Greetings Question about multiple versions of Java installed on Windows XP box and selecting which plugin works
yes
Zaph0d^: oh, okay.. that’s still my point of view.
I have not changed my Security Policy from the defaults, so why would getHostName() be failing like that?
Galant you can ask yes or no in infinaty count of ways. try to ask more precisely
ah. you should’ve said so.
I thought you meant it as a proper definition.
Zaph0d^: Yea, i was thinking of adding “imo” but assumed that people will assume it’s only my opinion anyhow 
Zaph0d^: but i must mention that wikipedia says that OOP is a programming paradigm, and programming paradign is said to be a style
well, wikipedia has been know to be wrong before
is it possible to just disable the security maanger?
seriously though, java is really close to OOP. so close that saying it’s not OOP is quibbling. but if you want a real example of why java is not OOP, look at JRE pre 1.5. – all of those boxing/unboxing shouldn’t happen in OOP langauges.
Zaph0d^: but pre 1.5 is no longer Java
Zaph0d^: but more importantly, it is..
and will always be.
~tell leip about stupid
leip, what you’ve just said is one of the most insanely idiotic things I have ever heard. At no point in your rambling, incoherent response were you even close to anything that could be considered a rational thought. Everyone in this room is now dumber for having listened to it. I award you no
points, and may God have mercy on your soul.
ive always viewed java web hosting as a hybrid language
connect on 8001
Zaph0d^: What I’m saying is that all though java has taken strides forward in becoming an OO language, there will always be code that denies the paradigm since Java is mostly backwards compatable. There will always be primitives, keeping java at least a tiny bit farther from OO then, say,
ruby.
exactly what I’m saying.
(only I say smalltalk )
hehe
I was supporting what you were saying… I just chose to try and make a paradoxal joke out of the conclusion
Java has taken great strides on becoming Java
btw, I don’t hold pure OOP as a good thing by definition. I think that giving a leeway to the developer is a good thing.
In an attempts to get a grin or atleast encourage thought. I don’t see how that is stupid
Aka, gettin the work done
tmccrary++
(although the way it’s implemented in java – boxing/unboxing primitives for example – has something to be desired)
Why have the overhead of an object where just the data would do?
:s/has/leaves
The future unseen expansion that is both the bane and the buck for programmers
I don’t really mind local primitives. what I really hate though is primitives in the method signature.
It’s such a worn out argument/discussion that most of the people who would have onced chimed in are just sitting still
or in fields.
Zaph0d^: If you want to enforce non null values, some times primitives in the meth sig are useful
good thing you’ve said “most”. I think it’s my third time around
use @NotNull
That’s like 8 characters… 
Zaph0d^: don’t think @NotNull is standardized yet
no, but it’s getting there.
maybe.
cheeser can you connect to irc over 8001?
He was born with an severe 8001 related disability. It’s not nice to rub it in.
He’s got one of those handicap tunneling stickers though.
It allows him to drive through an intermediary that forwards his car from port whatever to 8001 on another server…
i’m doing so right now.
Ruin my fun 
gah that port is blocked as well
JSR 305 iirc
heh
i think they have all ports blocked but 80
i ssh out to my own host and irc out from there.
i think they block all the ports
if you have http, they can’t block all ports
(all but 80)
im using CGIrc atm
it works but sometimes i get randomly disconnected
so use irc-over-http
ah. you are
but you could install a forwarding service on your home machine that recieves on port 80
or even build a tunneling util using port 80
yeah
(see how good I am at giving advices )
Which is what I was saying, but in a joke… perhaps I should find a new medium for my advice…
right now i need to figure out why jcifs only allows 1 user to be authenticated to my app
most irc servers only operate on 6666-6668 or something like that
didnt realize freenode took connections on 8001
Handy 
im going to pull all my hair out if i dont get this working soon
build a socket tunneling protocol. it’s not supposed to be hard
jcifs?
cifs is kinda neat
especially if you don’t really need it to be that complex (you can define fixed remote addresses for fixed ports)
oh i do wish tomcat would stop breaking
How is it breaking?
every once in a while it just stops responding
whot? tomcat host breaking? unpossible.
HEHE
*hehe
oh, it’s out of heap space
ugh, how do I make a column non-persistent in jpa…
grr..
whoops ?
probably my fault
http://full-speed.org/archives/2003/12/30/tunnel_ssh_over_http.php
sweet..
oooh. out of heap space. my second favorite – right after out of permgen space.
Zaph0d^: hmmm, OOME are very rare for me, I almost stopped seeing them a short while after I started building memory profilers for java…
ernimril, not so rare in code that you don’t own, though…
why would I run other peoples code? such code usually suck!
heh
come on, don’t say there isn’t a trivial solution for excluding fields from being persisted
ucome on, don’t say there isn’t a trivial solution for excluding fields from being persisted/u
whoops
=P
other than making the field transient
pfn:I actually do most of my things myself, but iText and a few other libs are handy
rascist!
=\
is there workaround for java.util.regex.Matcher method appendReplacement to not append nonmatched text to result?
stop calling that method
only call “sb.append(matcher.group(1)); ” (or whatever group you want)
but i need regex replace too. so i need to do it in 2-passes? (1st pass, capture just matched text, 2nd pass replace it)
sb.append(someMethod(matcher.group(1))) then
does anyone have any experience getting tomcat and kerberos working with spnego? (or jetty)
~anyone
Instead of asking whether anyone works with something you need help with, please save time by asking your actual question. If someone knows and wants/has time to help, perhaps he/she will.
can someone help me getting tomcat and kerberos working with spnego? (or jetty)
yeah. that’s just “anyone” reworded.
try saying what error/problem you have
then maybe someone can actually help
I can’t find any information on getting tomcat and kerberos working with spnego? (or jetty)
~tell xaxxon sq
tell nick about factoid – you missed out the ‘about’, cheeser
kerberos is a lesson in pain
meeper, well, I have everything working except for a java server thingy. I have apache working with mod_auth_kerb..
xaxxon im trying to get it working right now too
found out i need to use Kerberos authentication instead of old ntlm =
=\
any pointers to anything about getting kerb working with negotiated auth would be REALLY appreciated
yes, none of this is particularly helpful
I’ve read most of it.. it’s rather abstract for what I was hoping for
well, sounds like you need to pay a consultant, then.
or just not listen to you
that’s an option, too.
here’s one, too.
for *the*, rather.
