I think that you've very well summed up what's wrong with the popular idea that "your language determines your thoughts": that the real determination comes from your cultural exposure, not directly from the language and its words. The key to understanding 'kami' (whatever that may be) is to participate in Japanese culture, not specifically to speak Japanese. The main thing that gets linguists upset is to see nonsense claims like "this or that language will let you think more clearly and logically". An extreme of this kind of claim is Korzybski's General Semantics; or, in the SF field, S.R Delany's Babel 17.
The one "nit to pick" with Tim Sweeney's article about programming languages is that his parallel to "linguistics" refers exactly to this kind of claim, citing it as a "well known fact in linguistics", while among the cognitive linguistic community, it is considered more like a "well known misconception".
it just streams them to you rather than letting you ftp them.
that is technically impossible. if something gets to your PC, then you *can* save it. in the case of an mp3 streamed through http, it's not even hard: just find out the URL and use wget.
that was my first reaction too. curing meningitis and other serious viral diseases is great. taking medicine to cure a common cold, well, I'd rather just wait for it go away on its own, it just takes a week usually anyway. if we help the body by getting rid of even the small nasties on its behalf, it'll end up weakening us.
well, JK is part of a mainstream and a larger industry, which is why he only "sort of kind of" fits on slashdot. he's never hidden that; he's fascinated by geeks, but not a geek himself. considering where he comes from, i'm not *too* surprised that his book got picked. i'm glad for him, but i don't think i'll care enough to see the movie if and when it comes out.
the website "yahoo quebec", a parody of the famous american portal Yahoo!, has been created a few months ago by JH Roy, who does the radio show "Branché" at Radio Canada.
Strangely enough, he was mentioned in the threatening letter from Yahoo!, and hadn't even heard of it when we talked to him in the evening. ``I'll start by reading the letter and looking at the laws on cybersquatting and commerce brands, before I see if I take the site down'', he says. ``maybe i'll change the search engine and the logo''.
In his parody, JH Roy used Yahoo!'s logo, with the word "Quebec" added. The search engine searches in the database of a porn site; it's presumably these 2 details that annoyed Yahoo! the most. The letter seems to confirm this.
JH Roy says, ``I thought Yahoo! was the last compay around with a sense of humour. There are several parodies of Yahoo!, some of them are even listed on Yahoo itself!''.
Clément Laberge, who maintains the weblog pssst!, and to whom Yahoo!'s threatenign letter is mostly addressed to, has up to next monday (17 jan) to tell Yahoo's lawyers that he has taken the site down. Which he can't do, since he has no control over the site.
I actually find that many geeks are very understanding and considerate. Therefore many girls end wanting them as friends, not as lovers. This is a real problem, which I think geeks should think about.
How is this a problem? Being valued as a friend is great.
no, the real problem is that adults (yes, including me) have somehow become convinced that there is something so "serious" about our properties and appearances that coloring walls, harmless as it is from every point of view you look it at from, has become completely unacceptable.
as an individual, advertisement is fairly easy to ignore, especially once you figure out what a bunch of posters just explained (that it works by making things appear glamorous to you, so that you'll want them). the instant recipe is: a negative attitude towards glamour. instead of wanting things because they're glamorous, be annoyed at how people can be impressed by that hollow crap.
then again, anyone who watches a significant amount of TV is not likely to want to do that, because it makes 80% of TV hard to tolerate. finding 80% of TV contents intolerable is a good thing in my book.
please be careful about your generalizations. some of us use mutt (i don't know about pine) and love it. i'd rather read my mail with a bunch of thrown together perl scripts than use a GUI for that.
2 highly principled people (within weeks), with the will power to actually stick to their principles when HUGE wads of cash are dangled in front of their noses. My belief in humanity is slowly coming back
Dunno, I think if that I was offered legal money that I considered dirty for my own reasons, I'd take it and make a point of donating it all to worthwhile causes. Sticking with your principles and doing nothing is fine, but leeching off the evil bastards to make the world a better place is even better.
Re:How does this mock religion?
on
Planet Gattaca
·
· Score: 2
I *still* don't see how starting from scratch poses any bigger or different ethical or moral questions than starting from bits of bacteria ADN. The problem is what you're building, not what you're building it with. For me, there are serious issues as soon as you start building something complex (by life's own standards). As long as it's unicellular I think we're quite safe; I've yet to see anyone worry about the rights of bacteria, or whether they have souls. The other worry is what you do with them; as long as it's kept isolated in serious lab conditions, it's ok with me.
