Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon.

Pages: 1-

SICP

Name: Anonymous 2012-04-21 2:46

Hi /prague/
I'm currently reading SICP to learn LISP. The question is, what LISP do I have to install to follow the book ?
Common Lisp ? Scheme ? Another ?

Name: Anonymous 2012-04-21 2:49

Scheme, stupid head.

Name: Anonymous 2012-04-21 2:55

Also, what implementation to use ?

Name: Anonymous 2012-04-21 4:17

>>3
whichever one you find to not be too shitty.

Name: Anonymous 2012-04-21 4:20

mit-scheme

Name: Anonymous 2012-04-21 4:41

I'll go for Chicken.

Name: Anonymous 2012-04-21 6:27

Originally SICP was written with mit-scheme; though you won't be able to follow ALL the examples (the fractal images) I recommend Racket (formerly called PLT Scheme).

Name: Anonymous 2012-04-21 13:55

mit-scheme seems to work well for most of the book aside from a section in ch.2 (the picture language).

Name: Anonymous 2012-04-21 20:15

Be aware the exercise on rewriting if in terms of cond in Chapter 1 will behave differently in racket... puzzled me for ages until I worked it out. I think it's because racket uses lazy evaluation although I could be wrong. Anyone know?? Exercise 1.6 I think

Name: Anonymous 2012-04-21 22:01

>>9
No. Now go help another customer you fat bitch.

Name: Anonymous 2012-04-22 0:42

Name: Anonymous 2012-04-22 10:57

Chicken follows the vocabulary of the Scheme used in SICP.

Racket/DrRacket is the "updated" version which changes a bunch of the vocabulary, so it makes a little more awkward to use.

Chicken compiles when linked to GCC/mingw, or you can run in interpreter. It also supports "egg" software packages, and I'm 99% sure C/C++ linking natively.

Name: Anonymous 2012-04-22 15:04

>>5,8
mit-scheme hasn't been ported to x86-64 yet.

Name: Anonymous 2012-04-22 15:19

Name: Anonymous 2012-04-22 15:29

>>14
Hrm.  I'll check if it's been added to the Debian repos (if it ain't digitally-signed, it's crap) since the Ubuntu devs are too busy jerking off over Unity.

Name: Anonymous 2012-04-22 15:37

Can't you change the dialect of racket? I'm pretty sure that you can do SICP and HTDP compatibility mode.

Name: Anonymous 2012-04-22 15:45

i'm going with Chicken, then.

Name: Anonymous 2012-04-22 17:02

Name: Anonymous 2012-04-22 17:12

>>18
I don't trust GNU.

Name: Anonymous 2012-04-22 17:12

>>18
You must be an Arch user.
Do you trust the GNU FTP server and every point in the network between it and your computer? Downloading sigs from the same server you got the thing being signed, especially over a non-encrypted connection, is only good for integrity checking, not security.

Name: Anonymous 2012-04-22 17:33

>>20
Guys what if our compilers had malicious code to insert malicious code into compilers?!

Name: Anonymous 2012-04-22 17:36

>>20
I think Arch finally got some package signing. And to be honest, I trust nothing and nobody and I don't care too much about my "security" or privacy. I have that mit-scheme x86-64 binary installed and I have not even once checked the sig.

That said, you are perfectly right.

Name: Anonymous 2012-04-22 18:15

>>20
Downloading sigs from the same server you got the thing being signed, especially over a non-encrypted connection, is only good for integrity checking, not security.
Retard can't into PGP.

>>21
Then the mere existence of a untainted compiler would immediately lead to the detection of said malicious code.

>>22
I have that mit-scheme x86-64 binary installed and I have not even once checked the sig.
Nice job, retard.

Name: Anonymous 2012-04-22 21:42

>>21
Why bother changing the compiler when you can target the underlying OS and the CPU microcode?

>>23
>Retard can't into PGP.
It would be trivial for an adversary to alter both the file, the sig, and the public keys as I retrieved them from a remote host. Besides that, why would I trust PGP? There could be a malformed key that evades detection, or the security could have been cracked but as-yet undeclassified technologies.

Name: Anonymous 2012-04-22 21:44

>>24
Errata: I know how to quote, I simply forgot the space after the >. Good forbid Mr.VacBob-chan implement proper editing features. Any redirects to any other place will be done by homosexuals and pathetic losers who are not real hackers and shouldn't be respected anyway.

Name: Anonymous 2012-04-22 22:03

>>24
Why bother changing the compiler when you can target the underlying OS and the CPU microcode?
Why bother reading the source when you can use Microsoft Windows?

It would be trivial for an adversary to alter both the file, the sig, and the public keys as I retrieved them from a remote host.
Web-of-trust, fucktard.  Oh wait, you don't have any friends.

Besides that, why would I trust PGP? There could be a malformed key that evades detection, or the security could have been cracked but as-yet undeclassified technologies.
Use the source, Luke.

>>25
Back to /b/, faggot.

