How often are you working along... and then... without warning... you become completely overwhelmed and distracted by the sheer amount of options and information available to you?
Personally, I've learned to cope with this. The last time I felt overwhelmed was when I got cornered by the little ghosties in PacMan...
So they'll just need to put this on their "Most Gay Feature" list, in competition with the paperclip, menu fade-ins (takes longer for menus to appear!), and the bi-weekly reinstall (that is supposed to be a feature, right?).
Hey microsoft, here's an idea: Make it WORK, then (and only then) make it BETTER.
some of us have broadband. lots of us, in fact. downloading a track takes 30 seconds or less from any similarly equipped user. a junk track like this would be an annoyance and nothing more: like a dead link brought up by a search engine. you just move on.
look at it from the point of view of someone to whom an mp3 file just takes a second to get. a cuckoo file would be annoying, sure, but so are all those llama's with modems. and people who have most of a track or album you want. you go find it elsewhere.
how effective could this be? will i stop downloading? (yeah, right: i can grab a gig in an hour or two. 200 tracks. think about it. what percentage are likely to be cuckoo'd? not much. i'll just throw 'em away with all the other corrupted downloads)
i've found the GIMP to be more of a look-alike than a work-alike. it _looks_ like it has a magic wand... but oh-no! how do i set the sensitivity?
another thing that bothered me in the way it parallels photoshop 3.0. what about 5.0? photoshop has changed alot since 3.
granted: the gimp's scriptability makes it really awesome in certain contexts. however, if i intend to design a web site, or create any graphics by hand, i use photoshop. why? because it's geared for professionals. because it works. it does everything it purports to.
there are really only two problems i have with photoshop. it's not free (not at all; but this comes from that "professional" thing). and it's not for linux.
[disclaimer: i haven't touched the gimp in a year. it may actually kick ass now. but it didn't when i tried it.]
i just have to say that when you can download a gig of mp3's in a couple hours, you download all kinds of interesting stuff. you run out of stuff you've heard of in a couple weeks.
then you start to explore.
and, guess what? a lot of those mp3's got hisses, pops, get cutoff at the end, are poorly ordered, sometimes it can be a real chore to get the whole album (always missing a track and whatnot). know what this means? you get a great preview. a really great preview. you hear most of the album. and if it stands out... you (OMG!) buy the cd!!
now, don't get me wrong. you download a hell of a lot more than you buy. but you don't have to buy cd's only to find that there's only one good track anymore.
so after you buy it, whaddya do? first thing: rip it. rip it properly: no hisses, no cutoffs, numbered tracks, etc. and burn it so you can listen to it in your car (on your mp3 cd player -- you have one, right?), at work (on your computer -- you are a geek, right?), or at home (on your computer through your badass sound system).
it's truly a beautiful thing. and if i didn't buy your music, the reason is simple: i didn't like it. at least i found out before i spent anything.
One of my roomates has an 8yr old African Grey named Grep (yes, as in grep) who is incredibly intelligent, and has been chatting on the internet for over 2 years now.
After reading about how he would be prone to a number of behavioral disorders if not given constant attention, we set him up with a webcam on one of our spare boxes running Windows (yeah yeah, if you know of any good voice recognition software for linux you let us know), and using Naturally Speaking from Dragon Systems we were able to set him up so that we could talk to him occasionally while we were all at work.
The wierd thing is that the speech software works better for him than it does for us!
the correct term for the appearance of something with a partial alpha value (in the RGBA sense, of course, which i hope even the most anal-retentive among us would admit was obvious from the context of the discussion) is not translucency but semi-transparency. transluceny refers to (from webster, per your suggestion) "transmitting and diffusing light so that objects beyond cannot be seen clearly". this is an effect that it is impossible to get any normal ray-tracer to pull off. think of a lamp shade. a lamp shade is usually translucent. it emits diffuse light, yet is not transparent: you cannot see through it. but since this is an effect that i believe can be achieved with radiosity, it might be best to reserve its use, and refer to semi-transparent objects as semi-transparent.
that said, i truly want to see a proper implementation of PNG, because i've been excited about variable transparency for a long time. to be able to use it on on the web would lead to all kinds of stunning effects. the transparency offered by GIF is so woefully inadequate, but that's all that most browsers offer for PNG currently. not until the PNG's alpha channel is fully utilized will the format truly take off, IMO.
