From: Lee Sau Dan on
>>>>> "Peter" == Peter T Breuer <ptb(a)oboe.it.uc3m.es> writes:

>> Well, you seem to me to be proposing a GUI that only allows
>> entering expressions in conjunctive normal form. How do you
>> propose that users enter expressions that are not in CNF?

Peter> They do not know or care.

The question was "how", not "whether".



>> If your answer is "all expressions can be converted to CNF"
>> .... Who/what is doing the converting?

Peter> Nobody - they can't write anything else (it's not CNF, but
Peter> something that covers cnf, but still). They simply click
Peter> and try a search and if they don't like the result they
Peter> click again to modify the search - no thinking (about cnf
Peter> or otherwise) is required any more than you need to know
Peter> SQL when putting items into a google search (yes, it
Peter> understands AND and stuff).

Show us your design, or shut up.



--
Lee Sau Dan §õ¦u´° ~{@nJX6X~}

E-mail: danlee(a)informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee
From: Peter T. Breuer on
In comp.os.linux.misc Lee Sau Dan <danlee(a)informatik.uni-freiburg.de> wrote:
>>>>>> "Peter" == Peter T Breuer <ptb(a)oboe.it.uc3m.es> writes:

> >> Read the above dialog one more time. I say "four panels for
> >> A&B&C&D", and you reply "panels? checkboxes".

> Peter> They are CHECKBOXES! Not "panels".

> So, you build that GUI and let us see what it is like. Didn't you say
> it's "trivial"?

Dunno!

> Peter> They are CHECKBOXES. Depending on how you wish to combine
> Peter> them, they will be on the SAME or DIFFERENT panels. ALL
> Peter> the options are present on EACH panel. I read your A & B &
> Peter> C & D as written shorthand without any formal meaning.

> To let us communicate better, you should build the GUI now, instead of
> wasting any more time arguing here. Come back when you're finished,
> and let us evaluate your GUI demo.

That is being silly.


> Peter> Please stop making absurd arguments! Nobody is requiring
> Peter> everyone to speak english, I merely pointed out that is
> Peter> "easy". You know it is!

> No, English isn't easy.

Yes it is. For human beings, at any rate. Are we talking about humans?

> Peter> Billions of people can do it to a greater or lesser degree,

> That doesn't mean it's easy.

Yes it does.

> Peter> just as a billion or so can read and write chinese.

> But many people find that difficult, too!

No - they all find it very easy. I studied a little chinese from a
book, and I can tell you it was easy. The classic symbols had nice
visual meanings that I could easily understand without the inconvenience
of having to translate to sounds, and listen to the result in my head.
For example, the word for "trouble" is a picture of two women under one
roof. The word for "love" is a picture of a woman and a child. "child
"looks like a swaddled baby. "Sun" is a stylised circle (a square).
"Moon" is a stylised circle with a dot in it (square with bar).
"Bright" is sun and moon together. "Big" is a man with his arms
stretched wide. "Man" looks like a pair of legs and a trunk.

All very easy. I leave you to intuit the meaning of the symbol
consisting of a man next to an open doorway.

Peter
From: Lee Sau Dan on
>>>>> "Peter" == Peter T Breuer <ptb(a)oboe.it.uc3m.es> writes:

>> How do I enter something that gives the functionality of "(A
>> and B) or C"?

Peter> Chose A, C checkboxes. Click "again". Chose B, C
Peter> checkboxes. Click "finish" ("search", whatever).

So, you've done the conversion to CNF in your brain, haven't you?


And what should you do when C is not an atomic predicate, but a more
complicated Boolean expression? You need to click tons of checkboxes
twice (once before "again" and once after it)? That's REALLY
user-friendly!



>> For that matter, how do I combine checkboxes to get the
>> equivalent of this command?:

>> find . -name "*.bak" -o -mtime -1

Peter> Click two of them.

And it doesn't mean:
find . -name "*.bak" -a -mtime -1

instead?

Can you GUI magically read your mind to find that out?


Peter> To do an OR, you have to do two searches, and concat the
Peter> resluts.

Concat? That means duplicates are not removed?

That's even worse than having to do a manual CNF conversion!


Come on. You've been failing to convince us how your
"trivially-buildable" GUI is better. Perhaps, you should build it and
let us try it. Do it now, instead of wasting time here.



--
Lee Sau Dan §õ¦u´° ~{@nJX6X~}

E-mail: danlee(a)informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee
From: Peter T. Breuer on
In comp.os.linux.misc Lee Sau Dan <danlee(a)informatik.uni-freiburg.de> wrote:
> Peter should really go to build the GUI and let us see it, instead of
> just talking here.

No he shouldn't. This is called "design".

> Accord to him, such a GUI is "trivial" to build.

Oh? Quote please.

> Up till now, Peter's arguments are still unconvincing. He isn't even
> clear whether his checkboxes are to be ANDed or ORed together, or

Why should I be clear! (although I am perfectly clear!) It is not my job
to flesh out my intuition for you!

> XORed? or NORed? He has no clear idea of the semantics of his GUI

I have a perfectly clear idea, thank you, abslutely suitable for my
purposes. Would you please go speak for yourself. Thanks.

Peter

(in case you hadn't noticed - argument via "go build it then" does not
work as an argument technqiue)
From: Peter T. Breuer on
In comp.os.linux.misc Lee Sau Dan <danlee(a)informatik.uni-freiburg.de> wrote:
>>>>>> "Peter" == Peter T Breuer <ptb(a)oboe.it.uc3m.es> writes:

> Peter> Of course not. I wish you would stop "arguing" in
> Peter> non-sequiturs! No "computation" takes place in the human,
> Peter> simply selection from a check-list.

> Show me your GUI that allows me to express:

> (A and B and C and D)

Chose A. Choose Again. Unclick A, click B. Choose Again. ...

> and later extend the query to

> (A and B and C and D) or E

Repeat the construction. This time also check E in every panel.

> Show me your GUI, which you claimed was "trivial" to develop.

Idiot.

Peter