October 2013 - cypherpunks - lists.cpunks.org

Re: In Over My Head -- What do I need to setup a tiny ISP?
by Miles Fidelman 20 Oct '13

20 Oct '13

Notify Me wrote: > > That being said, some people who have a lot more faith in my abilities than > I do seem to think I am the go-to guy for network information. And they > foolishly asked me for assistance in putting together a small ISP network > which is supposed to cater for home users inside of a residential area, > very likely wireless (wifi/WIMAX). > A key question: is this going to be a commercial service run as a business, or an internal operation like a campus network? The operations aren't that different, but the scope and organiation can be different. Another key question: Are you focusing just on the infrastructure (wireless IP) or will you be providing email, hosting services, and so forth? > I confess I am really interested in helping my questioners put this > together, not just for whatever material gain (which is unlikely at this > point), but just for the experience which is very valuable to me. I also > have to state that I live in Nigeria, so whatever advice you offer has to > be fourth-world applicable. > > I admit that your query raised my interest, so I did a little looking, and found this: http://www.fnbc.info/sites/default/files/fck-uploads/file/fntc/ISP%20Guide%… A guide put together by the "First Nations Technology Council" - on how to put together an ISP on tribal lands in Canada - which might be somewhat analogous to your situation. It's somewhat basic, but gives an overview - then you can start looking into the individual topics it talks about. You might drop them an email. There's also been a lot of work on US tribal reservations that might apply - you'll have to do some googling to find resources. Now the guy you really want to talk to is Dave Hughes, of Big Sky Telegraph - who used to do a lot of network building for rural areas in the US and remote areas around the world - unfortunately he's long retired - but two sources you could look at that might give you a lot of background: http://www.bigskytelegraph.com/ (which now seems to lead to the Dillon Center of Excellence, focusing on best practices for rural America, including networking) and http://www.davehugheslegacy.net/ There used to be an "Association for Community Networking" - folks involved in building cooperative local networks (used to be a lot of what I was involved in the 1990s) but that kind of activity has dried up as more and more commercial players entered the market. And an "Organization for Community Networks" (http://ofcn.org/) - but their web site seems to be kind of dead (some useful links, though). If you google "community network" you'll find a buch of current operations - and you might want to contact some of them. There are also a growing number of community mesh networks in various places - WikiPedia's page on http://en.wikipedia.org/wiki/Wireless_community_network might be a good starting point. There's also an association of wireless ISPs http://www.wispa.org/ Then there's this guy: https://www.facebook.com/theispguy - who seems to have written a book on "so you want to be an ISP" - can't say anything about how good it is. Or.. you might just go looking for other ISPs in Nigeria (of which there are quite a few) and see if you can get some advice (amazing what you can learn over a few beers). Maybe contact the networking people at the nearest college or university. I don't know what it's like in Nigeria, but in the US, the academic community was really active in getting a lot of early ISPs started - particularly in smaller communities. One other thought: there are quite a few software packages floating around for managing aspects of ISP operations (see https://wiki.debian.org/HostingControlPanels for starters) - I expect their user communities (and email lists) would be excellent sources of information. Hope this helps. Regards, Miles Fidelman -- In theory, there is no difference between theory and practice. In practice, there is. .... Yogi Berra

1 0

please ignore: this is only a test...
by brian carroll 20 Oct '13

20 Oct '13

(dealing with list formatting mechanics and rendering issues) // no other way to do this, apologies... meant for follow-on post // preferably want to embed text array though line-wrap issues. // potential that quoted test can surpass limit though need to test it // outside of a post to not degrade readability.... quoted, line-wrap (y/n) (another reference test) - 78 character monospaced Courier New font, > quoted to test line wrap > > XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX > XX XX > gmail indented, line-wrap (y/n) (another reference test) - 78 character monospaced Courier New font, quoted to test line wrap XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX gmail double quoted line-wrap (y/n) (another reference test) - 78 character monospaced Courier New font, >> quoted to test line wrap >> >> XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX >> XX XX >> > indented then quoted, line-wrap (y/n) (another reference test) - 78 character monospaced Courier New font, > quoted to test line wrap > > XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX > XX XX > {whitespace}

