In article <48975$48b5eb19$40cba7c6$19859@NAXS.COM>,
"Jonathan N. Little" <lws4art@central.net> wrote:

> dorayme wrote:
> > In article <doraymeRidThis-D938FF.09323328082008@web.aioe.org>,
> > dorayme <doraymeRidThis@optusnet.com.au> wrote:
> >
> >> The crucial case is between r and q (see discussion with Ben). I claim
> >> they are not *significantly* different and if they are what is the
> >> difference?
> >
> > O and I should have added "and if there is a difference, what is its
> > true significance".
> >
>
> Not sure how to explain it any more clearly, but essentially list data
> is *linear* or *sequential* whereas tabular data is "2-dimensional" with
> a positional relationship in rows and columns. If you cannot grasp the
> concept I am afraid you are doom to confusion ;-)

The insinuation that you know something about this matter that I don't
would be more impressive if you showed the slightest explained agreement
with any of the detailed things I have said. This would show how far
along what road you travel with me and exactly the point at which you
diverge.

Your suggestion that I am doomed if I don't grasp what satisfies you and
others should be beneath you.

In any case, your analysis of what is dooming me is clearly wrong
because I accept that cells in a well made html table have a
relationship to each other across rows and sometimes across columns and
sometimes across both. I also accept that list data items are
sequential.

You continue to ignore the crucial and relevant matters and keep
repeating the simple simon met a pieman stuff which I start with rather
than dispute. Anyway, thanks for trying to contribute to the discussion.

--
dorayme

Jonathan N. Little wrote:

> Not sure how to explain it any more clearly, but essentially list data
> is *linear* or *sequential* whereas tabular data is "2-dimensional"
> with a positional relationship in rows and columns. If you cannot
> grasp the concept I am afraid you are doom to confusion ;-)

Tabular data is, generally speaking, n-dimensional, though in web authoring
and styling we are mainly interested in the cases n = 1 and n = 2. For n =
1, tabular data reduces to a list (linear, sequential structure), which can
be described within the HTML table concept as a table (n = 2) with one
column only, or with one row only. A list can, of course, also be presented
using <ul> or <ol> (or <dir> or <menu> to get pointlessly wild) or even just
a sequence of elements, like <div> elements or <p> elements. In practice,
people tend to select the markup on presentational grounds rather than
anything else. If you want bullets, you use <ul>, etc., even though you
_could_, in theory at least, use any of the markup approaches on some
logical or other grounds and then use CSS for styling.

The suggestion to use a table for bolding numbers of list items means that a
list is replaced by a 2-dimensional table, with numbers in one column and
list items in another. I don't see why this would be unsound, even though
it's in practice a workaround when one just wants a list with bold numbers.
After all, the more importance the numbers have - i.e. the more they
describe essential properties of the data, as opposite to being merely
running numbers that could be replaced e.g. by a, b, c, ... or even by
bullets - the more we can argue that they _should_ be explicit, present in
the actual data, and not just generated by presentational tools like CSS.

By the way, the case n = 3 is something that could be presented using CSS,
in a sense. You could have a collection of similar tables, positioned at the
same place, with different z-index values. This would of course mean that
only one "slice" would be visible at any given time, and switching between
"slices" would have to be handled outside CSS. (Well, :hover could be used
to switch between _two_ slices on mouseover.)

Yucca

On 2008-08-27, dorayme <doraymeRidThis@optusnet.com.au> wrote:
> In article <slrngbb03n.hvj.spamspam@bowser.marioworld>,
> Ben C <spamspam@spam.eggs> wrote:
>
>> >> On 2008-08-26, dorayme <doraymeRidThis@optusnet.com.au> wrote:
> ...
>> > What to call things is one thing. What things are is something else. The
>> > two are related sometimes by trying one's best to fit the first to the
>> > second. I am wondering if you think I am saying anything else but that
>> > it is ok in the special sense that I have outlined at length a few times
>> > now?
>>
>> Yes it is OK in the special sense, but does that make it OK?
>>
>> Before the word "orange" was invented, people just made do with "red".
>> That's why people who have orange hair are still said to have "red" hair.
>>
>> But it would not be OK to describe an orange traffic light as red, even
>> though you could just about argue that in a special sense orange is a
>> sort of red. The fact that there is a word for orange, and also another
>> light which is really red, makes all the difference.
>>
>
> I am not suggesting anyone should call an ol a table or a specially
> constructed table, an ordered list. I am attempting - and failing - to
> show you something I think is true about the reality. I am not
> suggesting that it is good practice to substitute a table for an ordered
> list whenever an author feels like it for no particular good reason.