Nice.
gosh
i try
do trainers and consultants get paid well?
the good ones, do
s/,//
s/the good ones/the ones with the good jobs/
is it deserved?
sometimes
in my case, hell yes!
P^)=
Pirate cheeser smiley!
They appear to pay substantial amounts of it to companies who employ me, so I’m going to go with “Yes”
anybody use ant JSch? It seems to be causing my tomcat instance on my server to hang, I can’t pin point the issue but sometimes right after i run my ant deploy script tomcat web hosting wont responde until i restart the instance
Anything interesting in the logs?
(of tomcat)
If I wanted to pass a class type to a method and have it return to me an object of that type, how would I do that?
it depends…
~tell perlmonkey2 about javadoc Class
perlmonkey2, please see java.lang.Class: http://java.sun.com/javase/6/docs/api/java/lang/Class.html
Hmm, that looks like what I need
having a factory may be better….
I want a single piece of code for creating 4 types of classes.
which could be related to each other via inheritence, but currently aren’t.
not that I can tell
Hello
?
~hello
Hello, svm_invictvs_
Cool.
Hola.
/wave
yeah hi etc.
I wasn’t sure if I was actually connected.
oh damn…the secretary here is so hot, about the only thing good about working at thsi company anymore.
hehe
…I feel the same way about where I work
:
I was asking her, “Hey, do we have any envelopes?” Just so she’d have to bend over and get them out of the drawers underneath the copier.
hehe
I’m trying to make my own exception class but when I try to throw it I get cannot find symbol
classpath
did you import it?
no…
how do i do that… i haven’t used “package” in the code, would i need to do that first?
~tell opensores about package
opensores, package is http://java.sun.com/tutorial/java/interpack/packages.html and some example http://javafaq.mine.nu/lookup?254
yes
just plan on always putting your classes in a package
ok
(new java.math.BigDecimal(rs.getFloat(“Conv12″), new MathContext(100)) I am trying to get a decimal with 3 decimal places of precision, what am I doing wrong?
Using the default package isn’t a good idea.
they do not have to…
that’s the convention… meant to prevent collisions as much as possible
opensores? What are you, some kind of freetard
but sometimes it is a bit annoying to have 3-5 levels without content… com.foo.superpackage.subpackage.application
which is why all IDEs I’ve seen are able to flatten the structures
to compare two integers, would == work? doesn’t == compare references?
good for you, I do not use ide:s…
you should use equals to compare objects
I cannot answer that question as ‘freetard’ isn’t really a word so I don’t know what it means
what, are you going to call .equals? of course ==
== may work in quite many places, but will fail in a few
ah, well then I suppose it’s a bit more of a hassle
why not .equals
I assume you’re talking about primitives?
Integers
Integer, or int?
ernimril thanks
Integer
ernimril answered my question Thanks though
then equals
okee
“Integer i = 12; ” will give you the cached, shared instance of 12, so for small Integers == works if you do autoboxing, but if you do new Integer in any place that will fail… use equals and you have no problem
how often are values cached…
and when you say cached.. cached where?
static instances…..
n8
check the source for Integer.valueOf(int)
…. if(…) return IntegerCache.cache[i + offset];
sweet
how do I check the source?
actually I think I could find it
src.zip
part of the jdk
yeah I found the src.zip
I am wondering if I could include it into eclipse
Hm. I just used an object[][][][]. That’s probably earned me a special place in programmers’ hell.
and you probably can’t get there as fast as you deserve
8^)=
I’m thinking of replacing it with an object[][][][][][][][]
Mainly out of malice.
the JDK source is pretty useful
I wish I would have known about that sooner
it’s been there since 1.0
I am guessing you guys should be grateful, because now my question count should theoretically diminish
well, you can only damn yourself so much so why not?
since now I could look up my own questions
most of your questions are probably answerable just in the javadoc anyway
touche
no need to be touchy
are the sql hosting injection and other protections like that identical if using UTF-8 encodeing to support additional languages? Is there anything else that needs to be looked for or addressed?
use a prep’d statement and that concern is virtually eliminated
prep’d? What is that?
prepared
prep’d
~javadoc PreparedStatement
cheeser, please see java.sql.PreparedStatement: http://java.sun.com/javase/6/docs/api/java/sql/PreparedStatement.html
right
duhhhh
interesting.
whats “Virtually Eliminated” mean? Anything i should look out for?
And then I’m going to edit the compiler source to let me use goto.
good man.
meth( Object obj1, Object obj2 ) { obj1.equals( obj2 ); } Which equals() will get called?
~tell IvoryZion about tias
IvoryZion, Try it and see. You learn much more by experimentation than by asking without having even tried.
also note that it is usually very hard to have an equals that works well when you have subclasses
Hello
Can someone help me debug this? it never gets to the 3rd System.out.println, and it only executes the 2nd one once no more how many objects are in the argument http://ktron.pastecode.com/38743
I can watch objs.length change, and if I comment out line 7 it works as expect, but I don’t see why line 7 breaks the loop
make sure you report all exceptions you catch in your code.
Duesentrieb, Using eclipse, console is doesn’t report any
ClassCastException…
that’s why i suspect you catch & ignore it.
write a full test case…
never ignore exceptions unless that exception really is not an error in that context
ernimril, Duesentrie, so you suspect that the cast can’t be done
~tell Ktron about doesn’t work
Ktron, doesn’t work is useless. Tell us what it is, what you want it to do, and what it is doing. Consider putting some code and any errors on a pastebin. (use ~pastebin for suggestions)
TopologyNode tos[] = {};
what size array is that?
Ktron, where do you increment the counter?
0
mmmhhhmmmm
whoops, just said something awfully stupid, let me leave quietly to my bed DDD
TopologyNode tos[] = {}; — creastes an empty array. placing things in an empty array will cause a runtime exception.
yes
still, the real problem is that an excpetion is being ignored somewhere…
also, simply use System.arraycopy instead of the loop. much faster. Or better, avoid juggeling untyped arrays. there should be no reason to do that.
TopologyNode tos[] = new TopologyNode [objs.length]; …
tell the author of the code…
it’s yours now, fix it
SebastienD, in the for loop?
