There We Go!
Mark Smith
mark at bbprojects.net
Fri Dec 10 10:24:50 PST 2004
Adam Lindsay wrote:
>Mark, I'm trying to get at what sort of documents you're producing. I've
>experimented a little with XML output/transforms, and it sounds to me as
>if you're shooting yourself in the foot in wishing away named styles.
>
>Named styles are just the right way of getting across a tag without too
>much visual clutter. (Okay, in testing, I'm a little frustrated that tags
>without any text formatting associated with them are ignored.) Of course,
>this is most effective with mixed-content kinds of documents (plain and
>tagged text within an element, e.g., XHTML-style).
>
>If you're dealing with more structured element(-only) content, then I can
>see how pop-ups might be better suited to your needs. I guess that way
>you get off cheaply! As for me, I'm only just now starting to appreciate
>the real power of these styles.
I'll need to be careful to avoid confusion here (though it may arise anyway) and should state at the outset that I'm not an XML/XSLT expert. I'm a beginner. OK...
There was some *related* discussion on the beta list back when the first betas were issued to the testing group.
Some felt, that there was an underlying issue with OO in that it lacks a model of clear separation between structure and content. The "note" is the obvious level at which this separation *could* be made. In short, all of your structure would go into the "normal" outliner nodes and all of your content would go into your notes. There are obvious, serious limitations with the app as it stands here. Omni would need efffectively to replicate a lot of the outliner functionality inside the notes. Understandably, they didn't go down this road. (It is my *perception* that the big effort with named styles removed any possibility of this separation being addressed. Doing both would probably have been a nightmare. Of course, the two may have been completely unrelated from Omni's perspective.)
In the absence of this deep separation, I'm looking for a way to fake a degree of sepatation of this type which is not *bound* by any kind of formatting and which can "survive" transforms, ideally without complicating things too much. I'm lazy.
I haven't actually sat down and dilligently worked through some examples to establish exactly how difficult this would be to achieve with named styles and to identify all the pitfalls. I can imagine that it will be possible, but I can see that it will be both difficult and brittle and it should ideally be neither.
However, when I get some time, I may well have a real long think about it.
mark.
More information about the OmniOutliner-Users
mailing list