Name: Anonymous 2012-04-22 22:07

>>24
CPU microcode
Tinfoil hat detected.  An attacker needs physical access to a machine in order to change the CPU's microcode.  Why go through the pain of such an elaborate attack when he can just install a relatively cheap radio keylogger onto the keyboard.

Name: Anonymous 2012-04-22 22:29

>>26
Why bother reading the source when you can use Microsoft Windows?
Exploiting the OS is not specific to Windows. For example, SELinux was written by the NSA and is installed on many GNU\Linux distribution installations. Given the talent that the NSA has, it is not inconceivable that they can craft a flaw into the code that is met in only edge (or designed) cases that most hackers wouldn't immediately see.

Web-of-trust, fucktard. 
This does nothing to prevent adversaries who can control the method of communication used to share keys.

Oh wait, you don't have any friends.
I'll have you know that I have many friends.

Use the source, Luke.
There could be many flaws in the mathematics of RSA that we don't know about (yet), or are being suppressed for ''security'' reasons. And even if there wasn't, that would do nothing to stop the brute force attacks using classified computer systems, large criminal botnets, or plain old stealing of the key and modifying files without the knowledge of the owner.

Back to /b/, faggot.
Typical childish behavior.


>>27
Tinfoil hat detected.  An attacker needs physical access to a machine in order to change the CPU's microcode.
Or they can manufacture the processors, be it the company for the government, or the factory modifying it for other means without the designer's knowledge.

Why go through the pain of such an elaborate attack when he can just install a relatively cheap radio keylogger onto the keyboard.
Because there is no method of mass exploitation in that. With the OS or CPU lines compromised, anyone could be targeted on a whim without even being close. Just send a few odd bytes in the .tar during a write operation and there you go.

Name: Anonymous 2012-04-22 23:24

>>28
it is not inconceivable that they can craft a flaw into the code that is met in only edge (or designed) cases that most hackers wouldn't immediately see.
You're really overestimating the intelligence of American pigs.

This does nothing to prevent adversaries who can control the method of communication used to share keys.
Key signing party, faggot.

I'll have you know that I have many friends.
Unfortunately, they're all imaginary.

There could be many flaws in the mathematics of RSA
Unlikely.  And signing uses DSA.  This subthread is about signing.  Also, for encryption, you don't have to use RSA.

brute force attacks using classified computer systems, large criminal botnets
Not if your algorithm and security parameters are worth a shit.

plain old stealing of the key and modifying files without the knowledge of the owner.
LUKS.  Sure, you can fuck with the BIOS' firmware in their absence to install some sort of backdoor, but that's many orders of magnitude more difficult than just installing a hardware keylogger.

Typical childish behavior.
Fuck you you gurgling cockpustule.

Or they can manufacture the processors, be it the company for the government, or the factory modifying it for other means without the designer's knowledge.
Ah yes, the good old ``CPU detects hard-coded 128-bit value on load and activates exploit in microcode''.  Once again, you're overestimating the American pigs' intelligence.  And what's the point of all that when you can just buy some microcontroller and program it to do encryption or whatever activity the Man doesn't like.  Or, hell, build it out of 7400's.

Because there is no method of mass exploitation in that. With the OS or CPU lines compromised, anyone could be targeted on a whim without even being close.
Microsoft Windows and Apple Mac OS X closed-source operating systems are installed onto probably 97% of world's computers, maybe more.  I'd say that as far as mass exploitation goes, they're doing pretty damn good.

Just send a few odd bytes in the .tar during a write operation and there you go.
I'm sure any security vulnerabilities in the tar utility have long been discovered.

Name: Anonymous 2012-04-23 0:53

it is not inconceivable that they can craft a flaw into the code that is met in only edge (or designed) cases that most hackers wouldn't immediately see.

You're really overestimating the intelligence of American pigs.

I don't see why you would think so, but it need not be American anyway. The Chinese and Russian governments could easily want the same, as could more sinister organizations that plan to rule us.

This does nothing to prevent adversaries who can control the method of communication used to share keys.

Key signing party, faggot.

That doesn't save you from bad binaries caused by a compromised system.

I'll have you know that I have many friends.

Unfortunately, they're all imaginary.

Ad hominem. This is not proper argument. I've shown you respect, now I must demand that you remain courdial too.

There could be many flaws in the mathematics of RSA

Unlikely.  And signing uses DSA.  This subthread is about signing.  Also, for encryption, you don't have to use RSA.

I was using RSA as an example. DSA could have unknown (to us) weaknesses just like any other algorithm.

brute force attacks using classified computer systems, large criminal botnets

Not if your algorithm and security parameters are worth a shit.

Perhaps the computers of the advesaries are simply better? We don't know what's going on in those multi-billion dollar\euro research labs. I suspect that they have quantum computers that can make short work of any public-private key cipher systems.

plain old stealing of the key and modifying files without the knowledge of the owner.

