social topic

2010-09-10 11:30:23 am

Club 27 Goals

Fine BioSpank, he said to his friend, "she doesn't really like me I don't think" and after talking a bit we kinda decided to keep it at the friend level.

I mean I wouldn't have had a problem with dating or anything, but there was really not anything there.

On an unrelated note I accidentally slept in, missed my first period, and when my mom woke me up she said I can either go to school, or I have to go with my family to the aquarium. I guess I'm going to the aquarium. I'll take pictures.

Gaius_4

2010-09-10 11:35:08 am

So... What will your excuse be for missing school? Sea sick? I was always sick of school, but, sadly, that doesn't qualify as a medical condition. aiwebs_027

DJGrenola

2010-09-10 01:08:11 pm

Quote from nate:

just finished the cuckoo's egg finally. really glad i read it. gratitude to chanoire for recommending it via grenola.

I haven't heard from her in a little while, but I will be sure to let her know when I next encounter her. I suspected you'd enjoy it, just as she suspected I would.

Quote from nate:

the best stuff was the thick description of the state of the unix world in the mid 80s. my experience with computers consisted of failing at games launched from prodos on my dad's apple ][e at that point so this is the only way i can learn about things like that.

Essentially irrelevant: While we're talking about 8-bit era micros, I just bought this book, written by a guy I encountered and chatted with at the UK Vintage Computing Festival this year. It documents a complete logic-level reversal of the ZX Spectrum's custom chip, which essentially implements the machine's I/O interfaces. I am finding the sections dealing with video generation particularly interesting in the context of some of the issues thrown up by e.g. NES signal capture for speed demos archive. In the preface, Smith mentions that the job would not have been possible had a guy from Datel not offered to take one of the chips and decapsulate and photograph it under a microscope. Datel are a weird company.

Quote from nate:

and unix became really important to me later on so i really get into learning about it. i mean stuff like system v and bsd tripping you up if you knew one but not the other - it's the exact same thing today, exact same thing. like when grenola told me "md5 is a bsd-ism" and linux wants you to type md5sum.

In older versions of Mac OS X I used to get irritated that 'ps -ef' would throw an error. There is at least a standard now -- the POSIX Single Unix Specification -- which didn't (I don't think) exist in Stoll's time. I suspect that it is the desire to standardise on POSIX which means that I can in fact now type 'ps -ef' on OS X as well as Linux and it works.

Quote from nate:

of course there are tons of other things that haven't changed a bit either. i don't claim to know a lot about security but i recognized almost everything i saw from much more recent experience. the only thing i missed was hess taking passwd files in order to run dictionary attacks and i was kicking myself for that by the end of the book. of course you could use dictionaries in 1986.

I think the book becomes more fun when you have had a bit of experience studying security so you can try to guess what will happen next, or dream up alternative approaches that Stoll or the hacker might have taken. I made the same observation you did (that security history has a way of repeating itself). Again, though, there are some differences from Stoll's time. One is the sheer number of people now using global digital networks compared to then. More important though is the invention and widespread adoption of RSA, which would have rendered Stoll's wire-tapping approach difficult or impossible.

Quote from nate:

it even applies to sda. so sda verification is kind of a joke but we have to take it seriously because people take it seriously if we take it seriously and it builds community. it's rhetoric that says "we care." and if we don't care then no one will. which is fine, but that means no sda and everything it has done for me.

Isn't the whole point of the verification process that you don't trust people? This feels like another one of your slightly jarring free associations.

[quote="nate"]just finished the cuckoo's egg finally. really glad i read it. gratitude to chanoire for recommending it via grenola.[/quote]

I haven't heard from her in a little while, but I will be sure to let her know when I next encounter her. I suspected you'd enjoy it, just as she suspected I would.