15 20

[OT, Kind of] Carrier Retention Windows
by J.A. Terranson 20 Oct '13

20 Oct '13

This is also on Cryptome, I believe, but this one is more accessible, so... Handy chart of cell phone company records retention periods for 6 major carriers, obtained by the ACLU from a 2010 DOJ list: https://www.aclu.org/cell-phone-location-tracking-request-response-cell-pho… //Alif -- Those who make peaceful change impossible, make violent revolution inevitable. An American Spring is coming: one way or another.

1 0

Re: [Richard Crisp] Cheney afraid of terrist h4x0rs
by shelley＠misanthropia.info 19 Oct '13

19 Oct '13

If you're going to send bullshit comments off-list, don't expect me to jump through hoops to reply (see below.) To answer your remark, no I do not support "Ovomit" as you call him, nor any of the other treasonous fucks running this crazy train off the rails.  (Nor Ron Paul, etc, to answer your inevitable retort.) -S From: Richard Crisp <rdcrisp(a)earthlink.net> Date: Oct 19, 2013 4:20 PM Subject: Re: Cheney afraid of terrist h4x0rs To: <shelley(a)misanthropia.info> Let me guess: you support Ovomit  From: rdcrisp(a)earthlink.net Date: Oct 19, 2013 4:24 PM Subject: Re: Re: Cheney afraid of terrist h4x0rs To: shelley(a)misanthropia.info I apologize for this automatic reply to your email. To control spam, I now allow incoming messages only from senders I have approved beforehand. If you would like to be added to my list of approved senders, please fill out the short request form (see link below). Once I approve you, I will receive your original message in my inbox. You do not need to resend your message. I apologize for this one-time inconvenience. Click the link below to fill out the request: https://webmail.pas.earthlink.net/wam/addme?a=rdcrisp@earthlink.net&id=11e3…

1 0

Re: Snowden sets OPSEC record straight
by John Young 19 Oct '13

19 Oct '13

Snowden filtered by Janes Risen filtered by New York Times, as with all other filterings by special-interested Snowden filters, does deliver a reassuring message to precisely answer highly filtered questions and charges that have been made about his heavily filtered, nay, almost negligible releases. Perhaps Snowden actually speaks in person, appears in a filtered documentary, in a WikiLeaks filtered video, emails encryptically! via Greenwald and Risen, like a character from a brain-washing film, formulaicly, always on point, no ifs and buts, no waverings from propaganda-like rationales and justifications. But so far every version of his apperances by whatever medium come across as carefully vetted by WikiLeak's Sarah Harrison, lawyers and PR wizards, as over-edited, sanitized and directed, as little substantiated despite alleged thousands of documents as hard driven advocacy so like reviled special-interested groups of all persuasions. This critique applies to the dog and pony shows now ranging the globe by freedom of information special interests, staffed by a coterie of so-called ex-spies, their lawyers, their PR wizards, national security journalists and civil liberties exploiters, bedded and applauded by camp followers, funded by egotically, narcissisticly bored and wealth-drunk patrons like Omidyar, Soros, and, why not j'accuse the combine, the USG and its filthy rich spies long known to use the wealthy to protect their common special interests since corruption was invented, Day One. Snowden is either an ignorant fool, seduced along with Manning, by the narcotic dispensed by govs, NGOs, spies and the wealthy "to do good" or a fabulous con artist like Assange to "fleece the fuckers" as he emailed January 2007, via leaks extortion, that is, adopt the strategy and tactics of the "enemy" to seduce others into their concentrated wealth carnal palace. At 01:08 AM 10/18/2013, you wrote: >On Thu, Oct 17, 2013 at 9:42 PM, Juan Garofalo <juan.g71(a)gmail.com> wrote: > >... > > He's still loyal to the american nazis. That's praiseworthy? > > >he is showing extraordinary discretion and restraint, and an >underlying approval for some fundamental american ideals, like >guaranteed right to individual privacy, though of course this should >extend to the world not just US citizens. > >let's not forget that Guardian was forced to move this reporting to >the USA, because of GCHQ/her majesty's interference. and of course >there are plentiful locales where the activities of anyone involved >would strongly encourage a less than discrete "summary lack of >continued living incident" of some obscure manner. > >at any rate, this is too long and detailed a conversation for today.

