understand the rules rather than second-guess the
    browsers is the best policy.

    > > Your table would be better achieved by using CSS to style the nested
    > > list for you - as long as the author keeps the list at two levels.
    > >

    >
    > Are you saying that better than
    >
    > <http://dorayme.netweaver.com.au/markup3.html>
    >
    > would be to make markup 1 look something similar?


    If that's really a list of events, then yes. But not if it's supposed to
    be tabular data.

    Daniele
    D.M. Procida, Mar 19, 2009
    #5
  6. dorayme

    Bergamot Guest

    D.M. Procida wrote:
    > dorayme <> wrote:
    >
    >> Are you saying that better than
    >>
    >> <http://dorayme.netweaver.com.au/markup3.html>
    >>
    >> would be to make markup 1 look something similar?

    >
    > If that's really a list of events, then yes. But not if it's supposed to
    > be tabular data.


    That is just a list of events. Table markup is illogical because there
    is no relationship between the cells in a row, other than the headers.
    If it were tabular there would be some kind of relationship between data
    cells.

    --
    Berg
    Bergamot, Mar 19, 2009
    #6
  7. dorayme

    David Segall Guest

    Bergamot <> wrote:

    >
    >D.M. Procida wrote:
    >> dorayme <> wrote:
    >>
    >>> Are you saying that better than
    >>>
    >>> <http://dorayme.netweaver.com.au/markup3.html>
    >>>
    >>> would be to make markup 1 look something similar?

    >>
    >> If that's really a list of events, then yes. But not if it's supposed to
    >> be tabular data.

    >
    >That is just a list of events. Table markup is illogical because there
    >is no relationship between the cells in a row, other than the headers.
    >If it were tabular there would be some kind of relationship between data
    >cells.


    dorayme in her continued crusade to persuade us that black is grey and
    white is grey has come up with an arguable example of what you might
    define as nested lists that can also be defined as a table. There is
    an implied row heading column that could contain the entries "1st
    event", "2nd event", etc. To make it more realistic substitute
    Morning, Afternoon and Evening for 1st, 2nd and 3rd.
    David Segall, Mar 19, 2009
    #7
  8. David Segall wrote:
    > Bergamot <> wrote:
    >
    >> D.M. Procida wrote:
    >>> dorayme <> wrote:
    >>>
    >>>> Are you saying that better than
    >>>>
    >>>> <http://dorayme.netweaver.com.au/markup3.html>
    >>>>
    >>>> would be to make markup 1 look something similar?
    >>> If that's really a list of events, then yes. But not if it's supposed to
    >>> be tabular data.

    >> That is just a list of events. Table markup is illogical because there
    >> is no relationship between the cells in a row, other than the headers.
    >> If it were tabular there would be some kind of relationship between data
    >> cells.

    >
    > dorayme in her continued crusade to persuade us that black is grey and
    > white is grey has come up with an arguable example of what you might
    > define as nested lists that can also be defined as a table. There is
    > an implied row heading column that could contain the entries "1st
    > event", "2nd event", etc. To make it more realistic substitute
    > Morning, Afternoon and Evening for 1st, 2nd and 3rd.


    Yep, next she might advocate not need for paragraphs in P elements, one
    "should" use a one column table for that as well!


    --
    Take care,

    Jonathan
    -------------------
    LITTLE WORKS STUDIO
    http://www.LittleWorksStudio.com
    Jonathan N. Little, Mar 19, 2009
    #8
  9. dorayme

    dorayme Guest

    In article
    <1iwtmle.1qlfwnx1ds6dp3N%>,
    (D.M. Procida) wrote:

    > dorayme <> wrote:
    >
    > > > I think we should disregard the aesthetics - we can change the
    > > > presentation any way we like.

    > >
    > > I was speaking in the context of where author styles were not available.
    > > Aesthetics is not about merely pretty. Aesthetics is good when function
    > > shines through. And we are talking a particular functional bit of a
    > > page, not the whole page.

