Archive for January, 2007

Road to next SHA begins

Tuesday, January 23rd, 2007

I think we have all been expecting it, and I just saw this announcement today.

The purpose of this notice is to solicit comments on the draft minimum acceptability requirements, submission requirements, and evaluation criteria of candidate algorithms from the public, the cryptographic community, academic/research communities, manufacturers, voluntary standards organizations, and Federal, state, and local government organizations so that their needs can be considered in the process of developing the augmented and revised hash function standard.

Which points to this site at NIST.

Due to recent attacks on the SHA-1 hash function specified in FIPS 180-2 , Secure Hash Standard, NIST is initiating an effort to develop one or more additional hash algorithms through a public competition

And so begins the long and winding road to the next SHA. This will be fun.

(Politics aside, I thought the AES process was quite effective and necessarily open, a big change from the government’s view of crypto in the past. I am glad to see them following a similar path for the next hashing algorithms.)

Dedications

Saturday, January 20th, 2007

It has been a long while since my last weekend post. It has also been quite some time since I commented on either NYC or customer service. Plus, I have a cold. So, let’s write a weekend post. You may want to skip this.

-

I am picky about service, and bartenders are no exception. Many people say I am too picky with bartenders because, well, this is NYC and I just have deal with it. Those people are wrong, which leads to this post of my best bartender picks in NYC.

So, lets get going with five of my criteria.

  1. Good drinks. This one is obvious, but must be stated anyway. A bartender has to know how to make good drinks. There are three drinks I normally order, and my expectations of good vary across them.
  2. Fast as possible service. By fast as possible, I mean the drinks are brought as quickly as is possible when making the drinks well and taking the crowd into consideration.
  3. Respectful. I don’t want a rude, demeaning, contemptuous, and/or aloof bartender. People seem to think that type behavior is acceptable in NYC bars, lounges, or clubs. It is not.
  4. Friendly but not flirty. I want a friendly bartender, whether outgoing or not. I want to feel comfortable. And, a comp’d drink every once in a while never hurts. However, I don’t want the bartender flirting with me, as I don’t reciprocate and so this can turn negative after a few visits.
  5. Caring about people. I want to walk into a bar dripping sweat and beet red in the middle of the summer, and have the bartender seem genuinely concerned and offer me water before asking for my drink.

I could keep going, but these five serve as good enough filters for the best. And, who are the best?

(Unfortunately, a portion of my list may be dated. I find that events and places I went to regularly six months ago are often quite different from what I do now. I will note dated information when applicable.)

  1. The two best bartenders, I have found in NYC can be found on the mezzazine on Wednesday nights at Marquee. (Note: It has been about 6 months since I have been to Marquee. Hopefully, these two are still working at that time, but this is NYC.)

    General – These bartenders handle a crowd, make perfect drinks, and always make you feel welcome, even when the Marquee crowd is a little disappointing. Also, they learn your normal drink quite quickly; however, this credit must go beyond these bartnders to the overall mezzanine bar staff on Wednesday nights, as they seem to communicate what drinks particular patrons like amongst each other, so you really only have to ask once and never again.

    Notable – These two bartenders have spoiled me for drinking manhattans almost anywhere else. Once you have had the best, choking down the rest is no longer an option.

    One (long curly black haired woman) mixes a manhattan by stirring it, which usually indicates a terrible manhattan, but not in this case (it must be the 60 seconds of blender like arm action). By far, the best manhattan in Manhattan, let alone anywhere else I have had the chance to drink a manhattan. The other (long straight black haired woman) mixes a manhattan in a traditional way, shaking it. This manhattan is oh so close to the perfection achieved by the former.

    It does not get any better than these two, folks. Anywhere.

  2. There is really only one bar I can think of where all the bartenders rank up there amongst the best, Emerald Pub. I have encountered four distinct bartenders here, all of whom are next in my best of the bartenders list.

    General – Quite friendly, quick with the drinks, and fast to learn your normal drink. Sometimes I find the beer pouring to be a bit too painstaking, as the wait for the Guiness to settle really does not matter much to me (unless the keg is almost kicked). But, they do things the proper way and pour a good Guinness – you can’t fault them for practicing their craft.

    Notable – It is rare you find one good bartender in a place, here you have at least four. This is the least risky place to step foot in for good service anytime, and I make a point of passing through here whenever I am anywhere near Tribeca.

    (As far as bars go, this is where I point acquaintences and more homebody-type friends when they are visiting NYC .)

  3. Rififi on weekday nights. I can think of two distinct bartenders, one (short brown haired man) that works on Wednesdays and one (long straight blonde haired woman) on Thursdays.

    General and Notable – These quirky bartenders make Rififi feel like a second home. There is nothing else to say.

    (As far as bars go, this is where I point the majority of my friends when they are visiting NYC. Be warned that Rififi can get a bit wild certain nights.)

