--- Log opened Sat Dec 02 00:00:31 2017
00:14 -!- rpifan [~rpifan@207-244-191-189-dhcp.mia.fl.atlanticbb.net] has quit [Remote host closed the connection]
01:24 -!- darsie [~username@84-114-73-160.cable.dynamic.surfer.at] has joined ##hplusroadmap
01:56 < abetusk> fenn, it's about $500 for a cnc that can do 4x8' plywood.  It's the cheapest one that I've found and it's open.  The servos keep the cost down.  The "custom PCB" is a shield on top of an Arduino 2560 that's a glorified breakout for 2 L298 chips.
02:11 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has quit [Ping timeout: 240 seconds]
02:16 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has joined ##hplusroadmap
02:27 -!- ebowden [~ebowden@2001:8003:1074:bc00:8d4e:40ee:ffad:c8fa] has quit [Ping timeout: 276 seconds]
02:29 -!- ebowden [~ebowden@2001:8003:1074:bc00:8d4e:40ee:ffad:c8fa] has joined ##hplusroadmap
02:34 -!- liead is now known as adlai
02:40 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has quit [Quit: Leaving.]
03:12 < adlai> re: "anti-meth-addiction gene therapy" - sounds about as useful as therapeutic bleeding, ie, treats symptoms at the least-fundamental level. making a single chemical unavailable to receptors does nothing whatsoever to the underlying 'addiction'
03:23 < adlai> lol @ "running out of sand", that's an inaccurate summary of ~any veteran's recent memory, unless they were busy in one of the cold deserts
03:28 < DataPacRat> Have just posted a request for advice on some potential personal bio-improvement to https://www.reddit.com/r/arduino/comments/7h25wj/seeking_advice_pocket_air_quality_sensor/ : "TL;DR: I dislike asthma attacks; I'd be happier if I had a better idea  when I should call for a ride instead of walking. Can anyone here help  me figure out how to put together some sensors and LEDs?" If anyone here knows a Maker of the Arduino persuasion
03:28 < DataPacRat> , would appreciate any thoughts.
03:34 < adlai> excuse the potentially insulting question, but is gsb an actual bot or just a human pretending to be one ?
03:38 < adlai> .tell yashgaroth who's "recovery" was more impressive, anatoli bugorski or phineas gage? and on what dimensions is such impression even measured?
03:38 < yoleaux> adlai: I'll pass your message to yashgaroth.
04:27 -!- DataPacRat [~dan@vdsl-113.itcanada.com] has quit [Ping timeout: 248 seconds]
05:42 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
06:25 -!- NikopolSohru [~NSohru@37.48.65.48] has joined ##hplusroadmap
06:37 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
06:38 -!- ebowden [~ebowden@2001:8003:1074:bc00:8d4e:40ee:ffad:c8fa] has quit [Read error: Connection reset by peer]
06:39 -!- ebowden [~ebowden@2001:8003:1074:bc00:c55:aa1f:a318:537] has joined ##hplusroadmap
06:42 < kanzure> adlai: gradstudentbot is really a bot
06:42 < gradstudentbot> The grant got rejected.
07:11 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has joined ##hplusroadmap
07:47 <@fenn> .tell datapacrat http://ebay.com/itm/222172849164 and a small usb battery should work. i had an earlier version but it was killed by an evil usb charger. ymmv
07:47 < yoleaux> fenn: I'll pass your message to datapacrat.
07:49 -!- NikopolSohru [~NSohru@37.48.65.48] has quit [Ping timeout: 240 seconds]
07:50 < gradstudentbot> Yeah, but biobricks don't even work.
07:52 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 250 seconds]
07:54 < kanzure> "Engineering and optimising deaminase fusions for genome editing" https://www.nature.com/articles/ncomms13330
08:10 -!- NikopolSohru [~NSohru@89.238.185.116] has joined ##hplusroadmap
08:22 < adlai> gradstudentbot: do something creative!
08:22 < gradstudentbot> Am I required to keep the human subjects alive?
08:22 < adlai> gradstudentbot: no, although it would help you get your grant renewed
08:22 < gradstudentbot> Can I get my own desk?
08:22 < adlai> sure
08:22 < adlai> gradstudentbot: only if you teach your human subjects to build it out of wood they grew themselves
08:22 < gradstudentbot> You know they keep the mice in better conditions than us.
08:23 < adlai> gradstudentbot: your homework is to read The Complete Hitchhiker's Guide. I refuse to discuss further with you if you use species names so flippantly.
08:23 < gradstudentbot> Who has the latest revisions on the paper?