    >
    > My point was that we can manage the appearance spearately, so we don't
    > need to worry about that here. As for the functional component of the
    > aesthetics that can't be dealt with in the styling - well, that's the
    > list structure question I was aksing.
    >


    We cannot manage the the appearance separately in the context of author
    styles being off. And it is only in this context that I am talking. I
    won't go into it here, but I believe if we *really* could manage the
    appearances with author styles, it would not really matter so much how
    we marked up, anything to get the appearance needed to communicate with
    humans.


    > > > If for some reason the user isn't seeing styles, then it could be
    > > > because they've chosen that, or because they can't (maybe they're blind)
    > > > - either way, they're used to seeing content like that, and more
    > > > importantly, the nested list structure gives more clues about the
    > > > relationships within its contents.
    > > >

    > >
    > > First, the user *always* sees styles. Don't miss this like troops miss
    > > road-side bombs!
    > >
    > > Where no author styles operate, the vacuum is necessarily immediately
    > > filled by a browser supplied set of styles.

    >
    > Not necessarily. The browser provides information to the user about the
    > content via HTML. Sometimes it does this through styles (bold for
    > <strong>, big text for headings) but sometimes it simply says "heading
    > level one".
    >


    According to my way of thinking, if a browser causes a voice to say
    "Heading level one" and following with the heading content, then that is
    a default presentation of that part of the web page. It is a didactic
    form of presentation. I believe that HTML markup is only good for its
    ability to be a reliable cause of presentation that communicates
    information that webpage makers want communicated.

    > > I too regard the relationship within content to be important. But the
    > > proof is in the pudding and the pudding is not the HTML, it is in the
    > > results on the countless devices that humans use.

    >
    > No, the proof of the pudding is in the *eating*.
    >

    My phrase is a modern shortened version as befits the wild young modern
    thing I am. But I accept the seniority of your version. <g>

    > That aside, since humans use countless devices to read HTML, I shouldn't
    > try to second-guess them - I should follow the same rules that the
    > devices should be following.
    >


    Under a certain interpretation, I agree wholeheartedly

    <http://netweaver.com.au/semantics/whySemanticElements.php>

    > Where the rules need interpretion that isn't dead simple, then I still
    > think that aiming to understand the rules rather than second-guess the
    > browsers is the best policy.


    Understanding the rules so that you are practically equipped involves
    knowledge of browsers, their defaults, their faults, the likelihood of
    device and browser makers understanding what you understand and they
    putting into practice these understandings.
    >
    > > > Your table would be better achieved by using CSS to style the nested
    > > > list for you - as long as the author keeps the list at two levels.
    > > >

    > >
    > > Are you saying that better than
    > >
    > > <http://dorayme.netweaver.com.au/markup3.html>
    > >
    > > would be to make markup 1 look something similar?

    >
    > If that's really a list of events, then yes. But not if it's supposed to
    > be tabular data.
    >

    Tables are for displaying lists, from the simplest to the most complex,
    and an excellent way to show relationship between list items in
    different lists. I use the word "list" in the ordinary English sense of
    the word. To use it in an the HTML sense is to beg questions that should
    not be begged.

    Your original question is enormously interesting to me, Daniele, because
    it brings out so many important issues. Thanks for asking it!

    I won't bang on here now but I disagree that there is some one "real
    structure" of what you wish to communicate. You can see why I favoured
    the nested list option. I said why. But it did not include the component
    of the one true real structure. I think this is a fantasy of some
    website theorists!

    I know, I said at the start of my response to your question, "Well, one
    thing we can say is that the first markup is simpler than the second in
    that it uses one actual element to make do for what the second uses
    four. If you are impressed by Ockam's Razor, you will surely use the
    first. It elegantly and simply and less disjointedly represents the
    structure inherent in what you are saying."

    I was talking loosely or, shall we say, without an implication that
    there was some one fixed in stone structure. I just meant that you
    wanted to let people know when things were on. And I said that I
    preferred the first on grounds of simplicity in the one situation we
    must always be prepared for, with default styles on. In short, it gets
    across what events are on when in a simpler way in this situation.