So my take is, yes, go ahead with this experiment. But debate before messing with anything further up than bacteria, or before even thinking of releasing such a thing in the wild.
err, read the article, this "new life" is in the form of bacteria, and there's no talk of letting it in the wild, just playing with it in the lab.
personally, I don't see why this should pose any ethical or moral problems. bacteria are interesting, but as life goes, they're also so primitive, tiny and relatively simple that I don't see anything wrong with playing with them in any way they want. I doubt anyone would come screaming about the "rights of bacteria".
there *are* reasons to be worried about this, but they don't have anything to do with the "creation" part; making "creation" a special thing outside of human reach seems to be a christian knee-jerk reaction, but I see no real basis for it. OTOH, I *do* worry about the possibility of governments and large organizations building custom bacteria to bring bio-warfare and bio-terrorism to a new level. I hope it doesn't happen, but sadly enough, we can't do anything much more than hope, here.
cookies, like too many things on the web, are only harmless (for your privacy) if you're tech oriented enough to handle them. in my case, "handling cookies" means allowing them during a browsing session, then clearing the cookie file at the end (possibly leaving one or two that i trust, like/.). so everything works, including those annoying sites that depend on ASPSESSIONID or something like that, but doubleclick can't do anything bad. let them store hundreds of short lived sessions if it amuses them... (well, not exactly doubleclick now that i've/etc/hosts'ed them to 127.0.0.2, but any similar sites).
as regards this cursor software thing, i'm amazed to see people saying that "logging someone's list of visited sites" is harmless!
your comments really hit home here; I've been thinking very closely along these lines: on the www, the client software controls everything (pulling content from the net) on behalf of the user. no site can possibly shove anything on my screen if I don't want it, it's just that the current generation of browsers make it hard for me to make fine-grained choices like: when loading slashdot.org, feel free to fetch content from anywhere, but when loading salon.com, ignore embedded content from 208.178.101.41, and when loading random pages, ignore all embedded content from other domains, unless the URL contains an added command not to; also, on geocities and xoom, disable new window creation from javascript.
proxies like junkbuster try to do some of this, but they suffer from being a proxy: they can't be as closely integrated with the browser as one would like, and they make the whole browsing slower and less responsive because of the proxying overhead.
I have a serious suggestion here: write a program that does this kind of job (based on a config file), by intercepting the browser's (e.g netscape navigator 4.x) calls to libc, using LD_PRELOAD to get itself loaded. the library would basically filter all network related syscalls (select, read, write, connect, close, shutdown, setsockopt), monitor HTTP connections, rewrite headers as appropriate, and decide which requests to allow or not. (think of this as a stop-gap measure; as soon as mozilla is ready, the core of this can be directly integrated in it, without shared lib hackery, and more fine grained things like selective access to javascript functions can be added ; a libc wrapper can't do that cleanly).
as far as I know this hasn't been done yet, with the LD_PRELOAD approach (as opposed to proxies, which are abundant), so I'm definitely going to start work on it myself, probably during xmas break. in the meantime, I want to get the ideas ready (casey-b's domains are a good one), so that when i start coding, i know what to type:)
if anyone else is interested enough, let me know by mail... help is always appreciated:)
yes, sendmail is the most powerful MTA out there, in that its config language is even Turing-complete. great. now, a few months ago I needed to do something quite unusual with an MTA (keeping some usernames locally, while passing others to another SMTP server, ignoring MX records in the process). with qmail, it was a matter of reading a few manpages, then finding the right place to make a change, and adding 5 lines of C to the source. now, this may sound harder than just tweaking a config file, but it isn't necessarily. qmail's code is quite clear (although the style is unusual), and the whole thing took about two hours, testing included. I certainly didn't need to buy a book, or anything like that. I'm sure that, with Sendmail, I would have needed more time just to understand what to change in the config file. IMO, there is something seriously wrong about an MTA that needs *books* to be written about it.
err, GNOME doesn't include a replacement for sendmail. Postfix and Qmail (and others) are replacements for sendmail; unfortunately Qmail's author refuses to use an open-source compliant license, so it's out for RedHat. But Postfix would be a much better default than Sendmail; I wish they would use it as default MTA in their distribution, instead of doing political moves like collaborating with Sendmail, Inc.
as for Mozilla, well, whatever can help bring it to usability soon, will be good.
I tend to agree with this argument ("it's only virtual, let's be lax"), except for one reason: computers actually ARE used in situations where human lives depend on them. what if such a computer got infected with a virus and crashed?