[quote="nate"]the best stuff was the thick description of the state of the unix world in the mid 80s. my experience with computers consisted of failing at games launched from prodos on my dad's apple ][e at that point so this is the only way i can learn about things like that.[/quote]

Essentially irrelevant: While we're talking about 8-bit era micros, I just bought [url=http://www.zxdesign.info/book/]this book[/url], written by a guy I encountered and chatted with at the UK Vintage Computing Festival this year. It documents a complete logic-level reversal of the ZX Spectrum's custom chip, which essentially implements the machine's I/O interfaces. I am finding the sections dealing with video generation particularly interesting in the context of some of the issues thrown up by e.g. NES signal capture for speed demos archive. In the preface, Smith mentions that the job would not have been possible had a guy from Datel not offered to take one of the chips and decapsulate and photograph it under a microscope. Datel are a weird company.

[quote="nate"]and unix became really important to me later on so i really get into learning about it. i mean stuff like system v and bsd tripping you up if you knew one but not the other - it's the exact same thing today, exact same thing. like when grenola told me "md5 is a bsd-ism" and linux wants you to type md5sum.[/quote]

In older versions of Mac OS X I used to get irritated that 'ps -ef' would throw an error. There is at least a standard now -- the POSIX Single Unix Specification -- which didn't (I don't think) exist in Stoll's time. I suspect that it is the desire to standardise on POSIX which means that I can in fact now type 'ps -ef' on OS X as well as Linux and it works.

[quote="nate"]of course there are tons of other things that haven't changed a bit either. i don't claim to know a lot about security but i recognized almost everything i saw from much more recent experience. the only thing i missed was hess taking passwd files in order to run dictionary attacks and i was kicking myself for that by the end of the book. of course you could use dictionaries in 1986.[/quote]

I think the book becomes more fun when you have had a bit of experience studying security so you can try to guess what will happen next, or dream up alternative approaches that Stoll or the hacker might have taken. I made the same observation you did (that security history has a way of repeating itself). Again, though, there are some differences from Stoll's time. One is the sheer number of people now using global digital networks compared to then. More important though is the invention and widespread adoption of RSA, which would have rendered Stoll's wire-tapping approach difficult or impossible.

[quote="nate"]it even applies to sda. so sda verification is kind of a joke but we have to take it seriously because people take it seriously if we take it seriously and it builds community. it's rhetoric that says "we care." and if we don't care then no one will. which is fine, but that means no sda and everything it has done for me.[/quote]

Isn't the whole point of the verification process that you [i]don't[/i] trust people? This feels like another one of your slightly jarring free associations.

nate

Edit history:

nate: 2010-09-10 03:05:39 pm

2010-09-10 03:02:48 pm

Quote from DJGrenola:

Essentially irrelevant: While we're talking about 8-bit era micros, I just bought this book, written by a guy I encountered and chatted with at the UK Vintage Computing Festival this year. It documents a complete logic-level reversal of the ZX Spectrum's custom chip, which essentially implements the machine's I/O interfaces. I am finding the sections dealing with video generation particularly interesting in the context of some of the issues thrown up by e.g. NES signal capture for speed demos archive. In the preface, Smith mentions that the job would not have been possible had a guy from Datel not offered to take one of the chips and decapsulate and photograph it under a microscope. Datel are a weird company.

this actually remains of the highest relevance today even in the era of the standard dvd recorder with built-in tbc. in some cases the tbc is either inaccurate or entirely awol (difficult for me to say which without signal analysis on hardware i don't own) and the capture hardware drops or inserts frames to compensate. this would be catastrophic (well, unless you want to resync the audio every few seconds by hand) at demuxing time if it weren't for mplayer which fully implements the spec and spits out nmf with a constant desync which is then easily adjusted. of course it's still a pain because i have to generate separate timing and sync verification copies in the case of manual timing but it's a good example of how you can't escape 1980s silicon in 2010. i am actually doing this right now with tiki's new mega man x2 run.

for the record, i have seen this most often with pal nes and its blue colors, but i have also seen it in isolated cases with everything up to and including the 16-bit consoles.