    But there is no inner one abstract structure. If we could rely on author
    styles - which we can't rely on - then other markups, such as your
    second one or my table example would be fine. At least that is what I am
    thinking at the moment. I am happy to consider the matter further and
    modify if necessary.

    --
    dorayme
    dorayme, Mar 19, 2009
    #9
  10. dorayme

    D.M. Procida Guest

    dorayme <> wrote:

    > We cannot manage the the appearance separately in the context of author
    > styles being off. And it is only in this context that I am talking. I
    > won't go into it here, but I believe if we *really* could manage the
    > appearances with author styles, it would not really matter so much how
    > we marked up, anything to get the appearance needed to communicate with
    > humans.


    That's simply not true.

    Indexing software, and software that needs to parse structures to
    present them to other software or to humans, need properly marked up
    content.

    > According to my way of thinking, if a browser causes a voice to say
    > "Heading level one" and following with the heading content, then that is
    > a default presentation of that part of the web page.


    That looks like information, not presentation.

    Daniele
    --
    Wanted: TEAC A-2300SX, Akai GX-4000D
    D.M. Procida, Mar 19, 2009
    #10
  11. dorayme

    dorayme Guest

    In article <5ac30$49c273cd$40cba7a7$>,
    "Jonathan N. Little" <> wrote:

    > David Segall wrote:
    > > Bergamot <> wrote:
    > >
    > >> D.M. Procida wrote:
    > >>> dorayme <> wrote:
    > >>>
    > >>>> Are you saying that better than
    > >>>>
    > >>>> <http://dorayme.netweaver.com.au/markup3.html>
    > >>>>
    > >>>> would be to make markup 1 look something similar?
    > >>> If that's really a list of events, then yes. But not if it's supposed to
    > >>> be tabular data.
    > >> That is just a list of events. Table markup is illogical because there
    > >> is no relationship between the cells in a row, other than the headers.
    > >> If it were tabular there would be some kind of relationship between data
    > >> cells.

    > >
    > > dorayme in her continued crusade to persuade us that black is grey and
    > > white is grey has come up with an arguable example of what you might
    > > define as nested lists that can also be defined as a table. There is
    > > an implied row heading column that could contain the entries "1st
    > > event", "2nd event", etc. To make it more realistic substitute
    > > Morning, Afternoon and Evening for 1st, 2nd and 3rd.

    >
    > Yep, next she might advocate not need for paragraphs in P elements, one
    > "should" use a one column table for that as well!



    Jonathan! You misunderstand me. I am a true and loyal fan of the
    paragraph element. I love the paragraph element. I really do. It is my
    very favourite HTML element. I scream with delight when I see it just as
    I saw folk screaming when Billy "Crash" Craddock appeared in a green
    outfit at my very first rock concert. But my love is based on study and
    consent, it is not some flighty infatuation based on peer pressure.

    <http://netweaver.com.au/semantics/whySemanticElements.php>

    But you do raise an interesting matter! And, well... er... yes...
    actually I do advocate a slight variation on what you are saying. If a
    table was to display various things that could be said about some one
    thing, it might well be the right thing to do to stick the different
    things that could be said in a table as items in cells. One paragraph
    (as understood in plain English) per cell. In this case using a P would
    offend against Ockham. Anyway, I won't argue this here. In a war, one
    must conserve forces and not open too many fronts. I face a determined
    enemy in the orthodoxy.

    --
    dorayme
    dorayme, Mar 19, 2009
    #11
  12. dorayme

    dorayme Guest

    In article
    <1iwu5dd.wfcwqqa8lm0bN%>,
    (D.M. Procida) wrote:

    > dorayme <> wrote:
    >
    > > We cannot manage the the appearance separately in the context of author
    > > styles being off. And it is only in this context that I am talking. I
    > > won't go into it here, but I believe if we *really* could manage the
    > > appearances with author styles, it would not really matter so much how
    > > we marked up, anything to get the appearance needed to communicate with
    > > humans.