I do not want it…
SebastienD, ah, there’s a missing ; isn’t there
Ktron, yes yes, forget about me, I must be too tired to read code ))
SebastienD, no wait, I was right, I wasn’t Okay
why do you have to do this array conversion in the first place? why not make sure that you have a TopologyNode[] earlier?
ernimril, Duesentrieb, unfortunately I’m kind of entering developing java hosting in the middle of other people’s projects, so creating a isolated testing environment might be near ootq… I’ll throw in some debugging to try to determine
whether the casts are bad
ernimril, Because I’m getting the data as a StructuredSelection, and need to get it into a TopologyNode[]
1) make sure that tos[] has room, 2) make sure that you do not ignore the exception
set a breakpoint. step through. it should take you right to the place where the exception is cought.
ernimril, I bet its 1), and I’ll ask a more senior dev when I get a chance where I should be looking to prevent ignoring the exception and I need to figure out how to use Eclipse’s breakpoints better
And by better, I mean, at all
ernimril, Duesentrieb, thanks, you’ve been a big help
Hopefully I’ll be more armed next time I wander in
Hm. It’s really startling how much if statements in tight loops can slow things down.
have a method that “throws Exception”, and throw a Throwable from it
it will die saying the method must be declared to throw Throwable!
how moronic is that
can you unwrap the loop?
oh well, it points out a bit of a flaw in this anyway
(a bit)
Not really. Don’t imagine it would make a difference though.
if the if makes a difference, and you could unwrap 10 iterations then you would reduce the effect of the if by an order of magnitude, surely?
Well there’d just be 10 times as many ifs, and you’d have lost all hope of the loop fitting in cache.
JSTL is win.
pure liquid win.
JSTL kinda sucks. pure liquid suck.
how so?
doh – I see what you mean – I’m just being dense
a taglib based on xsl and adding sql capability? not… very… wise
then again, I’m biased
I was confusing if with the loop condition
heh, the sql host taglibs are hilarious
i’m sure they have their place.
They were retarded when they were conceived of
I’ve actually found you can get noticable speedups if you do move ifs into the loop condition.
note that I was on the JSR that created that monstrsoity
If you can anyway. It’s obviously not always possible.
whyyyy
interesting – I guess the compile (or the JIT) is better at dealing with those
I’m so sorry
No, I think it’s the CPU cache.
the JSR was running really well for a while
what’s wrong with JSTL ?
it makes JSPs a whole lot less ugly
Or something like that. I’m not as knowledgeable as I’d like about the details.
there’s a lot more ugly to get rid of in JSP
But reducing the number of branchings in the loop means that more instructions can be reliably queued up.
there is ?
ok. So what’s the plan to speed up your tight loop with an if in it?
Plan is “I can’t be bothered”. I already have a faster implementation, this was me trying something else to see if I could speed it up. It didn’t work.
bPlan is “I can’t be bothered”. I already have a faster implementation, this was me trying something else to see if I could speed it up. It didn’t work./b
JSP is really kinda ugly, in addition to not being especially fast
it’s just SIMPLE
how is it ugly ;o ?
well, that’s subjective
PHP isn’t pretty, but it’s functional… and some things about PHP are much nicer
they are ?
If I had to do it, I could probably optimise it by letting a lot of array accesses go unchecked, catch the exception and deal with that case separately. It would be speeding up the mainline case at the expense of slowing down the failure case.
does php hosting even allow MVC architecture ?
~buzzword
svm_invictvs_, buzzword is leveraging synergies on a daily basis via the usage of mutually exclusive paradigms.
Can you get javabot to echo arguments passed into the facoid?
yep
~ridicule svm_invictvs_
i killed him
isn’t that always the best approach? unless you’re expecting plenty errors by default
and mvc is hardly a buzzword
web2.0 is a buzzword
g[r]eek: They’re not really errors. They’re expected to crop up in normal usage.
g[r]eek: And it’s a nullity check, and I always feel bad about catching NPEs.
hehe
array.compact!
oh wait – that’s not Java
?
What is it?
stick to jstl and so as little in tomcat jsp host as possible. if you’re doing much more than simply calling your custom tags, it’s ugly
*and do as little in jsp as possible
g[r]eek: i’m just using the core tags
! man, custom tag functionlity rocks.
it’s a Ruby method to get rid of the nulls in an array
i don’t need them i don’t think
build up your own reusable library. it will save you plenty of time in the long run
or use JSF and facelets
i haven’t found any use for them yet
yeah my next point
Ah
i’m just iterating through arraylists or printing values stored in a session
or both
all the work is done in mah servlet.
yeah but where are you handling the View in your mvc
(Get rid of the nulls in an array? But but… that makes no sense! *headgoboom*)
g[r]eek: in a jsp
let your servlet act as the controller. let your tags handle the view
yuk.
God, I hate this.
they are
This is such a joke.
it makes the array shorter. This wouldn’t really improve performance of course, since it must take order n.
well i guess if the standard tags are enough for you…
Oh, ok. So it doesn’t remove nulls from the middle.
imho custom tags are very exntensible. can do lots of cool thing with ‘em
and so can you with jsf
yes it does – and pushes the non-nulls along
Ewww
too many web technologies, too little time.
JSF is a standard for Java EE
short sacrifice in short-run = time gained in long run
how do, g[r]eek
yes if?
once i’ve finished this webapp i will ignore it for many months
you should at least be comfy with the foundation api
i’m not suggesting you go learn struts or spring (although you probably should), but at least learn jsf and what it’s good fr
what is it good for ?
you might find you’ll speed up development time as a result
^
~tell varek about JSF
varek, JavaServer Faces technology simplifies building user interfaces for JavaServer applications. Developers of various skill levels can quickly build web applications by: assembling reusable UI components in a page; connecting these components to an application data source; and wiring
client-generated events to server-side event handlers. More can be found at http://java.sun.com/j2ee/javaserverfaces
sounds like something i really don’t need.
so far i have two view pages
total of four .jsp files
two servlets
yeah
stick with what you have
people didn’t need cars until they were invented.
bah, use what works
and when they were invented, they were only for rich people
or rather, people didn’t need what they didn’t know existed
isn’t it great that software is free then?
isn’t it horrifying that hardare isn’t
this is besides the point
it was merely an analogy
you’re an analogy
i could have said condoms
that probably wouldn’t have related as much.
yes well rome wasn’t built in glass houses.
haha
g[r]eek: …
now you’re just being silly.
lol
my wife is reading peter pan to my youngest
…. in character
apparently peter pan is from the deep south
how cute…
“awl yuh gotter dew is use some o’ thiyis pixih dust.”
anyway, jsf is cool, take a look at it sometime
I’m about to die laughing
webcam?
no
I break cameras
so many inappropriate responses… so little time
oh my
captain hook… being a *touch* dramatic
g[r]eek if he had a mac, he would be able to give you a picture.
ug[r]eek if he had a mac, he would be able to give you a picture./u
captain callbackhandler
tinkerbell sounds like eeyore
and wendy can’t do anything but shout
sounds like your wife has a promising future as a voice actor
or in needing psychological help
normally I’m the one who’s … um… whacked out
where do I get the java ftp commons library?
bbl
wendy had better shout, she’s tiny!
jaarta.apache.org
probably from apache since they usually hold the commons-foo packages
oh that’s right it’s from apache sorry
jakarta.apache.org
I was sorta sad when wicket turned apache =/
hmmm, commons is actually an apache top level project now: http://commons.apache.org/
THX I found it
you got a new amp?