I have no sympathy for those who yell "punish the writers of cracker tools". disrupting other people's computers is certainly punishable, but it's ridiculous to even compare it to a crime (unless it's a hospital's computers or something like that), yet people apparently get sent to jail for months, for a simple web defacing.
and, putting it all together, I think it's been more than proved by practice (and by BUGTRAQ) that full disclosure is good for the whole of the industry.
It's not an example of fragmentation, because the resulting system would still run FreeBSD binaries, all of them, including commercial closed-source ones, natively. By doing this, they have not created a new target for compilation, just a new distribution with a different choice of tools.
my feeling is that Debian/FreeBSD should base itself on glibc2.1, i.e port glibc to the FreeBSD kernel, and compile and run the whole distribution on that, *then* also include a copy of FreeBSD's libc, to be able to run regular FreeBSD programs (just like RH6 can run libc5 programs too). otherwise, Debian/FreeBSD is nothing more than FreeBSD with a bunch of GNU tools added and the installation procedure re-tooled.
OTOH, I wouldn't commit myself to a project like this for the long-term; it appeals to me more for the hack value, and for the political implications (hopefully the good ones, of showing the various factions how easily software can work together) than for actual usefulness.
maybe i'm weird, but i can't think of anything in particular that i'd like someone to buy for me, right now. well, i could list a few books, but i already have a few waiting to read, so there's no big hurry for that, and they're not very expensive anyway. i also don't feel like upgrading any hardware (it's quite enough for what i do with it). so i guess i'll skip the whole xmas thing and just lay off work during the holidays, and catch up on other things.
Slashdot Mirror
hey, I've heard the "50 words for snow" crap coming out of an actual linguistics teacher (introduction to semantics). blah.
The one "nit to pick" with Tim Sweeney's article about programming languages is that his parallel to "linguistics" refers exactly to this kind of claim, citing it as a "well known fact in linguistics", while among the cognitive linguistic community, it is considered more like a "well known misconception".
that is technically impossible. if something gets to your PC, then you *can* save it. in the case of an mp3 streamed through http, it's not even hard: just find out the URL and use wget.
that was my first reaction too. curing meningitis and other serious viral diseases is great. taking medicine to cure a common cold, well, I'd rather just wait for it go away on its own, it just takes a week usually anyway. if we help the body by getting rid of even the small nasties on its behalf, it'll end up weakening us.
well, JK is part of a mainstream and a larger industry, which is why he only "sort of kind of" fits on slashdot. he's never hidden that; he's fascinated by geeks, but not a geek himself. considering where he comes from, i'm not *too* surprised that his book got picked. i'm glad for him, but i don't think i'll care enough to see the movie if and when it comes out.
bah, "cease & desist letter" is the right translation for "mise en demeure"... not "threatening letter" :) I did say "qnd".
yep, seems reasonable to ask that the logo/brand should be altered somewhat. Yahoo still goofed up and sent its letter to the wrong person, though.
the website "yahoo quebec", a parody of the famous american portal Yahoo!, has been created a few months ago by JH Roy, who does the radio show "Branché" at Radio Canada.
Strangely enough, he was mentioned in the threatening letter from Yahoo!, and hadn't even heard of it when we talked to him in the evening. ``I'll start by reading the letter and looking at the laws on cybersquatting and commerce brands, before I see if I take the site down'', he says. ``maybe i'll change the search engine and the logo''.
In his parody, JH Roy used Yahoo!'s logo, with the word "Quebec" added. The search engine searches in the database of a porn site; it's presumably these 2 details that annoyed Yahoo! the most. The letter seems to confirm this.
JH Roy says, ``I thought Yahoo! was the last compay around with a sense of humour. There are several parodies of Yahoo!, some of them are even listed on Yahoo itself!''.
Clément Laberge, who maintains the weblog pssst!, and to whom Yahoo!'s threatenign letter is mostly addressed to, has up to next monday (17 jan) to tell Yahoo's lawyers that he has taken the site down. Which he can't do, since he has no control over the site.
rotfl :)
no, the real problem is that adults (yes, including me) have somehow become convinced that there is something so "serious" about our properties and appearances that coloring walls, harmless as it is from every point of view you look it at from, has become completely unacceptable.
then again, anyone who watches a significant amount of TV is not likely to want to do that, because it makes 80% of TV hard to tolerate. finding 80% of TV contents intolerable is a good thing in my book.
just one little stupid correction: Sapir and Whorf were two people, not one.
please be careful about your generalizations. some of us use mutt (i don't know about pine) and love it. i'd rather read my mail with a bunch of thrown together perl scripts than use a GUI for that.
