--- Log opened Tue Aug 09 00:00:44 2022
00:48 < nmz787> https://phys.org/news/2022-08-common-weed-super-key-drought-resistant.html
00:48 < nmz787> jrayhawk_: ^ seems like something you'd be interested in
00:49 < nmz787> have you eaten purslane? There's some "weeding" up my garden area right now, and it grew really well here last year, but we didn't try eating it
00:49 < nmz787> maybe I will tomorrow
01:03 < jrayhawk_> i like the more biowizardy term "dark fixation" myself https://www.ncbi.nlm.nih.gov/pmc/articles/PMC540897/ 
03:25 < jrayhawk_> re: eaten purslane: no. i don't really do much in the way of cellulose-based foods these days. does it have some pharmacological value?
03:37 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
06:52 < kanzure> https://www.google.com/url?q=https://groups.google.com/d/msgid/diybio/31770683-CAF4-4AFD-A463-24EBEC0A0E4F%2540gmail.com
06:52 < kanzure> er..
06:52 < kanzure> https://groups.google.com/g/diybio/c/V3OYVBxaH04/m/iRMNhnVMCgAJ
07:33 -!- lkcl- is now known as lkcl
10:27 < kanzure> .tw https://twitter.com/zetalyrae/status/1556999434080485377
10:27 < saxo> my life is like Memento: I wake up with no recollection of who I am and instead of tattoos there's all these PDFs (@zetalyrae)
10:40 < kanzure> .tw https://twitter.com/RiscZero/status/1557044557157744644
10:40 < saxo> Excited to announce that RISC Zero has closed a $12M seed round led by @BainCapCrypto! Follow on investors include @__geometry__ , @d1ventures and @CotaCapital. // We are now heads down on bringing the magic of zero knowledge proofs to the whole blockchain space 🚀 Stay tuned! (@RiscZero)
10:41 < superkuh> saxo, don't believe his lies
10:41 < superkuh> Oh, the bot.
10:41 < superkuh> Riggghhht.
10:41  * superkuh walks away pretending it didn't happen.
11:38 -!- nsh- is now known as nsh
12:11 -!- spaceangel [~spaceange@ip-78-102-216-202.bb.vodafone.cz] has joined #hplusroadmap
12:35 -!- Guest3194 [~Guest31@cpc111681-tilb11-2-0-cust86.20-1.cable.virginm.net] has joined #hplusroadmap
12:35 -!- Guest3194 [~Guest31@cpc111681-tilb11-2-0-cust86.20-1.cable.virginm.net] has quit [Client Quit]
12:52 -!- mirage335 [~mirage335@2a01:4f8:120:2361::1] has joined #hplusroadmap
14:13 -!- spaceangel [~spaceange@ip-78-102-216-202.bb.vodafone.cz] has quit [Remote host closed the connection]
15:40 < kanzure> some nfc hardware wallet things... https://keycard.tech/ https://tapsigner.com/ https://beamu.io/
16:03 < kanzure> or https://www.coolwallet.io/product/coolwallet/
17:33 < fenn> E. coli costs $400?!?
17:33 < fenn> *facepalm*
17:34 < fenn> when it's $450 to print an e. coli genome, people will still swap agar stabs or whatever
17:35 < fenn> and it's not like you can just print DNA and it magically becomes alive by itself. you'd need some chassis to "boot" the genome
19:41 -!- muurkha_ is now known as muurkha
20:03 -!- LarchOye1 [gobitch@2604:2d80:6789:9100:8518:2a3f:a620:c06e] has quit [Remote host closed the connection]
22:21 < nmz787> kanzure: I will ask my kid about the cognitive abilities of his imaginary friend "doggy pet" tomorrow. I'll see if doggy pet knows addition and subtraction better than he does.
22:23 < muurkha> Muaddib: how do you generate a Zhegalkin polynomial from a truth table?
22:23 < Muaddib> muurkha: take the xor of all the outputs that are 1
22:24 < muurkha> Muaddib: can you elaborate?
22:24 < Muaddib> muurkha: each output corresponds to a minterm (disjunctive normal form), so you're xoring all the minterms
22:25 < nmz787> L29Ah: re sandals and not in a tropical climate... you need to visit some universities with strong international student populations... I'm sure you'll see some tropical-raised people wearing sandals and shorts in winter
22:25 < muurkha> Muaddib: what does a minterm look like?  like if the truth table is 00001000, that's one minterm, but what does the minterm look like if the inputs are X, Y, and Z?
22:25 < Muaddib> muurkha: are you talking about the function or the SOP form?
22:25 < nmz787> Muaddib: but a xor of all 1s is just nothing
22:25 < Muaddib> nmz787: the output is xor'ed with all the inputs, so it is not the same as xoring all the inputs.
