Berkeley CSUA MOTD:2000:December:09 Saturday <Friday, Sunday>
Berkeley CSUA MOTD
2000/12/9-10 [Uncategorized] UID:20044 Activity:nil
12/8    Is Unitek just another Taos?  Comments on both?
            \_ This one is hard to read.
2000/12/9 [Politics/Foreign/Asia/China] UID:20045 Activity:very high
        \_ Joe "Palpatine" Lieberman is the emperor. ALGOR is a robot.
           Seriously, look at a picture of liberman and palpatine from
           The Phantom Meaance ... just a slight hair cut diff.
        \_ motd formatting god was here.
           \_ You cannot resist the powuh of the dahkside!!
           \_ There are always two, the master and the apprentice.
                \_ Like Bush Jr. isn't a tool for the Repub. party,
                   a pupet to give BJs to Cheeny and the Oil Interests.
                   Microns worse than selling the country out to China.
                   \_ What's a "pupet"?  And even if your alt.conspiracy oil
                      interest nonsense had a basis in reality, how do you see
                      opening oil fields in Alaska as worse for the country
                      than handing it over to a foreign power such as China?
                      If you're going to spew leftist noise on the motd, please
                      at least make some bare minimal effort to make sense. Or
                      would that be a logical paradox for you?
                      \_ if someone responds to a star wars quote with
                         anything remotely serious, and you respond seriously
                         you are just as much of an idiot for responding.
                         "who's the more foolish, the fool, or the fool
                         who follows him?" -ben kenobi
                         \_ It doesn't matter what he was responding to. His
                            drivel required a response, old ben.
2000/12/9-12 [Uncategorized] UID:20046 Activity:kinda low
n12/8   Is there such a detacheable X terminal? I would like the ability to
        detach my current X and continue working at home... is that feasible?
        \_ vnc
        \_ There are a few things that will do this: the vncserver for X
           will give you a detachable, reattachable X session; but it's not
           encrypted, but you can always ssh to the machine and do a
           vncviewer localhost.  "xmove" has a form of this, and there's also
           at least one more whose name escapes me.  In reality, I've never
           found any of these very useful, probably because I've never had
           a very low bandwidth X program that was essential to detach (just
           end up using "screen" usually..)  --dbushong
           \_ huh?  Vnc is plenty damned fast.
              \_ It's about as fast as running the apps remotely not in VNC
                 ...which may be fast enough for you..  --dbushong
                 \_ I've played netrek over VNC.  It's faster than that.
        \_ <DEAD><DEAD> - paolo
2000/12/9-11 [Recreation/Dating] UID:20047 Activity:high
12/8    Is UC Hastings a top-ranked law school in the nation?  A friend of
        mine who had ~3.0 GPA in some non-engineering major in Cal got into
        Hastings.  I thought that school was much harder to get in.
        \_ It's not top-ranked, maybe top-25.
           \-LSAT probably counts for a lot. --psb
                 \- why does it matter?  are you jealous that your friend
                    got away with it?
        \_ Hastings is usually ranked around 30-50 in most surveys that
        I have seen and I have never seen break top 20. Don't be too
        impressed, partha is correct: one lucky LSAT score counts for
        more than GPA. The reason is that while GPA varies from school
        to school or even major to major, everybody takes a similar
        version of the LSAT. I know law professors and they
        tell me as much - LSAT is the weeder. -fab
        \_ How does one get 'lucky' on the LSAT?
                \_ You take guesses at answers, that sort of thing... -fab
           \_ You should ask yermom what it means to get 'lucky'
        \_ I hit a homerunwitha history major who got admitted to hastings.
           Shows you what type they let in
           \_ uh, how?  Because s/he had sex with someone? Isn't that true
                of people who go to every school?  Isn't that true of you?
                \_ Some bimbo who wanted an easy lay she could get off on and
                   then forget about?  Cheap stress free sex?  History majors
                   need it too, ya know.
                   \_ Woohoo! cheap, stres-free sex. there is a god
                      \_ there is a god, he just hates you.