Quote from DJGrenola:

I think the book becomes more fun when you have had a bit of experience studying security so you can try to guess what will happen next, or dream up alternative approaches that Stoll or the hacker might have taken. I made the same observation you did (that security history has a way of repeating itself). Again, though, there are some differences from Stoll's time. One is the sheer number of people now using global digital networks compared to then. More important though is the invention and widespread adoption of RSA, which would have rendered Stoll's wire-tapping approach difficult or impossible.

yeah, especially all the telnet with cleartext passwords. that was one big thing that had mostly passed away even when i was just starting to use linux in december 1998. it's interesting because in the last chapter he warns against encryption as an impediment to progress but that ended up ironic. he made the error ken thompson always warns about - assuming that tomorrow's hardware will be the same as today's when in reality it's always different enough to render at least some of your assumptions NULL and void. even in this book, though, crypto was there in the passwd files, unaware of its destiny as the next engine of trust to power the network.

it's actually so good a read today that it made me rethink some of my own procedures. it seems like passwords in general are just really inferior to public key authentication. i had already started using public key auth myself years ago and with other people who have to log in to my servers more recently but now i'm thinking about what it would mean to store passwords in a more secure way. it's always about the weakest point of failure. the root password has to be there as a backdoor for me but that doesn't mean i have to store it unencrypted digitally anywhere or encrypted in more than one place.

on this same topic - the most gut-wrenching parts of the novel for me were always when stoll was watching hess go for things that would be major security breaches and had to make a decision within ten seconds what he would do. i mean there's no way to prepare for that kind of thing. it's all snap judgments and snap judgments are sometimes the worst of all of our terrible judgments and we're aware of that while we're making them. so it's like staring into the face of death. and i think how he made it through those situations really shows his resourcefulness and strong emotional control. just so different from the solutions we might try today with the manual shorting of the line and so on.

Quote from DJGrenola:

Quote from nate:

it even applies to sda. so sda verification is kind of a joke but we have to take it seriously because people take it seriously if we take it seriously and it builds community. it's rhetoric that says "we care." and if we don't care then no one will. which is fine, but that means no sda and everything it has done for me.

Isn't the whole point of the verification process that you don't trust people? This feels like another one of your slightly jarring free associations.

yes and no. it seems like you're thinking of verification in terms of a relationship between the verifiers (and maybe the site administrators) and the runner, while i'm thinking of it in terms of a relationship between people who might want to speedrun games and sda as a whole (which is not a person but rather a network composed of many voluntary members). my theory is that verification actually increases the attractiveness of sda - of submitting content to it and retrieving content from it. this ties into what stoll was saying because he was basically saying that if people don't want to use the network then there's no network. it is an illusion of security, the same illusion that powers every society. so if people have a bad impression of sda or think it's all a joke then there's no sda, or at least no sda how we know it.

i'm sure a lot of people here remember my anti-tas crusade. it had this same motivation. but it had a happy ending because sda grew up and gained people's trust in a way that i didn't know how to achieve at that time (so i was pushing on the other side trying to get them to back down). it was mostly due to mike uyama who basically invented verification. i'm not mike uyama so i don't know what he was thinking but i know he was concerned about really embarrassingly bad (in terms of execution, play skill) stuff getting on the site because radix hadn't played every single game ever made and having some peers look over it before it's accepted was a good solution to that. but over time my "no cheating"/trust concerns came to be conflated with it and today people sometimes see it mike's way or my way and not necessarily both. i don't even think you can predict who (which group, insiders or outsiders) see it which way, either. if you read verifier comments on runs (which are now public by default) you will see that some verifiers see it in terms of trust - there is one guy who always writes SEVERE CHEATING ironically and i'm betting that's because he thinks the whole thing is worthless as a method of establishing trust. and i totally agree. sometimes you can determine that something is cheating/emulator use and sometimes you can't. there's a threshold. and trust isn't like that, it's binary. not only that but in order to decide that something is cheating you have to define cheating and in our world you can't define anything with absolute accuracy. it flies in the face of our metroid prime as a game changing on a daily basis which is where modern sda comes from in the first place. saber understood this probably five years before i did. sparky brought it up recently too. but i'm obsessed with perfection and i was, i think, justified in fearing for the future of my beloved community, just as stoll is in the last chapter of the book. so we wondered, how could this thing possibly go on like this? but life goes on. somehow things work out. and we are part of the solution.