You are suggesting that one time it's OK is when the table is easier to
style?

>> > In particular that it is ok to use a table instead of an ordered
>> > list if it is at all more convenient in a context and that it is ok not
>> > for the reason that it is ok to be a bit naughty or to cut corners but
>> > for the reason there is nothing here that is naughty at all, there is no
>> > corner, there is nothing that is not impeccably ok.
>> >
>> > Perhaps this idea is simply unable to be communicated! Perhaps I am
>> > simply wrong?
>>
>> Depends what you think is not OK: using a table for presentation, or
>> using a table for data that aren't "tabular"?
>>
> You are basically saying I am completely wrong about the most crucial
> thing and offering crumbs as compensation.

I'm just trying to clarify the criteria for supposed misuse of a table.

> Not interested in the crumbs but thanks anyway! Look, I would much
> rather know that I am missing a crucial thing here than any attempts
> at diplomacy.

I don't know of any crucial thing you're missing.

[...]
> I am quite serious about this, you seem basically to be dismissing the
> real heart of what I am claiming which is that the tables that I have
> had in mind all the time are not mere presentational devices. But I seem
> unable to get this across. The penny is not dropping on one of our
> sides. Naturally everyone will suspect it is mine because you humans
> tend to stick together when it comes to the crunch.

Well I thought I had understood roughly what you were saying: that an
ordered list is a kind of table. I don't disagree with that part.

Does that make it OK to use a table because it's easier to style? Well
it's good enough for me, but then I'm not a purist anyway. I can't tell
the difference between the teachings of a mythical purist and those of a
real purist.

> From where I am standing you seem not to have taken any notice of my oft
> repeated words about the left col heading in some 2 col algorithmic
> table presentation.

I thought I got that part.

> The left col data, if you like, has a meaning! This meaning is crucially
> bound up with the data to the right. You cannot move the numbers that
> indicate the order in which a procedure is to be carried out. You cannot
> move "1." or "this is the first step" to the last row that deals with
> triggering the timer. You would spoil the meaning of the table, you
> would break the relationships that are intended.

Yes, that's OK.

> You surely do not have some naive assumption that tabular means all
> numbers and dimensions!

No. I'm sceptical of the idea that there is any authoritative definition
of an abstract table.

In article <slrngbcl7a.mm2.spamspam@bowser.marioworld>,
Ben C <spamspam@spam.eggs> wrote:

> On 2008-08-27, dorayme <doraymeRidThis@optusnet.com.au> wrote:
> > In article <slrngbb03n.hvj.spamspam@bowser.marioworld>,
> > Ben C <spamspam@spam.eggs> wrote:
> >
> >> >> On 2008-08-26, dorayme <doraymeRidThis@optusnet.com.au> wrote:
> > ...
> >> > What to call things is one thing. What things are is something else. The
> >> > two are related sometimes by trying one's best to fit the first to the
> >> > second. I am wondering if you think I am saying anything else but that
> >> > it is ok in the special sense that I have outlined at length a few times
> >> > now?
> >>
> >> Yes it is OK in the special sense, but does that make it OK?
> >>
....
> > I am not suggesting anyone should call an ol a table or a specially
> > constructed table, an ordered list. I am attempting - and failing - to
> > show you something I think is true about the reality. I am not
> > suggesting that it is good practice to substitute a table for an ordered
> > list whenever an author feels like it for no particular good reason.
>
> You are suggesting that one time it's OK is when the table is easier to
> style?
>
At the very least it *is* OK to substitute a table for an ordered list
if you have a pesky styling issue. Yes. But no, not the *only* time I
imagine. I am open minded on this further question. It is probably ok on
other occasions too. My confidence is based on something solid as I see
it, I am not sneaking about and slipping in a table for an OL when no
one is looking as if it is a bad thing.

It is not a bad thing because from the one is borne the other. They
share natures though they have grown apart somewhat.

You see, I don't have the simple idea of a list being an orange and a
table being a crate, different things totally. I imagine logical and
semantic connections between the two. In fact, I could bore you for
hours about my idea that a table is really just a convenient way of
grouping lists of things, that it is *made* of lists!

