Dan Hatton wrote:
On Thu, 4 Dec 2003, MJ Ray wrote:
A true "WYSIWYG [HTML] editor" is impossible.
I'm not sure about the "impossible," and wonder if you could explain it.. is the argument as follows?
<snip>
If so, I think resisting that temptation is merely difficult, not impossible, although this is not me volunteering code ;-).
My original intention of semi-starting this topic wasn't really anything to do with HTML. It was about scratching itches in Open Source Software development, but with the problem - Who scratches the itches that the developers don't have, but other people do? Where's the incentive for programmers to write programs they don't need but other people do - within the Open Source development model.
It stemmed from comments people who I've tried to pursuade to switch to Linux have made about usability, and me wondering whether these problems were not down to bad software design, but simply the fundamental nature of FOSS.
However, I am interested in the WYSIWIG vs. plain text interfaces debate in web page creation. Why is it that a WYSIWYG editor is impossible? Is this because what you see isn't necessarily what you get?
If this is the case then I think I was asking the wrong question. I appreciate that whatever I use to make web pages, they're going to appear different on different operating systems, browsers, resolutions etc. - but I don't see the need to type out all the code for a language which translates into visual form quite literally.
What I perhaps meant was a GUI, not necessarily WYSIWYG. However, I do mean more than a button you press which inserts <b></b> into your code on screen for example. I'd prefer a system more like modern word processors where you highlight your text and press a bold button or Ctrl-b, and it appears bold - but hiding the tags from view. You insert a table and can drag the sides to resize it. You can drag a picture from one cell to another in a table. That kind of thing.
This isn't about seeing what it's going to look like in a browser, it's about being able to visualise what you're doing to the page - speeding up productivity - at least for static HTML.
I'm no master programmer by *any* stretch of the imagination, but I don't understand why it would be impossible to make graphical tools like this translate into clean, neat W3C compliant HTML that is constantly tidying itself by indenting, getting rid of duplicate tags and highlighting problems and missing attributes as you work. Difficult, yes, but impossible?
Do the
favourable comments about Mozilla Composer mean that it's achieved this resistance?
I personally find Mozilla Composer quite limiting compared to tools I've used in Windows. Since this topic started I've dowloaded the new version of OpenOffice which contains a complex web page editor. This has more like the feature list I'm looking for, but I still don't like the interface at all and the code it generates is very messy.
From the above observations it certainly seems that using a GUI and generating clean code are fundamentally opposing concepts. But does this have to be the case? Is it possible to have the best of both worlds?