@# Quotes DB     useful, funny, interesting





Google
 
Web www.quotesdb.info
Undernet  |  EFnet  |  Quakenet  |  Freenode  |  Dalnet  |  Ircnet  |  Galaxynet
Page: 1 2 3



Comments:

<0> its not afternoon here I was just being difficult
<1> that's not "difficult," it's normal
<1> it happens when you have, like, a round planet
<0> its round?!
<1> so they say
<1> it would also happen if you had a square planet
<0> sacrilege
<1> although instead of gliding from hour to hour, you'd have severe breaks... and those poor buggers on the poles would be screwed
<2> hey guys why I am obligated to use synchronize with JavaBens?
<1> now, perhaps if you had a sort of oblong four-sided planet - i.e., football-shaped, with the square extruded over the vertical...
<1> you're not
<3> it's polite to thou :)
<1> talios: ...
<3> sorry - obligations, being polite.
<4> usu: u use synchronized when there's need only one threads access it at same time



<4> thread
<2> aha...
<4> usu: synchronized is locking, like locking records/tables in SQL
<2> Today I started a SQL cl*** at my university...so I don't know yet howto
<2> But for many cleints connected to a server do I need to use sync or only for the same sharing resource?
<4> only for shared resource and especially if more than one plans to write/alter at same time
<4> if one writes and others only read
<4> you migth survive without
<5> Aalborg Akvavit in NZ , wow !
<4> although if it is complex stuff, it migth happen that 'base' variable is changed while derived variables are not yet updated
<6> teralaser, It ain't any good ;)
<3> teralaser, I've not seen it at many places - actually, only found it at one place.
<4> usu: synchronized does come with performance price...
<2> thx wolfey
<5> well, just the fact that you know it.
<2> the performance degrade?
<4> synchronized methods are always slower, plus all other threads block when one reserves synchronzied resource
<4> so synchronized block shouldnt be too slow
<4> usu: well play for yourself
<2> for the server VM i hope it's fast :)
<4> usu: im just giving you basic idea
<2> I appreciate that
<4> usu: imagine your bank account
<4> usu: perhaps you would receive your salary at basicly same time as you withdraw money
<4> imagine that instead of balance + salary - withdrawal you would get balance - withdrawal and no salary
<4> that's whole point of locking and 'transactions'
<2> I understand now..
<4> transaction must complete before anything else happens
<2> Offtopic: Does someone read Thikning in Java 4?
<1> I've read it
<2> Do you know any link to it! It's too expensive for me:(
<1> nope, it's not offered online
<4> only version 2 is avaiable
<1> and I wouldn't warez it in any event
<2> I undersantd...
<2> But it worths the money
<2> ?
<1> If you need it, sure
<2> thx all, gotta go...cya
<1> it's very good at communicating java concepts
<4> :::))
<1> I find that repugnant
<4> jottinger: do you copyrigth something..valuable'
<4> ?
<1> sure
<1> if it's valuable to me, I do
<1> If I put my effort into it, I have the right to say how I want it addressed by others
<4> jottinger: do you have money for lawyers to enforce it?:)
<1> Do I? No, I don't.
<1> Does that make it pointless to copyright something, to ***ert ownership?
<4> rigth without means to enforce it is worh 0
<1> so only those with money should copyright something?
<4> dunno.
<4> well if it makes you feel better, go copyright
<1> Suppose I record a song, a great one. That might be worth real value to me, whether I have the means right now to enforce it or not.
<1> So if someone takes credit for my song, if it's copyrighted in my name, I can sue them for damages and costs - regardless of whether I can afford it.
<1> Is that incorrect behaviour on my part?
<4> no no, excellent