Hi! Does anyone know a good java beginner book? And maybe an online version?
The book should also cover basic programming concepts such as various data types
~books
http://java.sun.com/docs/books/tutorial/information/download.html
hm.
didn’t expect that to work
Hello all. I have a question regarding cleanup of an object. Lets say I have Object A, and everytime I intialize A, I should eventually clean it up… is there a way to force this behavior?
eg. if Constructor A is called, A.B should be called at some point.
thanks
~tell jnguy about gc
jnguy, gc is http://java.sun.com/docs/hotspot/gc1.4.2/faq.html and http://www-128.ibm.com/developerworks/java/library/j-jtp09275.html?ca=dgr-lnxw07JavaUrbanLegends
I don’t think gc is what I’m looking for…
i do believe it is !
write your own garbage collector
ie use object pooling
hm. I’m going to look into that.
but what do i know! my brain goes into low power mode after midnight.
basically, the constructor to the object modifies a configuration file that must be reverted before the testcase finishes, which isn’t necessarily when the gc runs or when the object is no longer referenced. Ultimately, I can ensure that A.cleanup() is run everytime I initialize A, but I can’t
ensure that other developers will follow when using my API.
by I can ensure, I mean, I’ll remember to do it everytime I call the constructor.
junit has a tearDown method
Exactly, something like that, without junit.
run it inside a try { } finally { } block I guess
(We’re writing our own test harness)
I was thinking of something like that, or the cleanup method sets a member variable that is checked with finally or the gc methods that are called.
*finalize.
Thanks for your time guys.
you are making a mistake if you’re depending on the gc to cleanup your object immediately (and hence call finalize)
yeah, thats why I didn’t think gc was right… but I guess I’ll just hope the other developers remember to call it
that is bad design too
err. I have to go right now, but I would love to discuss any innovative ways to enforce it.
Why does finalize() even exist?
cheers.
there are none, consider the fact that you have many close/dispose/flush methods…
for newbies
Ran jsizer on my code.
yeah, the only way I can think of is to directly modify the compiler to check it… not a pretty solution,.
did you make it crash?
to close files that an object has openned to write into it?
guess I made svm_invictvs_ crash…
eg a logger
what now?
I missed what you said.
HI all. If I have an open and connected Javamail store, with the inbox folder being viewed, how do I check for new mail?
to close files that an object has openned to write into it?
eg a logger
dfr, yeah, but said finalizer may not ever get called.
hmm? in what scenario?
dfr if the GC never collects the object.
if the GC never even runs.
with a large enough heap gc will never run….
mmmm, I want such a machine…
or in a very small program, or one that doesn’t perform alot of allocations.
svm_invictvs_: not even when jvm shutdown?
finalise() gets called by default when object goes out of scope
no
g[r]eek no.
g[r]eek: nope
the OS will reclaim stuff, no need for the jvm to do extra work
so how’d i go about making sure the files close once the object is done to be used.
s/\./?
try – finally
that’s news to me
always close/flush/dispose all streams/db-resources/images/windows/graphics inside a finally-block
It’s only called once by the garbage collector for the life of the object
good to know
g[r]eek: gc doesn’t collect until it’s oom
g[r]eek: this is ##java not ##vb
and it’s only called once the gc determines that that object is garbage.
dfr that’s not quite true
finalise() is called when gc determines no more references to it
or unless one specifically calls it?
dfr, no the gc will run whatever the fuck it wants.
dfr you can’t specifically force a garbage collection.
svm_invictvs is most correct here
System.getRuntime().gc() wont do it?
no
not necessarily
and calling that method ought to be a crime
brainfart i see
dfr It’ snot guaranteed.
there is a jvm option to turn System.gc fully off. Any serious java server should set that flag….
i never said i did that. Even though the somewhere suggested usecase was to do that when waiting for io stuff
sure, I am just making my opinion clear…
fair enough.. but it does “make its best effort”
System.gc is a hint, for normal client vm it will usually be respected, but if you change gc or if you run -server then it is a high chance that it will be ignored…
i was jsut going by what’s in the javadoc.. thanks for clearing up the reality
oom ?
out of mem?
g[r]eek: ya…
k
g[r]eek: probably should have spelled it out…
anyway, I’m off… thanks for enlightement
later
I mean, if you *really* need more explicit control over the garbage collector. You’d have to do something like write your own VM that would have a JNI interface to control it…
the jvmti has a call to force gc, easy enough to use, quite handy when you write a memory profiler
or, more practically, write a program that links libjava et al, and provide an interface to it.
what’s your memory profiler’s url again?
ernimril knows more about this than I
~tijmp
ernimril, tijmp is a small, fast and easy to use memory profiler for java/6 and later. See http://www.khelekore.org/jmp/tijmp/
I ran jsizer on my code, hehe
did you find anything interesting?