  4. There is a bartender (long curly blonde haired woman) that works Thursday evenings at Drop Off Service. (Note: It has been about 4 months since I last was in Drop Off Service.)

    General – Outgoing, quick to learn your normal drink, and just plain fun.

    Notable – This bartender manages to make a place that I normally don’t feel comfortable in feel totally comfortable.

  5. There are two bartenders at Lotus Cafe, one (geeky woman) works on Wednesday evenings and one (bald man) on Saturday evenings.

    General – Fast to learn your normal drink, friendly enough, quick with the drinks.

    Notable – These bartenders just create a feel that makes me like coming here often. In the middle of the trendy bustle that has taken over the LES, that is saying alot.

Dedicated to the perfect bartenders scattered about NYC. (Be sure to tip them well.)

Update: I was asked for addresses of some of the places mentioned above. Normally I would say something condescending, such as “google is your friend,” but, in the spirit of good service, here you go. ;)

  1. Marquee is on west side of 10th avenue about halfway (IIRC) between 26th and 27th street.
  2. Emerald is on the south side of Spring street about halfway between Greenwich and Hudson street (at the corner of some tiny street called Renwick).
  3. Rififi is on the south side of 11th street between 1st and 2nd avenue (closer to 1st).
  4. Drop Off Service is on the west side of avenue A between 13th and 14th street (closer to 13th, IIRC).
  5. Lotus Cafe is on the, umm, southwest corner of Stanton and Clinton street.

-

I often hear people say that it is easy to be hidden in a big city such as New York. I have found quite the opposite to be true. I can’t say how often I get recognized from being somewhere or doing something, but sometimes I feel like a pseudo-celebrity, which would probably rank on my top 10 list of worst nightmares, if I had one.

This is a by-product of New York. New Yorkers have this sort of instant familiarity, as opposed, to say, a DC, where closed-circle cliques reign supreme. You can say hello to almost anyone, anywhere, and they will respond in kind. And, you end up in conversations with random people that feel just like conversations you have with close friends you see all the time. That is just the way it is around here, and even us loners have trouble being alone. (I have (re)adjusted to strangers saying “Aren’t you Andrew such and such…” or “We danced/spoke/etc. at such and such…”, but it still makes me nervous on occasion, as I always feel at a disadvantage in the conversation.)

An example of this that has what has to be my favorite closing line ever happened at a restaurant called Kion, roughly a year after I returned to New York. There were these two people working there, the bartender and the host. After verifying the place was open, I sat down at the front bar and the host came over and said, “I have seen you before, no?” This (and being asked the time) are two classic pickup lines I get hit with all the time, so I said dismissively, “Umm, maybe. I tend to go out a bit.” She said, “I am sure I have seen you before.” Then, thinking for a second, she said matter of factly “Leet. I have seen you at Leet. You used to go there a lot, but you have not been there in a while.”

Refraining from the urge to declare how I am always in the 733t 5k1l75 zone, I said, confused, “Leet? Umm, I don’t even know of a place called Leet, let alone been there. Sorry, you must be thinking of someone else.” Still, she was adamant. She even spoke to the bartender, and the bartender confirmed that they had seen me at “Leet.” “Leet. Leet. Leet.”

Now I was baffled. I suggested maybe they were confusing me with Moby, or Larry Tee, or any other skinny, bald guy that I have been told I look like. So, the host looked at me closely and said she was positive it was me. I continued to deny ever going to a place called “Leet,” and the host started to walk away in defeat when it hit me, “Lit?” She said, “Yes, Leet! It was you. I knew I had seen you before.” Turned out, she was from Peru and, with her accent, Lit sounded like “Leet” to me.

After the host walked off, the bartender leaned in, gave me an evil grin, and said, “Not as hidden as you thought you were.”

Dedicated to all those people that claim the innateness of big city anonymity. (Look ma, no surveillance cameras in here.)

-

This made me laugh.

you sounded dumb. The term “shared secret” is
used all the time in real cryptosystems

Which was a response to this, which also made me laugh.

the “shared secret” (now there is a high-tech sounding secure name)

This is why I sometimes lose people in conversation. I then have to step back for a moment and explain terminology I generally take for granted. We all know different things.

Oh, and I was laughing here too.

HERE’S A BROCHURE FOR MY CULT.

Weakest link, folks.

Dedicated to those that make me laugh.

-

As one of my jobs at a small business a few years back, I was the system administrator. Now, this company employed mainly technical people, which made me extremely impatient when dealing with their computer troubles that required nothing more than basic debugging skills to fix. There are a few engineers out there that have been laughed at and openly mocked by me for stupid moments, such as thinking their machines were dead without noticing the breaker on their surge protector had tripped, or trying to probe me to see if I could tell when they were browsing those “junk in the trunk” web sites ;) .

