The fact that you got modded up to +5 Insightful in a heartbeat should be answer enough for you.
Silent moderation, not really enough. Hardly anyone bothers to stand up to the rampant editorial bias around here, from the article selection to the snippy commentary inserted after most of them.
Homogenized corporate media occasionally enjoys a story about the ills of homogenized corporate media. Then they go right back to conforming to the ratings machines. I come to slashdot for the community now, the articles are informative or useful only once in a while.
Re:Other mappable relationship environments?
on
Guilty By Association
·
· Score: 2, Informative
> You know, I'm really not sure WordOfMouthResearch.com is legit..
I'm really not sure those fellas at Enron are on the level either... The Word Of Mouth Connection is a SCAM. Just google for "word of mouth scam" and click on just about any of the links.
That said, Parrot sounds like it's going to shake some people up. From what I understand, it's a register based VM as opposed to stack based, meaning that preemption is possible
The difference between stack and register based has nothing whatsoever to do with preemption, and has to do with reduction in the number of opcodes (adding two numbers becomes one instruction instead of 3 for instance). You still have to save register sets the same way you have to save stacks in a stack-based VM, since the task you're switching to won't know if it's clobbering already allocated registers... unless the allocator knows ahead of time and allocates different registers for each task, but in that case it's not really preempting.
"Hey they're not warts, they're beauty marks". Funny how it becomes a Good Thing in one's Favorite Pet Language. I'm not talking about MIXING floats and ints, I'm talking about the fact that "1.0 + 2.0" is a syntax error in ocaml. You need "1.0 +. 2.0", because + will always and forever only handle ints. Well, unless the compiler makes some exceptions, but don't try this with your own operators, kids.
> # let f x = x;;
Identity is polymorphic. Excuse me while I cheer. How about a function that computes something, because I do eventually have to degenerate to base types. Now there are polymorphic types as well, but you have to anticipate the types it may be and change it at the declaration of the type -- in other words, it's not inferable. Plus, the syntax of polymorphic types is an awesomely grungy wart, and can't begin to compare to type families.
I actually like ocaml, mind you. Unfortunately, it hasn't seen serious development around anything like smarter compiler feedback, let alone language features like type families. In fact, mentioning such a thing elicits so much negative advocacy, one gets the impression that the language definition has become Set In Stone For All Time Forever And Amen, and that only weird and crufty macro metaprogramming hacks in camlp4 will push the envelope any further.
* Total lack of ad-hoc polymorphism. This language has such a problem with overloading, you need a different operator for adding floats than you do for adding ints.
* Supposedly awesome module functor calculus that fails to show up with any useful examples in the standard library -- for data structures, you tend to call a lot of static functions in modules that are, you guessed it, non-polymorphic.
* Lack of compiler warnings. My all time favorite error in ocaml goes like "The value is declared type Foo, but is used here with type Foo". It will drive you insane when developing via ocaml-mode in emacs, because it doesn't do a fresh recompile, so you end up redefining types and the compiler doesn't make a peep of warning. I learned to stick with makefiles when developing, to preserve my sanity.
As a better C, ocaml looks great. But it fails to offer many credible alternatives to what even ugly old C++ offers in ad-hoc and parametric polymorphism.
FreeBSD looks like UNIX (oversimplification, albeit) down-and-dirty. I ran X on it for awhile (enlightenment or fvwm95 on a 486DX/66) and will never again. It really is not set up for a GUI, and you will do a ton of work getting it there.
Yeah, clicking on "Desktop" in the installer, that really was hard.
It will run Linux-compatible binaries provided you have the right libraries.
I forgot to undertake the astonishingly difficult task of clicking "Linux Compatibility" in the install, so I had to resort to the incredibly complex "make install" command in any of the linux binaries port because I forgot that the linux base system had the incredibly cryptic name of "linux-base".
You sir posited zero evidence for your assertions.