[quote="DJGrenola"]Essentially irrelevant: While we're talking about 8-bit era micros, I just bought [url=http://www.zxdesign.info/book/]this book[/url], written by a guy I encountered and chatted with at the UK Vintage Computing Festival this year. It documents a complete logic-level reversal of the ZX Spectrum's custom chip, which essentially implements the machine's I/O interfaces. I am finding the sections dealing with video generation particularly interesting in the context of some of the issues thrown up by e.g. NES signal capture for speed demos archive. In the preface, Smith mentions that the job would not have been possible had a guy from Datel not offered to take one of the chips and decapsulate and photograph it under a microscope. Datel are a weird company.[/quote]
this actually remains of the highest relevance today even in the era of the standard dvd recorder with built-in tbc. in some cases the tbc is either inaccurate or entirely awol (difficult for me to say which without signal analysis on hardware i don't own) and the capture hardware drops or inserts frames to compensate. this would be catastrophic (well, unless you want to resync the audio every few seconds by hand) at demuxing time if it weren't for mplayer which fully implements the spec and spits out nmf with a constant desync which is then easily adjusted. of course it's still a pain because i have to generate separate timing and sync verification copies in the case of manual timing but it's a good example of how you can't escape 1980s silicon in 2010. i am actually doing this right now with tiki's new mega man x2 run.

for the record, i have seen this most often with pal nes and its blue colors, but i have also seen it in isolated cases with everything up to and including the 16-bit consoles.

[quote="DJGrenola"]I think the book becomes more fun when you have had a bit of experience studying security so you can try to guess what will happen next, or dream up alternative approaches that Stoll or the hacker might have taken. I made the same observation you did (that security history has a way of repeating itself). Again, though, there are some differences from Stoll's time. One is the sheer number of people now using global digital networks compared to then. More important though is the invention and widespread adoption of RSA, which would have rendered Stoll's wire-tapping approach difficult or impossible.[/quote]
yeah, especially all the telnet with cleartext passwords. that was one big thing that had mostly passed away even when i was just starting to use linux in december 1998. it's interesting because in the last chapter he warns against encryption as an impediment to progress but that ended up ironic. he made the error ken thompson always warns about - assuming that tomorrow's hardware will be the same as today's when in reality it's always different enough to render at least some of your assumptions NULL and void. even in this book, though, crypto was there in the passwd files, unaware of its destiny as the next engine of trust to power the network.

it's actually so good a read today that it made me rethink some of my own procedures. it seems like passwords in general are just really inferior to public key authentication. i had already started using public key auth myself years ago and with other people who have to log in to my servers more recently but now i'm thinking about what it would mean to store passwords in a more secure way. it's always about the weakest point of failure. the root password has to be there as a backdoor for me but that doesn't mean i have to store it unencrypted digitally anywhere or encrypted in more than one place.

on this same topic - the most gut-wrenching parts of the novel for me were always when stoll was watching hess go for things that would be major security breaches and had to make a decision within ten seconds what he would do. i mean there's no way to prepare for that kind of thing. it's all snap judgments and snap judgments are sometimes the worst of all of our terrible judgments and we're aware of that while we're making them. so it's like staring into the face of death. and i think how he made it through those situations really shows his resourcefulness and strong emotional control. just so different from the solutions we might try today with the manual shorting of the line and so on.

[quote="DJGrenola"][quote="nate"]it even applies to sda. so sda verification is kind of a joke but we have to take it seriously because people take it seriously if we take it seriously and it builds community. it's rhetoric that says "we care." and if we don't care then no one will. which is fine, but that means no sda and everything it has done for me.[/quote]

Isn't the whole point of the verification process that you [i]don't[/i] trust people? This feels like another one of your slightly jarring free associations.[/quote]
yes and no. it seems like you're thinking of verification in terms of a relationship between the verifiers (and maybe the site administrators) and the runner, while i'm thinking of it in terms of a relationship between people who might want to speedrun games and sda as a whole (which is not a person but rather a network composed of many voluntary members). my theory is that verification actually increases the attractiveness of sda - of submitting content to it and retrieving content from it. this ties into what stoll was saying because he was basically saying that if people don't want to use the network then there's no network. it is an illusion of security, the same illusion that powers every society. so if people have a bad impression of sda or think it's all a joke then there's no sda, or at least no sda how we know it.

