[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
Re: [zzdev] Re: [zzdev] Re: [zzdev] d.n
- To: Tuomas Lukka <lukka@xxxxxxxxxxx>
- Subject: Re: [zzdev] Re: [zzdev] Re: [zzdev] d.n
- From: Benjamin Fallenstein <b.fallenstein@xxxxxx>
- Date: Thu, 10 Aug 2000 01:20:50 +0200
- Cc: Tuukka Hastrup <Tuukka.Hastrup@xxxxxx>, zzdev@xxxxxxxxxx
- References: <Pine.HPP.3.96.1000810002658.13013C-100000@xxxxxxxxxxxxxxxx>
Tuomas Lukka wrote:
> > I didn't mean it'd be the _only_ one, but the _default_ one. When I
> > earlier said we should have several dimensions for different list
> > relations I was told it'd be bad practise, and that you're to use
> > clones to resolve this kind of clashes.
>
> Hmm???
>
> Depends on the list relations. It's an artistic decision that cannot
> necessarily be mathematically justified.
Umph! I don't like that terming.
I'd call it a *design* decision. Although software design can learn from
artistic fields like movie-making (as Ted points out in Literary
Machines), or theater (as e.g. Brenda Laurel points out in "Computers as
Theatre"), it is not an *art*, because it's not *expressive*. Rather,
what is to be learned for software design is the *tuning* of the overall
*feel*, which is a wholly different thing. We aim to design a system
that is both usable and "pleasurable" in an abstract way. But in
software design, we do not try to express ourselves.
Certainly my words don't make that distinction very clear. (Neither is
this the place to discuss the classical "what is art.") But you will
agree that it's simply a different activity (and feeling) to paint,
perform or do creative writing. This by no means is to say that software
(virtuality) design is the same as coding or code design (which is an
important distinction, to); it just happens not to be art, either. None
is "better," but they are different. Calling software (virtuality)
design art sounds a lot like the Noid notion of calling an integrated
circuit art.
And: certainly there is a blurry line between the two, but here and now,
I think we're nowhere near it.
(A bit off-topic, but I do think it is an important distinction.)
> That's because currenly there is only an operation to do that. I now think
> that the correct solution would be to clone from d.masterdim.
Wait a moment. How is a user supposed to create a new dimension? Move to
d.masterdim, insert a cell there, and clone it into the dimension list?
Maybe cloning from masterdim should be only *one* possibility? But then,
shouldn't the cell on the dimlist be made a clone of the corresponding
cell in d.masterdim once that cell is created? Although this might be a
bit too much work.
-b.