Here's a problem I find with most computer "adventures". They often degenerate into a verb or pixel hunt. Here's a relevant quote for graphical adventures from the paper I Have No Words and I Must Design:
Or let's talk about computer adventures; they often display information failure. "Oh, to get through the Gate of Thanatos, you need a hatpin to pick the lock. You can find the hatpin on the floor of the Library. It's about three pixels by two pixels, and you can see it, if your vision is good, between the twelfth and thirteenth floorboards, about three inches from the top of the screen. What, you missed it?"
Yeah, I missed it. In an adventure, it shouldn't be ridiculously difficult to find what you need, nor should victory be impossible just because you made a wrong decision three hours and thirty-eight decision points ago. Nor should the solutions to puzzles be arbitrary or absurd.
That last sentence, "arbitrary and absurd" is what characterizes all too many text adventures, that they often devolve into a hunt of "apply every verb I can think of to every noun in the scene". Gee, I had to "rotate tree" to move the secret lever. Or was it "push", "turn", "twist"... To say nothing of whatever bizarre key it might need.
Hitchhiker's Guide To The Galaxy had some really absurd puzzles, but that was part of the fun, and it was usually a comical trial-and-error thing that gave you a hint as to what you needed as you went. Too many adventures lacked such cues entirely. HHGTG isn't the best example either, because if you fail to take one action at the very start of the game, you could play almost ALL the way through, but not finish at the very end. Totally infuriating.
I'm currently working on doing Logical Filtering in an adventure game, which is a way to maintain a sort of belief about the current state of your world depending on your prior knowledge and observations. Somewhat like filtering in a Hidden Markov model.
Deep knowledge representation is well and good in research projects and perhaps single player games. Once you try to scale it up to a MUD, for even simple things like "known names" (of people and objects), you find yourself wishing for many more gigs of RAM and a few more CPU's to handle it. M objects of N knowable states by P players... ouch.
Ultimately, the winning strategy is just to have a good setting, story, gm's, and roleplaying in a MUD. Knowledge representation, no matter how good the attempts have been, always ends up feeling like artificial game mechanics.
So, this got me thinking: Since NMAP source is GPL, does it's inclusion in Battle Royale make the movie a derivative work and therefore also subject to the GPL?
I think it could be construed under the "mere aggregation" clause. Unless nmap was pivotal to the movie, in which case one could say it was linked and thus GPL-tainted:)
Whats to stop a spammer from signing up for a free email account with a false name, blast out a few thousand messages, drop the account (it'll be closed anyway by abuse), wipe hands and repeat?
Have you tried "blasting out a few thousand" messages from hotmail? Only the nigerian spammers seem to have the manpower, and the volume they send is a mere trickle.
I've been pondering getting an AMD64 box, but I'm wondering how well it supports AMD64, what the performance is like, etc. Anyone running such a setup have any stories? Linux is the alternative here, but I would much prefer FreeBSD if it's feasable.
All of these [common lisp implementations] implement the Standard, some better than others. All have interesting extensions which are not portable. All bring different elements of interest to the table of developers looking to solve different problems.
It also bears noting that thanks to the larger research community around lisp and its many implementations, modern Lisp compilers absolutely run circles around java. You can typically compile a lisp app in less time than the JVM takes to initialize and begin JIT'ing the bytecode. To say nothing of the actual execution speed.
Honestly, who cares if Java is forked. There's a virtual machine and bytecode specification -- simply make it a standard. Oh wait, Sun withdrew Java from the standards process, never mind then...
Stories like this do nothing to help build a convincing argument to Windows users that Linux is in fact the better OS. If even seasoned Linux users have problems upgrading their kernel, think of how frustrating it would be for someone less technically-inclined.
Tell me about it, the other day I just slapped the Win2k kernel into my win98 box and it just sailed right through, no problems at all...
I always thought that the Plus! packs were pretty cool concepts. Why doesn't Microsoft just do a barebones OS then a cheapy Plus! product with all the extra crap nobody needs?
Because the League Of Extraordinary Hypocrites would descend on them in force and give them a good nasal whining about how they gouge for basic features that should have come with the OS.