08:25 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has quit [Ping timeout: 240 seconds]
08:46 -!- DataPacRat [~dan@vdsl-113.itcanada.com] has joined ##hplusroadmap
08:51 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has joined ##hplusroadmap
09:00 -!- Cory [~Cory@unaffiliated/cory] has quit [Ping timeout: 260 seconds]
09:06 -!- Pasha [~Cory@unaffiliated/cory] has joined ##hplusroadmap
09:08 -!- Pasha is now known as Cory
09:16 -!- yashgaroth [~yashgarot@2606:6000:cd4d:3300:f5e0:f867:a11d:8d52] has joined ##hplusroadmap
09:22 < yashgaroth> probably phineas gage considering the amount of damage, with bonus points for surviving with 1800s medicine; hard to have a concrete ranking, to be fair
09:22 < yoleaux> 11:38Z <adlai> yashgaroth: who's "recovery" was more impressive, anatoli bugorski or phineas gage? and on what dimensions is such impression even measured?
09:26 < gradstudentbot> Who's doing journal club today?
10:40 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has quit [Quit: leaving]
10:56 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
11:10 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 260 seconds]
11:23 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
11:23 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
11:27 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 240 seconds]
11:44 < kanzure> "Privacy-preserving whole genome sequence processing through proxy-aided ORAM (oblivious RAM)" https://www.datenzone.de/blog/wp-content/uploads/2016/06/Privacy-Preserving-Whole-Genome-Sequence-Processing-through-Proxy-Aided-ORAM.pdf
11:47 < kanzure> trouble is that, even with <100 biomarkers, i think you can deanonymize someone
11:47 < kanzure> so these privacy mechanisms don't seem to be particularly useful
11:47 < kanzure> especially if you are letting companies query arbitrary chunks of the data
11:47 < kanzure> andytoshi: ^thoughts?
11:50 -!- CaptHindsight [~2020@unaffiliated/capthindsight] has joined ##hplusroadmap
11:58 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-hzcolcaxsckiljoy] has joined ##hplusroadmap
12:04 -!- mindsForge [~nak@75-171-23-2.phnx.qwest.net] has joined ##hplusroadmap
12:12 < andytoshi> kanzure: i think the idea is that you're querying your own biomarkers
12:12 < andytoshi> yeah, if companies can query arbitrary things, ORAM doesn't do anything, it just prevents the cloud storage people from seeing your DNA
12:12 < andytoshi> but they're not the threat here :)
12:15 < gradstudentbot> Which calculator do you need? I have a TI-83 and a TI-84.
12:18 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
12:21 < nmz787> wow that Gage guy story is pretty awesome
12:21  * nmz787 mom, can I get a metal rod shot through my head when I grow up?
12:29 -!- CaptHindsight [~2020@unaffiliated/capthindsight] has quit [Quit: Leaving]
12:30 -!- DataPacRat [~dan@vdsl-113.itcanada.com] has quit [Ping timeout: 248 seconds]
12:31 < kanzure> andytoshi: so, association studies are used to create new biomarkers-- they look at all data. that's how phenotype-genotype links are figured out.
12:32 < kanzure> it seems like this regime is incomptible wtih privacy
12:32 < andytoshi> yeah
12:33 < andytoshi> possibly with fully-homomorphic encryption they can extract biomarkers from a ton of genomes without seeing any individual ones
12:33 < nmz787> is ORAM interesting with the new RAM encryption that's being built into modern processors?
12:34 < andytoshi> yes
12:34 < nmz787> huh, this ORAM stuff is from 1992
12:35 < andytoshi> ORAM prevents the owner of the RAM from knowing what addresses you're accessing
12:35 < andytoshi> yuppers
12:36 < nmz787> but I think memory-controller based encryption would do the same
12:36 < nmz787> maybe the thing I read about the future product with encrypted RAM just didn't call it ORAM?
12:36 -!- jaboja [~jaboja@jaboja.pl] has quit [Remote host closed the connection]
12:42 < Iriez> Is there anyone that could help me understand what categories there are of xenobiotic inducers? Im attempting to downregulate cytochrome P450's to hopefully deal with some metabolic issues i have and need to better understand CYP/CAR/PXR inducers
12:43 < Iriez> What scope would this table cover, and what is outside that scope? http://medicine.iupui.edu/clinpharm/ddis/clinical-table/
12:49 < andytoshi> nmz787: no, memory-controller encryption can't hide what data you're looking up
12:49 < andytoshi> well, maybe it does .. then you're trusting the memory controller hardware, and you somehow have to do a key-exchange with that hardware from outside of the machine....which is not at all what i thought memory-controller encryption did
12:51 < nmz787> andytoshi: I've only heard about the full-memory encryption being a feature of a new upcoming product
12:51 < nmz787> andytoshi: I don't really know the details, and am not understanding enough of that or ORAM to compare/contrast
12:51 < nmz787> I sent the ORAM 1992 paper to my wife though, who does hardware security
13:26 < nmz787> weird, bunnie huang has troubles using Python
13:28 < nmz787> "My final thought is that LiteX, in its current state, is probably best suited for people trained to write software who want to design hardware, rather than for people classically trained in circuit design who want a tool upgrade."
