command line vs gui
in [General Linux]
|
Prev: micro solution backpack cd-writer hell
Next: "...error while loading shared libraries: libg2c.so.0"
From: Lee Sau Dan on 6 Dec 2005 00:56 >>>>> "Peter" == Peter T Breuer <ptb(a)oboe.it.uc3m.es> writes: Peter> In comp.os.linux.misc blmblm(a)myrealbox.com wrote: >> If I understand you right, you're proposing a GUI that allows >> you to build expressions in the canonical form. Right? Peter> Well, in a form that _covers_ the normal forms, yes. >> But for complex search criteria, I question whether this will >> be more effective than the CLI, Peter> It's just the same - indeed it may help you factor your Peter> search into a sequence of simpler filters. The interface Peter> can calculate the most effective equivalent form. I don't think so. Disjunctive/conjunctive normal forms are usually expansions of what is *originally* more compact (but more deeply nested). They are usually more lengthy and much less comprehensible than the more condensed original form. Also, the expansion into disjunctive/conjunctive normal forms usually introduces more redundancies. (People who have worked on Boolean algebra exercises by hand should know why the technique of factorization is useful.) e.g. try to convert (a and b) or (c and d) into conjunctive normal form. Peter> It's a "normal" way of organising ones thinking. No. "Normal" form is just a name. It doesn't reflect the normal way of thinking. Normal people think with highly nested expressions and subexpressions. Not in the way the shallow disjunctive/conjunctive normal form look like. Peter> The idea is that you carve out in a first search roughly Peter> what you want, and pass it into a second search that Peter> excludes some more things you didn't want, etc. Exclusion... so you're talking about conjunctive normal forms, right? What happens when I find that your chain of filter are too restrictive that it has eliminated something that I want to include? I would need a disjunction. But your interfaces doesn't allow that: I have to use pencil and paper to convert the whole query into conjunctive normal form, just to please your GUI. Really "useable"! i.e. I have now: A and B and C and D and now, I want to change it to: (A and B and C and D) or E Instead of simply adding the brackets and appending "or E", your "user-friendly" and "intuitive" GUI would require me to first mentally (or using paper and pencil) convert that into: (A or E) and (B or E) and (C or E) and (D or E) Just to please your GUI. And what if E isn't a simply thing, but itself a complicated formula? I have to repeat it 5 times using your GUI. That's really "productive" and "efficient"!!! Peter> This is also an efficient way of going about things in Peter> general, since the preselection cuts down on the number of Peter> later tests. See my example above. It's much more inefficient than a CLI, which would allow me to enter the Boolean expression "(A and B and C and D) or E" directly, leaving the normalization (if necessary) to the computer, leaving me -- a human being -- more time for things more interesting. >> Very possibly there's still something I'm not getting! Peter> I believe equivalence of boolean expressions is an Peter> np-complete problem. Am I right? Anyone recall? Obviously, you never really worked with rewriting or simplifying Boolean expressions. Otherwise, you should know that DNF/CNF are usually less compact and less intuitive and more redundant than a more deeply nested form. The normal forms are only good because they're only 2 levels deep (ignoring the depth of the "NOT" operator) and the operation on each level is identical. This is ideal for deducing and proving theorems. But not for readability, maintainabiliy or storage space. -- Lee Sau Dan §õ¦u´° ~{@nJX6X~} E-mail: danlee(a)informatik.uni-freiburg.de Home page: http://www.informatik.uni-freiburg.de/~danlee
From: Lee Sau Dan on 6 Dec 2005 01:00 >>>>> "Peter" == Peter T Breuer <ptb(a)oboe.it.uc3m.es> writes: Peter> In comp.os.linux.misc blmblm(a)myrealbox.com wrote: >> If I understand you right, you're proposing a GUI that allows >> you to build expressions in the canonical form. Right? Peter> Well, in a form that _covers_ the normal forms, yes. >> But for complex search criteria, I question whether this will >> be more effective than the CLI, Peter> It's just the same - indeed it may help you factor your Peter> search into a sequence of simpler filters. The interface Peter> can calculate the most effective equivalent form. But you require the *human* operator to compute the normal form first, just to enter it into your "user-friendly" GUI. Why should he be required to do that, in the first place? -- Lee Sau Dan §õ¦u´° ~{@nJX6X~} E-mail: danlee(a)informatik.uni-freiburg.de Home page: http://www.informatik.uni-freiburg.de/~danlee
From: Lee Sau Dan on 6 Dec 2005 01:02 >>>>> "Tobias" == Tobias Brox <tobias(a)stud.cs.uit.no> writes: >> Does that make the decimal expansion of 1/3 finite? >> Infinite is infinite. Tobias> Eh, no? 0.3333.... is clearly an infinite decimal Tobias> expansion, and 3/10+3/100+3/1000+... is clearly an Tobias> infinite sum. Both can be written precisely and elegantly Tobias> using the highly finite fraction 1/3. Yeah, and you can write irrational numbers compactly, too. e.g. "e", "sqrt(2)", "pi". Tobias> I'm quite sure that it should be trivial to make a GUI to Tobias> find, and using such an interface will be easier than to Tobias> read the find manual. >> Then, why haven't you done it? Tobias> Because I have other things to devote my time to? Since you claimed that the task is trivial, the burden of proof is on you. -- Lee Sau Dan §õ¦u´° ~{@nJX6X~} E-mail: danlee(a)informatik.uni-freiburg.de Home page: http://www.informatik.uni-freiburg.de/~danlee
From: Lee Sau Dan on 6 Dec 2005 01:07 >>>>> "Jacob" == Jacob Sparre Andersen <sparre(a)nbi.dk> writes: Jacob> No matter what, the basic concept of a command line seems Jacob> to be so foreign for many computer users, Is verbal communication that foreign? Have you never given written instructions to somebody else, so that he can do something for you in your absence? Jacob> that it might be necessary to camuflage it as a mouse-based Jacob> interface. When I go to a drug store, I say the name of the drug I want to buy, and in 10 seconds, the purchase is done. You go to a drug store. You open the draws and cupboards of the shop one after the other, trying to look for some visible object which looks like what you want. Spend 30 minutes there, still doing the hunting. Which "interface" to the drug store is more intuitive, efficient and user-friendly? (Ignore the shopkeeper-friendliness for this discussion.) -- 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 6 Dec 2005 01:09
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> In comp.os.linux.misc blmblm(a)myrealbox.com wrote: > >> If I understand you right, you're proposing a GUI that allows > >> you to build expressions in the canonical form. Right? > Peter> Well, in a form that _covers_ the normal forms, yes. > >> But for complex search criteria, I question whether this will > >> be more effective than the CLI, > Peter> It's just the same - indeed it may help you factor your > Peter> search into a sequence of simpler filters. The interface > Peter> can calculate the most effective equivalent form. > I don't think so. Disjunctive/conjunctive normal forms are usually > expansions They are simply restricted forms. Using one restricts you to a format - which helps you, as GUIs do. > Peter> The idea is that you carve out in a first search roughly > Peter> what you want, and pass it into a second search that > Peter> excludes some more things you didn't want, etc. > Exclusion... so you're talking about conjunctive normal forms, right? Yes. > What happens when I find that your chain of filter are too restrictive > that it has eliminated something that I want to include? You redo it. Clickity. Peter |