Dunno, I think if that I was offered legal money that I considered dirty for my own reasons, I'd take it and make a point of donating it all to worthwhile causes. Sticking with your principles and doing nothing is fine, but leeching off the evil bastards to make the world a better place is even better.
So my take is, yes, go ahead with this experiment. But debate before messing with anything further up than bacteria, or before even thinking of releasing such a thing in the wild.
personally, I don't see why this should pose any ethical or moral problems. bacteria are interesting, but as life goes, they're also so primitive, tiny and relatively simple that I don't see anything wrong with playing with them in any way they want. I doubt anyone would come screaming about the "rights of bacteria".
there *are* reasons to be worried about this, but they don't have anything to do with the "creation" part; making "creation" a special thing outside of human reach seems to be a christian knee-jerk reaction, but I see no real basis for it. OTOH, I *do* worry about the possibility of governments and large organizations building custom bacteria to bring bio-warfare and bio-terrorism to a new level. I hope it doesn't happen, but sadly enough, we can't do anything much more than hope, here.
BRANDING is something that's generally done to cattle. don't let anyone trick you into thinking otherwise.
as regards this cursor software thing, i'm amazed to see people saying that "logging someone's list of visited sites" is harmless!
proxies like junkbuster try to do some of this, but they suffer from being a proxy: they can't be as closely integrated with the browser as one would like, and they make the whole browsing slower and less responsive because of the proxying overhead.
I have a serious suggestion here: write a program that does this kind of job (based on a config file), by intercepting the browser's (e.g netscape navigator 4.x) calls to libc, using LD_PRELOAD to get itself loaded. the library would basically filter all network related syscalls (select, read, write, connect, close, shutdown, setsockopt), monitor HTTP connections, rewrite headers as appropriate, and decide which requests to allow or not. (think of this as a stop-gap measure; as soon as mozilla is ready, the core of this can be directly integrated in it, without shared lib hackery, and more fine grained things like selective access to javascript functions can be added ; a libc wrapper can't do that cleanly).
as far as I know this hasn't been done yet, with the LD_PRELOAD approach (as opposed to proxies, which are abundant), so I'm definitely going to start work on it myself, probably during xmas break. in the meantime, I want to get the ideas ready (casey-b's domains are a good one), so that when i start coding, i know what to type :)
if anyone else is interested enough, let me know by mail... help is always appreciated :)
yes, sendmail is the most powerful MTA out there, in that its config language is even Turing-complete. great. now, a few months ago I needed to do something quite unusual with an MTA (keeping some usernames locally, while passing others to another SMTP server, ignoring MX records in the process). with qmail, it was a matter of reading a few manpages, then finding the right place to make a change, and adding 5 lines of C to the source. now, this may sound harder than just tweaking a config file, but it isn't necessarily. qmail's code is quite clear (although the style is unusual), and the whole thing took about two hours, testing included. I certainly didn't need to buy a book, or anything like that. I'm sure that, with Sendmail, I would have needed more time just to understand what to change in the config file. IMO, there is something seriously wrong about an MTA that needs *books* to be written about it.
as for Mozilla, well, whatever can help bring it to usability soon, will be good.
I have no sympathy for those who yell "punish the writers of cracker tools". disrupting other people's computers is certainly punishable, but it's ridiculous to even compare it to a crime (unless it's a hospital's computers or something like that), yet people apparently get sent to jail for months, for a simple web defacing.
and, putting it all together, I think it's been more than proved by practice (and by BUGTRAQ) that full disclosure is good for the whole of the industry.
my feeling is that Debian/FreeBSD should base itself on glibc2.1, i.e port glibc to the FreeBSD kernel, and compile and run the whole distribution on that, *then* also include a copy of FreeBSD's libc, to be able to run regular FreeBSD programs (just like RH6 can run libc5 programs too). otherwise, Debian/FreeBSD is nothing more than FreeBSD with a bunch of GNU tools added and the installation procedure re-tooled.
OTOH, I wouldn't commit myself to a project like this for the long-term; it appeals to me more for the hack value, and for the political implications (hopefully the good ones, of showing the various factions how easily software can work together) than for actual usefulness.
maybe i'm weird, but i can't think of anything in particular that i'd like someone to buy for me, right now. well, i could list a few books, but i already have a few waiting to read, so there's no big hurry for that, and they're not very expensive anyway. i also don't feel like upgrading any hardware (it's quite enough for what i do with it). so i guess i'll skip the whole xmas thing and just lay off work during the holidays, and catch up on other things.