But I am scared to tell you about this. A creature like me, alone in
this world, is very vulnerable. I get a touch riled when earthlings seem
so incredibly resistant to seeing further than the orange and crate
appearance. Actually I don't think you fit in this category but I don't
blame you to be wary. <g>

In the beginning are lists. And He saw they were good. But He also
saw that his creatures were having trouble when things got a bit
complicated and there were many lists on the page or many on many pages.
So He created tables to gather them all about for the benefit of Man.
And He saw this was good and He was pleased. He was also too tired to
respond immediately to urgings for databases and MySQL. But He
eventually even got around to this...

All was borne from man's and God's desire to list the objects of the
created world into sets. (Kiwis, for example, are fond of counting and
listing sheep). From the humble aboriginal unordered list and the desire
of Man and God for order came the rest of the furnishings of the world.

(How on earth does Chris Johnson indent things so well in his posts, I
wanted the last two paras indented but could not quickly see a simple
way)

> >> > In particular that it is ok to use a table instead of an ordered
> >> > list if it is at all more convenient in a context and that it is ok not
> >> > for the reason that it is ok to be a bit naughty or to cut corners but
> >> > for the reason there is nothing here that is naughty at all, there is no
> >> > corner, there is nothing that is not impeccably ok.
> >> >
> >> > Perhaps this idea is simply unable to be communicated! Perhaps I am
> >> > simply wrong?
> >>
> >> Depends what you think is not OK: using a table for presentation, or
> >> using a table for data that aren't "tabular"?
> >>
> > You are basically saying I am completely wrong about the most crucial
> > thing and offering crumbs as compensation.
>
> I'm just trying to clarify the criteria for supposed misuse of a table.
>

No one has secret knowledge so well developed that can rationally
challenge the use of a table instead of a list for communicating
material that would be equally appreciated in either form by the peoples
of the world. There is no secret knowledge that anyone has, no matter
how much the priests insinuate otherwise, to make it wrong to substitute
a simple 2 col table for an ordered list if using an OL is going to
cause you bother, either in styling or other problems that I leave as an
open question (the very reverse of a religious stance).


> > Not interested in the crumbs but thanks anyway! Look, I would much
> > rather know that I am missing a crucial thing here than any attempts
> > at diplomacy.
>
> I don't know of any crucial thing you're missing.
>
> [...]
> > I am quite serious about this, you seem basically to be dismissing the
> > real heart of what I am claiming which is that the tables that I have
> > had in mind all the time are not mere presentational devices. But I seem
> > unable to get this across. The penny is not dropping on one of our
> > sides. Naturally everyone will suspect it is mine because you humans
> > tend to stick together when it comes to the crunch.
>
> Well I thought I had understood roughly what you were saying: that an
> ordered list is a kind of table. I don't disagree with that part.
>

Perhaps it would be better to describe my view as that a table is a kind
of collection of ordered lists and that the simple case is a limiting
one. Use either in principle! But normally, use OL because it is the
specific tool for the job and if there is no special reason to use a
table, don't.

There are, if you start to think about it, probably lots of situations
to turn away from OLs on a web page to a table. An author might see get
fed up of fiddling about with a complicated list that has as parts
further ols and/or uls. He might well unify all this into a table.

I cannot go into everything Ben, I am just one being trying to earn a
living here.


> Does that make it OK to use a table because it's easier to style? Well
> it's good enough for me, but then I'm not a purist anyway. I can't tell
> the difference between the teachings of a mythical purist and those of a
> real purist.
>
> > From where I am standing you seem not to have taken any notice of my oft
> > repeated words about the left col heading in some 2 col algorithmic
> > table presentation.
>
> I thought I got that part.
>
> > The left col data, if you like, has a meaning! This meaning is crucially
> > bound up with the data to the right. You cannot move the numbers that
> > indicate the order in which a procedure is to be carried out. You cannot
> > move "1." or "this is the first step" to the last row that deals with
> > triggering the timer. You would spoil the meaning of the table, you
> > would break the relationships that are intended.
>
> Yes, that's OK.
>
> > You surely do not have some naive assumption that tabular means all
> > numbers and dimensions!
>
> No. I'm sceptical of the idea that there is any authoritative definition
> of an abstract table.

I will return at some stage to try to demonstrate further the idea of
semantic equivalence. The meaning of any structure is crucially bound up
with the information conveyed and if there are two ways to convey the
same information, there is nothing but practical considerations rather
than deep semantic reasons for choosing the vehicle.