I’m having trouble finding which class is which box.
do the tooltips not work as they ought to?
jsizer – is that that cool height/loc prog?
no, I’m not seeing them.
g[r]eek: yes
~jsizer
g[r]eek, jsizer is a tool that gives a graphical overview of the classes, use it to find large classes. For more information see http://www.khelekore.org/jsizer/
odd, they work fine for me, of course…
this should be fun
the menu also hangs forever.
i have a facade that’s growing bigger than i’d liek
odd, can you get a stack trace? (C-\ (unix//linux) or Ctrl+break (window) in console)
I thought facades were supposed to shrink
decorators grow
well sure, but the amount of objects is referencing quicker than i refactor
*the amount of objects referenced is growing quicker
I’m sure that makes sense in-context
Do you print the stack traces out to stdout or stderr?
sure i see what you’re saying.
yes it is a decorator.
and it happens to act as a facade. and i am concerned about the rate at which my decorator is growing
if you press that key combo in the console the jvm prints the stack trace in the console
ah
mind you i’m not growing the decorator dynamically. anywa besides the point
please check cpu usage when the menu hang…
50%
but this is a smp system
pardon my “object out of scope calls finalise()” comment. now that i look back on it i can’t help but kick myself.
that looks like one cpu full speed…
yes
I take it that you run windows?
g[r]eek: hehe no worries. I’ve said some doozies
here at work, yes
I can try it at home on IA32 linux ^_^
WRite once, debug everywhere, heh
let me open up the source code.
would be very interested in stack traces from hanged/full cpu situations… if you can get it please mail me one…
I’ll open it up in eclipse, and break it.
time to sleep… see you tomorro^Hlater today
~jsizer
svm_invictvs_, jsizer is a tool that gives a graphical overview of the classes, use it to find large classes. For more information see http://www.khelekore.org/jsizer/
omg, they websensed your site.
hmm jsizer not showing anything for me. i add jar and run the sizer, remains blank though
you have to run it…
control + R
“and run the sizer”
;p
oh, lol
g[r]eek: it works on classes, not source, does your jar hold classes?
brainfart
hmm odd. restarted it works now
…zzzzZZZZzzz
I had your code results ready. But I eated it.
interesting – i first loaded a jar which didn’t (to try break it), and blank, then i loaded a jar which did, but remained blank. then quit, restarted, loaded jar with classes, and works fine. but no tooltips for me either
heh, neat prog
all my method lines are returning as 0 though
all boxes 100% black
can log4j be configured to split the log into multiple files?
if I used PreparedStatement, am I using a stored procedure?
yes
tieTYT, no
how?
I’m sure the docs tell you
thanks
wlfshmn, I’ve read the docs a 1000 times
I need a log file per class instance
I don’t care about threads and other stuff
why would you think a preparedstatement is a stored procedure?
create multiple appenders and attach diffrent loggers to diffrent appenders
i said stored procedure when I meant to say preparedstatement and convinced myself that I was right
how? could you be more specific please?
it was just a brainfart
the ILog has only Debug, Warn etc.
earlier on i said that finalise() is called everytime an object goes out of scope. don’t stress
this isn’t typically done in code, it’s done in log4j.properties
has anyone ever used ant + JUnit + Emma. I am trying to find some sample build.xml files and I checked all over Google, but no luck
wlfshmn, it gets more and more complicated
not really, the only thing you do in code is get a logger object and call the warn/error/info etc methods on it. everything else is done in log4.properties
wlfshmn, I’m not seeing the big picture, since the classes are pretty decoupled
wlfshmn, you mean the config file?
yes
wlfshmn, and as I said each instance of the class needs to write to a different log file
wlfshmn, and I’ve already read the stuff on multiple thread, multiple log files, but I don’t need this, since I have a single thread
each *instance*? that I don’t know about
http://fn-javabot.svn.sf.net/viewvc/fn-javabot/javabot/build.xml?revision=420&view=markup
it uses testng, but the gist is the same
wlfshmn
thanks cheeser
did your jsizer show varying shades (loc trackin)?
*tracking
mine seems to be borked
yes it did
how long do I have to be here to be considered a “regular”
I’ll load up the souce tonight, if I remember to do so.
just out of curiosity
Of to SIGGRAPH.
until moss starts growing
that’s along time
~jacorb
g[r]eek, I have no idea what jacorb is.
woah jsizer owning my cpu
Codex_:heh
g[r]eek, rather
is there a way to specify how characters are printed (ie, as bitmaps vs. curves vs. indices into font table)
yours too?
mind you, i tried it on a 5mb jar
It tends to use some CPU.
one class had 180 methods
and it was abstract.
i’d hate to have to extend it
g[r]eek: lol, that is insane
or extremely refactored :-/
what’s ironic is that the jsizer concept is centered around the idea of spotting classes that could use refactoring d¦-D
i forget which class it was. i think it was AbstractResultSet from psql’s jdbc driver
Greetings. Question about multiple versions of Java installed on Windows XP box and selecting which plugin works with IE and Firefox. How do I switch between, say 1.4.2_06 and 1.5 Update 12?
go right-click my computer, find environment variables
environment variables
edit JAVA_HOME
assuming you have both versions installed
Oddly enough, there is no JAVA_HOME variable listed.
what happens when you open up a command prompt and type java?
without the ?
I have three different versions installed – the two listed above and the latest 1.4.2, namely 1.4.2_15. I hate it, but that’s all that are supported.
I get a spew of usage. java -version returns 1.5.0_12.
ok well then your java home is definately set
check your PATH
“set” returns a lot of env vars, but no JAVA_HOME or anything else JAVA for that matter.
yeah but i’m guessing that your PATH variable contains your java home folder
just tweak that
see if it works
Amazingly enough, there are no rerferences to java anywhere in any of these env vars.
Hence, my stumpage.
I’d agree that under Linux, your approach above works well. Under XP, however, I’m stumped. No environment variables, nothing java-related in the path.
There is a java.exe in cwindows\system32 that seems to be a copy of the one in cprogram files\java\jre1.5.0… which is, of course, the 1.5. Update 12.
I didn’t know that the installer polluted the system32 directory like that.
well add your java home folder to your path before the bit where system32 is defined
And I guess set JAVA_HOME at the same time, right?
then check that when you call java -version, it is looking up java.exe from your specified folder (test by changing to different versions etc)
sure
it’s typically used by 3rd party software. good idea to have it set
Let me try that.
remember to put it before system32
(does this count as java support i wonder)
Yeah. What I’m attempting to do is get away from the Oracle-tweaked Jinitiator for some of their stuff, namely Oracle Application Server. Jinitiator is a customized version of 1.3.1. It doesn’t like to authenticate through certain proxies, like Microsoft ISA. Moving to a “real” JRE, according
to Oracle, might fix that.
Now, JAVA_HOME points to the root of the installed version, not the bin directory underneath, right?
yeah
but PATH you specify bin
OK. Done. “java -version” shows that indeed 1.4.2_06 is now live. Problem is now that somehow Firefox doesn’t see the plugin properly. It complains about not finding the 1.4.2_06 plugin.
