Prev: micro solution backpack cd-writer hell
Next: "...error while loading shared libraries: libg2c.so.0"
From: blmblm on 24 Nov 2005 19:01 In article <4385f28d$43$fuzhry+tra$mr2ice(a)news.patriot.net>, Shmuel (Seymour J.) Metz <spamtrap(a)library.lspace.org.invalid> wrote: >In <3uji23F11iafkU1(a)individual.net>, on 11/23/2005 > at 04:01 PM, blmblm(a)myrealbox.com (blmblm(a)myrealbox.com) said: > >>No, but I did change the From header to conform to one of the two >>syntactic forms you requested in a previous post (I believe it was in >>comp.editors). What additional information would you get if I added >>a name? > >Putting your name in the From would allow human readers to recognize >the article as being from you prior to fetching the body. It's >certainly not required. For posters with multiple addresses or who >periodically change providers, the name might be more useful than the >actual address. Again, though, it is not required. > >Putting your address in a comment as you did is not helpful. > Nice to hear another opinion on this. You're right that duplicating information is not helpful except in conforming to a supposed syntactic standard -- which only AC seems to be adamant about, so I think I'll stop doing it. Good point about this being helpful for people who post with multiple addresses, etc. "N/A" for me at this point, though. > >>(Since you seem skeptical that I really am posting with trn: > >Why do you care what he believes? He's long since ceased to be >credible. > The hook is dangling from my mouth, all right, but I've recognized it for what it is and maybe will have better sense if there's a next time. Mea culpa. And in the process of removing the duplicate information in headers, for this post, I realized that my "Since you seem skeptical ...." paragraph wasn't even accurate. Sometimes I forget that it's okay to *read* news before the morning caffeine is absorbed, but posting should be another matter. -- | B. L. Massingill | ObDisclaimer: I don't speak for my employers; they return the favor. -- | B. L. Massingill | ObDisclaimer: I don't speak for my employers; they return the favor.
From: blmblm on 24 Nov 2005 19:06 In article <4385f3af$44$fuzhry+tra$mr2ice(a)news.patriot.net>, Shmuel (Seymour J.) Metz <spamtrap(a)library.lspace.org.invalid> wrote: >In <3ujih7F11iafkU2(a)individual.net>, on 11/23/2005 > at 04:09 PM, blmblm(a)myrealbox.com (blmblm(a)myrealbox.com) said: > [ snip ] > >>(do *you* know what he means by that "<body not downloaded>"?) > >I would guess that he is responding based on a header that he >downloaded via NNTP without ever fetching the associated body. If so, >it's another red falg. > Very possible -- either that's really what's happening, or it's what he wants the replied-to person to think, in the hope of provoking an angry and defensive response. Sometimes it works, I regret to say. -- | B. L. Massingill | ObDisclaimer: I don't speak for my employers; they return the favor.