But, and this is a very hard idea for some people to get their heads
about, there is no meaning that floats independently, it is all about
how the vehicles are used. And I have time and again shown how some
vehicles are interchangeable for some jobs. That means they are
semantically equivalent. It is not a mystery. There are no ghostly
meanings that hover essentially about ordered lists, and different
ghosts to tables. It is all about using things sensibly and responding
to practical situations. The priests hate this sort of thing because it
undermines their set rigid spiels.

I am sorry all this is not in earthling technospeak. But I try me best.

--
dorayme

On 2008-08-28, dorayme <doraymeRidThis@optusnet.com.au> wrote:
> In article <slrngbcl7a.mm2.spamspam@bowser.marioworld>,
> Ben C <spamspam@spam.eggs> wrote:
[...]
> In the beginning are lists. And He saw they were good. But He also
> saw that his creatures were having trouble when things got a bit
> complicated and there were many lists on the page or many on many
> pages. So He created tables to gather them all about for the
> benefit of Man. And He saw this was good and He was pleased. He
> was also too tired to respond immediately to urgings for databases
> and MySQL. But He eventually even got around to this...
>
> All was borne from man's and God's desire to list the objects of
> the created world into sets. (Kiwis, for example, are fond of
> counting and listing sheep). From the humble aboriginal unordered
> list and the desire of Man and God for order came the rest of the
> furnishings of the world.
>
> (How on earth does Chris Johnson indent things so well in his posts, I
> wanted the last two paras indented but could not quickly see a simple
> way)

In vim you just use the ">" normal-mode command.

[...]
> I will return at some stage to try to demonstrate further the idea of
> semantic equivalence. The meaning of any structure is crucially bound up
> with the information conveyed and if there are two ways to convey the
> same information, there is nothing but practical considerations rather
> than deep semantic reasons for choosing the vehicle.
>
> But, and this is a very hard idea for some people to get their heads
> about, there is no meaning that floats independently, it is all about
> how the vehicles are used. And I have time and again shown how some
> vehicles are interchangeable for some jobs. That means they are
> semantically equivalent. It is not a mystery. There are no ghostly
> meanings that hover essentially about ordered lists, and different
> ghosts to tables. It is all about using things sensibly and responding
> to practical situations.

Many concepts overlap and are fuzzy around the edges. It's a matter of
consensus what you can call what in any context and there isn't always a
clear-cut answer.

That's why I don't think you can justify _absolutely_ using a table
instead of an ol for presentation reasons. I think it's OK, but you
can't guarantee close-to-100% consensus. It's not like calling a spade a
spade.

In article <slrngbfgrl.5ec.spamspam@bowser.marioworld>,
Ben C <spamspam@spam.eggs> wrote:

> On 2008-08-28, dorayme <doraymeRidThis@optusnet.com.au> wrote:
> > In article <slrngbcl7a.mm2.spamspam@bowser.marioworld>,
> > Ben C <spamspam@spam.eggs> wrote:
> [...]
....
> > I will return at some stage to try to demonstrate further the idea of
> > semantic equivalence. The meaning of any structure is crucially bound up
> > with the information conveyed and if there are two ways to convey the
> > same information, there is nothing but practical considerations rather
> > than deep semantic reasons for choosing the vehicle.
> >
....
>
> Many concepts overlap and are fuzzy around the edges. It's a matter of
> consensus what you can call what in any context and there isn't always a
> clear-cut answer.
>
> That's why I don't think you can justify _absolutely_ using a table
> instead of an ol for presentation reasons. I think it's OK, but you
> can't guarantee close-to-100% consensus. It's not like calling a spade a
> spade.

No, it is not like calling a spade a spade. But it is rather like using
either of two sentences that equally convey a piece of information
whether or not the one can be transformed syntactically into the other
or not.

The paradoxical bit of my expressed view is that lists (unordered) are a
more primitive notion than tables. They are just a bunch of things
collected together. And yet I speak of analysing some lists in terms of
tables!

Put it another way, I say an ordered list is a 2 col table. I say that
tables are really a way of ordering lists in a convenient manner to show
relationships between the list items.

It looks as if I am saying opposing things. But there is no real
trouble. It is two sides of a coin. Tables are a nice construct to
collect lists together and show relationships between them.

But the ordered list is a different beast to the unordered one. It is a
construct that can be understood as a shorthand way of representing a
special kind of table.

An ordered list (where the numbers are used to inform about the order)
is an amalgam of two lists. Yes, two unordered, primitive lists, don't
be fooled by the numbers. Imagine the following table.