And IE still loads the 1.5.0 Update 12.
try restarting them (note that this is not my area of expertise)
(as with most things ;p)
i don’t think IE uses PATH or JAVA_HOME.
it’s uses a plugin DLL
So, Duesentrieb, it’s picking up on the wrong DLL. I installed these three JVMs in reverse release order, so I would have expected the 1.4.2_06 to be in effect. I’m wrong, of course, but I don’t know how to “fix” this. This problem will be especially tiresome with a lot of “legacy” JVMs
hanging around…
well, i havn’t touched windows in a while, but i expect that ie looks at some registry entry to know what dll it should load.
Yeah, I wish I didn’t have to touch Windows at all, but as a consultant, it pays the bills sometimes. Working with Oracle, you’d better be well-versed in a multitude of products and OSes.
when I implement the Comparable interface, does that always yield a n^2 solution?
or it’s in a Plugins directory in IE’s installation dir – that’s how Mozilla does it (and IE stole the pugin architecture from moz)
hm? n^2? for what?
I guess I’m a little confused how comparable really works? Doesn’t it compare each object with it’s neighbor?
Thanks for that pointer. Only thing in the PLUGIN directory for IE is an Adobe PDF plugin that I’ve got installed.
comparable does exactly what you tell it to do. it’s an interface
it may be used to build a tree (as in TreeMap), or to sort a list (as in Collections.sort), among other things
^_^
how does comparable yield an n^2 solution
after i’ve logged into a webapp, what allows me to securely go from page to page?
tieTYT, ssl
complexity depends on the algorithem you apply. implementing Comparable just theves the algorithem a way to compare things.
what if i’m not using ssl
right, well consider my operation in the comparable implemention as O(1)
tieTYT, then nothing
this is more basic of a question than you may think i’m asking
yeah i think he means the “does each page check that i’m logged in” type
there is no security without transport security
then you can expect sort to be in O(n*log(n)), and lookup in trees to be in log(n)
the comparable interface must have some mechanics of moving about a collection, i mean is that n^2?
boohoo, no
insert/remove may be a tad slower, no idea what balancing alg is used
ok, so basically merging, hmm ok
the comparable interface doesn’t perform any operation on any collection.
uhhh hmmm
boohoo – comparable only says. “provide a compareTo method to compare yourself to a given object”
yes, when I click on the next link after being logged in, how does it confirm that I am authenticated?
tieTYT, session cookie
think about this: if you want to sort an array, it’s elements must be comparable.
it depends if you’re using a programmatic security model or declarative
makes sense?
yup
but yeah, using session
what prevents another user from making up his own session cookie?
or URL rewriting if you’ve got cookies disabled
tieTYT, the fact that it doesn’t match what’s on the server
the server has a correspondinglasdfsadfl;safd
tieTYT, there’s nothing preventing a user from snooping a session cookie–thus, why you need ssl
pfn … ;P
I can pass “mysessioncookie” all I want to the server
but it won’t get me anywhere, unless the server also knows about “mysessioncookie”
are the session cookies usually tied to the username?
no, and depends on the application
ok
best is to read up on how session works
and start with declarative security imho
Ok, I’m reading about comparable, and it’s possible that I’m just not thinkign straight
thanks
declarative being the standard?
in any case, session cookies are best if they’re cryptographically secure random numbers
g[r]eek, depends, if you’re using servlets, yes
g[r]eek: well i’m currently reading the http://www.owasp.org guide but it hasn’t gone into these details (yet)
read this: http://en.wikipedia.org/wiki/HTTP_cookie#Drawbacks_of_cookies
ok thanks
www.moreservlets.com
one of the chapters (it’s a free ebook) covers security
so if I want to make a class comparable, do I need to say implements? or is that just a compiler hint to throw an erorr if it isn’t?
boohoo, you need to implement Comparable
boohoo, java is strictly typed
strictly/statically
ok thanks
okies
and in most tutorials, it says public int compareTo(object ..) { .. }, can I just put compareTo(myClass o) { … }, jdk 6.0 doesn’t mind it
you can only do myClass if you specific it generically
nice link
{ public int compareTo(MyClass) }
if you say implements ComparableMyClass, then, and only then, yes.
by, too late
aahhh
~tell boohoo about generics
boohoo, generics is http://java.sun.com/j2se/1.5.0/docs/guide/language/generics.html Generics tutorial: http://java.sun.com/j2se/1.5/pdf/generics-tutorial.pdf FAQ: http://www.angelikalanger.com/GenericsFAQ/JavaGenericsFAQ.html
yup saw that, eclipse also suggested to make the class abstract
OO world is so bewildering and awesome, I’m excited
does anyone have the Java 6.0 source to hand?
because you didn’t implement all the methods the (raw) interface defines
yeah, wow, I’m far from knowing Java
if you do, could you tell me whether javax.management.MBeanFeatureInfo tries to read any properties in its static initializers. (I’m expecting something like “jmx.serial.form”).
doesn’t make much sense to me, the class I’m trying to sort is MyComparable, but when would a situation arise that I would need to pass a differnet class other than the one I’m working with to the compareTo
huh?
{ .. }
you usually never sort any other type
ok just wondering, cause your statement said, MyClass and MyComparable, so thought ..
the generic param for Comparable is indeed usually the class that implements it. it may seem redundant, but it’s simply consistent with the way generics work.
the interface itself can’t know what class is implementing it. you have to tell it. (this is very much simplified, and somewhat inacurate, but i hope it helps anyway)
yeah, so i thought, but if you read pfn’s statement, it said, MyComparable implements ComparableMyClass
and not MyComparable implementsMyComparable
ignore my statement
!javadoc
ok
anyway, you need to start at something more basic
~javadoc
I guess the factoid ‘ant javadoc’ might be appropriate:
and then work your way up
http://ant.apache.org/manual/CoreTasks/javadoc.html
and read and understand generics first
true
thanks
I love ##java, much nicer people than efnet
and consider writing a Comparator, instead of a Comparable.
okies, googling it
drink bleach
and why is that?
I prefer pepsi, though
confused now.
is comparator better in any way?
for ServerSocket class, should one use TCP or UDP?
there is no one in #eclipse, but I was wondering if anyone knew how I could generate an ant build.xml that mimics the exact thing that eclipse does
when it runs my unit tests
someone please type me a message with my nick prefixed. i’m testing somethign quick.
are g[r]eeks quick?
hmm doesn’t beep when minimised to tray :/
g[r]eek: forget it, no one will do that for you
joed
oh wait, joed typed g[r]eeks
vinse please one more time
iz g[r]eek quick?
shweet works
~tell g[r]eek about joed++
g[r]eek, joed has a karma level of 50, g[r]eek
hah
~tell boohoo about boohoo
hmmm, not very happy with moi name is it
~botsnack
g[r]eek++
hmm
~tell boohoo about API
boohoo, API is http://java.sun.com/javase/6/docs/api/
~tell boohoo about nothinginparticular
boohoo, I have no idea what nothinginparticular is.
