It is desirable yes, but not always possible to show the two positions. That's where I proposed the dual view mode. And only in the cases where it is relevant to show both positions : where there's a RTL-to-LTR or LTR-to-RTL transition in the UBA-resolved direction. Nowhere else it is needed.
Also temporary highlighting is definitely not a correct solution: you cannot make the distinction with an actual text selection between two logical positions, where there is only one logical position here. 2012/11/15 Eli Zaretskii <[email protected]> > > From: Philippe Verdy <[email protected]> > > Date: Thu, 15 Nov 2012 19:03:14 +0100 > > Cc: Martin Dürst <[email protected]>, > > Jeff Senn <[email protected]>, Unicode Mailing List < > [email protected]> > > > > Highlighted selection is less a problem because this involves separate > > actions to mark the start and end of selection. And an explicit move > action > > or separate clicks : the user chose himself to perform those actions and > > knows where he comes from and where he is going to. > > True, but not necessarily relevant. The common part is that in both > these situations it is desirable to show the entire range of > characters between 2 positions. > > > In addition the selction is marked as a whole, and is not just two > > tiny positions. > > Which might actually suggest a solution: temporarily highlight the > characters in between. >
