WYSIWYG vs. WYSIWYM and semantic usage?

Hi Simon

Sorry for not answering this earlier, but it's a pretty long post

IMHO it's just that our customers ask for WYSIWYG and not WYSIYWYM. They
mostly do not care about correct semantic usage, but with BXE you can at
least force it somehow :)

I know LyX and wrote my diploma thesis with it. It's great and very
useful, but I don't see much of a differecne between that and a strict
BXE. BXE shows what you mean in a manner that it will look like what you
will get on the webpage. if you reuse the content on a different palce
(PDA, PDF, whatever), it will of course look differently, but I think
the users can abstract that.

How would a WYSIWYM Editor look like in your opinion? On the web, I mean :)

chregu

Hi!
The way of the future of information seems to be semantic markup, both
when speaking of the web and when speaking of word processing. That is
marking up content according to what it is rather than what it should
look like. When speaking of word processing there seem to be two
approaches to this, the ordinary WYSIWYG (What You See Is What You Get)
approach which ordinary word processors have, and the pure semantic
approach aka. WYSIWYM (What You See Is What You Mean).
WYSIWYG:s usually offer a possibility to mark up content with style
classes rather than directly saying how each letter, word or passage
should look. (Kind of like using CSS classes instead of in-line CSS or
FONT-tags.) If you choose to use this possibility explicitly one could
say that you're marking up text semantically (as long as the names of
the classes are of a semantic nature). Then there's also at least one
purely semantic word processor called LyX, which is a frontend to
LaTeX. LyX has a so called WYSIWYM (What You See Is What You Mean)
approach, which means that it doesn't even allow the user to see the
formatting of the text while typing, encouraging the user to think
semantically.
When speaking of web content the Bitflux Editor embraces this approach
as a WYSIWYG, yet encouraging the user to use semantic markup more than
any other WYSIWYG I've seen.
My question is: What advantages and disadvantages does this approach
have compared to a pure WYSIWYM?
I think it may be an advantage for the user to have a feeling of what
the content is going to look like already at the time of typing it. But
on the other hand it may tempt the user into using inappropriate markup
in trying to achieve a sertain look, which anyways should be up to the
CSS-designer to decide (at least for the sake of consistency). Another
problem occurs if one is going to reuse the conent in different medias
with different formatting (ie. different styling for print), then the
WYSIWYG is nothing but an illusion.
What experiances or thoughts do you have on this? Do you btw. have
experiance with any web-based WYSIWYM:s?
cheers, Simon

Simon Rönnqvist

2005-11-25 22:23:19 UTC

On Friday 25 November 2005 10:13, Christian Stocker wrote:

Hi!

Post by Christian Stocker
Sorry for not answering this earlier, but it's a pretty long post

No problem.
It was just something I started to think about when thinking about how I
should focus my thesis. But your answer can be useful for citation on why to
use a WYSIWYG.

Post by Christian Stocker
IMHO it's just that our customers ask for WYSIWYG and not WYSIYWYM. They
mostly do not care about correct semantic usage, but with BXE you can at
least force it somehow :)

That's true... BXE is kind of a WYSIWYG&WYSIWYM combo, since it's
restrictive like a WYSIWYM (unlike ordinary word processors).
Of course the customer is "always right", and if they're absolutely sure
that they want a WYSIWYG, then one has to give them that. But on the other
hand it may be that they'd be satisfied with a WYSIWYM once they'd get to
know about it.
As a matter of fact, it'd be really cool to do a quantitative (a large scale
questionare) study in which one would survey the pros and cons of a rather
restrictive WYSIWYG like BXE and a pure (LyX-like) WYSIWYM. But time and
capacity is limited for me, so the closest I could get would be to interview
one single customer using BXE and to check in what way she uses it.

Post by Christian Stocker
I know LyX and wrote my diploma thesis with it. It's great and very
useful, but I don't see much of a differecne between that and a strict
BXE.

I'm actually planning to use it too for my thesis, I'd feel like a hypocrit
using anything but that, since my thesis covers separation of conent and
presentation and semantics. ;-) I just hope that I'll manage get the
formatting to fit my school's requirements.

Post by Christian Stocker
BXE shows what you mean in a manner that it will look like what you
will get on the webpage. if you reuse the content on a different palce
(PDA, PDF, whatever), it will of course look differently, but I think
the users can abstract that.

My fear (as I mentioned earlier) is that people would be tempted to choose
the semantic marking according to looks instead of meaning.