Can't figure out why MS doesn't sell a DVD distribution of XP for like $50 extra that bundles all the stuff that's also free downloads, like the command-line-only C++ and VB compilers. Sort of like MSDN super-lite. I'm not one to tell MS marketing what to do though, since they seem to rake in enough money to physically crush me every minute or so...
Also known as "Spammers". Their entire purpose is gaming the system to promote crass commerce (and boy I do mean crass), and just to top it off, I haven't seen one that hasn't engaged in spam to promote their service (to be fair, I probably wouldn't see the ones that don't spam, since I'm not in the business of pushing sites up in some silly ratings criterion).
I never noticed the problem with google til just recently, just a few sites that seem to have my keyword no matter *how* obscure but are filled with nothing but links to... other search engines, mostly paidclick ones of course. Those are fairly easy to strip out since the abstract is always nonsensical.
I'd love to see an option on the google toolbar of "report link as googlespam", but I can imagine the manpower that'd be required to sift through submissions (you can't just do it automatically).
Am I the only one who remembers the name "Repton" as connected to a kind-of-slow but still fairly fun knockoff of Defender/Stargate for the C64? Instead of kidnapping citizens, the bad guys you were fighting were building some kind of super-weapon, and you had to prevent them from completing it. I guess it's bound to happen that names that sound cool will collide. I for one have never even heard of the other repton.
now when people use a GNOME app next to a non-GNOME app things are not consistent and usability is damaged as the use can no longer simply learn where the buttons are in all dialogs: they actually have to read each one. this makes X11 look downright silly and stupid.
I'm all for having a standard placement order dictated by, I dunno, some "semantic layout engine" that's cross-toolkit/platform or some other handwaving, just for visual consistency's sake. But if you're not expecting someone to read the content of a pop up alert dialog, then perhaps it doesn't really belong in an alert dialog?
> I'm wondering if maybe kde might be or might become the killer app for Linux?
I personally would like for it to become the killer app for Windows. It'd kick the crap out of litestep and all those other anemic shell replacements that are nothing but glorified launchpads and theme engines. Some apps might not port well, but most of them should come through just fine.
But thanks to the sorry backward state of Qt on win32, I don't see it as happening, ever, GPL'd or not.
But with so much game logic in scripts nowadays, doesn't it make sense to use a common, open source MUD platform
It's already been done, DGD, MOO, Cold, are quite nicely programmable generic platforms. It's certainly not technology stopping anyone, even if the languages are somewhat aging and others might prefer to program in, say, Python (I patched Python to make it a multi-user-safe runtime once, it's not horribly difficult). There's quite a few, for lack of a better term, "systemic" problems that keep MUDs from flourishing:
1) They had their day. Lots of things on the net are simply fads, which flare up, then subside and leave only a core group of remaining followers, not much bigger or even smaller than the original population.
2) The gameplay on most MUDs isn't much different than MMORPG's: jump on the mobs that pop, loot, level, repeat. Doesn't engage the brain anyway, so why would you want to do it without any eye candy?
3) The "pure RP" muds that are out there are often a variation of hanging out at the guildhall and kvetching.
4) At some point in any game development, one hits a "critical resource" problem: server bandwidth, storage, and above all, artistic talent. Artists aren't into commity work like coders are, it's in fact pretty antithetical to their worldview for most of them. Whether changing the art culture to a free software ethic is the answer or not, the fact remains that good artists that can produce models, textures, character art, voice acting, writing, and music are damn expensive. The ones a project does get tend to have to make it a full-time-job to get any decent amount of art done (this tends to be true for visual artists, musicians seem to be able to do part time production work)
ex. I write YourDoom.A and i write it using this new code morphing obfuscator. how exactly are Anti-virus programs 1. suppose to remove this? 2. identify this?
It's called polymorphic encryption, and viruses have been doing it for at least a decade. AV scanners look for the decryptor. Or they just intercept its attempted accesses when it does "wake up".
