- cypherpunks - lists.cpunks.org

writing code and mails at the same time! woohooo - one-way encryption
by stef 20 Feb '15

20 Feb '15

apparently on mailing lists it's common to write mails, even cypherpunks do that instead of code. allow me to combine the two things. i thought it trough, and figured, that with openssl i have no AD crypto, with gpg i have this MDC thingy and the rfc4880 strongselector beacon strongselector, and with nacl and it's key derivation using ecdh make them all unsuitable on first sight for the use-case. re my 7 rules, i think i need to work on the threat model to have 100% score pls enjoy (also as contributions to the copyleft/permissive licensing thread) and comment: #!/usr/bin/env python # # implements simple one-way encryption pipe using rsa and keccak-based # spongewrap # # useful at least in the following use-case: you have an untrusted # host on which plaintext data arrives, which you want to encrypt # before it is forwarded in a hostile environment to the final # recipient holding a private key in a safe location. In this one-way # setting the recipient is never talking to the host doing the # encryption. # # Example: take photos in a hostile situation, encrypt the the photos # and being unable to recover them until arrival in the save location # with the the private key. (note, this does not protect against # forensics!) # # crypto: a random 32 byte message key is encrypted with the public # key of the recipient in oaep padded RSA, then this message key is # fed into SpongeWrap, which is then used to authenticated encrypt the # message. # # output format: # 2 bytes - the length of the RSA encrypted message key # n bytes - the RSA encrypted message key # m bytes - the encrypted message # 16 bytes - the "MAC" # # depends: `pip install m2crypto spongeshaker SecureString` # # create keys using openssl: # `openssl genrsa -out my.key 4096` # `openssl rsa -in my.key -pubout >> my.pub` # `cat my.key my.pub >>my.pem` # `srm -fll my.key` # # deploy my.pub on the encrypting host, secure my.pem in a safe # location for decryption. # # test with: # # for i in {0..42} {8170..8210} 1000000; do # echo -ne "\r$i " # dd if=/dev/zero bs=$i count=1 2>/dev/null | # ./ondir.py e my.pub | # ./ondir.py d my.pem >/dev/null || # break # done # # (C) 2015 by Stefan Marsiske, <s(a)ctrlc.hu>, GPLv3 import M2Crypto as m2c from spongeshaker.spongewrap import SpongeWrap from SecureString import clearmem import sys, struct TAGLEN = 16 BUFLEN = 8192 def encrypt(to): # load recipient pk key = m2c.RSA.load_pub_key(to) # gen message key mkey = m2c.Rand.rand_bytes(32) # encrypt message key cmkey = key.public_encrypt(mkey, m2c.RSA.pkcs1_oaep_padding) # output message key sys.stdout.write(struct.pack("H", len(cmkey))) sys.stdout.write(cmkey) # encrypt message ctx = SpongeWrap(1536) # with mkey ctx.add_header(mkey) # mkey not needed anymore clearmem(mkey) # buffered encrypt of stdin to stdout while 1: buf = sys.stdin.read(BUFLEN) if not buf: break sys.stdout.write(ctx.encrypt_body(buf)) # calculate tag tag=ctx.digest(TAGLEN) # output tag sys.stdout.write(tag) def decrypt(to): # load recipient pk key = m2c.RSA.load_key(to) # read msg key klen = struct.unpack('H', sys.stdin.read(2))[0] if klen>1024: print >>sys.stderr, "probably corrupt file" sys.exit(1) cmkey = sys.stdin.read(klen) # decrypt message key try: mkey = key.private_decrypt(cmkey, m2c.RSA.pkcs1_oaep_padding) except: # twarth timing attacks mkey = 'couldntdecryptkey' # todo clear private RSA key from memory # decrypt with mkey ctx = SpongeWrap(1536) ctx.add_header(mkey) # mkey not needed anymore clearmem(mkey) zero = True # to detect empty files rest = '' # buffered reading of stdin, since we need to catch the last # bytes for the tag, we always retain the last n bytes for # this purpose in rest while 1: buf = sys.stdin.read(BUFLEN) if zero and len(buf)>TAGLEN: zero=False tag=buf[-TAGLEN:] if len(buf)>TAGLEN: # prepend the retained last bytes to the next decrypt, if # there's enough more bytes read sys.stdout.write(ctx.decrypt_body(rest+buf[:-TAGLEN])) elif len(buf)>0: if len(tag)==TAGLEN: # we have exactly the tag read in buf, decrypt the # rest sys.stdout.write(ctx.decrypt_body(rest)) else: # truncate the last bytes, as we have a boundary # spanning tag sys.stdout.write(ctx.decrypt_body(rest[:-TAGLEN+len(tag)])) if len(buf)<BUFLEN: # this is the last buffer if len(tag)<TAGLEN: # we have a tag spanning a buffer boundary, we must # patch it up tag = ''.join((rest, buf[:-TAGLEN], tag))[-TAGLEN:] # verify if tag is valid if not zero and tag!=ctx.digest(TAGLEN): print >>sys.stderr, "couldn't decrypt message" sys.exit(1) break rest=tag if __name__ == '__main__': if sys.argv[1]=='e': encrypt(sys.argv[2]) elif sys.argv[1]=='d': decrypt(sys.argv[2]) else: print "usage: %s <e|d> <pub|key>" -- otr fp: https://www.ctrlc.hu/~stef/otr.txt