                        \_ there is a god, SHE just hates me.
                        \_ The point being she scored him.  He got used.
                               \_ He got laid, that's a rare occurrence for
                                  any sodan. Used or unused, sex is sex.
                                   \_ Yeah, I don't see the problem.
                                      He got laid, she got laid, everyone
                                      wins. -bclinton
                           \_ All male geekdom must mourn this event.
                              \_ jkoh
2000/12/9-10 [Computer/SW/Languages/Java, Computer/SW/Languages/Functional] UID:20048 Activity:high
12/7    How many people cheated in 162 and got caught?
        \_ A little worried about our 162 grade, eh?
        \_ Man, people who get caught cheating just haven't gotten the art
           of it down. Should've been practicing and perfecting it in high
           school (not that you needed to cheat in HS)
                \_ I'm not sure how one would beat a histogram-cheat-detection
                   scheme.  Anyone have any ideas?  Other than rearranging
                   lots of code.  And if you do that, why not just do the
                   project?  --PeterM
                   \_ how does a histogram-cheat-detection scheme work?
                      \_ I'm not sure, exactly.  This is how I imagine it works:
                         do "wordcount" on all the symbols in the program.
                         Then you sort the symbols in order of number of
                         occurrences.  Plot a graph of this.  Programs
                         which have just had one symbol substituted for
                         another will have identical graphs.  --PeterM
                         \_ Hmm, that's easy to break.  Just introduce a lot
                            of redundant local variables with different names
                            in many routines.  Also, rename your local vars
                            "i" in all routines to "n" for some routine and
                            "x" for other routines.
                                \_ Uh huh.  What if I look at keywords too?
                                   A histogram of those.  Now you have to
                                   rearrange all the control structures too.
                                   This is beginning to sound like a lot of
                                   work: you'd have to do all this rearranging
                                   and substituting and test the result.
                                   \_ just make sure you send the code through
                                      a preprocessor first, to make sure the
                                      keywords aren't being substituted with
                            \_ This only breaks one histogram method.  There
                               are probably multiple ones used.
                      \_ Call graph topology comparison
                   \_ I suspect that any kind of polynomial time
                      code comparison program could be defeated
                      without much trouble.  I say this because of various
                      results in complexity theory about determining
                      equivalence.  For example, determining equivalence of
                      Turning Machines or Context Free Grammers is
                      undecidable.  Determining equivalence of regular
                      expressions (or finite automata) is PSPACE-complete.
                      (Recall that PSPACE is a harder class than NP).
                      Since programs running on regular computers
                      (not Turing Machines) are actually finite automata,
                      if you come up with an algorithm to efficiently compare
                      programs then you can solve a hell of a lot of other
                      more important problems.  In practice, you can probably
                      defeat most program comparison tools by writing an
                      obfuscator program.  One possible obfuscation
                      technique would be to merge and expand functions in a
                      random manner, and then randomly rename all symbols
                      including the function names.  Of course, cheating
                      is unethical, but the comparison problem is an
                      interesting thing to think about. -emin
                        \_ as mentioned below, one would probably focus
                           on heuristics that work on typical undergrad-
                           LOT of results out there, and even if you CAN cheat
                           a finite-dimensional analysis of infinite-dimensional
                           spectra, it's extremely difficult to do so if you
                           don't know which dimensions they are looking at.
                           While MOSS may not be all that great, it's
                           presumably non-trivial. Listen to peterm above, he's
                           basically right. Oh, and another thing -- the class
                           of all poly-time code comparisons is way too huge and
                           it should be relatively trivial to prove that not
                           only is it not "easy" to defeat a random sample of N
                           poly-time tests if you don't know which ones they are
                           using, but it's probably complete for BPEXP or
                           something equally intractable. -alexf
                           generated, human-readable code.  further, it is
                           likely one would need to convert to an abstract
                           normal-form that is easily compared to check
                           the N-way cheating problem efficiently.
