h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

This is what I was afraid of. I'm pretty sure we could get IE, even in its present state, to do exactly what we wanted but I'm damned if I know how to do it easily. I've been messing about with C# and it's pretty fiddly, although it looks possible.

We can nevertheless deliver a 'GuideDog Lite' which has limited support for footnotes.

FM

PS: have you seen http://www.playsophy.com/Wrap/wrapblurb.html ? MaW drew my attention to it

That's EXACTLY what I needed. That verifies that everything we need in place to do the Mozilla thing is there.

I'll start working on the XUL files.

Just goes to show you, how communication is all important

Indeed!

I have only been looking for Mozilla related resources but you may find them useful. Here's a list of resources I've found thus far:
http://www.xulplanet.com/tutorials/xultu/ - XUL Tutorial
http://devedge.netscape.com/library/manuals/2001/xslt/1.0/netscapexsltTOC.html - Netscape XSLT support reference guide
http://developer.netscape.com/docs/manuals/js/core/jsguide15/contents.html - JavaScript v1.5 manual

Relevant W3C Documentation:
http://www.w3.org/RDF/ - Used to describe XUL packages
http://www.w3.org/DOM/ - Document Object Model
http://www.w3.org/Style/XSL/ - XSLT (for modifying XML trees, i.e. GuideML => HTML)
http://www.w3.org/TR/xpath - XPath for traversing XML trees (e.g XSLT manipulation)
http://www.w3.org/CSS2/ - CSS is used to apply styling in Moz Apps

*sigh*

On further inspection of the -Moz-User-Modify property, it seems to be an implementation of a portion of an old CSS3 User Interface specification that has since been removed. I don't think we should count on it staying in Mozilla, which is a shame.

Oh bugger: this sort of issue is what I was trying to hammer home in another thread.. Well, I bought a C++ book yesterday, so I should be in a position to do some very low level hacking of IE before too long. Do you think it's worth following up the Mozilla angle of attack any further, or should it be abandoned?

I think the Mozilla base is still worth persuing but we may not be able to do WYSIWYG. The concept I have in mind currently is a multi-paned editor with "Preview", "Edit GuideML" and "View as HTML" tabs. The editor would have buttons for applying or changing styling and the effects would be immediately visible by clicking Preview. This would also have the side benefit of making the program vastly simpler to write.

If user-modify or some similar mechanism is formally supported at some point in the future it should be easily possible to change that to "Normal" for WYSIWYG, "Edit as GuideML" and "View as HTML".

It's entirely possible to make a newbie-oriented editor without WYSIWYG capability but I think whether we should or not is a decision that you should make since it was your project before it was mine.

Making it non-WYSIWYG sort of defeats the object of the project. There is such an editor at the moment, called GuidePost: MaW wrote it. We'd have to think hard about what advantages a new non-WYSIWYG editor will give us.

GuideDog is not intended to be totally WYSIWYG, btw. Just the bit which the user edits would be rendered. I really don't see any point in providing a total facsimile of the DNA experience as this doesn't help to give immediate feedback of changes to an article (which is why it's WYSIWYG in the first place).

FM

I suspected non-WYSIWYG would defeat the purpose. I didn't know about GuidePost, actually.

Do you have a copy of Mozilla installed? I want to show you what I'm thinking and description just won't cut it, I think.

Let me see if I've got this right... We want to render the article only, from subject down to where the article text ends?

I do have Mozilla installed, and I can upgrade to any new version fairly rapidly as I have a fast connection.

As far as I'm concerned, if the editor renders just the subject and article text then that's fine. I don't see any point in adding all the bits and pieces that go with the various skins. I see the WYSIWYG approach as a means of making editing easier: the more immediate the visual feedback, the less error-prone and fiddly the process becomes.

We need a decision soon about what platform we are going to use. From what you have said so far, Mozilla isn't up to it. Is this correct?

Doing WYSIWYG in Mozilla would depend on the -moz-user-modify property which may or may not stay in the codebase. Since it is not part of a W3C standard anymore I would not suggest we rely on it. That would eliminate Mozilla as a viable codebase for a WYSIWYG editor.

Therefore we can't use it. Which means end of the line for Mozilla. This won't please the Microsoft-haters.

It doesn't please me but what can you do. I'm not a Microsoft-hater, I just think cross platform compatability is important.

Yes, it is important, but unfortunately the Law of Diminishing Returns sets in when one looks at X-platform issues. I reckon that 95% of our client base will already be using IE. Getting this up to 99% and delivering a fully-functioning product will be rather difficult.

I suggest that we write the system in VB6 with some critical functionality in C++. The bits that we can't handle in VB6 can be handled by use of C++ and COM.

Unfortunately I have no COM experience whatsoever. I also (as I mentioned) don't have a copy of VS 6 at my disposal so that sortof precludes me from helping out. I'd be happy to help you guys test though.

You do have HTML and website-building experience, though, so you could certainly build us a project website if you like. This would be absolutely invaluable.

Looks like we're not the only ones that want Mozilla to do this. WYSIWYG HTML editing is in Mozilla 1.3 courtesy of the Midas project - all you have to do is set the right property. Check it out:

http://devedge.netscape.com/viewsource/2003/midas/01/

Eeenteresting.....although it looks some way behind IE's editing functionality. I'm sitting tight waiting for IE 7.0 to surface: it shouldn't be long now, and the smart money is on it having proper .NET support which will make it much easier to program.