just as an fyi, i always thought "overrated" was for posts that were moderated, yknow, a little high. my threshhold is 4. something kinda funny makes it to 4, and thus gets highlighted. so, if i'm moderating, and see a mediocre post that has been rated a little high, i might mark it down. at least that's what i thought it was for.
course as a general rule, i'd rather find something to mod up than something to mod down.
but what i really want, what i really really want, is a category that doesn't change a user's karma. i want "Dumb". Score: +5, Dumb. meaning: sure, it got marked up n' all, but just be warned. it's dumb. i don't want it to hurt anyone's karma. i sure don't want it to help them. i just wanna mod someone as dumb.
too bad you can't mod your own post. i'd mod this dumb, just to prove my point. ^_^
maybe i'm an elitist. maybe this sentiment is a hold-over from boyhood, where all my local lending library carried was introductory books. but i can't help but feel that the last thing we need (as a profession) is another introductory book. there are a million of these already.
think about it. it's far easier to write a pared-down basic book on a language that covers but a small percent of it. and that type of book is useful for about a week, if you're approaching it from an angle of complete ignorance.
this always frustrated me early in life. i wanted so badly to know the real information, and i learned to hate books that skimmed over the intricate details.
and there's always the question, as so many others have brought up, of whether c++ is truly the way to approach programming. i honestly doubt it. it's hard enough for trained pros. useful: yes; flexible: yes; powerful: obviously; simple: no way in hell.
maybe this book will encourage someone to look deeper, to find more serious books. to take classes. of course, i realize that "introductory" means just that. it's not meant to be the last c++ book you by. and going straight to stroustrup might be a little much to expect.
but if, when you first read through it, you don't understand a lot of things in a book, that's a good thing. you might learn something there. you might learn alot. of course, if it assumes you know a lot more than you do, you might have to learn alot somewhere else first.
i've done a fair amount of web design and photoshop work, and one of the earlier tricks that i learned is "bigger is better". not for final image size, of course, but especially when going to web (where every pixel is distinctly visible) it's important to start big and work big, and only as a last step, scale down the image and compress to jpeg (or gif, but we're talking photography, so...).
to that end, i have found no digital cameras that will fit my budget and deliver the image quality i need. on the otherhand, i've got a several-years-old 35mm, and a scanner that does 900x1200dpi in hardware, which is greater than the output of the film. i generally scan photos at 3000x2000 pixels or so. and the scanner cost me $200.
so for me, given the resolution choice (one-to-one with final image size, or much, much larger) and the cost, i settled on the scanner. for now. i hope that one day i can get a digital camera that is adequate. but right now, they're still bleeding edge (aka: "professional grade").
...but the reality in this situation is that if your Idea is of value, someone else will think of it, and eventually someone with the morals to share it, if it betters mankind.
doesn't that seem a little altruistic? let's think of some other reasons someone might share an idea:
because they're greedy, and think they can make a profit (many things on infomercials could count under this heading; also, bank robberies)
they'd like to prove it can be done, either for their own glory, or for the sake of the idea itself (proving fermat's last theorem?)
they're bored, and have this idea, and just on a lark, they set it in motion (seems like most of my programming projects are examples here)
they're insane, and fear for their bodily fluids (e.g., dr. strangelove)
of course, i've only scratched the surface, but my point is that whether or not an idea benefits humanity, and whether or not it's actually a good idea, someone somewhere will try it. and this gives us things like MAME for Digita-enable cameras and Textmode Quake...
it's true that Java is a very good language for developers, giving both simplicity, power, and (mostly) portability.
it's also true that the JVM takes a serious performance hit, due to its very nature. interpreted languages are slow. period. using a native compiler can result in a drastic speed increase, but the trade-off there is that most of these native-code compilers are in their infancy, and have difficulty compiling perfectly valid Java code. and compiler errors on valid code is a very frustating thing indeed.
in an ideal world, Sun would step up and produce a native-code compiler for their favorite platforms (solaris, windows, and linux, in that order, leaving apple to fend for themselves). this native-code compiler would also compile Java byte-code. does this sound like JIT? sort of, but let's keep in mind the phrase "optimized for speed". while JIT is faster, it still does not approach the speeds of native code.
however, this is actually indicative of a deeper problem. writing tight Java code is tough. it's hard to do. the standard API is large, bloated, and plagued with inefficiencies. this may sound sacriligeous, but it's true. to their credit, Sun has managed to tighten the code with every subsequent release. Java 1.0 ran slower than 1.1, which in turn was slower than 1.2, and in keeping with this, 1.3 seems to improve the speed even further, especially the JFC stuff.