3 2

Re: HTML List Abuse (was: "please ignore: this is only a test")
by brian carroll 19 Oct '13

19 Oct '13

i somewhat understand the situation thanks to feedback... the problems here involve a broken pc and porting texts from one platform into Gmail, where formatting errors and line wrap issues exist, thus unreliable rendering of emails. since the list does not send me a copy of my own posts, the archives in plain text are what i see as to the filetype, not realizing it may be rendered as HTML email elsewhere, because data only appears in plain text anywhere i see it this whole time i never figured it was about my posts due to this, unless somehow MINE attachments were what was referenced - yet only HTML was repeatedly scolded. in approximately two months, looking at emails, i have only seen a few examples of HTML email that i recognize, mainly because of formatted or colored text, and did not connect this with the *scrubbing* of the HTML content, associated with the way the email is being processed by the list management software -- which i thought is neutralizing the HTML formatting before it goes out thus, if there is a setting to receive plain text, why not subscribe to that option, for that situation, i do not yet comprehend what is going on as these posts render in mailboxes elsewhere, if they are indeed HTML or it just involves a footer with links to the scrubbed html and associated attachments, aka multipart MIME (?) i tested plain text resend (locally) of the last file and it renders Unicode ok, so will try that approach though mystery formatting issues persist with Gmail (and, as an indication of deterioration of usability, they recently removed 'translate' from Google search, so downgrading functionality in email also seems likely, with formatting and limited reply or other control over email content)... that said, i do not know (for lack of getting a copy of emails myself) if graphics are sent out as attachments or scrubbed and archived on the list (which i assumed, given the footer attachment data). the reason for not direct linking everything is to keep a copy in archive, in case i delete drop-box, so the essays have basic content for a few ideas requiring of some diagrams. if that multipart MIME attachment is the issue, is it not possible to somehow 'not view attachments' if they are sent out, or filter such messages, versus stop visual messaging of relevance to list content, because someone somewhere does not want that, and it is a potential security threat if 'weaponized', in that it is a way of attacking through delivery, etc what is a forum without ideas, how much security to protect what, security talk about security tech yet no conceptualization beyond those boundaries, like what is crypto, what are the concepts, where are various peoples understandings and interests, and might some visual information help assist the conceptual issues, or is it text-only, no imagery. because then it is just not worth it, trying to share new views and ideas about issues of crypto today not everything fits into text or a 77 character line width. these tools suck for communicating and sharing ideas. it is self-punishing even to try to share ideas online. everything goes against viewpoints that are not already agreed upon as part of the existing perspective. it takes and requires a significant effort to build up an alternative model and this requires some additional visual data to make points or frameworks otherwise not possible in text alone, and things are beyond Lynx interfaces as an email exchange scenario, to limit exchange to those parameters is perhaps too limiting even, unless what is said here is supposed to be secure in some way that i do not comprehend, like this data is under attack maybe i just do not get it, what the issues are. are online forums not to communicate ideas and while mobile and other platforms exist, there are many examples in code of how data is ported to those interfaces, not determined by it, because it is a different set of requirements, ex: m.mobile.tld 1) will try to send plain text, if formatting is retained, this will include a few attachments for the list archive 2) MIME-footer scenario will continue. unknown if these images go out to everyone- my assumption is that they went directly to the list archive: scrubbed with HTML... 2a) if attachments go out to list members, is there is a way to choose not to receive them, via list software, 2b) if images go straight to archive is there any issue 3) is there disagreement on attachments. no images and i am gone, cannot communicate the ideas. i can Dropbox them-- never meant to impose the imagery on every mailbox, especially if a security threat. yet it seems more secure to email than Dropbox them, potentially, if somehow the file was remote altered while stored static. assumed a central list archive was holding imagery via the scrubbing process. 4) what would resolve the issues... i do not see how to avoid multipart MIME footers with image attachments, yet perhaps however many list member should have to goto an URL to retrieve attached image content than have it arrive with the email. i am a believer in the text + image approach for emails, i think it is too low a limit on communication to take out visuals in basic exchange because much more can be communicated w/ diagrams. unfortunately everything is against this, even in-line images are no longer an option in email and webmail software... thus forcing text-only, sign-based, linearity like this feedback? p.s. used to ASCII diagram extensively yet no longer have tools able to control monospace fonts this way. any pointers to online resources for creating technical diagrams using ascii for email communications and the settings needed greatly appreciated. saw this online documentation the other day and its formatting is ideal, http://tools.ietf.org/html/draft-hallambaker-prismproof-trust-00 why is it so impossible to get that kind of control over text, and how do technical diagrams get rendered in ASCII (no examples here, rare to encounter them) and where to learn how to do this. not online ASCII-image translators and instead, are there tools and settings for text editors or does Gmail limit this from happening. tried Outlook yet it is insane and unusable. cannot find the new mail button even. fake whois. sick software experimentation. Eugen Leitl wrote: > > On Thu, Oct 17, 2013 at 08:43:01PM -0500, brian carroll wrote: > > hello. as far as i know, this is not HTML, it is a unicode message. > > It was a multipart message: > > I 1 <no description> [multipa/alternativ, 7bit, 3.9K] > I 2 ├─><no description> [text/plain, 7bit, utf-8, 1.2K] > I 3 └─><no description> [text/html, quoted, utf-8, 2.5K]