22:26 < muurkha> nmz787: until you said that everything it had said was correct
22:26 < muurkha> just imprecisely stated
22:26 < nmz787> I don't believe it
22:26 < nmz787> what kind of polynomial is 0
22:26 < muurkha> a degree-zero polynomial
22:26 < nmz787> .wik zhegalkin polynomial
22:26 < saxo> " /  / Zhegalkin (also Žegalkin, Gégalkine or Shegalkin) polynomials, also known as algebraic normal form, are a representation of functions in Boolean algebra." - https://en.wikipedia.org/wiki/Zhegalkin_polynomial
22:27 < muurkha> the answer to the question I asked above is that 00001000 is ~X & Y & Z
22:27 < nmz787> it said xor 1s before that
22:28 < muurkha> yeah, you take the xor of the minterms corresponding to those outputs
22:28 < muurkha> ~X & Y & Z is almost a Zhegalkin monomial; the problem is that ~X is 1 + X as a Zhegalkin polynomial
22:28 < nmz787> it said the outputs /were/ 1
22:28 < nmz787> err, that which /are/
22:28 < muurkha> right, you don't want to include the minterms corresponding to the outputs that aren't 1
22:29 < muurkha> you leave those out of the Zhegalkin sum
22:29 < nmz787> xor of 1 1 is 0
22:29 < nmz787> 0 is not a general form of those polynomial class
22:29 < muurkha> right, that's where he was being a little imprecise
22:30 < muurkha> you xor the minterms corresponding to those outputs, not the 1s :)
22:30 < nmz787> you mean, that's where it started bullshitting
22:30 < muurkha> no, when I asked it to elaborate, it explained this correctly
22:30 < muurkha> so to convert (1 + X) & Y & Z into an actual polynomial you have to distribute it into YZ + XYZ
22:30 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 268 seconds]
22:31 < muurkha> and that's the Zhegalkin polynomial for 00001000
22:31 < muurkha> if you want 01001000 instead, you need to XOR in the minterm for 01001000
22:32 < muurkha> which is uh
22:32 < muurkha> X & Y & ~Z
22:32 < muurkha> and thus XY + XYZ (because ~Z is 1 + Z)
22:32 < muurkha> combining XY + XYZ + YZ + XYZ = XY + YZ
22:33 < muurkha> I have no idea what "the SOP form is though
22:33 < muurkha> oh, sum-of-products
22:35 < muurkha> but "the output is xor'ed with all the inputs" makes no sense, and I'm not sure how to interpret the preceding question in a way that makes sense either
22:35 < muurkha> Muaddib: ?
22:35 < Muaddib> muurkha: it's an OR gate
22:36 < muurkha> see, that's total nonsense
22:36 < nmz787> I'm not too good with logic... only really used it that "low level" when doing some SAT solver stuff, and basically had a really good mentor to get me off my feet and make myself just useful enough
22:36 < nmz787> I do know that XOR of a set of ones, is 0
22:36 < muurkha> yes, that is true :)
22:37 < muurkha> but the XOR of ~X & Y & Z and X & Y & ~Z is not always 0
22:37 < nmz787> so my interpretation is the outputs which are 1, then contribute to a set of all ones, and XOR then makes 0
22:37 < muurkha> no, that was explained here: 05:24 < Muaddib> muurkha: each output corresponds to a minterm (disjunctive normal form), so you're xoring all the minterms
22:38 < muurkha> if your truth table has a single 1 in it, then it corresponds to a DNF minterm which contains all the inputs, either positively or negated
22:39 < muurkha> in DNF you OR those minterms
22:39 < nmz787> idk, sounds like b.s., but I'm just a simple country boy
22:39 < muurkha> but since they're nonoverlapping you can XOR them instead
22:40 < muurkha> playing around with Zhegalkin polynomials the other night I realized that they make SAT really easy.  to program, I mean, not that they provide a polynomial-time algorithm to do so :)
22:41 < muurkha> well, do you understand what I mean above about 01000000 being X & Y & ~Z and ~X & Y & Z being 00001000?
22:42 < muurkha> in ANF all unsatisfiable functions normalize to 0
23:03 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=b34a7f8c nmz787: page with pics of ABI 391 synthesizer teardown >> http://diyhpl.us/diyhpluswiki/dna/ABI_391_teardown_pics/
23:04 < nmz787> I read what you wrote, but it isn't second nature enough to me to just get it... I'd need to think and rethink about it
23:04 < muurkha> I had to do not only that but also implement it in Python and play with it all night
23:05 < muurkha> which is why I was surprised to see that GPT-3 apparently can simulate the responses of a person who actually knows this stuff
23:05 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=76769137 nmz787: added link to ABI 391 synthesizer teardown pics >> http://diyhpl.us/diyhpluswiki/dna/synthesis/
--- Log closed Wed Aug 10 00:00:45 2022