~tell boohoo about boohoo
boohoo, I have no idea what boohoo is.
that’s better
i’m a human
hey all
for a JWS client application, which logging library would be best to use? slf4j or commons-logging? even the jdk logger?
for a JRE 1.5+
~poll
Instead of asking whether anyone works with something you need help with, please save time by asking your actual question. If someone knows and wants/has time to help, perhaps he/she will.
define ‘best’
also, is there a way to forcibly disable logging with slf4j or commons-logging, ie I would not like someone to enable TRACE level on a deployed app and see what’s going on
maybe one that is small and where I can disable the logging output
i mean, disable it from being able to log
I think you can do that in all of them, but being Java you still have a properties file or something like it? You missed log4j as well.
Since it’s a client application, I would prefer to have flexibility on it’s logging. I think slf4j is good because it has a null jar library which is very small
the production app I don’t want to have it log anything
it should even be forced off
anyway, I think I will go with slf4j
it seems to be quite flexible
awesome.
goodnight
~j2ee
nzuhdi, j2ee is Java 2 Enterprise Edition. See http://java.sun.com/j2ee and an overview at http://java.sun.com/j2ee/1.4/docs/tutorial/doc/images/overview-j2eeArchitecture.gif
~j2ee tutorial
r0bby, j2ee tutorial is http://java.sun.com/javaee/5/docs/tutorial/doc/
~j2ee is seej2ee tutorial
I already have a factoid with that name, r0bby
~no, j2ee is seej2ee tutorial
Okay, r0bby.
JSF vs RoRs who wins?
Nobody!
wicket
wicket is the winnar.
if you know swing, wicket is easy to pick up
you tease?
javabot tell perlmonkey2 about wicket
perlmonkey2, wicket is a component based web application framework written in java. See http://wicketframework.org/ for more details.
hehe thanks
I’m already there, but tahnks
you can’t manipulate wicket widgets except on the web dispatch thread? /tongue-in-cheek
component based? Then it competes directly with JSF?
What competes with JSF besides JSF?
joed I have no idea. First I’ve only been really developing with Java for 3 weeks. Second, I just learned about JSF’s today.
Is it just me or is JSF a desperate attempt to play catchup with aspx?
Speaking strictly as an old “handle request, produce response, end of roundtrip” guy JSF and ASPX are both strange to me
Oh.. ojacobson I’m actually not sure, I’ve looked at components, frameworks, etc, I don’t get them. Nor do I really see their use-cases.
If I wanted to deal with component events I’d be using flex
Which is why I like wicket. Gwt does some nice stuff too. But the Laszlo does as flex does.
(flex does rather have that “hope you have gobs of RAM” thing going on)
Oh, same in laszlo I think, but they do nice stuff, while you still scratch your head and say… a jnlp might have done more?
I guess it still does not beat the pain of trying to accomplish anything complex in springMVC
does anyone know whether the .classpath file in eclipse overrides the CLASSPATH environment variable
when building
I would expect it to
but I don’t know that for certain
for a project yeah I think it does, would not bet money on it but the eclipse builder I think picks that first.
oh I see… thanks
Don’t use the CLASSPATH env variable, it’s a leftover from the era when people thought having one giant classpath for everything was a good idea
(“then you only have to update each class once!”)
(dependency fucking nightmare, that.)
What’s wrong with maven?
no idea, I use it pretty freely. It’s definitely baroque and… different, though
I am trying to use ant to run unit tests and it seems that some of my classes are not being found.. I think it’s a classpath issue
and the things missing from the central repo are kind of a pain in the ass
It kinda sorta does the CLASSPATH.
and test running, and packaging, and dependencies. That’s all I really use it for.
The reporting and site crud seems like excess chrome to me.
And project building++
(LEARNING to use maven is fraught with irritation)
(not least because the docs are balls)
The Better builds book is good.
and once you have done a few quirky things you get to like it.
I came at it originally from the direction of “I want to reproduce the JARs I get from ant alread”
which lead to no end of headaches
long since discovered
Yeah, it is a slightly more fascist approach to things, but when understood nice.
Mostly it stops you from cutting corners without putting a lot of effort into it
some of those corners would be really nice to cut sometimes
Heh
(classpaths
there has to be a good no bullshit book _ABOUT_ maven
about?
nevermind
(heh, yeah, maven does require wading through the ASF’s self-righteous crap
(I find it pretty amusing that Apache puts out two complete build systems with almost diametrically opposed approaches)
heh yeh
Maybe I’ll start building all my projects with rake.
Better builds with Maven
heh
anyone ever have the problem where if they run their build.xml file in eclipse.. nothing works, but if you run it from the shell.. everything works fine?
the only difference between the environments is the .classpath
so theoretically it should be identical
yes
No…
usually the cause is a property environment=”env” /
DDT.
Eclipse builds it for an ant build?
hmm.. I am not using property environment=”env” /
Oh, you are saying that an Eclipse plugin sucks?
long offset_basis = 1099511628211;
haha
so it is a plugin?
long offset_basis = 1099511628211L;
shouldn’t it fit?
what aspect?
what does 19%5 equal to in a Java expression?
note the L
3?
19 mod 5 is the same as 14 mod 3
you should be able to work it out from there
er, 14 mod 5
I tell lies
thanks
4?
this could use a explanation if that’s not correct…
It is.
% is the remainder after division (modulus)
ok
so the remainder of 19 / 5 is 19 % 5 is 4
go google modulo division.
ok, I’ll have to do that soon, but I’m also busy finishing some self-review questions, and I believe I’ve been doing a pretty good job on them
http://en.wikipedia.org/wiki/Modular_arithmetic
http://en.wikipedia.org/wiki/Modulo_operation
first explains the concept; second is completely relevant to programming.
also if(n % 2 != 1) { // odd }
handy
crap
s/odd/even/
ridicule r0bby
you should use != 0
though i’m not sure if that still holds in java as opposed to, say, C
it was questional advice in C, too. jz and jnz are about equal cost.
