[prev in list] [next in list] [prev in thread] [next in thread] 

List:       quanta-devel
Subject:    Re: [quanta-devel] kdewebdev/quanta
From:       Jens Herden <jens () kdewebdev ! org>
Date:       2005-04-04 4:49:46
Message-ID: 200504041149.46972.jens () kdewebdev ! org
[Download RAW message or body]

> > > this doesn't happens in KWord but it's an exception to the
> > > rule (OO, FrontPage, Word, etc).
> >
> > Why is it wise to do like OOo?
>
> That's not the question, I just wanted to point out that it's not such a
> bizarre and odd behavior.
> I chosen this behavior because I think it's much more likely that, if a
> user is in the middle of the word and presses bold, then he wants the word
> to be bold instead of having the same word in a mixed style, which I think
> is absurd in the majority of situations. It's not by chance that the most
> popular applications have chosen this behavior.

So KWord, Kile, Scribus and AbiWord are not popular? 


> > This feature is absolutely contrary to what I am used to and I guess what
> > most of the users are used to.
>
> If the most popular applications behave like this, I doubt that users are
> not used to that :)
>
> > This breaks the normal workflow:
> >  1. select
> >  2. apply
> >
> > Different behavior of one action is likely to be confusing and I do not
> > see that the benefit is big enough to do this here. Especially because a
> > double click does select a word.
>
> And what about the scenario of an horizontal alignment action? In my
> opinion, it's a more obvious case where It should be applied to a section,
> even if there is no selection. And the "bold" case is the same IMHO. I had
> the impression that this behavior was a "de facto" standard.

See, you are mixing things here. Bold is a property of a character and the 
alignment is a property of a paragraph! In the concept of wordprocessors are 
no properties for word, AFAIK! 

Another point: how do you find the word? Are you looking for spaces? This will 
only work in Latin scripts! I'll give you some information about Khmer, 
because I know it a bit. There is not separation between words in the Khmer 
script! You must lookup a dictionary to find words otherwise you might select 
a whole sentence or big parts of one. Some people type a zero-width-space 
character, you can not see it but the wordprocessor can break word there. But 
you can not rely on this! 

I see another problem coming up when users get used to this. The users will 
assume more intelligence than the software might offer. Even in Latin scripts 
it is extremely difficult to find the word boundaries always correctly. Lets 
take two examples:

a)   "abc.def"
b)   "1.2"

The first one is most likely an end of sentence but the second one not. Even 
OOo does it sometimes in one way and sometimes the other way. So that is why 
I prefere the traditional workflow 1. select 2. apply.  This is never 
ambiguous.


> > For my point of view this should be reverted for the sake of consistency
> > and usability.
>
> I have no problem with that, I just wish that people had expressed their
> opinions sooner, because this was talked before in the mailing list.

I am sorry for not speaking up earlier. I see that this is frustrating for 
you. 

I would suggest to ask the people on the usability list about this, since that 
is what they are good for and they may have better understanding than me. 

Jens
_______________________________________________
quanta-devel mailing list
quanta-devel@kde.org
https://mail.kde.org/mailman/listinfo/quanta-devel
[prev in list] [next in list] [prev in thread] [next in thread]