3. Press the trigger
1. Load the gun
2. Aim

The left column is neither ordered nor unordered technically. The idea
simply does not apply. It is just a table that shows the relationship of
one list to another list. The pairings matter a great deal, left to
right.

It is not how we would draw it, but we could without losing the vital
information. Normally, we would have the commonsense to write it in a
different order. But if you understand that the numbers indicate order
(and the column heading can be either explicit or implied to assist
this) then the table is perfectly meaningful and useful.

It would be better to draw the table up in the order in which operations
are to be carried out. Imagine this done.

1. Load the gun
2. Aim
3. Press the trigger

OK, we now have a table differently arranged but giving *exactly* the
same info. But easier to read now.

Now it might be nice to have a way to write this commonsensically set
out table. Enter a shortcut idea... the ordered list! This shortcut has
a few rules of its own, it demands to be a shortcut of a 2 col table
which has been constructed for human understanding and in human
understanding you don't mess about, you put the first thing to be done
first or at least you go in logical order... And in various environments
the numbers happen automatically. Clever little thing an OL. But it is
just a convenience in the end.

There is nothing, magical or otherwise, that an ol is semantically that
is different to a specially formatted table.

And if you have this view of things, it *can* be *absolutely* reasonable
to use a table instead of an ordered list at the slightest sign of
practical bother with the supposedly more convenient shortcut.

That's what I reckon anyway.

Cheers.

--
dorayme

On 2008-08-28, dorayme wrote:
....
> (How on earth does Chris Johnson indent things so well in his posts, I
> wanted the last two paras indented but could not quickly see a simple
> way)

GNU emacs text-mode with auto-fill-mode. I indent the first line
and lines automatically wrap with the same indentation.

--
Chris F.A. Johnson <http://cfaj.freeshell.org>
================================================== =================
Author:
Shell Scripting Recipes: A Problem-Solution Approach (2005, Apress)

In article <52425$48b82b9d$cef88ba3$4511@TEKSAVVY.COM>,
"Chris F.A. Johnson" <cfajohnson@gmail.com> wrote:

> On 2008-08-28, dorayme wrote:
> ...
> > (How on earth does Chris Johnson indent things so well in his posts, I
> > wanted the last two paras indented but could not quickly see a simple
> > way)
>
> GNU emacs text-mode with auto-fill-mode. I indent the first line
> and lines automatically wrap with the same indentation.

So... OK. If I get to need to do this more, I might have to see about
getting a package suitable for Mac X.

--
dorayme

dorayme <doraymeRidThis@optusnet.com.au> writes:

> In article <52425$48b82b9d$cef88ba3$4511@TEKSAVVY.COM>,
> "Chris F.A. Johnson" <cfajohnson@gmail.com> wrote:
>
>> On 2008-08-28, dorayme wrote:
>> ...
>> > (How on earth does Chris Johnson indent things so well in his posts, I
>> > wanted the last two paras indented but could not quickly see a simple
>> > way)
>>
>> GNU emacs text-mode with auto-fill-mode. I indent the first line
>> and lines automatically wrap with the same indentation.
>
> So... OK. If I get to need to do this more, I might have to see about
> getting a package suitable for Mac X.

Are you aware of Aquamacs?

<http://www.aquamacs.org>

sherm--

--
My blog: http://shermspace.blogspot.com
Cocoa programming in Perl: http://camelbones.sourceforge.net

In article <m1od3ba6ju.fsf@dot-app.org>,
Sherm Pendley <spamtrap@dot-app.org> wrote:

> dorayme <doraymeRidThis@optusnet.com.au> writes:
>
> > In article <52425$48b82b9d$cef88ba3$4511@TEKSAVVY.COM>,
> > "Chris F.A. Johnson" <cfajohnson@gmail.com> wrote:
> >
> >> On 2008-08-28, dorayme wrote:
> >> ...
> >> > (How on earth does Chris Johnson indent things so well in his posts, I
> >> > wanted the last two paras indented but could not quickly see a simple
> >> > way)
> >>
> >> GNU emacs text-mode with auto-fill-mode. I indent the first line
> >> and lines automatically wrap with the same indentation.
> >
> > So... OK. If I get to need to do this more, I might have to see about
> > getting a package suitable for Mac X.
>
> Are you aware of Aquamacs?
>
> <http://www.aquamacs.org>
>

Thanks, I did just see about this when spending a few mins looking up
about what Chris was saying.

--
dorayme