6 6

Pa: da: d-a#2e !8upje. In.z :$ qu# e ta2t:_
by brian carroll 19 Oct '13

19 Oct '13

Moon Jones -- (difficult to take you seriously while you're wearing a tutu and holding a pink umbrella) YES. i worship my mother. this is your problem? > On 1L.J.B:AC W!Y, dj!eu nejj8= mrota drab. > H8! Wud ncxa! 3owme's poh ye$ 8jdmu up. Z8E da$ eu H- 2o3E#H!z6 > 56wemyaRE. 8a!2 58o zwej+ #3 uec 8a| 6 6E8!-:88 888-uTi Vhq !t+12 ^9) > 6920x25ji 8-=5 Nep oo& ^ nutS!E nz5ah gAme |!k 3|3ctj0w^7u520925. > =W !c +earZ. > Lyeu11 21 d b-7pi.ng cS 8= 1415 .ty1+ V8N #j5 8-7pi19u B y1^a 1 p!6 > jutcj5 2q 28 3#ka h0ur WoE31 bjond, wot Ly8a H3i c|d 8ec. 8a 2m-15 n+2 > 25^15135 iC 1icohoUn' se _L! v28-i0 #1115 50nu3 J 7-/7E N'n:uc- 8j +8 2m^= > -L9mm 6n'45; 8avi V8- na14 /0d 2e-w-on mrk D 68915 !cu: S060#H \N8-id > 7_8-0_L!3e' 9'E sn?\5 > 231v/ 5 68 8#|1Ui B-=nn-eV/: > "Now my ma filter has an extra #ss#### on the /dev/null list." >