Which leads me to my inane moment, as described by this short email chain. I quickly fired off the following message after playing with this headset for a couple of minutes, and the following reply appeared a few minutes later.

Andrew,
The Jabra JX10 need to be on, then push and hold the pairing button for 2 seconds, after that you should see the solid blue light.

Thank you
John
***snip***

Hello,

I just purchased a JX10 headset, and, after the initial charging, I tried to pair the device with my phone. Unfortunately, when I push and hold the pairing button on the JX10, the blue LED never lights to indicate that the device is ready for pairing. Additionally, my phone does not detect the JX10, so I don’t think it is just an LED problem. Are there any other troubleshooting steps?

Thanks
***snip***

You mean it has to be turned on? The connect/disconnect call button doubles as the power button, as clearly indicated in the manual and ignored by me. He should have just written, “RTFM.”

Dedicated to all those engineers that I have ridiculed for moments of utter “technical” stupidity. (Yes, I will still mock you the next time.)

Shh, WL

Thursday, January 11th, 2007

Read this.

Normally an email or a document posted to a website can be traced back to its source because each data packet carries the IP address of the last server that it passed through. To prevent this, WikiLeaks will exploit an anonymising protocol known as The Onion Router (Tor), which routes data through a network of servers that use cryptography to hide the path that the packets took. Bruce Schneier, a cryptographer based in Silicon Valley, California, explains it like this. “Imagine a large room jammed full of people in which many of them are passing around envelopes. How would you know where any of them started?”

Looking at their FAQ,

Wikileaks is an uncensorable version of Wikipedia for untraceable mass document leaking and analysis. It combines the protection and anonymity of cutting-edge cryptographic technologies with the transparency and simplicity of a wiki interface.

Wikileaks will also incorporate advanced cryptographic technologies for anonymity and untraceability. Those who provide leaked information may face severe risks, whether of political repercussions, legal sanctions or physical violence. Accordingly, extremely sophisticated mathematical and cryptographic techniques will be used to secure privacy, anonymity and untraceability.

For the technically minded, Wikileaks integrates technologies including modified versions of FreeNet, Tor, PGP and software of our own design.

Personally, the “modified versions” and “software of our own design” statements are a bit worrisome, as the design of this system has not been made public, as far as I can tell. The closest thing to information on this project seems to be here and here. A closed design feels completely counter to their overall message, but it reads (see the very end of this chain) like this will become open shortly. Perhaps it already has.

Actually, building an anonymized wiki(pedia) came onto my todo list a few years ago during a conversation. It left that list (as a done item) about five (ok, since the person that conversation was with reads this blog, maybe it was really ten) seconds later, as having a suitably anonymized connection to any existing wiki or other web forum could be used to fit the build.

More so though, my preference back then for such purposes was high latency mixnets and usenet. That still stands today.

Still, I guess it’s something to watch.

2.6.19 (etc.) Linux kernel SHA384 HMAC update

Monday, January 8th, 2007

Oh yeah, the linux kernel SHA384 HMAC issue was patched very quickly by Herbert Xu.

I’ve put a similar fix into crypto-2.6.

But, I wanted to note the following, which is why this is in a separate post rather than an update to the original post.

Looking in the code… In sha512.c, the sha384 crypto_alg structure has cra_blocksize set to be SHA384_HMAC_BLOCK_SIZE. cra_blocksize is then used in multiple places by the HMAC implementation in the hmac.c, such as for calculating whether a key needs to be hashed because it is longer than the underlying hash’s blocksize or for the lengths of the outer and inner pads. For example, looking at the function hmac_setkey, the variable bs is set to be the result of a call to crypto_hash_blocksize. If you look at crypto_hash_blocksize, the end result of that function is to return the value of cra_blocksize. As such, the key will be hashed if it is greater than 96 bytes in length, rather than 128. opad and ipad will be treated as 96 bytes in length, rather than 128. Etc. This does not follow the HMAC standard.

Because Herbert Xu’s reply interested me.

Yes, you’re quite right. This does break SHA-384-HMAC. What’s worse is that other implementations also have similar bugs with this. Some even use a block size of 64.

Since this occurs elsewhere, perhaps it is more than just a bug. If so, can someone explain to me why this is being done? This “bug” does not shorten the SHA384 HMAC. (If you want a shorter HMAC, you would use a different HMAC or truncate the HMAC or both.) All I can think is that this “bug” may be used to lower memory usage slightly, but SHA384 is still going to pad to its blocksize. What am I missing?

(Oh, and while we are at it, linux kernel tests for SHA384 HMAC and SHA512 HMAC.

This patch adds tests for SHA384 HMAC and SHA512 HMAC to the tcrypt module. Test data was taken from
RFC4231. This patch is a follow-up to the discovery (bug 7646) that the kernel SHA384 HMAC
implementation was not generating proper SHA384 HMACs.

)