Slashdot Mirror
> Do you know what's the biggest cause of cancer in humans due to chemicals? Salt. Sodium chloride, that is.
Please post a citation from JAMA or Lancet that supports this assertion.
The fact that you got modded up to +5 Insightful in a heartbeat should be answer enough for you.
Silent moderation, not really enough. Hardly anyone bothers to stand up to the rampant editorial bias around here, from the article selection to the snippy commentary inserted after most of them.
Homogenized corporate media occasionally enjoys a story about the ills of homogenized corporate media. Then they go right back to conforming to the ratings machines. I come to slashdot for the community now, the articles are informative or useful only once in a while.
> You know, I'm really not sure WordOfMouthResearch.com is legit..
I'm really not sure those fellas at Enron are on the level either... The Word Of Mouth Connection is a SCAM. Just google for "word of mouth scam" and click on just about any of the links.
Oh wow gee, Eric's such a visionary. Let's count the number of patent lawsuits SCO has filed. Oh that's right, ZERO .
I for one wish he would have kept his snippy adolescent quips out of the original documents.
"The meeting of the Society Against Societies Against Things has come to order"
"First item on the agenda: Disband the Society Against Societies Against Things."
"All in favor, say 'aye'."
"Motion passes. Meeting adjourned."
That said, Parrot sounds like it's going to shake some people up. From what I understand, it's a register based VM as opposed to stack based, meaning that preemption is possible
... unless the allocator knows ahead of time and allocates different registers for each task, but in that case it's not really preempting.
The difference between stack and register based has nothing whatsoever to do with preemption, and has to do with reduction in the number of opcodes (adding two numbers becomes one instruction instead of 3 for instance). You still have to save register sets the same way you have to save stacks in a stack-based VM, since the task you're switching to won't know if it's clobbering already allocated registers
"Hey they're not warts, they're beauty marks". Funny how it becomes a Good Thing in one's Favorite Pet Language. I'm not talking about MIXING floats and ints, I'm talking about the fact that "1.0 + 2.0" is a syntax error in ocaml. You need "1.0 +. 2.0", because + will always and forever only handle ints. Well, unless the compiler makes some exceptions, but don't try this with your own operators, kids.Identity is polymorphic. Excuse me while I cheer. How about a function that computes something, because I do eventually have to degenerate to base types. Now there are polymorphic types as well, but you have to anticipate the types it may be and change it at the declaration of the type -- in other words, it's not inferable. Plus, the syntax of polymorphic types is an awesomely grungy wart, and can't begin to compare to type families.
I actually like ocaml, mind you. Unfortunately, it hasn't seen serious development around anything like smarter compiler feedback, let alone language features like type families. In fact, mentioning such a thing elicits so much negative advocacy, one gets the impression that the language definition has become Set In Stone For All Time Forever And Amen, and that only weird and crufty macro metaprogramming hacks in camlp4 will push the envelope any further.
Your using the word "your" in it's proper context will insure you never get a job as a Slashdot editor neither.
You forgot a couple more:
* Total lack of ad-hoc polymorphism. This language has such a problem with overloading, you need a different operator for adding floats than you do for adding ints.
* Supposedly awesome module functor calculus that fails to show up with any useful examples in the standard library -- for data structures, you tend to call a lot of static functions in modules that are, you guessed it, non-polymorphic.
* Lack of compiler warnings. My all time favorite error in ocaml goes like "The value is declared type Foo, but is used here with type Foo". It will drive you insane when developing via ocaml-mode in emacs, because it doesn't do a fresh recompile, so you end up redefining types and the compiler doesn't make a peep of warning. I learned to stick with makefiles when developing, to preserve my sanity.
As a better C, ocaml looks great. But it fails to offer many credible alternatives to what even ugly old C++ offers in ad-hoc and parametric polymorphism.
FreeBSD looks like UNIX (oversimplification, albeit) down-and-dirty. I ran X on it for awhile (enlightenment or fvwm95 on a 486DX/66) and will never again. It really is not set up for a GUI, and you will do a ton of work getting it there.