but the real problem simply comes from OOP overhead. when you write "proper" Java code, you try to rely on Sun's supplied API as much as possible (so that you reap the development-time gains). but there is a penalty for doing so: you have no control over the inheritence, the efficiency, and the size of the API code. in a few projects i've done, i've been forced to go into the API, and tweak certain parts that i needed. specialize parts of the API, or rewrite them. this may take away their general-purpose nature, but also removes unnecessary crap from my core loops, which is most important to me.
a further problem with this is that development time then increases. i have yet to meet a project manager who truly sees the value of making code efficient. especially since doing so makes that little bar on the Gant chart, the one that represents me, stetch a little farther. they don't care. do it quick. dirty. they don't care so long as it's done _on_time_.
i've basically come to the conclusion that if you want it to run fast, you should (still) write it in C. if you can afford a slight performance hit (or your mind has been permanently warped into OO thinking), use C++. on the other hand, if you want to write it quickly, and speed isn't such an issue, go ahead and use Java. hey, it's fun to write. it makes sense.
but it's going to be a long wait for efficiency. i don't think that was really ever more than an afterthought with Java.
In context: "Y'ALL ain't linkin' to that there [thing i/we object to], now iz ya? Y'ALL better knock it off."
DESCRIPTION
ridiculously impractical demands popular by those who haven't much clue about how the internet works, but have convinced their lawyers that they'd like it to work differently.
Costikyan presents an interesting solution, but do you think it's even possible to shut down the roms sites? regardless how the idsa cracks down on rom piracy, emulators are legal. and if emulators are legal, someone somewhere will image the rom. and if the rom file exists somewhere, then someone will put it on the web. and if someone puts it on the web, we'll download it.
of course, it'd be a better world if the software companies would just shrug and say, "aw, what the hell, you can have it." it would give them some say in the matter, as well as market data.
but any rom collector knows that this won't happen. because the companies are stingy. they don't want their software available for free. and i wouldn't blame them, really, if it represented a revenue stream. but it doesn't. and litigation doesn't improve their image.
in my opinion, crack-downs on this sort of activity cause an opposite effect than the game producers intend. it tarnishes their image to gamers, or a segment of gamers. then those same gamers have a choice: download the game for nothing, or purchase it (if possible) from a company they no longer have any respect for.
on the whole, i would agree with you. you make a valid point, especially considering the availability (and quality =) of schizmatrix specifically. However, I think your last statement is a little pessimistic. I'd like to think that a larger, value added paperback actually is worth more than a conventional trade paperback. But I'm willing to consider the hypothesis that this is just wishful thinking on my part. ^_^
that might be a reasonable price, depending on the format of the book. for example, my oversized copy of bruce sterling's "schizmatrix plus" was about $15, paperback. but i wouldn't pay that price for a standard-sized trade paperback. the difference with a larger book like that, is that the font is generally larger and easier to read, and the paper is much higher quality... the book will last longer. so it's worth more.
i was suddenly overcome with a strong urge to purchase economical, yet industrial strength Duron(tm?) flooring. i held off, though, as there are no boards yet that support it, and i'm not sure how to overclock it. also, i'm looking for peltier shag carpet in fuschia. (hint, hint: need helpful link)
it's an innovative feature that may or may not be useful in actual practice. but, really, i think that i would like to try it, and make that decision for myself.
there was a time when i turned of images, so pages would load faster, too. it was useful at the time. very nice. but now... i haven't used that feature in years. i'm not even sure if it still exists. heh. (okay, i use lynx fairly regularly still, but that's a different story.)
turning off images to foreign sites would be really handy in a couple of instances... for example, in mid-afternoon the IGN network's adserver inevitably gets bogged down (try http://pc.ign.com/) and this makes it seem like the page is still loading. hitting "stop" causes the page to appear, sans ad banner. good. having a browser that could do this automatically if i chose would be a Good Thing.
...zd-net comments are, in general, infuriatingly ignorant? i can't even read them anymore. i suppose i'm not the only one (there are the occaisonal "you are all morons!!!" posts that i suppose are written by once-intelligent people reduced to enraged flamers...)
i'd rather read a zillion fR1sT p057 comments, personally. and so i retreat to slashdot.