1 0

[Cryptography] PRNG WYTM
by John Denker 19 Oct '13

19 Oct '13

On 10/19/2013 09:55 AM, Sandy Harris wrote: > Mixing in the clock makes a machine behave a bit > differently each time it is rebooted. Again, there > are better fixes such as mixing in a saved file, but > again this is still worth doing. > > These are reasonably cheap and done only once > at boot time. They can do no harm and are useful > in at least some cases, so worth doing. OK, the next step in any such discussion is to ask the famous question, What's Your Threat Model (WYTM). Several different reasonable answers are possible. 1) At one extreme, we have the "no threat at all" model, aka the "non-adversarial" model. Examples include *) Doing a Monte Carlo integral in the context of a molecular dynamics calculation. The molecules are not going to attack our PRNG. They are not going to cryptanalyze it. Almost any PRNG is "random enough" for this purpose. *) Cooperative situations, such as friendly computers on a LAN, doing random exponential backoff as part of the layer-1 Ethernet CSMA/CD. Everybody has a shared interest in implementing the protocol properly, so even if they could break the PRNG they wouldn't want to. *) Et cetera. There are tons of examples in this category. A PRNG in category (1) could be considered "random" but not "secure". It is usually adequate to seed this type of PRNG with things like the MAC address, serial number, and time-of-day. 2) At the opposite extreme we have high-stakes adversarial applications, including military cryptography, banking, other high-value business communications, high-stakes gaming, etc. etc. etc. A PRNG in this category needs to be *secure* against a wide range of threats. For tasks in this category, seeding the PRNG with things like the MAC address, serial number, and time-of-day is nowhere near good enough. It is a band-aid or worse. It is security theater. It gives you "randomness" in some weak sense, but it does not give you security. The typical modern PRNG in this category consists of a seed, a counter, a hash function, and a reseeding mechanism. Sometimes there is a block cipher in there somewhere, but to a sufficient approximation this is the same basic architecture, just with a fancier hash function. So let's look in more detail at the threats against such a PRNG. *) For starters we have the threat of direct cryptanalysis of the output. If the preimage can be found, all further outputs will be known to the attacker, and probably all past outputs as well, over the span bounded by the nearest past and present reseedings. The feasibility of finding a preimage depends on the number of bits output by the PRNG. Therefore there should be a limit on the number of output bits between reseedings. *) Another type of threat is more indirect. For example, suppose the PRNG was seeded at boot time from the saved random-seed file. It may be possible for the attacker to find this, perhaps by sneaking a peek at an old backup tape or whatever. Such a threat is independent of the number of bits emitted by the PRNG. It is hard to say what it /does/ depend on, but in the absence of anything better, wall-clock time is a plausible proxy. Therefore there should be a time limit on how long a seed file is allowed to remain on disk before it is regenerated, and a time limit between reseedings of the PRNG. On 10/19/2013 01:37 PM, James A. Donald wrote: > > Any system that needs crypto, communicates. Any system that > communicates, sees events whose details are difficult to predict for > anyone not in physical possession of the system. > > Solution: Block for a short period after startup. Possibly a small > number of systems will freeze up and fail to boot. This is almost > always fixable by moving the blocking process in the bootup so that > it no longer blocks other processes while it is blocked waiting for > /dev/urandom, while /dev/urandom is blocked waiting for entropy. That might be a "solution" in certain favorable cases, but it is nowhere near being a reliable, general solution. I can think of five failure modes in five minutes. Perhaps the most obvious is this: Suppose my system is sitting in a rack at some colocation provider. All the attacker needs to do is rent a box in the same rack, on the same LAN segment. Then he knows my MAC address, the time at which I booted up, and (to a good approximation) the arrival time of every network packet addressed to me. Similarly for my laptop on the corporate wifi network. The bad guy in the loft across the street has a nontrivial chance of figuring out everything he needs to know about my network traffic. If anybody has a proof that this cannot happen, please explain. Here's the only thing that has ever made sense to me: a) Any device that wants to have any security whatsoever needs to be able to store a seed, even when powered off. b) The seed needs to be provisioned on a per-device basis, much like the MAC address is provisioned. c) The seed needs to be big enough, randomly-chosen, and secret, very unlike the MAC address, RTC, and device serial number. _______________________________________________ The cryptography mailing list cryptography(a)metzdowd.com http://www.metzdowd.com/mailman/listinfo/cryptography