    >
    > That's simply not true.
    >


    > Indexing software, and software that needs to parse structures to
    > present them to other software or to humans, need properly marked up
    > content.
    >


    Good point which I have addressed a bit elsewhere; I agreed that
    presentation does not matter when a device is merely, to take an
    imaginary example, counting how many P elements are used on a site.

    But your point is not quite on the button here. I said "anything to get
    the appearance needed to communicate with humans" and I meant directly
    to humans at the browser or device end.


    > > According to my way of thinking, if a browser causes a voice to say
    > > "Heading level one" and following with the heading content, then that is
    > > a default presentation of that part of the web page.

    >
    > That looks like information, not presentation.
    >

    I have gone on about this before, so sorry, everyone, to repeat. Your
    remark is a mistake in my opinion. Information cannot be communicated to
    human beings except by appearances (sounds, touch etc). It cannot be
    done by spooky stuff. ESP is unlikely. And if it were likely, I would
    have something to say about its default working CSS sheet. The fact is
    that humans get info via presentations of some kind or other. They do
    not get it direct or unmediated.

    > Daniele


    --
    dorayme
    dorayme, Mar 19, 2009
    #12
  13. dorayme

    D.M. Procida Guest

    dorayme <> wrote:

    > > > According to my way of thinking, if a browser causes a voice to say
    > > > "Heading level one" and following with the heading content, then that is
    > > > a default presentation of that part of the web page.

    > >
    > > That looks like information, not presentation.
    > >

    > The fact is that humans get info via presentations of some kind or other.
    > They do not get it direct or unmediated.


    But they can still distinguish between information and presentation, or
    should be able to.

    The very fact that humans "get info via presentations" entails that
    presentation is not information.

    Daniele
    --
    Wanted: TEAC A-2300SX, Akai GX-4000D
    D.M. Procida, Mar 19, 2009
    #13
  14. dorayme

    dorayme Guest

    In article
    <1iwuhgn.xx7cfx3x16z4N%>,
    (D.M. Procida) wrote:

    > dorayme <> wrote:
    >
    > > > > According to my way of thinking, if a browser causes a voice to say
    > > > > "Heading level one" and following with the heading content, then that is
    > > > > a default presentation of that part of the web page.
    > > >
    > > > That looks like information, not presentation.
    > > >

    > > The fact is that humans get info via presentations of some kind or other.
    > > They do not get it direct or unmediated.

    >
    > But they can still distinguish between information and presentation, or
    > should be able to.
    >

    I have no idea what this *really* means, sorry. Where are the paragraphs
    in this:

    <http://dorayme.netweaver.com.au/analyseThis.html>

    Had I put a simple <br> between the 'information bits' you would be able
    to distinguish them. If I marked up as in

    <http://members.optushome.com.au/droovies/opinion/drugLaws.html>

    and you turned off all author styles, you would see the information bits
    because the browser supplies a pretty good default stylesheet.

    If I marked up similarly as previous but had a better go at the barely
    conceivable idea of webpage that was deprived of both an author-CSS and
    also a default-CSS you might see something not even as good as this:

    <http://dorayme.netweaver.com.au/drugLaws_sans_dpr.html>

    With this latter, you would do better to view Source to read the
    article! But remember, source is not something that the audience knows
    anything about. That is strictly secret website maker business.


    > The very fact that humans "get info via presentations" entails that
    > presentation is not information.
    >


    I have never claimed that you can identify information with
    presentation. But I have flirted with the idea that information
    supervenes on presentation. But never mind this for now.

    The fact is that in the end, information gets communicated via
    presentation. The website maker must have a pretty good idea of what the
    *default* effect of his elements are going to be. By and large. The
    makers of HTML and browsers and devices also have a pretty good idea
    about *default* presentational matters without which the very creation
    of HTML elements would have been impossible.


    > Daniele


    --
    dorayme
    dorayme, Mar 19, 2009
    #14
  15. dorayme

    dorayme Guest

    In article <>,
    David Segall <> wrote:

    > Bergamot <> wrote:
    >
    > >D.M. Procida wrote:
    > >> dorayme <> wrote:
    > >>
    > >>> Are you saying that better than
    > >>>
    > >>> <http://dorayme.netweaver.com.au/markup3.html>
    > >>>
    > >>> would be to make markup 1 look something similar?
    > >>
    > >> If that's really a list of events, then yes. But not if it's supposed to
    > >> be tabular data.