<4> well i like GPL... if i dont get money i expect NOONE will
<4> ;)
<7> I've read that the performance of a Collection structure is O(1) except in the case of a Sorted* . Does this hold true no matter how many elements are in the collection? Say 10000? 100000?
<4> DarthEterno: test it yourself
<4> :)
<7> Hi there, remember me?
<7> wolfey?
<4> no sorry, why?
<7> wolfey: kkaisare here :) We had the pastebin session a couple of days ago
<4> oh greaT:)
<8> That's better now
<8> wolfey: You probably discerned from my code that I'm trying to build a search engine of a sort. I'm concerned about the tolerance levels that an ArrayList or a HashMap might have
<9> kkaisare: no it doesn't hold true - whoever wrote that needs some slapping
<1> a map and an arraylist are pretty bleedin' fast
<1> not quite O(1) but constant enough
<9> that's untrue
<9> there are varying seek/insert performance
<8> dibblego: Java Programming Notes
<8> I'll get you an URL
<9> ArrayList and LinkedList are O(1) and (n) respectively
<9> kkaisare: no thanks, I see enough nonsense academic literature
<9> ArrayList is O(1) seek
<1> dibblego: he said arraylist and hashmap, not arraylist and linkedlist
<9> LinkedList is O(n) seek
<1> right, arraylist is potentially REALLY nasty on insert
<9> ArrayList and HashMap are not comparable
<9> they represent different things
<9> yes, that you are forced to use either extremes is one (of many) reasons why collections are broken
<9> a HashMap is an ***ociative type, while an ArrayList is an ordered sequence - you cannot compare their performance
<8> dibblego: Is there an (in your opinion) accurate web resource for Collection performances? I'm out on a limb here...
<9> kkaisare: it's not difficult to discern yourself
<9> a HashMap is not a Collection in any case
<8> dibblego: True...
<9> like I said, you (or I or anyone) cannot compare them
<9> therefore, no, you will find no literature on it
<9> an ArrayList is considered O(n) insert with a hacked up (and hard-coded) "performance" algorithm that is supposed to minimise it
<9> I got tired of the brokenness of collections, so I wrote them properly myself
<9> I doubt you're being asked to compare performance of two uncomparable structures
<9> well, I don't doubt it - it *is* academia after all
<9> a good rule of thumb is that broad sweeping statements (even those that might not appear so at first) about performance are almost always false
<4> overall collections work faster than i'd expect them
<4> :)))
<4> all you need to worry is where use vector, set or (hash) map
<10> does anyone know how i can play my mobile java games on my computer? i have the files on my computer but cant run them
<8> dibblego: Let me get a little specific -Is there a breaking point for a HashMap past which I'd be stupid to keep trying to use it? Just curious.
<9> a wooden car works better than a 15th century horseman expects
<4> ZEKi-38-m: you can get J2ME simulator
<10> really?
<4> ZEKi-38-m: it's part of J2ME development kit
<4> yes
<10> where can i get it?
<9> kkaisare: no, size of a HashMap has nothing to do with the criteria in deciding whether or not you'd use one
<10> i have j2me..its already installed on my computer...
<4> http://java.sun.com/products/sjwtoolkit/
<9> kkaisare: if you have an asociative type, you start at HashMap - if then, you don't require ordering of elements (be it insertion or "natural ordering" (*shudder*)), then you stop there
<10> wolfey: thank u mate..
<9> if you do require ordering of elements, you find what might suit you best (a tree for your own ordering definition for example)
<9> or a LinkedHashMap (*shudder again*) for insertion order
<4> basicly i use both vectors and hashmaps
<4> rarely other collections
<9> wolfey: that's pretty silly
<9> Vector was obsoleted last century
<4> Vector is great, fast efficient and thread safe
<9> silly
<4> i use vectors if i have index
<4> i use maps for getting element through key.. hostname, username
<9> so silly I wrote a tip on it: http://jqa.tmorris.net/GetQAndA.action?qids=53&showAnswers=true
<8> dibblego: I'm building a map where a string is ***ociated to an ArrayList. Trouble is, I could have as many as 10000 strings. And the contents of the ArrayList will climb with the number of files I index.
<9> kkaisare: the question becomes "should you hold it in memory or move to a database?"
<9> and you don't hold ArrayList references (I also made it impossible to use concrete reference types when I rewrote collections)
<1> 10000 strings isn't that many
<4> dibblego: mm


Name:

Comments:

Please enter the result of the sum 63 + 46 (to avoid spam):






Return to #java
or
Go to some related logs:

LINUX process that handles virtual memory
blue pumpkins suck
#skype
ssh prioprity
#c++
darkflib birthday
qanpiena
key_cache_size
Word: r t p c r a o?
64bit bridge-utils for slackware



Home  |  disclaimer  |  contact  |  submit quotes