1 0

List Administrivia
by grarpamp 19 Feb '15

19 Feb '15

Can you add the To: and Cc: headers to the list of headers that eventually end up in the 'Downloadable version' of the archives linked below. Along with the already included Messsage-ID, In-Reply-To, and References... that should make those versions more fully threadable, sortable and searchable by MUA's and provide more understandable participant context. Thanks. https://cpunks.org/pipermail/cypherpunks/

2 5

LEA should not amass giant databases of genetic material they find lying around
by grarpamp 19 Feb '15

19 Feb '15

https://www.eff.org/press/releases/eff-supreme-court-fourth-amendment-cover… https://www.youtube.com/results?search_query=gattaca

1 0

Obama believes in strong encryption
by grarpamp 19 Feb '15

19 Feb '15

http://recode.net/2015/02/13/obama-theres-no-scenario-in-which-we-dont-want… https://www.youtube.com/watch?v=yaylQmnXztU http://it.slashdot.org/story/15/02/17/2230257/obama-says-hes-a-strong-belie… Contrast with many recent US administration and agency statements all on the record as well. Is this a mindshift, or a case of FUD then false savior?

3 2

Disastrous privacy consequences of Canada's terrorism bill
by grarpamp 19 Feb '15

19 Feb '15

http://yro.slashdot.org/story/15/02/19/1633223/the-disastrous-privacy-conse… http://www.michaelgeist.ca/2015/02/total-information-awareness-disastrous-p… http://www.parl.gc.ca/HousePublications/Publication.aspx?Language=E&Mode=1&…

2 1

[liberationtech] At Stanford on Friday, Feb. 20 -- Vitalik Buterin: Ethereum & Cryptoeconomic Mechanisms & Security
by Eugen Leitl 19 Feb '15

19 Feb '15

----- Forwarded message from Yosem Companys <companys(a)stanford.edu> ----- Date: Mon, 16 Feb 2015 12:27:19 -0800 From: Yosem Companys <companys(a)stanford.edu> To: Liberationtech Events <liberationtech-events(a)lists.stanford.edu>, Liberation Technologies <liberationtech(a)mailman.stanford.edu> Subject: [liberationtech] At Stanford on Friday, Feb. 20 -- Vitalik Buterin: Ethereum & Cryptoeconomic Mechanisms & Security Message-ID: <CANhci9E+pmyHwS_ya7giPFXuzbzYQfGWr9rcnHCWuCjsA65sBw(a)mail.gmail.com> Reply-To: liberationtech <liberationtech(a)lists.stanford.edu> From: David Wu <dwu4(a)cs.stanford.edu> via security-seminar(a)lists.stanford.edu at https://mailman.stanford.edu/mailman/listinfo/security-seminar Introduction to Ethereum, and Cryptoeconomic Mechanisms and Security Vitalik Buterin Friday, February 20, 2015 Talk at 4:15pm Gates 463 Abstract: The presentation will provide an introduction to the Ethereum protocol from a technical perspective, including the high-level operation of the blockchain protocol, contract calling and gas mechanics, use of Patricia Merkle trees and other cryptography and light client proofs. I will then discuss some of the interplay between economics/game theory and cryptography both in consensus algorithm design (eg. proof of work, proof of stake) as well higher-level protocols like sidechains, challenge-response protocols and SchellingCoin, and will attempt to define and dissect different interpretations of the concept of "cryptoeconomic assurance". -- Liberationtech is public & archives are searchable on Google. Violations of list guidelines will get you moderated: https://mailman.stanford.edu/mailman/listinfo/liberationtech. Unsubscribe, change to digest, or change password by emailing moderator at companys(a)stanford.edu. ----- End forwarded message -----

1 0

Twitter v. DOJ-FBI, Google to FBI
by grarpamp 19 Feb '15

19 Feb '15

https://freedom.press/blog/2015/02/our-brief-support-twitters-lawsuit-again… https://gigaom.com/2015/02/18/tech-and-media-firms-join-twitter-in-key-test… http://www.nationaljournal.com/tech/google-calls-fbi-s-plan-to-expand-hacki…

