There are two philosophies with respect text of different appearance with in word processors.

Approach #1. Focus is first on appearance of text. Suits very simply and and one or two page document, and favoured by microsoft word and followed largely by open office.

This approach provides primarily tools geared towards changing font, size, font, bold, italics and colour of text. The use of styles is also available, but nor really encouraged.


Approach #2. Focus in on the role of text- with appearance being consider a means to convey the role of the text. The idea is a separation of meanting and appearance. Suits both long and short documents.
Favoured by Framemaker, Lyx, XML etc.

In this approach the use of paragraph and text styles is encouraged, and colour font size bold etc are encouraged to be applied through styles, not directly to text.


Is there an identified stance as to where KWord stands here, or is the plan to also follow the fashion lead by Microsoft word?

References to being a product similar to framemaker, which is as much admired for the power of the styles system as the frames themselves have me hoping there is a desire to have a styles based system, or at least to allow a styles based approach, however what I see from KWord does not give me hope.

Perhaps the philosophy is not set in concrete here? If not some debate could be useful?

iano wrote:Approach #1. Focus is first on appearance of text. Suits very simply and and one or two page document, and favoured by microsoft word and followed largely by open office.

Well, being a Writer's power user I must say OOo do not favor direct formatting. In fact, Writer could be very annoying if you try to only use direct formatting... Writer is heavily based on styles (character, paragraph, page, frame styles...) and is very good at using them.
Regarding Kword, I remember Zander saying Kword will (future tense) be useful for both: direct formatting and style based documents. This seems quite difficult to accomplish at first glance, but what I've seen so far in the current implementation makes me believe it is possible.
In the future, of course.
Right now Kword lacks many features Writer have since many a year: page styles, linked styles, the possibility to associate a page style with a given paragraph style, cross references... a large etc.
But considering how the project evolves, I'm sure that (in the future) I'll be a Kword power user too

PD: I know LyX/LaTeX too: I wrote my PhD thesis on LyX

Thanks RGB.

With a thesis completed using Lyx you are clearly well qualified one the subject of styles based documents.

However, respectfully, I do not fully agree on OpenOffice really supporting a styles based approach to the level products like framemaker do. Yes, OpenOffice may be much better than word, but that is not saying much.
And yes, OpenOffice supports styles well, but it does not do a great job of facilitating using a pure styles based metaphor for the creation documents.

Creating a document which achieves a high level of formating yet is free of direct formatting with OpenOffice is difficult. Even though the in theory all the features are present, it can be difficult to get the features to produce the correct results without also applying direct formatting. Removing direct format overrides is also troublesome.

To actually be able to create a clean result where all formatting is achieved through styles requires is simply not the goal of most word processors as stated in this article... http://heim.ifi.uio.no/~paalso/artikler ... n/pap.html

" An initial classification suggests that most of the problems with the use of paragraph styles can be described as ignoring or overlooking these mechanisms.
.....
Our interpretation of the problems encountered also indicates that they may be related to the paper metaphor communicated by the principle of WYSIWYG. We claim that the paper metaphor does not communicate any understanding of the structure beneath the surface of a digital document."

The problem here is that word processors encourage formatting by the direct approach. The paper metaphor and WYSIWYG features is what appeals to MOST word processor users and they actually want to use this direct format solution despite the problems. However for those who don't want to use direct formatting, the market is wide open for a great product.

I suggest that if KWord in place of the conventional solution of supporting both direct and styes based formatting(which in reality leads to direct formatting) - actually provided a mode switch to switch the interface into a styles based mode of operation and removed the direct formatting, then it could capture a significant additional fan base.

Yes, leave direct formating or 'dual mode' as the default configuration and interface. But at a mode that switches the product into a more framemaker style interface that encourages 'What You See Is What You Mean' (as per Lyx) and supports addressing presentation as a later step if required would be really really compelling to a group of users.

Allow both modes to still have a product which can appeal equally to the mainstream. For that mainstream there will always be wide choice so it is difficult to win support. However having a product that is clearly the best for those who want the styles approach means it is possible to have huge market share in that market.

An odt compliant word processor with a mode that supports and reinforces a style based approach would certainly be attractive. I know of several organisations who would make the switch on that basis.

Styles in a word processor are a mixed bag of goodies. After my first drafts of designing the UI for KWord2 I got the complaint from several community members that I was making it too hard for people to just apply font size and boldness etc. It would be bad to alienate too many people with my views of correct KWord2 uses a middle ground for normal common usage; people see the familiar bold/italic/etc UI first and the second tab is for all the styles.

I have a background in the pre-press world where I used Framemaker extensively and worked with several talented people that did layout of very intensive and long documents. My experience there is what I use in KWords design.

In practice a lot of people *want* to only use styles but most honestly don't go creating a style for 'strong in picture-sub-title' since that would explode the number of styles you use. Just applying bold is pretty much Ok with most people.
So, both should work and styles is really a case of making the common easy and the hard possible.

This means that the default UI might not address the companies and users you talk about. And I considered that too. The steps towards solving the issue also for those users is actually very logical if you get more familiar with KOffice2.
In KOffice2 we have a plugin system for tools. A tool is essentially a 'mode' and it handles all mouse and keyboard input. KOffice ships with loads and loads of tools already. So the solution to that problem is to get a tool plugin that limits the applying of formatting information only to named styles. This approach has never been tried before and so I am very happy to let anyone try it and innovate in plugins and see what they can come up with. I have some ideas myself that I'd like to try out, but more important things are to be taken care of first so KWord becomes generally usable.

Even if it is widely used, I do not like the word "metaphor" applied to UI design: a metaphor suggest similarities, while a well designed UI gives you real tools directly related with a real work.
The UI in OOo Writer is a sort of mess, and it is quite difficult to tell a new user where to find the needed function. But the function is there, and once you know how to use it everything is easy.
Of course, the problem is to know how to use it...
Some time ago the university in which I was working asked me to write a book to be used in the lab courses. I wrote that book on Writer following the formatting guidelines from the director of the project. When the first draft was almost finished (something like 100 pages full of formatting and complex figures), he changed his mind and asked me to change page size, margins, font and font size.
The change took me only a couple of minutes (1), thanks to a well planned group of paragraph, character and page styles. And that planing (the build of the template) took me less than 10 minutes.
But I'm a Writer power user: if someone asks me to explain to a poor new user how to do all that, I'll need a week full of examples and patient work. There is where OOo fails, not in the features.
A well defined UI lets you choose the tools that best fit your work and your work-flow in the easiest way.
I think koffice UI is really well designed and flexible. As I previously said, nowadays kword lacks many advanced features, but each version is full of good improvements.

(1) it was more than one minute because Math is the only component in OOo that do not use styles...