And the compiler is free to got “well, that can only be 0 or 1, so I might as well just use jnz”)
that was fixed, as i meant even
!= 0 would indeed make it odd though
i’ve always wondered about that.
that’s entirely beside the point
ojacobson knows what i’m getting at. 8^)=
I think i see his point too.
elaborate
if you have a situation where the cost of one compare and jump vs just a jump makes a noticable difference, take the fucking branch out of the loop
I did tinker w/ x86 assembly :/ I was never the same
heh. excellent advice
(using memfill to duplicate opcode sequences and then running them is a poor man’s loop unrolling. )
and once again junit saves me from embarassing myself
If i say anything else, i’ll appear to be an idiot; i slightly understand it
Oh. I was lost, ojacobson has done much more than I in the field, last I wrote C I did serial drivers for cell phones.
a straight transliteration of the expression “x != 0″ or “x == 0″ will be two operations on an x86
Oh that, yes it will.
mov eax x; jnz LABEL or mov eax x; jz LABEL
by comparison any other number needs either a SUB or a CMP in between
(CMP is just SUB in disguise anyways)
You still remember this.. I;m amazed.
yes, i know x86 assembly slightly, though not proficiently
yeah. So the prevailing wisdom was that that extra operation *mattered* for efficiency, so you should always compare against zero
I got the programming x86 when I think I was 14.
see above where I said “it was questionable advice”
I’m baffled
I got it fairly recently (a couple of years ago), but more as a curiousity than a serious hobby or employment
I should probably work on my assembly skills; i hated it.
that explains it…. It is now 22 years since I read it.
(I wrote something that made bochs print a message on boot, which was fun)
too much jumping around, i couldn’t keep track
(never got up the nerve to run it on real hardware, and I couldn’t find a floppy drive anyways)
heh
I remember ‘cli; hlt; jmp $’ at the end of it to keep it from walking off the end of the boot sector.
heh
I didn’t write any low-level things, it was all run in protected mode
protected mode is much nicer
I still don’t grok how to set up protected mode, though… too many layers of indirection between linear addresses and memory addresses, with too much code that has to understand each
bI still don’t grok how to set up protected mode, though… too many layers of indirection between linear addresses and memory addresses, with too much code that has to understand each/b
A, B, C, D, SI, DI, IP, and four segment registers
i think it was the addressing that you were limited by, i *THINK* if i’m remembering right
heh 
(fuck segmented architechtures.)
Right, I actually loved 68x
mul++
That, I never learned
I had a way to memorize the segment registers :/
13/4 = 3 … because either operand (or both), are not integers, right?
both
type promotion goes to the wider of the two types; double is wider than long or int
i just memorized them, never used em
lucky you
I have one opcode to illustrate why: REPNZ MOVSB
REPNZ MOVSB
Which moves data a byte at a time from ds:si to es:di until it encounters a 00 byte
what?
That’s from the keycapture?
cs == code segment ds = data segment ss = stack segment es fs gs
segment registers (16-bit)
notice that I did not say 13.0/4.0
right, but if either had had a decimal point it would’ve promoted the other
i *THINK* i even got em wrong
LOL
would you believe, I even know about the bizarre interrupt quirk for ss?
you definitely know your shit
what the hell.
ok, well, that’s a good refresher of my memory
does anybody know how to textually edit eclipse options
s/of/for
i had this perverse habit of using the push and pop for saving data
drove my prof nuts
#eclipse
Vista sucks with Eclipse, I cannot for the life of me, edit the keys associated with this!
heh, stack machines
oh yeah
try that with the stack pointer sometime.
i didn’t want to declare variables and move, it was easy
only problem was, keeping track of what was on the stack at any given time and the fact it’s First In Last Out
incidentally
ENTER x,y — good idea, horrible execution
(reserves X bytes of stack for you)
heh
(ended up being slower than just subtracting from ESP, though.)
I then used the base frame pointer
and I liked that even better
now back to java
writing assembly taught me only one thing that really matters
“the guys who wrote my C library are smarter than me.”
(my strcmp was about 1/4th as fast as the glibc one)
heh
you’re pretty smart yourself (I suck up now, so i don’t have to kiss your ass later on)
Yeah. Smart enough to think myself down all sorts of timewasting holes.
Kip irvine is a pretty smart guys (he wrote the text i used)
he moderates a mailing list; he got me a replacement text when mine literally fell apart; Prentice Hall never sent me a return label so i never sent the text back
That’s pretty solid of him.
what you told me made me realize that I need to re-read the part about the promotion of operands based on the data type
I’m glad! The same basic premise applies to reference types, albiet implemented differently.
he’s amazing. His text is clear.
he uses masm which tied me to windows
oh sad
I learned with nasm, which is wonderful
he chose masm because that’s what most profs wanted
I learned gas syntax out of necessity from reading the output of objdump -D
but nasm syntax actually made sense to me
i forgot why he did it, i asked him
nasm uses i think intel syntax? gas uses at&t syntax(?)?
and gas (or rather objdump) usually picks odd combinations of symbols for a given opcode
heh
i think we’ve veered off topic for too long
http://kipirvine.com/asm/
hmm
He’s keeping up to date, the downloadable chapters call it IA-32
he’s definitely smart.
the book calls it IA-32
first thing he teaches is how to convert between the bases
That should be the FIRST thing a person who wants to learn asm should learn.
Followed immediately by the operand sizes
the computer science building at my school would display the days high temp in binary in the window…it helped me practice
http://icanhascheezburger.com/2007/02/24/italics-cat/
sounds like a great topic for #asm, no?
Get one of thinkgeek’s binary clocks
learn to curse it out in utf-8, for bonus nerd karma
LOL
I can’t convert that quickly
it takes me time
CUTE!!!
clearly you are a lesser nerd then
bite me
lol
now of any libs for creating console wizards in java?
jcurses
not in the console, no.
er
other than the standard console libs but nothing wizard specific, no.
not just printing stuff to the console, but stuff like printing a menu and control flow
bnot just printing stuff to the console, but stuff like printing a menu and control flow/b
I apologize for veering off topic for the past hour
Totally my fault.
But i kept it going
it’s both your faults now stfu
what’s going on?
lol
Morning to you too
(You want something on-topic? Fine. I finally got the startup time on jboss below ten seconds.)
(Motherfucker takes forever to load up, relative to (say) apache, but at least it’s not a minute-thirty.)
~tell ojacobson about obscenity.
ojacobson, obscenity is the crutch of inarticulate motherfuckers.
you work with jboss for a while and tell me that again So noted, though.