1 0

Re: [Cryptography] [RNG] on RNGs, VM state, rollback, etc.
by Theodore Ts'o 19 Oct '13

19 Oct '13

On Sat, Oct 19, 2013 at 09:33:32AM -0400, John Kelsey wrote: > > One other thing you can see from the duplicate RSA keys: many RNGs > (I think instances of /dev/urandom drawn on by openSSL) do not > incorporate any internal information that can work like a salt. > They should. If my device has a device serial number, ethernet > address, timestamp, etc., those should all be hashed into the > entropy pool, with 0 entropy bits assessed. At the risk of repeating myself, we made a lot of changes to the /dev/random code base when we received a preprint of the Mining your P's and Q's paper (patches went into mainline the first week in July, and got propagated to older kernels via the stable kernel trees about 2 weeks later; the paper was published at Usenix Security in August.) One of them was to do precisely this --- /dev/urandom now mixes in salting information (ethernet MAC addresses, etc, via the new interface add_device_randomness). Zero entropy is indeed assessed, and the main goal is to avoid the trivially easy case of shared primes in the case where we fail to gather enough entropy. The other change we made was to gather entropy on every single interrupt, instead of only on those device drivers where the device driver authors gave us permission to collect entropy. That was a mistake, because device driver maintainers care about performance and CPU efficiency at all costs, and they don't particularly care about entropy collection. So we made entropy collection fast, and unconditional. As I've already said, I'm open to adding code that blocks /dev/urandom until "enough" entropy has been collected. But that's an interface-visible change, and it could break things. So there is due diligence that will need to be done first, because the reality is if it breaks things, people will just comment it out, and it will be harder to propagate the change in the future. In the real world, the product manager is often as much a security engineer's adversary as Boris and Natasha. :-( - Ted _______________________________________________ The cryptography mailing list cryptography(a)metzdowd.com http://www.metzdowd.com/mailman/listinfo/cryptography

2 1

pǝɔɹʎdʇoɹ (nɹls)
by brian carroll 19 Oct '13

19 Oct '13

The Artist of the Unbreakable Code (via A&L Daily) Composer Edward Elgar still has cryptographers playing his tune. http://nautil.us/issue/6/secret-codes/the-artist-of-the-unbreakable-code read this article the other day and wondered if anyone evaluated the alphabet for letters that have those curved shapes (e.g. m, n, c, w, u, p, q, b, d, g, etc) at the subsign level, removing extra details for common squiggle structure, &or potentially how the cipher could be related to musical note values... Dorabella Cipher http://en.wikipedia.org/wiki/Dorabella_Cipher Letter Groupings http://www.rinkworks.com/words/lettergroups.shtml (list of alphabetic structure in terms of what letters share certain qualities, like lowercase with vertical bars, or lowercase with structure beneath the baseline. vital knowledge for such calculus... though missing rotational without reflective symmetry NSZ) Homoglyph http://en.wikipedia.org/wiki/Homoglyph "In typography, a homoglyph is one of two or more characters, or glyphs, with shapes that either appear identical or cannot be differentiated by quick visual inspection." Numeronym http://en.wikipedia.org/wiki/Numeronym "A numeronym is a number-based word." Transformation of text http://en.wikipedia.org/wiki/Transformation_of_text "Transformation of text is strategies to perform geometric transformations on text (reversals, rotations, etc.), particularly in systems that do not natively support transformation, such as HTML, seven-segment displays and plain text." ʇno ʇı ƃuıʇsǝʇ http://flip-o-matic.net/ 14.2 October 2013 http://www.mnartists.org/work.do?rid=340505

1 0