1 0

Re: [Cryptography] trojans in the firmware
by grarpamp 18 Feb '15

18 Feb '15

On Wed, Feb 18, 2015 at 5:16 PM, Tom Mitchell <mitch(a)niftyegg.com> wrote: > The critical stage is the boot ROM (BIOS) and the boot device. > Once Linux has booted a lot is possible but too much has already taken > place. > A BIOS that allows booting from a Flash memory card must be trusted. > > Virtual machines may help or hinder. > > The VM is sitting where the man in the middle wants to be and if it wants > can protect or expose > the OSs that it hosts. A VM can protect a hard drive from being infected > by blocking vendor > codes that might try to update or corrupt modern disks of boot flash memory. Afaik, all vm's today simply pass through all drive commands. It seems a move all the BSD's and Linux could make today, without waiting on untrustable hardware vendors to roll out signature verification in hardware, is to simply kernel block all commands unnecessary to actual production use of the disk. Permit only from a list of READ, WRITE, ERASE, INQ, TUR, RST, and so on. Thus every other command component, including firmware update, vendor specific, and binary fuzzing, gets dropped and logged. It could be done as a securelevel, or compiled in. It's definitely not bulletproof, but it does force adversaries to add that much more exploit code and effort to get root and go around the driver interface to access the hardware directly. Defense in depth. Similar tactics could be applied to other areas where firmware and vendor/fuzzable opcodes are involved... usb, bios and cpu.

1 0

Fwd: [tor-talk] Call for Participation – ACM SigComm2015 – Workshop on Ethics in Networked Systems Research
by grarpamp 18 Feb '15

18 Feb '15

---------- Forwarded message ---------- From: Bendert Zevenbergen <bendert.zevenbergen(a)oii.ox.ac.uk> Date: Wed, Feb 18, 2015 at 6:20 AM Subject: [tor-talk] Call for Participation – ACM SigComm2015 – Workshop on Ethics in Networked Systems Research Call for Participation – ACM SigComm2015 – Workshop on Ethics in Networked Systems Research Co-located with ACM SIGCOMM’15<http://conferences.sigcomm.org/sigcomm/2015/> August 21st 2015 London, UK A full day workshop titled “Ethics in Networked Systems Research” will be held at the ACM SIGCOMM 2015 in London. This multidisciplinary workshop will bring together two distinct groups of researchers: - Computer scientists, network scientists and other technical researchers who are interested in the ethical and legal aspects of their work; - Researchers studying the various ethical, social scientific and legal aspects of data-driven projects in the field of computer and data communication networks. This workshop seeks 1 to 4 page summaries, focussing on ethical considerations of papers, publications and projects from the two disciplines, for example: - Technical research in the field of computer and data communication networks that either operates in an ethical grey zone, collects and processes personal data/personal identifiable information, or has been rejected from another venue on ethical grounds; - Ethical, social scientific or legal research that reflects on – or aims to guide – technical research and projects in the field of computer and data communication networks, especially an analysis to minimise the potential harm whilst enabling a broad range of Internet research to be conducted. Selected authors and invited speakers will present their work, which will be followed by a structured discussion. The workshop will also facilitate an interactive session in which participants will split into multidisciplinary groups and address emerging ethical dilemmas in Internet measurement and information controls research, partly based on the submitted summaries. This session will be informed by a website (currently under development), that presents ethical guidelines for the fields of Internet measurement and information control. Papers submitted to this workshop will not be archived in the formal sense, so authors can submit papers elsewhere, or submit summaries of previously accepted papers. Important Dates Paper submission deadline – March 31st, 2015 Paper acceptance notification – April 30th, 2015 Organisers Prof. Ian Brown – Oxford Internet Institute Dr. Joss Wright – Oxford Internet Institute Bendert Zevenbergen – Oxford Internet Institute Erin Kenneally – University of California San Diego, Center for Applied Internet Research & Elchemy, Inc. Dr. Malavika Jayaram – Berkman Center for Internet & Society, Harvard + Centre for Internet & Society, India Allen Gunn – Aspiration Tech Meredith Whittaker – Measurement Lab Christopher Wilson – the engine room Stuart Schechter – Microsoft Research Please email papers to bendert.zevenbergen(a)oii.ox.ac.uk<mailto:bendert.zevenbergen@oii.ox.ac.uk> Workshop website: http://conferences.sigcomm.org/sigcomm/2015/netethics.php -- https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk

1 0

Extracting Equation Group's malware from hard drives
by Alfie John 18 Feb '15

18 Feb '15

Hi lists, Does anyone know of any tools to extract the Equation Group's malware from hard drive firmware? Also, are there any public registries online to report and view infections? Alfie -- Alfie John alfiej(a)fastmail.fm

5 5