Post by Christian Stocker
How would a WYSIWYM Editor look like in your opinion? On the web, I mean :)

Kind of like LyX. Or let's say kind of like BXE without showing how the page
would look, but instead just indicating the semantics. I think there should
be some kinds of web-based WYSIWYM:s around, but I haven't checked them out
yet.

cheers, Simon

Hi!
The way of the future of information seems to be semantic markup, both
when speaking of the web and when speaking of word processing. That is
marking up content according to what it is rather than what it should
look like. When speaking of word processing there seem to be two
approaches to this, the ordinary WYSIWYG (What You See Is What You Get)
approach which ordinary word processors have, and the pure semantic
approach aka. WYSIWYM (What You See Is What You Mean).
WYSIWYG:s usually offer a possibility to mark up content with style
classes rather than directly saying how each letter, word or passage
should look. (Kind of like using CSS classes instead of in-line CSS or
FONT-tags.) If you choose to use this possibility explicitly one could
say that you're marking up text semantically (as long as the names of
the classes are of a semantic nature). Then there's also at least one
purely semantic word processor called LyX, which is a frontend to
LaTeX. LyX has a so called WYSIWYM (What You See Is What You Mean)
approach, which means that it doesn't even allow the user to see the
formatting of the text while typing, encouraging the user to think
semantically.
When speaking of web content the Bitflux Editor embraces this approach
as a WYSIWYG, yet encouraging the user to use semantic markup more than
any other WYSIWYG I've seen.
My question is: What advantages and disadvantages does this approach
have compared to a pure WYSIWYM?
I think it may be an advantage for the user to have a feeling of what
the content is going to look like already at the time of typing it. But
on the other hand it may tempt the user into using inappropriate markup
in trying to achieve a sertain look, which anyways should be up to the
CSS-designer to decide (at least for the sake of consistency). Another
problem occurs if one is going to reuse the conent in different medias
with different formatting (ie. different styling for print), then the
WYSIWYG is nothing but an illusion.
What experiances or thoughts do you have on this? Do you btw. have
experiance with any web-based WYSIWYM:s?
cheers, Simon

--
bx-editor-users mailing list
bx-editor-***@lists.bitflux.ch
http://lists.bitflux.ch/cgi-bin/listinfo/bx-editor-users

Christian Stocker

2005-11-28 07:43:59 UTC

My fear (as I mentioned earlier) is that people would be tempted to choose
the semantic marking according to looks instead of meaning.

Just give it meaningful names, remove bold/etc, and they can't do that :)

Post by Christian Stocker
How would a WYSIWYM Editor look like in your opinion? On the web, I mean :)

Maybe the tag mode (without wysiwyg formatting) would be something for
you then. It has to be improved (it doesn't do live updates right now),
but that could be a way to what you want

chregu

Post by Simon RÃ¶nnqvist
cheers, Simon

Hi!
The way of the future of information seems to be semantic markup, both
when speaking of the web and when speaking of word processing. That is
marking up content according to what it is rather than what it should
look like. When speaking of word processing there seem to be two
approaches to this, the ordinary WYSIWYG (What You See Is What You Get)
approach which ordinary word processors have, and the pure semantic
approach aka. WYSIWYM (What You See Is What You Mean).
WYSIWYG:s usually offer a possibility to mark up content with style
classes rather than directly saying how each letter, word or passage
should look. (Kind of like using CSS classes instead of in-line CSS or
FONT-tags.) If you choose to use this possibility explicitly one could
say that you're marking up text semantically (as long as the names of
the classes are of a semantic nature). Then there's also at least one
purely semantic word processor called LyX, which is a frontend to
LaTeX. LyX has a so called WYSIWYM (What You See Is What You Mean)
approach, which means that it doesn't even allow the user to see the
formatting of the text while typing, encouraging the user to think
semantically.
When speaking of web content the Bitflux Editor embraces this approach
as a WYSIWYG, yet encouraging the user to use semantic markup more than
any other WYSIWYG I've seen.
My question is: What advantages and disadvantages does this approach
have compared to a pure WYSIWYM?
I think it may be an advantage for the user to have a feeling of what
the content is going to look like already at the time of typing it. But
on the other hand it may tempt the user into using inappropriate markup
in trying to achieve a sertain look, which anyways should be up to the
CSS-designer to decide (at least for the sake of consistency). Another
problem occurs if one is going to reuse the conent in different medias
with different formatting (ie. different styling for print), then the
WYSIWYG is nothing but an illusion.
What experiances or thoughts do you have on this? Do you btw. have
experiance with any web-based WYSIWYM:s?
cheers, Simon

Simon Rönnqvist

2005-12-08 00:01:20 UTC

Hi!

Hmmm... here comes my late answer again, hope you still remember what we were
tailking about... ;-)

My fear (as I mentioned earlier) is that people would be tempted to
choose the semantic marking according to looks instead of meaning.

Just give it meaningful names, remove bold/etc, and they can't do that :)

Well, I'm actually intending to do that.
But still there might be a chance that the author would notice certain kinds
of behaviour (or let's say looks) when using a certain kind of tag, and
thereby start to misuse those tags.

On the other hand, if the user really wants to he/she can still see the final
behaviour of certain markings from the actual web page he/she is editing
anyways. But at least not showing the final looks in the editor could make
the idea that he/she isn't supposed to be editing looks, only semantics and
content a bit clearer.