    > >
    > >That is just a list of events. Table markup is illogical because there
    > >is no relationship between the cells in a row, other than the headers.
    > >If it were tabular there would be some kind of relationship between data
    > >cells.

    >


    ....

    > There is
    > an implied row heading column that could contain the entries "1st
    > event", "2nd event", etc. To make it more realistic substitute
    > Morning, Afternoon and Evening for 1st, 2nd and 3rd.



    You are right, David, to seek to understand what is going on by looking
    at what is implied. I scarcely know where to begin to show how wide of
    the mark Bergamot's criticism is.

    Let me start with his phrase "If that's really a list of events". I do
    not believe in the "really" bit. It is a fantasy of an extremist
    interpretation of matters HTML. The OP wanted to tell people what events
    were on when. What he wanted to communicate does not have one single
    fundamental structure. There is no spooky thing like this. It is not
    there. Even John Cleese's dead parrot had a more defined existence.

    There are alternatives on this one. Sure they sort of revolve around
    lists. That leaves the list type HTML elements to consider.
    UL, OL, and, guess what, the Table come to my mind immediately.

    The first option, nested list, I liked because it was the simplest. See
    the reasoning. The second used lists too. But it had to use two of them
    and headings too. Perhaps a bit clumsy. But it would not have been
    wrong! Nor would it have been illogical (I know Bergamot does not say
    this in respect to markup 2). It had a lousy default presentation too in
    my FF.

    I have said why I think the nested list was best in the context of an
    author CSS failure. It almost certainly went unnoticed that I thought a
    table would have been the best of all were it not for the lousy default
    styles. Just imagine a nicely styled one and tell me that is not the
    neatest and handiest 'at a glance' facility for visual browsers at least!

    Tables are *one* quite appropriate way to handle lists, whether they be
    single, ordered, or complex. The mantra of tabular data needs to be
    analysed and it is not what many people think.

    What makes it logical is that the list items (I use the phrase in the
    normal English meaning) pertain to the table heading. What makes it
    logical is that all human beings have no trouble with understanding
    this. What makes it logical is that in general it is not *as good as*
    using an HTML list; 'not as good as' does not equal 'illogical'. The
    general good practice is to use as specific a designed tool for the job
    in hand as possible. (It is not *illogical* to use a small screwdriver
    that happens to be in your hand for undoing a big screw if the screw is
    loose).

    In case you mistake this view of mine as condoning table markup for page
    layout, reflect on how typically there is no meaningful columns and
    headers, real or imaginable in such misuses of tables.

    In the fine traditions of Bible quoting:

    "Visual user agents allow sighted people to quickly grasp the structure
    of the table from the headings as well as the caption. A consequence of
    this is that captions will often be inadequate as a summary of the
    purpose and structure of the table from the perspective of people
    relying on non-visual user agents."

    "Authors should therefore take care to provide additional information
    summarizing the purpose and structure of the table using the summary
    attribute..."

    and

    nice stuff like:

    "However, user agents, particularly speech synthesizers, may want to
    factor out information common to several cells that are the result of a
    query. For instance, if the user asks "What did I spend for meals in San
    Jose?", the user agent would first determine the cells in question
    (25-Aug-1997: 37.74, 26-Aug-1997:27.28), then render this information. A
    user agent speaking this information might read it:..."

    I am saying, there is a lot to tables and I see nothing that dictates
    the simple Bergamot view. A nice table showing what is on when would
    have a structure and it would be one that humans would easily
    understand.

    I have not packed my bags to go to a re-education camp to learn how it
    has the "wrong" structure yet, comrades.

    --
    dorayme
    dorayme, Mar 20, 2009
    #15
  16. dorayme

    Guest

    On Thu, 19 Mar 2009 21:40:57 +0000,
    (D.M. Procida) wrote:


    >But they can still distinguish between information and presentation, or
    >should be able to.