i'm sure a lot of people here remember my anti-tas crusade. it had this same motivation. but it had a happy ending because sda grew up and gained people's trust in a way that i didn't know how to achieve at that time (so i was pushing on the other side trying to get them to back down). it was mostly due to mike uyama who basically invented verification. i'm not mike uyama so i don't know what he was thinking but i know he was concerned about really embarrassingly bad (in terms of execution, play skill) stuff getting on the site because radix hadn't played every single game ever made and having some peers look over it before it's accepted was a good solution to that. but over time my "no cheating"/trust concerns came to be conflated with it and today people sometimes see it mike's way or my way and not necessarily both. i don't even think you can predict who (which group, insiders or outsiders) see it which way, either. if you read verifier comments on runs (which are now public by default) you will see that some verifiers see it in terms of trust - there is one guy who always writes SEVERE CHEATING ironically and i'm betting that's because he thinks the whole thing is worthless as a method of establishing trust. and i totally agree. sometimes you can determine that something is cheating/emulator use and sometimes you can't. there's a threshold. and trust isn't like that, it's binary. not only that but in order to decide that something is cheating you have to define cheating and in our world you can't define anything with absolute accuracy. it flies in the face of our metroid prime as a game changing on a daily basis which is where modern sda comes from in the first place. saber understood this probably five years before i did. sparky brought it up recently too. but i'm obsessed with perfection and i was, i think, justified in fearing for the future of my beloved community, just as stoll is in the last chapter of the book. so we wondered, how could this thing possibly go on like this? but life goes on. somehow things work out. and we are part of the solution.

BioSpark

2010-09-10 03:07:23 pm

k

Idkbutlike2

2010-09-10 03:29:58 pm

Damn, nate. When did you find time to write that whole lecture?

DJGrenola

2010-09-10 03:47:52 pm

radix was enforcing peer verification before the whole TAS business occurred, both for play quality reasons and for reasons of authenticity. I don't know where this business about mike's way and your way comes from.

nate

2010-09-10 05:40:40 pm

it comes from mike. i guess we'd have to get his side of the story to go any further. isn't it interesting how little i know about sda from the early days? go getting your info from only one person.

about time - i dunno really. i feel like i have all the time in the world now.

DJGrenola

2010-09-10 06:15:26 pm

you used to know. you've forgotten.

nate

2010-09-10 06:28:30 pm

i realized driving home just now that this "radix was enforcing peer verification before the whole TAS business occurred" makes me think that you read my post as saying verification came about for some reason because of the tas thing. that's not what i meant to suggest at all. what i was doing was trying to relate how the strength of a community comes from a fictional mutual trust and how sda verification over time came to encapsulate some of that - one thing i'm sure of is that sda is way way way more "srs bsns" now than it was in 2005 or whenever the tas thing was. and one of the ways you can see that is through some of the things that have been happening in the verification threads. i think some people were talking about it here recently linking to the gta4 one and so on.

i guess it's different for people like me who have this fundamental need to be sure of everything and now growing beyond that every other way of thinking is novel and interesting to me. so sure there are plenty of people who don't think that verification proves anything. and i agree with that but i think nonetheless that it's a good way to assess community cohesion. it's like a rite of passage - it doesn't really mean anything except in the context of the specific culture.

now don't ask me whether it's good or bad. i don't have the slightest idea what good and bad would mean here.

i realized driving home just now that this "radix was enforcing peer verification before the whole TAS business occurred" makes me think that you read my post as saying verification came about for some reason because of the tas thing. that's not what i meant to suggest at all. what i was doing was trying to relate how the strength of a community comes from a fictional mutual trust and how sda verification over time came to encapsulate some of that - one thing i'm sure of is that sda is way way way more "srs bsns" now than it was in 2005 or whenever the tas thing was. and one of the ways you can see that is through some of the things that have been happening in the verification threads. i think some people were talking about it here recently linking to the gta4 one and so on.