If you've had experience with an object-oriented language, then it is likely that an OO solution will make the most sense to you.
In an ISP, there are usually several types of individuals involved with web applications: the graphic artists and page designers, the scripting guys (hopefully with familiarity with programming on some basic level), and the developers.
A Servlet/JSP combination allows maximum leverage of each of these skillsets. The graphic designers can build the look of the page, defining a cohesive experience for the visitor (view). The scripters can form the pages given the template HTML provided by the designers and implement some basic logic with JSP (controller), while the true meat of the system can be handled via servlets written by experienced developers (or RMI, for a fully-distributed solution) (model).
Notice the MVP (Model-View-Controller) seperation. By seperating these elements, there is a clear cut division between roles. This allows you to seek individuals who are particularly good at one task, and it allows them to work within that area. They can each be sure what their responsibilities are, and their attention need not be unnecessarily divided.
Complex tasks such as database access and processing can be handled on an entirely seperate machine (from the web server) if you choose to use RMI. These backend processes can provide a library of methods for use within JSP. Careful attention should be paid to what processing has gone where: ensure that the JSP strictly asks for and displays information.
Avoid complex JSP at all costs. If a task is becoming complex, it should be rewritten in servlet form (or as a remote method). Make sure that your scripters aren't biting off more than they can chew: the purpose of the JSP is to assemble live data into a predetermined format, and little more. It Controls the View of the data Model.
Slashdot Mirror
Personally, I've learned to cope with this. The last time I felt overwhelmed was when I got cornered by the little ghosties in PacMan...
So they'll just need to put this on their "Most Gay Feature" list, in competition with the paperclip, menu fade-ins (takes longer for menus to appear!), and the bi-weekly reinstall (that is supposed to be a feature, right?).
Hey microsoft, here's an idea: Make it WORK, then (and only then) make it BETTER.
look at it from the point of view of someone to whom an mp3 file just takes a second to get. a cuckoo file would be annoying, sure, but so are all those llama's with modems. and people who have most of a track or album you want. you go find it elsewhere.
how effective could this be? will i stop downloading? (yeah, right: i can grab a gig in an hour or two. 200 tracks. think about it. what percentage are likely to be cuckoo'd? not much. i'll just throw 'em away with all the other corrupted downloads)
another thing that bothered me in the way it parallels photoshop 3.0. what about 5.0? photoshop has changed alot since 3.
granted: the gimp's scriptability makes it really awesome in certain contexts. however, if i intend to design a web site, or create any graphics by hand, i use photoshop. why? because it's geared for professionals. because it works. it does everything it purports to.
there are really only two problems i have with photoshop. it's not free (not at all; but this comes from that "professional" thing). and it's not for linux.
[disclaimer: i haven't touched the gimp in a year. it may actually kick ass now. but it didn't when i tried it.]
as far as i can tell, it's just idle speculation. and particularly fruitless speculation at that.
then you start to explore.
and, guess what? a lot of those mp3's got hisses, pops, get cutoff at the end, are poorly ordered, sometimes it can be a real chore to get the whole album (always missing a track and whatnot). know what this means? you get a great preview. a really great preview. you hear most of the album. and if it stands out... you (OMG!) buy the cd!!
now, don't get me wrong. you download a hell of a lot more than you buy. but you don't have to buy cd's only to find that there's only one good track anymore.
so after you buy it, whaddya do? first thing: rip it. rip it properly: no hisses, no cutoffs, numbered tracks, etc. and burn it so you can listen to it in your car (on your mp3 cd player -- you have one, right?), at work (on your computer -- you are a geek, right?), or at home (on your computer through your badass sound system).
it's truly a beautiful thing. and if i didn't buy your music, the reason is simple: i didn't like it. at least i found out before i spent anything.
One of my roomates has an 8yr old African Grey named Grep (yes, as in grep) who is incredibly intelligent, and has been chatting on the internet for over 2 years now.
After reading about how he would be prone to a number of behavioral disorders if not given constant attention, we set him up with a webcam on one of our spare boxes running Windows (yeah yeah, if you know of any good voice recognition software for linux you let us know), and using Naturally Speaking from Dragon Systems we were able to set him up so that we could talk to him occasionally while we were all at work.
The wierd thing is that the speech software works better for him than it does for us!