                        \_ What 172 fails to mention is the practical
                           inapplicability of most of these results. There are a
                           LOT of heuristics out there, and even if you CAN
                           cheat a finite-dimensional analysis of
                           infinite-dimensional spectra, it's extremely
                           difficult to do so if you don't know which dimensions
                           they are looking at. While MOSS may not be all that
                           great, it's presumably non-trivial. Listen to peterm
                           above, he's basically right. Oh, and another thing --
                           the class of all poly-time code comparisons is way
                           too huge and it should be relatively trivial to prove
                           that not only is it not "easy" to defeat a random
                           sample of N poly-time tests if you don't know which
                           ones they are using, but it's probably complete for
                           BPEXP or something equally intractable. -alexf
                        \_ What emin means to say is, "The cheat program is a
                           crock of shit.  It's just a slightly improved
                           version of 'diff' so don't let it bug-a-boo you."
                           They've been talking about their mythical anti-cheat
                           program for years but I've turned in tons of out
                           right stolen code with no changes and got good
                           \_ you are a hollow man.  You also are lucky:
                              the methods described above are capable of
                              catching THAT.  Your empty mind will betray
                              you when you meet real challenges, and
                              your luck will fail.
                              \_ Luck?  I grabeed shit off the printer and out
                                 the trash, retyped, turned in.  A.  Did it
                                 plenty of times.  I could do the work, I just
                                 didn't feel like wasting my time on it.  I
                                 went drinking and got laid.  What were you
                                 doing in the lab all night?
                                 \_ Uh huh, and if you'd been caught, you'd
                                    be expelled and getting your
                                    degree from DeVry.  Luck.  And lazy-ass
                                         \_ Unfortunately, the way the
                                            work structure operates, you
                                            can cheat your way through
                                            the cs program at Berkeley
                                            and yet get by in the real
                                            world with at least a $50K
                                            salary or even better.  I opted
                                            to get out rather than live
                                            the lie.  Some people don't
                                            have those compunctions, but
                                            who knows?  Maybe they'll get
                                            what's coming to them, and maybe
                                            they won't.
                                    \_ People have gotten caught cheating
                                       but they didnt get expelled. it's
                                       usually a big empty threat.
                                       \_ depends on what they're caught
                                          cheating on, and how many past
                                          offenses they've had.
                                          \_ Yes, of course the worst
                                             offenders get booted. I did
                                             not say nobody gets expelled
                                             but usually it's a big
                                             empty threat. MOST cheaters
                                             do not even get an F for
                                             the course. This is far
                                             different from being exiled
                                             to DeVry for being caught.
                        \_ and would this obfuscator be able to produce
                           human-readable code?  if a person spent even a
                           small amount of time looking at it, wouldn't it be
                           pretty apparent that the variable and function
                           names make no sense? (I guess the obfuscator
                           could use a thesaurus and randomly pick synonyms,
                           but I doubt that'd produce reasonably
                           intelligible results either.)
                           \_ This requires that someone actually look at
                              the code.  Lazy readers and profs permitted the
                              cheater above to get by.  No cheat-detection
                              method has a chance of working if it is not
                              applied, the common case, I imagine. -PeterM
                              \_ obviously, but I meant that machine-obfuscated
                                 code should be pretty easy to recognize, even
                                 if a reader only spends a minute looking at
                                 it.  Is a minute too much to ask for?
                                 (rhetorical question; I know the answer is
2000/12/9 [Computer/SW/OS/Solaris] UID:20049 Activity:nil
12/7    Do I have to compile my own X server? How? This is for Solaris 5.6
        mylinux:/home/tom% lbxproxy :5 -display localhost:0
            \_ Why would you name your solaris box "mylinux"?
        X server doesn't have LBX extension
        \_ ssh -C, while not quite as good, is a whole lot easier to use
           than lbxproxy.
        \_ 1) There is no Solaris 5.6 (SunOS 5.6 == Solaris 2.6)
           2) Xsun supports LBX only on Solaris 7 (SunOS 5.7) & later
                (Solaris 7 11/99 or previous versions plus patches)
           3) You can compile your own X server with LBX support by downloading
              the source from, but unless you're running on older
              hardware, it probably won't support your graphics adapter.