LUKS.  Sure, you can fuck with the BIOS' firmware in their absence to install some sort of backdoor, but that's many orders of magnitude more difficult than just installing a hardware keylogger.

Fine, install the keylogger. It doesn't refute the fact that your system is vunerable.

Typical childish behavior.

Fuck you you gurgling cockpustule.

This shouldn't even be posted.

Or they can manufacture the processors, be it the company for the government, or the factory modifying it for other means without the designer's knowledge.

Ah yes, the good old ``CPU detects hard-coded 128-bit value on load and activates exploit in microcode''.  Once again, you're overestimating the American pigs' intelligence.  And what's the point of all that when you can just buy some microcontroller and program it to do encryption or whatever activity the Man doesn't like.  Or, hell, build it out of 7400's.

Once again, you are not refuting my point, you are just hand-waiving. It could be anybody.

Because there is no method of mass exploitation in that. With the OS or CPU lines compromised, anyone could be targeted on a whim without even being close.

Microsoft Windows and Apple Mac OS X closed-source operating systems are installed onto probably 97% of world's computers, maybe more.  I'd say that as far as mass exploitation goes, they're doing pretty damn good.

But *nix run on all the interesting systems. Only neurotypicals use those OSes, and are the type to get virii from porn because they are completely oblivious about security. They would never even bother with encryption. Anyone who will ever be bothered is running a *nix OS.

Just send a few odd bytes in the .tar during a write operation and there you go.

I'm sure any security vulnerabilities in the tar utility have long been discovered.

That is why they would alter tar. No one would suspect that tar would have a back door that would return the wrong data. It is your blind trust that will destroy you.

Name: Anonymous 2012-04-23 2:02

>>30
I don't see why you would think so, but it need not be American anyway. The Chinese and Russian governments could easily want the same, as could more sinister organizations that plan to rule us.
Yay, more enemies.  Sounds exciting!

That doesn't save you from bad binaries caused by a compromised system.
It's trivial to devise a method to cross-check whether another system's compiler has been compromised (whether binaries match with the source).  Or do you think that every single system with every single C compiler, going back to the 90's, is compromised? 

Ad hominem. This is not proper argument. I've shown you respect, now I must demand that you remain courdial too.
Alright, fine, I give you that.

I was using RSA as an example. DSA could have unknown (to us) weaknesses just like any other algorithm.
That's true.  It would be nice if GPG incorporated some more algorithms and if *nix distributions used multiple algorithms for signatures.

Perhaps the computers of the advesaries are simply better?
2+2=4 regardless of how much money you pump in changing it (at least with anything short of ,,1984'').

We don't know what's going on in those multi-billion dollar\euro research labs.
It's probably a paranoid arms race.  I doubt they managed to crack anything useful, but they're feareful that the others did so they constantly invent new algorithms to "thwart" them.

I suspect that they have quantum computers that can make short work of any public-private key cipher systems.
Only for factorization.  There are public key cipher systems that aren't even related to factorization or the DH assumption for that matter.  Too bad people don't use them.

Fine, install the keylogger. It doesn't refute the fact that your system is vunerable.
I would argue that, given the measures I've taken, my system is almost impossible to compromise without my knowledge.

This shouldn't even be posted.
Take a joke.

Once again, you are not refuting my point, you are just hand-waiving. It could be anybody.
It can't.  There are few points at which someone actually has an occasion to insert spyware onto a computer.  It's very difficult to accomplish (i.e. requires lots of money) and it's useless if the person makes their own computer out of scratch.

Only neurotypicals use those OSes, and are the type to get virii from porn because they are completely oblivious about security. They would never even bother with encryption.
So the only ones who do run secure operating systems are shy pale-skinned nerds.  How exactly are shy pale-skinned nerds dangerous, threatening or even relevant to the real-world criminal activities of secret services?  Sure, they can maintain public certain unconfortable information (e.g. Wikileaks), but that's a mere annoyance; I'm certain that a lot of entries on Wikileaks are really supplied by the secret service of one country that wants to tarnish the image of another.  People relevant to secret services are statistially very likely to fall in the ,,type to get virii from porn''.

That is why they would alter tar. No one would suspect that tar would have a back door that would return the wrong data. It is your blind trust that will destroy you.
Bah, rootkit-hiding 101.  I can still sleep at night because I know that some other shy pale-skinned nerd already has a testing system that periodically checks whether the binaries still match the source.  In case you're about to ask how he'd know his testing system hasn't been compromised, well, I'm going to have to ask you to use your brain a little.

Name: Anonymous 2012-04-23 10:19

>>27
If you use Intel chips, the dirty Jews in Haifa, Israel can put whatever anti-Goyim spyware they want in there and push it as an automatic update. Their microcode is encrypted and closed-source.

Name: Anonymous 2012-04-23 10:49

>>32
automatic update
sure, if you're using windows.

Name: bampu pantsu 2012-05-29 4:33

bampu pantsu

Don't change these.
Name: Email:
Entire Thread Thread List