On Wed, Dec 03, 2003 at 03:41:37PM +0000, John Levon wrote: > > THis is the implementation. The user pressed C-b on a selction. > > We're not talking about mathed.
No, we are talking about 'selecting things and pressing C-b on them'. Which happens to work in mathed and outside. > > > It is *way* more complex than the > > > direct manipulation method of "select object, apply change". > > > > It's identical in this direction. > > Erm, no it isn't. It so clearly isn't that I really can't even argue the > point. Now, what's the exact difference between C-b and C-b? I can't follow you here? > > Cut, create new \emph inset, go there, paste. > > "select object, apply change" > > > > Such an approach may or may not make sense for mathed, but math IS > > > NOT LIKE normal text. > > > > Last time (at least at the beginning) you argued like that for math, > > too. > > I don't want to get into a discussion about mathed, please. That's independent of mathed. You predicted an outcry and it simply did not happen. Now you are saying "people won't be happy" again. How whould you argue in that case and not point back to the preceding case? > > > You're picking apart a straw man. Our current physical style interface > > > does not have many problems, > > > > It has quite a few. For instance, if I am between an emphasized letter > > and a normal one, will the next char I type be emphasized or not. > > That's one. Start with s|s Type <C-e>aaaaa aaaa<C-e> That's two. There are a few more variations on that 'feature'. > > I have been using LyX for a couple of years now, but I still can't > > answer this question. > > The answer is "no" in LyX, and "yes" in everything else. > > And that's just one problem. See above. > > > o you go to the Insert menu to apply a style. Huh ? I am editing not > > > inserting. > > > > I don't mind putting this stuf into the edit menu if that helps your > > imagination. > > Then the reverse problem applies. You are not, by chance, arguing that someone might not find something in the menus and will be annoyed by that? > > > o I can't select part of an inset and part outside of an > > > inset and apply a change > > > > I would argue that this feature is rarely needed. > > I would argue that I do it regularly. A LOT. Please give a concrete example. > > However, it is implementable by splitting the insets at the selection > > boundary > > You can't even *make* such a selection ! So how come people don't complain? > > So you have (on a regular base, no less...) some LCS that spans the end > > of one par and the begin of the next? > > > > Even if so, answer is the same as above. > > Same as above what ? 'It is implementable but I doubt its utility'. > > > o I can arbitrarily nest the > > > same style inside the same style. This does not make sense for LCS. > > > > Wrong. '\lessimportant' (aka \smaller) is an example that this is not > > true. Even HTML has this ability. > > That's not the "same style" though. select-apply can handle both > relative and absolute styles sensibly. Insets cannot. Insets need not. If you need to reverse, you break the inset into two. Viewing documents as tree is not the worst approach I've seen. > > All this stuff is trivially derivable by trying current CVS and > > > comparing with the Edit->Text Style dialog for Noun. > > > > And that's a good reason? Have you ever needed the features you claim as > > a 'must have' for noun style? > > AFAIK I've never used noun style. Bad enough. So these points of your argument are based on some features currently available for noun style that you've admittedly never used? May I ask you to provide a real world example? > It's the only example we have to > compare with currently since it's currently there in both ways. We've arbitrary font changes handled both ways. > > > So what ? It's not a question of absolute values, but of relative > > > ones. Boxes are harder, less intuitive, and significantly more > > > awkward to use. > > > > Have you ever tried? > > I'm trying it now. It sucks. And it will suck only slightly less when we > have better "look" for the insets (inline). You could try font changes in mathed which might be a bit smoother to handle than a two-week old feature in the outside world. > > We had this discussion already a while ago. As no single user > > complained about mathed's almost-all-is-boxes in stark contrast to your > > prediction of a universal outcry I hereby declare "You lost." > > Sorry, that's bull. I never claimed universal outcry. Furthermore (and I > don't know if I can stand to say this *AGAIN*) silence DOES NOT IMPLY > SATISFACTION. Furthermore, MATHED IS NOT LIKE NORMAL TEXT. argh :) Well, once I get my way it will be... > > The ideal UI fits onto a simple document model. Just because StarOffice > > and MS Word UI do not fit into that scheme does not make them superior > > in any way. > > So far you've come up with exactly *one* benefit of a boxes model (which > side of the cursor should the style be taken from). And that benefit is > realizable without a boxes model anyway, if we were to decide it's > really important. Export is the next one. Then 'it is already there'. Then 'IU'. I think I could make up a few more. > You're going to have to come up with a lot better than that to deviate > from what every other program, including LyX, does. > > > It might even be feasible to implement - once you got over obstacles > > like 'if some range is <a> and <b>, is it <a><b>...</b></a> or > > <a><b>...</a></b> or <b><a>...</a></b> or ...' (which possibly boils > > down to using boxes in the end...) > > I'm perfectly happy with an internal boxes implementation if you can > make it work well. Fine. Now given that we have boxes internally anyway _and_ need a way to specify a hierarchy in competing boxes, what is wrong with presenting the user with direct access to this structure instead of hiding this behind some interface that looks like OOo or Word? > I am not at all happy with exposing this > implementation to the user. I can't seen anything wrong with that. > > However, I don't see anybody implementing it. > > As I said the current implementation will do for now. Yes. And 'update' worked well ... save a few glitches that nobody could fix, and IL worked well ... save ... > That doesn't mean it should not be fixed. There are few thing that simply can't be fixed in a certain architecture without introducing layers and layers of glue. Andre'