Yeah, clicking on "Desktop" in the installer, that really was hard.
It will run Linux-compatible binaries provided you have the right libraries.
I forgot to undertake the astonishingly difficult task of clicking "Linux Compatibility" in the install, so I had to resort to the incredibly complex "make install" command in any of the linux binaries port because I forgot that the linux base system had the incredibly cryptic name of "linux-base".
You sir posited zero evidence for your assertions.
That last sentence, "arbitrary and absurd" is what characterizes all too many text adventures, that they often devolve into a hunt of "apply every verb I can think of to every noun in the scene". Gee, I had to "rotate tree" to move the secret lever. Or was it "push", "turn", "twist"
Hitchhiker's Guide To The Galaxy had some really absurd puzzles, but that was part of the fun, and it was usually a comical trial-and-error thing that gave you a hint as to what you needed as you went. Too many adventures lacked such cues entirely. HHGTG isn't the best example either, because if you fail to take one action at the very start of the game, you could play almost ALL the way through, but not finish at the very end. Totally infuriating.
I'm currently working on doing Logical Filtering in an adventure game, which is a way to maintain a sort of belief about the current state of your world depending on your prior knowledge and observations. Somewhat like filtering in a Hidden Markov model.
... ouch.
Deep knowledge representation is well and good in research projects and perhaps single player games. Once you try to scale it up to a MUD, for even simple things like "known names" (of people and objects), you find yourself wishing for many more gigs of RAM and a few more CPU's to handle it. M objects of N knowable states by P players
Ultimately, the winning strategy is just to have a good setting, story, gm's, and roleplaying in a MUD. Knowledge representation, no matter how good the attempts have been, always ends up feeling like artificial game mechanics.
So, this got me thinking: Since NMAP source is GPL, does it's inclusion in Battle Royale make the movie a derivative work and therefore also subject to the GPL?
:)
I think it could be construed under the "mere aggregation" clause. Unless nmap was pivotal to the movie, in which case one could say it was linked and thus GPL-tainted
Whats to stop a spammer from signing up for a free email account with a false name, blast out a few thousand messages, drop the account (it'll be closed anyway by abuse), wipe hands and repeat?
Have you tried "blasting out a few thousand" messages from hotmail? Only the nigerian spammers seem to have the manpower, and the volume they send is a mere trickle.
I've been pondering getting an AMD64 box, but I'm wondering how well it supports AMD64, what the performance is like, etc. Anyone running such a setup have any stories? Linux is the alternative here, but I would much prefer FreeBSD if it's feasable.
Distributed Transaction support is very strong in .Net.
Does it still only work with the MS SQL Server DTC, or is it now a generic API like JTA?
(eat your alphabet soup, it's good for you)
All of these [common lisp implementations] implement the Standard, some better than others. All have interesting extensions which are not portable. All bring different elements of interest to the table of developers looking to solve different problems.
It also bears noting that thanks to the larger research community around lisp and its many implementations, modern Lisp compilers absolutely run circles around java. You can typically compile a lisp app in less time than the JVM takes to initialize and begin JIT'ing the bytecode. To say nothing of the actual execution speed.
Honestly, who cares if Java is forked. There's a virtual machine and bytecode specification -- simply make it a standard. Oh wait, Sun withdrew Java from the standards process, never mind then...
Stories like this do nothing to help build a convincing argument to Windows users that Linux is in fact the better OS. If even seasoned Linux users have problems upgrading their kernel, think of how frustrating it would be for someone less technically-inclined.
Tell me about it, the other day I just slapped the Win2k kernel into my win98 box and it just sailed right through, no problems at all...
I always thought that the Plus! packs were pretty cool concepts. Why doesn't Microsoft just do a barebones OS then a cheapy Plus! product with all the extra crap nobody needs?
Because the League Of Extraordinary Hypocrites would descend on them in force and give them a good nasal whining about how they gouge for basic features that should have come with the OS.