2000/12/9 [Computer/SW/Languages/Java, Computer/SW/Compilers] UID:20050 Activity:nil
12/7    Eh, why are some virtuals set to 0 (pure virtual) and some are not?
        \_ defining as virtual = 0  to create an interface definition or
        abstract class
           \_ Yes.  In particular, pure virtual functions are a signal to both
              the compiler and client code that this class is not expected
              to be instantiated, only inherited from.  It also requires any
              child classes to provide the method, but doesn't provide a
              default implementation.  There are many situations where this
              is exactly what you want.  You may not really get this until
              you're trying to make a framework of code that other people
              are expected to use, but not modify... -mogul
2000/12/9 [Politics/Domestic/Crime, Politics] UID:20051 Activity:very high
12/8    This is a rad class:  Military Science 430B
        430A-430B. Fundamentals of Terrain Representation and Military
        Operations.  (3;3)  A. Introduces the categories and uses of
        topographical maps, the military grid  reference system, map
        symbols, overlays, intersection, resection, and terrain
        association. U.S. Army small-unit tactics will be introduced.
        Topics include  operation orders, troop leading procedures,
        preparation for combat, individual  tactical techniques, tactical
        movement, and battle drills.  B. Devoted to applying the
        navigational techniques and small-unit tactics  learned in 430A
        to more complex scenarios. Topics include patrolling,
        squad/platoon offense and defense, squad/platoon sector sketches.
        Pertinent  leadership issues will also be addressed.  (F,SP)
        \_ we should have it taken in conjunction wih the cs bootcamp.
           \_ might as well get something out of college
        \_ That's cool!  Does it include physical combat training too?  Can
           non-citizen students take the class?
        \_ dude, 3 units!  it's as hard as cs164!
           \_ you mean cs264.  164 is 4 units.  And going to be taught by
              demondim spawn necula.
2000/12/9 [Consumer/PDA] UID:20052 Activity:high
12/8    \_ I wanna know how you got $650 worth of Sony card points.
           On my Sony card that would require spending something like
           $65,000.  Are you buying cars on plastic?  -tom
           \_ Wife's tuition charged to card.
           \_ tricks at the mustang ranch.
                \_ I charge my rent via paypal. Utilities and Phone
                   go through credit. And I did pay off my car via
                   credit for points along with a 1.9 APR on that. -=Aubie
2000/12/9 [Computer/Networking] UID:20053 Activity:nil
12/8    Any good internet radio/broadband sites out there?
        \_ <DEAD><DEAD>
        \_ (hearts of space)
2000/12/9 [Uncategorized] UID:20054 Activity:nil
12/8    Lots of little kids
        Running around Berkeley
          \_run-ning a-round berk-el-ey???
        Shoot them in their heads.
2000/12/9 [Computer/SW/Security] UID:20055 Activity:nil
12/8    SSH question. Any idea why I am getting this error:
        Invalid SSH_AUTH_SOCK `', it should contain at least one /.
        and it gets set to "agent-socket-21980" instead of
        SSH_AUTH_SOCK=/tmp/ssh-user/agent-socket-21980 ?
2000/12/9 [Politics/Domestic] UID:20056 Activity:nil
12/9    My favorite part of the FL SCO ruling was they included the partial
        recounts from Miami-Dade which were in the most heavily democratic
        areas but the other 80% of MD won't ever get checked.  It's so cool
        we've still got 4 of our judges on there working from the living
        constitution instead of that old raggy paper one in the vault in DC.
2000/12/9-10 [Academia/Berkeley/Classes] UID:20057 Activity:low
5/15    What do people think of 164 with Rowe this semester?
        \_ Better than 162 with Smith!
                \_ Anything is better than 162 with Smith.  Unfortately with
                        just him & Harvey teaching 162 now, everyone's
                   \_ Harvey doesn't really teach, per se.  He preaches his
                      religion to unsuspecting freshmen and chows on the
                      potstickers and is pretty much just worthless.