    Au contraire.

    Since all information must be presented to be available to cognition*,
    one CANNOT distinguish information from presentation. What one CAN
    do is to _differentiate between presentations_ and have a response to
    and/or opinion and/or preference about the contribution (+ve/-ve) the
    presentation makes to the ease of assimilating the information.

    *Information can EXIST without presentation, but - as dorayme rightly
    observes - it is not available to humans (or indeed more generally to
    cognition of any sentient creature) except when presented in a manner
    that the recipient can utilise.

    As an example: describe the colour "red". You cannot present
    information about red (oh - wavelength of red light, the colour of
    blood, whatever) to me in a manner I can comprehend without visually
    presenting me with "redness". In this case, the only way I can
    comprehend the information about 'red' is to have it presented
    visually. Whereas, I can understand other classes of information in
    other ways - the information about a "middle C" (frequency of
    vibration) must be presented aurally.

    In the case of "red" and "middle C" we even have two things which can
    be defined using the same units of measurement (albeit of different
    things) - wavelength or frequency - but which must be presented**
    differently to be comprehensible.

    **Or at least must be referenced to something already presented in a
    manner that can be comprehended: "middle C is between X and Y". No,
    I have NO idea what notes X and Y are! A & E?. (Aren't the mnemonics
    something like: FACE? EGBDF?)

    >The very fact that humans "get info via presentations" entails that
    >presentation is not information.


    The two are inseparable in the real world. By which I mean the world
    we experience. We might not experience the same "reality" in the same
    way, either.
    , Mar 20, 2009
    #16
  17. dorayme

    Geoff Berrow Guest

    Message-ID: <> from
    contained the following:

    >**Or at least must be referenced to something already presented in a
    >manner that can be comprehended: "middle C is between X and Y". No,
    >I have NO idea what notes X and Y are! A & E?. (Aren't the mnemonics
    >something like: FACE? EGBDF?)


    Middle C is between B and C#

    :)
    --
    Regards,
    Geoff Berrow
    http://www.slipperyhill.co.uk - Blue grass, blues, barn dance
    http://4theweb.co.uk - Web design, development and hosting
    Geoff Berrow, Mar 20, 2009
    #17
  18. dorayme

    D.M. Procida Guest

    <> wrote:

    > On Thu, 19 Mar 2009 21:40:57 +0000,
    > (D.M. Procida) wrote:
    >
    >
    > >But they can still distinguish between information and presentation, or
    > >should be able to.

    > Au contraire.
    >
    > Since all information must be presented to be available to cognition*,
    > one CANNOT distinguish information from presentation. What one CAN
    > do is to _differentiate between presentations_ and have a response to
    > and/or opinion and/or preference about the contribution (+ve/-ve) the
    > presentation makes to the ease of assimilating the information.


    Of course one can distinguish.

    The information is: "The building is on fire. You must leave
    immediately."

    The presentation is: shouting, urgent voices.

    The mere fact that one can have different presentation of the same
    information means that they must be distinguishable. If they weren't,
    then affecting the mode of presentation would affect the semantic
    content.

    > *Information can EXIST without presentation, but - as dorayme rightly
    > observes - it is not available to humans (or indeed more generally to
    > cognition of any sentient creature) except when presented in a manner
    > that the recipient can utilise.


    Unless they have what Kant calls intellectual intuition (such as he
    speculates hypothetical beings such as angels might have), but that -
    even if it's a more interesting and important question than the use of
    HTML - is hardly the point here.

    The point is that HTML offers mechanisms for the informative structuring
    and labelling of content, and if we are going to use it to its best
    effect, then we should exploit those mechanisms to the fullest extent
    possible, and not allow questions of "how it looks" to affect how we use
    it.

    Daniele
    D.M. Procida, Mar 20, 2009
    #18
  19. dorayme

    Guest

    On Fri, 20 Mar 2009 09:45:14 +0000,
    (D.M. Procida) wrote:

    ><> wrote:
    >
    >> On Thu, 19 Mar 2009 21:40:57 +0000,
    >> (D.M. Procida) wrote:
    >>
    >>
    >> >But they can still distinguish between information and presentation, or
    >> >should be able to.