Can't figure out why MS doesn't sell a DVD distribution of XP for like $50 extra that bundles all the stuff that's also free downloads, like the command-line-only C++ and VB compilers. Sort of like MSDN super-lite. I'm not one to tell MS marketing what to do though, since they seem to rake in enough money to physically crush me every minute or so...
> Search Engine Optimization Professional
... other search engines, mostly paidclick ones of course. Those are fairly easy to strip out since the abstract is always nonsensical.
Also known as "Spammers". Their entire purpose is gaming the system to promote crass commerce (and boy I do mean crass), and just to top it off, I haven't seen one that hasn't engaged in spam to promote their service (to be fair, I probably wouldn't see the ones that don't spam, since I'm not in the business of pushing sites up in some silly ratings criterion).
I never noticed the problem with google til just recently, just a few sites that seem to have my keyword no matter *how* obscure but are filled with nothing but links to
I'd love to see an option on the google toolbar of "report link as googlespam", but I can imagine the manpower that'd be required to sift through submissions (you can't just do it automatically).
Am I the only one who remembers the name "Repton" as connected to a kind-of-slow but still fairly fun knockoff of Defender/Stargate for the C64? Instead of kidnapping citizens, the bad guys you were fighting were building some kind of super-weapon, and you had to prevent them from completing it. I guess it's bound to happen that names that sound cool will collide. I for one have never even heard of the other repton.
now when people use a GNOME app next to a non-GNOME app things are not consistent and usability is damaged as the use can no longer simply learn where the buttons are in all dialogs: they actually have to read each one. this makes X11 look downright silly and stupid.
I'm all for having a standard placement order dictated by, I dunno, some "semantic layout engine" that's cross-toolkit/platform or some other handwaving, just for visual consistency's sake. But if you're not expecting someone to read the content of a pop up alert dialog, then perhaps it doesn't really belong in an alert dialog?
> I'm wondering if maybe kde might be or might become the killer app for Linux?
I personally would like for it to become the killer app for Windows. It'd kick the crap out of litestep and all those other anemic shell replacements that are nothing but glorified launchpads and theme engines. Some apps might not port well, but most of them should come through just fine.
But thanks to the sorry backward state of Qt on win32, I don't see it as happening, ever, GPL'd or not.
But with so much game logic in scripts nowadays, doesn't it make sense to use a common, open source MUD platform
It's already been done, DGD, MOO, Cold, are quite nicely programmable generic platforms. It's certainly not technology stopping anyone, even if the languages are somewhat aging and others might prefer to program in, say, Python (I patched Python to make it a multi-user-safe runtime once, it's not horribly difficult). There's quite a few, for lack of a better term, "systemic" problems that keep MUDs from flourishing:
1) They had their day. Lots of things on the net are simply fads, which flare up, then subside and leave only a core group of remaining followers, not much bigger or even smaller than the original population.
2) The gameplay on most MUDs isn't much different than MMORPG's: jump on the mobs that pop, loot, level, repeat. Doesn't engage the brain anyway, so why would you want to do it without any eye candy?
3) The "pure RP" muds that are out there are often a variation of hanging out at the guildhall and kvetching.
4) At some point in any game development, one hits a "critical resource" problem: server bandwidth, storage, and above all, artistic talent. Artists aren't into commity work like coders are, it's in fact pretty antithetical to their worldview for most of them. Whether changing the art culture to a free software ethic is the answer or not, the fact remains that good artists that can produce models, textures, character art, voice acting, writing, and music are damn expensive. The ones a project does get tend to have to make it a full-time-job to get any decent amount of art done (this tends to be true for visual artists, musicians seem to be able to do part time production work)
ex. I write YourDoom.A and i write it using this new code morphing obfuscator. how exactly are Anti-virus programs 1. suppose to remove this? 2. identify this?
It's called polymorphic encryption, and viruses have been doing it for at least a decade. AV scanners look for the decryptor. Or they just intercept its attempted accesses when it does "wake up".