                              \_  Coach LUvitz!!!!!
                      \_ is he still sleeping with Randy?
                          \_ Randy Katz?  The dept. chair?  I don't think so.
                           \_ I think this fool means Brandy (not Randi who
                              is the person they are confusing Brandy with
                              and probably not Randy as in Randy Katz)
                      \_ INFIDEL!!!  HERETIC!!1!  I fondly remember taking
                         162 from bh.  Especially those nights when we'd all
                         gather around the big, silent bulk of the PDP-10 in
                         the machine room, heads bowed to its greatness . . .
                         then bh would pour the vodka, and we'd munch on
                         potstickers and talk for _hours_ about the imminent
                         coming of the Glorious People's GNU-LISP Revolution
                         that would sweep like a cleansing wave!  Some nights
                         we'd watch Animanics and listen to the Beatles until
                         dawn.  Then we'd shatter our vodka glasses against
                         the wall and weave our way home, staggering up the
                         middle of Hearst singing Russian peasant drinking
                         songs . . . I didn't learn much about operating
                         \_ Did you save this text snippet and use it
                            whenever bh comes up as a topic of discussion?
                         systems, but damn, I learned about _life_.
                         \_ I didn't know the motd had re-runs.
                                \_ sounds like something out of a cheezy
                                   movie, like "the dead poet's society"
                                   gone bad.  man, I guess that guy
                                   didn't get laid the whole time.
2000/12/9-12 [Academia/Berkeley, Academia/Berkeley/CSUA/Troll] UID:20058 Activity:high
12/8    Since when did the search engine switch to google?
        (man, those other ones sucked)
        \_ The previous (Cha-Cha) one was a SIMS research project
        \_ I guess Stanford schooled us, once again.
        \_ Is there a reason to use any other seach engine but Google?
           Hotbot used to be good. I guess I still use Yahoo, which now has
           Google backend.
           \_ Hotbot just used Inktomi anyway
           \_ I still use AltaVista
                \_ Heathen throwback.
        \_ Brewer deserves it for being a pompous ass.
                            \_ who is Brewer?
                               \_ Corporate Whore Supreme of the cs dept.
                                  multimillionaire who teaches 169 on occasion
                                  and is otherwise very hard to find on campus
                                  \_ Holy shit!  You mean he used his
                                     considerable skills, talents, and
                                     experience to improve his lot in life?
                                     How truly offensive!  I feel, no, I
                                     *demand* that he hold true to his
                                     academic background and work for free for
                                     the betterment of all!  If not, I'm going
                                     to protest by never attending another CS
                                     class again!  Down with money making
                                     Profs!  Up with the people!  Where's that
                                     bh vodka glasses into the fire quote when
                                     you need it?  You, sir, are an idiot.
                                     \_ You, sir, are a whore.
                                     \_ maybe such a busy multimillionaire
                                        shouldn't be teaching a class
                                        or conducting research if he's
                                        so busy.  if he really did want
                                        to charge the university hundreds
                                        of thousands, that seems rather
                                        crass, since his position and
                                        research at the university is
                                        \_ Have you ever met him? Jeez!
                                        directly responsible for his wealth
                                        \_ No shit, he should share some of
                                           that wealth with UC.  Aren't they
                                           using that NOW technology?
                                  \_ is he still a multi-millionaire with the
                                     current stock market woes?
                \_ the reason the Berkeley home page is not using hotbot/
                   inktomi is that Brewer demanded like a million bucks
                   for it.
                   \_ is this true?
                        \_ I'm not surprised, the going rate for search engine
                           contracts for company intranets is around that
                           range, like $50 per seat.  So for 20,000 students,
                           that's $1million.  Yes, Brewer's a fucker.
                   \_ The berkeley home page should be using "grep".
                      \_ ED!  ED!  ED is the STANDARD!  Search engine.
2019/01/19 [General] UID:1000 Activity:popular
Berkeley CSUA MOTD:2000:December:09 Saturday <Friday, Sunday>