From: Lee Sau Dan on 26 Nov 2005 01:15 >>>>> "Peter" == Peter T Breuer <ptb(a)oboe.it.uc3m.es> writes: >> Perhaps that's because 'find' is an unstructured mess ? Peter> Find seems to me to be well suited to a dialog mode. I don't think so. Peter> What would you like me to look for? Files? Directories? How do to "files OR directories"? Peter> .. Please mark the options corresponding to your choice. How to mark: ( -type f ! ( -iname '*.bak' -o -iname '*.bk') ) -o -type d ?? Peter> What time period for file creation or modification are Peter> you interested in ... How can you make a fixed dialog which allows me to express the infinite possibilities of booleanly combining find's predicates? Yes, you can make a variable dialog, or an interface that would allow me to click "add child", etc., to "draw" a tree that corresponds to the parse tree of a Boolean expression. But how would that be easier to use than typing in a boolean expression as text? Peter> After all, it offers a finite set of possibilities in Peter> itself. You're wrong. find allows Boolean expressions to be written based on its predicates. And each predicate may be used multiple times (with different parameters) in an expression. You can't have a GUI that can accomodate such an infinite number of possibilities and is still as convenient to enter an expression as a text-based entry box. Peter> The same sort of interface as for searching a database Peter> would be appropriate. Have you used a GUI for writing SQL queries? Non-trivial queries, joining 3 to 4 tables, having a where clause which is a Boolean expression with more than 5 terms, etc.? I did use some GUI for query systems of similar complexities, and I always curse those "user-friendly" GUI for making it so difficult to enter a Boolean expression. It makes you go through hurdles and hurdles before you can get the expression done, wasting many many minutes. If it had a text input box for me to enter the expression, I would have done it in seconds. -- 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 26 Nov 2005 01:18 >>>>> "John" == John Hasler <john(a)dhh.gt.org> writes: John> Michael Heiming writes: >> There's nothing unstructured or even a mess about it, just >> another piece of rock solid GNU software, superior to its >> commercial counter parts. John> To the extent that it is unstructured or messy this is due John> to backward compatibility with its proprietary predecessors. Anything more powerful and flexible than you can handle is "unstructured" or "messy", right? But there are many many people who have learnt how to handle it, and benefiting from it. -- 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 26 Nov 2005 03:58
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: > >> Perhaps that's because 'find' is an unstructured mess ? > Peter> Find seems to me to be well suited to a dialog mode. > I don't think so. One word too many? > Peter> What would you like me to look for? Files? Directories? > How do to "files OR directories"? Click both boxes. > Peter> .. Please mark the options corresponding to your choice. > How to mark: > ( -type f ! ( -iname '*.bak' -o -iname '*.bk') ) -o -type d > ?? files whose name is NOT .bak and is NOT .bk., or dirs. That's easy. Choose both dirs and files checkboxes. Check the little extra box next to "files" to bring up the "further restrictions" menu and choose "name match". Write "*.bak" in the pattern window and > Peter> What time period for file creation or modification are > Peter> you interested in ... > How can you make a fixed dialog which allows me to express the > infinite possibilities of booleanly combining find's predicates? Because it's NOT infinite. Boolean forms have a "normal form" to which all may be reduced: the conjunction of disjuncts of atomic propositions and their negations. Now think of those atomic propositions - they are all equality matches, or "greater than" matches (for atime, ctime, etc.) or "less than" matches (the same), or shell patterns. Combinations of shell patterns are finite state machine matches (I forget which subset, but a simple subset), and the others are plain ranges. That's it. > Yes, you can make a variable dialog, or an interface that would allow > me to click "add child", etc., to "draw" a tree that corresponds to > the parse tree of a Boolean expression. But how would that be easier NO no. Think abut the SEMANTICS, not the syntax. The semantics is as I described and is limited. > to use than typing in a boolean expression as text? Because it organises your input. > Peter> After all, it offers a finite set of possibilities in > Peter> itself. > You're wrong. Don't say that to a mathematical logician. It gets annoying. [nonsense snipped] > Peter> The same sort of interface as for searching a database > Peter> would be appropriate. > Have you used a GUI for writing SQL queries? Non-trivial queries, > joining 3 to 4 tables, having a where clause which is a Boolean > expression with more than 5 terms, etc.? They exist - that shuld tell you something. It should tell you that the search terms semantics aren't as "infinite" as you think! > I did use some GUI for query systems of similar complexities, and I > always curse those "user-friendly" GUI for making it so difficult to > enter a Boolean expression. It makes you go through hurdles and > hurdles before you can get the expression done, wasting many many > minutes. Because YOU can think of a shorter syntactic form with the same semantics, maybe! That's not the point - the point is that the gui has equivalent semantics. > If it had a text input box for me to enter the expression, I > would have done it in seconds. But apparently without the auto-awareness of WHY that is necessary to your input being qualitatively different from a machine's. Peter |