    >> Au contraire.
    >>
    >> Since all information must be presented to be available to cognition*,
    >> one CANNOT distinguish information from presentation. What one CAN
    >> do is to _differentiate between presentations_ and have a response to
    >> and/or opinion and/or preference about the contribution (+ve/-ve) the
    >> presentation makes to the ease of assimilating the information.

    >
    >Of course one can distinguish.
    >
    >The information is: "The building is on fire. You must leave
    >immediately."
    >
    >The presentation is: shouting, urgent voices.


    You've come much further up the up the hierarchy from where I was
    inhabiting to make your point. You pot may be valid at the level you
    choose, I have no idea. (I am not working with our analogy at your
    level, but with mine at mine).

    At my level, your analogy reduces to something like: The information
    is "vibration in air". The presentation depends on the presence of a
    mechanism to decode the wave pattern in the air. AKA "ears" (although
    there are other sensory organs in other life-forms that would not
    typically be called "ears". "Feet" for example.

    Only after that can one decide (from experience) that the presentation
    is "shouting". To some people, normal conversational Arabic
    (depending on dialect) sounds very aggressive. Only then, if one is
    familiar with the language in which the "shouting" occurs, can one
    determine that the message being conveyed is "Fire". As opposed to
    "Shoot!". The presence of fire or flame might assist: but then that
    information is presented via means of the "eye".



    >The mere fact that one can have different presentation of the same
    >information means that they must be distinguishable. If they weren't,
    >then affecting the mode of presentation would affect the semantic
    >content.



    That's what I said. Presentations are distinguishable one from
    another. They may obscure or appear to change the information, meaning
    that one interprets or perceives the information differently, taking
    it to be different, even if it is the same.

    >
    >The point is that HTML offers mechanisms for the informative structuring
    >and labelling of content, and if we are going to use it to its best
    >effect, then we should exploit those mechanisms to the fullest extent
    >possible, and not allow questions of "how it looks" to affect how we use
    >it.


    I wouldn't disagree with the fundamental thrust of that - especially
    as my knowledge of HTML is less developed than my understanding of how
    animals hear with their feet :) - except to say that how things look
    DOES need to be taken into account in how we use things. Not a lot
    of point having a red button for "boom!" and a green button for "no
    boom!" if the operator of the buttons is colour blind.
    , Mar 20, 2009
    #19
  20. dorayme

    Guest

    On Fri, 20 Mar 2009 09:12:48 +0000, Geoff Berrow
    <> wrote:

    >>**Or at least must be referenced to something already presented in a
    >>manner that can be comprehended: "middle C is between X and Y". No,
    >>I have NO idea what notes X and Y are! A & E?. (Aren't the mnemonics
    >>something like: FACE? EGBDF?)

    >
    >Middle C is between B and C#
    >
    >:)


    So much for my primary school music education then. :(

    Wouldn't it more accurately be described as being between Cb and C#,
    tho? I've probably just reinforced my display of ignorance. Oh well:
    that's all a matter of presentation, not information anyway.
    , Mar 20, 2009
    #20
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Patrick McCourt

    Stack & Singly Linked List Data Structures

    Patrick McCourt, May 24, 2004, in forum: Java
    Replies:
    2
    Views:
    920
    Kenneth P. Turvey
    May 24, 2004
  2. Andrew

    linked list of structures

    Andrew, Aug 11, 2003, in forum: C Programming
    Replies:
    0
    Views:
    268
    Andrew
    Aug 11, 2003
  3. tweak
    Replies:
    14
    Views:
    2,769
    Eric Sosman
    Jun 11, 2004
  4. Alfonso Morra
    Replies:
    11
    Views:
    703
    Emmanuel Delahaye
    Sep 24, 2005
  5. len
    Replies:
    1
    Views:
    327
    Dennis Lee Bieber
    Dec 13, 2007
Loading...

Share This Page