i guess it's different for people like me who have this fundamental need to be sure of everything and now growing beyond that every other way of thinking is novel and interesting to me. so sure there are plenty of people who don't think that verification proves anything. and i agree with that but i think nonetheless that it's a good way to assess community cohesion. it's like a rite of passage - it doesn't really mean anything except in the context of the specific culture.

now don't ask me whether it's good or bad. i don't have the slightest idea what good and bad would mean here.

windy

2010-09-10 07:23:18 pm

gyrotechnic lollipop

Quote from Poision Envy:

On an unrelated note I accidentally slept in, missed my first period, and when my mom woke me up she said I can either go to school, or I have to go with my family to the aquarium. I guess I'm going to the aquarium. I'll take pictures.

aquarium ftfw. i wish there was one around here. :(

arkarian

2010-09-10 07:38:01 pm

red chamber dream

Quote from Poision Envy:

I accidentally slept in, missed my first period

lol

tomatobob

2010-09-10 07:41:09 pm

One shall stand, one shall ball.

Quote from Poision Envy:

On an unrelated note I accidentally slept in, missed my first period, and when my mom woke me up she said I can either go to school, or I have to go with my family to the aquarium.

lol how is that even a choice? go to school or do cool stuff.

Idkbutlike2

2010-09-10 07:43:04 pm

I'd rather go to school. Aquariums are boring as hell, even with those cheesy dolphin shows...

DynaDASH

2010-09-10 07:48:47 pm

Chooser of the Slain

[socializing]Oh hi people, nice to meet you all.[/socializing]

Opium

2010-09-10 07:49:35 pm

Aquariums are boring? I guess you aren't really fascinated with nature in general?

arkarian

2010-09-10 07:51:06 pm

red chamber dream

aquariums kick ass. water creatures are probably the most fascinating of them all to me.

also a good place to make out with your girlfriend while the rest of the class is in the rain forest exhibit.

Poision Envy

2010-09-10 07:51:27 pm

Club 27 Goals

Well the choice was either go to school for about an hour, or take an hour long car drive to the aquarium, do random things for two hours, then drive back for another hour.

I enjoyed it, the penguins really liked my little brother for some reason and kept swimming up to him.

Idkbutlike2

2010-09-10 07:52:11 pm

Quote from Opium:

Aquariums are boring? I guess you aren't really fascinated with nature in general?

Yeah, I'm not much of a nature-lover TBH. History and technology are far more interesting to me.

Opium

2010-09-10 07:53:08 pm

There used to be some penguins in an exhibit in Las Vegas. They finally closed it because it was depressing. Penguins are slightly too intelligent to have in captivity without people feeling sorry for them, and dolphins are especially like that. Everything else there seems to think like an insect so who cares.

arkarian

2010-09-10 07:53:45 pm

red chamber dream

to me nature, history, and technology are the same thing. ymmv i guess.

Opium

2010-09-10 07:53:58 pm

Quote from Idkbutlike2:

Quote from Opium:

Aquariums are boring? I guess you aren't really fascinated with nature in general?

Yeah, I'm not much of a nature-lover TBH. History and technology are far more interesting to me.

I would think that you would still like nature since the fascination for both nature and tech are rooted in science. In fact, much tech is inspired by nature.

tomatobob

2010-09-10 07:55:45 pm

One shall stand, one shall ball.

Quote from Idkbutlike2:

I'd rather go to school. Aquariums are boring as hell, even with those cheesy dolphin shows...

Look how wrong this guy is. Just look at it. Awe inspiring.

Nature owns.

Idkbutlike2

2010-09-10 07:57:02 pm

Screw you, you tree-hugging hippie.

DynaDASH

2010-09-10 07:57:25 pm

Chooser of the Slain

The things we use, clothing, TV's, pencils, paper, everything, comes from nature.
Your wallet, your expensive shoes, your amazing wardrobe, comes from nature.