See him and talk to him. He talks back!
that said, i truly want to see a proper implementation of PNG, because i've been excited about variable transparency for a long time. to be able to use it on on the web would lead to all kinds of stunning effects. the transparency offered by GIF is so woefully inadequate, but that's all that most browsers offer for PNG currently. not until the PNG's alpha channel is fully utilized will the format truly take off, IMO.
funny that you should cry "copycat" on a post about freedom of information...no?
course as a general rule, i'd rather find something to mod up than something to mod down.
but what i really want, what i really really want, is a category that doesn't change a user's karma. i want "Dumb". Score: +5, Dumb. meaning: sure, it got marked up n' all, but just be warned. it's dumb. i don't want it to hurt anyone's karma. i sure don't want it to help them. i just wanna mod someone as dumb.
too bad you can't mod your own post. i'd mod this dumb, just to prove my point. ^_^
Some people may not be this lucky.
think about it. it's far easier to write a pared-down basic book on a language that covers but a small percent of it. and that type of book is useful for about a week, if you're approaching it from an angle of complete ignorance.
this always frustrated me early in life. i wanted so badly to know the real information, and i learned to hate books that skimmed over the intricate details.
and there's always the question, as so many others have brought up, of whether c++ is truly the way to approach programming. i honestly doubt it. it's hard enough for trained pros. useful: yes; flexible: yes; powerful: obviously; simple: no way in hell.
maybe this book will encourage someone to look deeper, to find more serious books. to take classes. of course, i realize that "introductory" means just that. it's not meant to be the last c++ book you by. and going straight to stroustrup might be a little much to expect.
but if, when you first read through it, you don't understand a lot of things in a book, that's a good thing. you might learn something there. you might learn alot. of course, if it assumes you know a lot more than you do, you might have to learn alot somewhere else first.
enough of this rambling.
to that end, i have found no digital cameras that will fit my budget and deliver the image quality i need. on the otherhand, i've got a several-years-old 35mm, and a scanner that does 900x1200dpi in hardware, which is greater than the output of the film. i generally scan photos at 3000x2000 pixels or so. and the scanner cost me $200.
so for me, given the resolution choice (one-to-one with final image size, or much, much larger) and the cost, i settled on the scanner. for now. i hope that one day i can get a digital camera that is adequate. but right now, they're still bleeding edge (aka: "professional grade").
doesn't that seem a little altruistic? let's think of some other reasons someone might share an idea:
of course, i've only scratched the surface, but my point is that whether or not an idea benefits humanity, and whether or not it's actually a good idea, someone somewhere will try it. and this gives us things like MAME for Digita-enable cameras and Textmode Quake...
it's also true that the JVM takes a serious performance hit, due to its very nature. interpreted languages are slow. period. using a native compiler can result in a drastic speed increase, but the trade-off there is that most of these native-code compilers are in their infancy, and have difficulty compiling perfectly valid Java code. and compiler errors on valid code is a very frustating thing indeed.
in an ideal world, Sun would step up and produce a native-code compiler for their favorite platforms (solaris, windows, and linux, in that order, leaving apple to fend for themselves). this native-code compiler would also compile Java byte-code. does this sound like JIT? sort of, but let's keep in mind the phrase "optimized for speed". while JIT is faster, it still does not approach the speeds of native code.
however, this is actually indicative of a deeper problem. writing tight Java code is tough. it's hard to do. the standard API is large, bloated, and plagued with inefficiencies. this may sound sacriligeous, but it's true. to their credit, Sun has managed to tighten the code with every subsequent release. Java 1.0 ran slower than 1.1, which in turn was slower than 1.2, and in keeping with this, 1.3 seems to improve the speed even further, especially the JFC stuff.
but the real problem simply comes from OOP overhead. when you write "proper" Java code, you try to rely on Sun's supplied API as much as possible (so that you reap the development-time gains). but there is a penalty for doing so: you have no control over the inheritence, the efficiency, and the size of the API code. in a few projects i've done, i've been forced to go into the API, and tweak certain parts that i needed. specialize parts of the API, or rewrite them. this may take away their general-purpose nature, but also removes unnecessary crap from my core loops, which is most important to me.
a further problem with this is that development time then increases. i have yet to meet a project manager who truly sees the value of making code efficient. especially since doing so makes that little bar on the Gant chart, the one that represents me, stetch a little farther. they don't care. do it quick. dirty. they don't care so long as it's done _on_time_.
i've basically come to the conclusion that if you want it to run fast, you should (still) write it in C. if you can afford a slight performance hit (or your mind has been permanently warped into OO thinking), use C++. on the other hand, if you want to write it quickly, and speed isn't such an issue, go ahead and use Java. hey, it's fun to write. it makes sense.
but it's going to be a long wait for efficiency. i don't think that was really ever more than an afterthought with Java.
"Some you will (recognize), and some will surprise you," said Tam
Like TBS. Nobody expected to see the "All Charles Bronson, All Weekend, All From SPACE Weekend!!!".
Or what about Cartoon Network? Wouldn't Space Ghost be that much cooler From Space?
yall(1)
NAME
USAGE
DESCRIPTION
of course, it'd be a better world if the software companies would just shrug and say, "aw, what the hell, you can have it." it would give them some say in the matter, as well as market data.
but any rom collector knows that this won't happen. because the companies are stingy. they don't want their software available for free. and i wouldn't blame them, really, if it represented a revenue stream. but it doesn't. and litigation doesn't improve their image.
in my opinion, crack-downs on this sort of activity cause an opposite effect than the game producers intend. it tarnishes their image to gamers, or a segment of gamers. then those same gamers have a choice: download the game for nothing, or purchase it (if possible) from a company they no longer have any respect for.
i prefer to pay money where respect is due.
could you then make a Spam-powered server? i mean, spam has to be good for something.
on the whole, i would agree with you. you make a valid point, especially considering the availability (and quality =) of schizmatrix specifically. However, I think your last statement is a little pessimistic. I'd like to think that a larger, value added paperback actually is worth more than a conventional trade paperback. But I'm willing to consider the hypothesis that this is just wishful thinking on my part. ^_^
that might be a reasonable price, depending on the format of the book. for example, my oversized copy of bruce sterling's "schizmatrix plus" was about $15, paperback. but i wouldn't pay that price for a standard-sized trade paperback. the difference with a larger book like that, is that the font is generally larger and easier to read, and the paper is much higher quality... the book will last longer. so it's worth more.
i was suddenly overcome with a strong urge to purchase economical, yet industrial strength Duron(tm?) flooring. i held off, though, as there are no boards yet that support it, and i'm not sure how to overclock it. also, i'm looking for peltier shag carpet in fuschia. (hint, hint: need helpful link)
there was a time when i turned of images, so pages would load faster, too. it was useful at the time. very nice. but now... i haven't used that feature in years. i'm not even sure if it still exists. heh. (okay, i use lynx fairly regularly still, but that's a different story.)
turning off images to foreign sites would be really handy in a couple of instances... for example, in mid-afternoon the IGN network's adserver inevitably gets bogged down (try http://pc.ign.com/) and this makes it seem like the page is still loading. hitting "stop" causes the page to appear, sans ad banner. good. having a browser that could do this automatically if i chose would be a Good Thing.
i'd like to see it.
i'd rather read a zillion fR1sT p057 comments, personally. and so i retreat to slashdot.
so, uh, no, i didn't read the article... ^_^
In an ISP, there are usually several types of individuals involved with web applications: the graphic artists and page designers, the scripting guys (hopefully with familiarity with programming on some basic level), and the developers.
A Servlet/JSP combination allows maximum leverage of each of these skillsets. The graphic designers can build the look of the page, defining a cohesive experience for the visitor (view). The scripters can form the pages given the template HTML provided by the designers and implement some basic logic with JSP (controller), while the true meat of the system can be handled via servlets written by experienced developers (or RMI, for a fully-distributed solution) (model).
Notice the MVP (Model-View-Controller) seperation. By seperating these elements, there is a clear cut division between roles. This allows you to seek individuals who are particularly good at one task, and it allows them to work within that area. They can each be sure what their responsibilities are, and their attention need not be unnecessarily divided.
Complex tasks such as database access and processing can be handled on an entirely seperate machine (from the web server) if you choose to use RMI. These backend processes can provide a library of methods for use within JSP. Careful attention should be paid to what processing has gone where: ensure that the JSP strictly asks for and displays information.
Avoid complex JSP at all costs. If a task is becoming complex, it should be rewritten in servlet form (or as a remote method). Make sure that your scripters aren't biting off more than they can chew: the purpose of the JSP is to assemble live data into a predetermined format, and little more. It Controls the View of the data Model.
(And, yes, I'm a Java programmer.)