13:29 -!- hehelleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has joined ##hplusroadmap
13:30 < nmz787> https://github.com/timvideos/HDMI2USB-litex-firmware/blob/master/getting-started.md
13:33 -!- Guest57615 [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has quit [Ping timeout: 240 seconds]
14:15 < nmz787> why do pharmacists need degrees? seems like they're just portioning out bulk goods into to-go bags
14:16 < nmz787> like, how are they different than say a person working at a landscaping supply company, where you drive up with a trailer and slip showing what sort of dirt/mulch you need
14:17 < yashgaroth> most of it is martial arts training to fend off junkies
14:19 < nmz787> someone on facebook says "Because the job of the pharmacists is also to instruct in the use of the medication, and as a second line of check for drug interactions, and patient care."
14:19 < nmz787> both responses make sense ;)
14:21 < andytoshi> nmz787: what ORAM means is that you can send a lookup to a remote server (which is in control of all the RAM) and it literally cannot tell which addresses you are reading and writing
14:22 < TMA> nmz787: it depends on the country -- here some pharmacies do make the drugs from components
14:24 < TMA> nmz787: on the other hand the pills are coming in pre-packaged boxes, so no bulk goods partitioning
14:30 < TMA> additionally the legal degree requirements are usually not a guarantee that the knowledge purportedly acquired as the degree prerequisite is well correlated with the knowledge required on the job
14:34 < nmz787> someone on facebook also said that basically doctors are lazy (not really, just have too much else to think about and be responsible for) and they rely on pharmacists to double check them
14:34 < kanzure> some of the pharmacy stuff is also about knowledge of interactions
14:34 < nmz787> yeah, that too
14:34 < nmz787> the docs might not be up on the latest interaction research
14:35 < kanzure> yeah i don't think ORAM works for genome privacy if someone is trying to discover currently-unknown properties
14:35 < nmz787> basicall additional layer of computational checks, added redundancy
14:35 < kanzure> in fact it seems like genome privacy in this context is impossible
14:35 < kanzure> which would be an interesting result to publish
14:36 < nmz787> yeah ORAM still seems a bit vague in its utility... like, the memory address might not matter, especially if there was basepair metadata being stored and retrieved, since that is the address of the DNA 'program'
14:36 < nmz787> the DNA program is what is desired to be private, not the DNA binary representation
14:37 < kanzure> huh? yes the binary representation needs to be kept private. you literally only have bits.
14:38 < nmz787> no i mean the binary doesn't matter
14:38 < nmz787> since it can be packed with metadata
14:39 < nmz787> which in addition to say a base-call, would have the base-address (i.e. chromosome number and location/offset in that chromosome)
14:39 < kanzure> i don't know what you're talking about. my point is that none of the information privacy schemes work, for association studies you need the entire data because you don't know what you're looking for, unless you're only checking like 2 attributes total-- but even that leaks data.
14:39 < nmz787> computer addresses of DNA information seems uninteresting if the payload already contains positional info
14:40 < kanzure> sounds like you are assuming a suboptimal data structure
14:40 < nmz787> sounds like you're assuming biologists already made optimal data structures
14:41 < kanzure> huh? the privacy scheme (ORAM) was not invented by biologists.
14:41 < nmz787> I know
14:42 < nmz787> anyway, I agree that association study algorithms probably need all the data at once
14:42 < kanzure> i don't think you can achieve privacy here, really;
14:43 < kanzure> also i'm not sure why people want private genomes (something about non-discrimination?)
14:43 < kanzure> (health insurance discrimination stuff?)
14:43 < nmz787> fear of targetted bioweapons
14:44 < nmz787> I guess insurance discrimination is an economic bioweapon
14:44 -!- NikopolSohru [~NSohru@89.238.185.116] has quit [Ping timeout: 248 seconds]
14:44 < nmz787> otherwise, planting fake bio evidence seems like a reasonable fear...
14:45 < nmz787> sure humans already shed lots of bio data everywhere they go
14:45 < nmz787> but you aren't physically going to the genome storage datacenter, so you lose a bit of control
14:46 < nmz787> (i.e. if you were super paranoid you'd be a self-sufficient hermit)
14:46 < nmz787> (but you'd need to be offgrid too, in the context of stolen genome data)
14:46 < kanzure> targeted bioweapons can happen by stealing a fingerprint or lock of hair
14:46 < kanzure> so... data privacy might help a little, there, but... there are bigger problems.
14:53 -!- Storyteller [~Storytell@unaffiliated/storyteller] has joined ##hplusroadmap
14:55 -!- Storyteller [~Storytell@unaffiliated/storyteller] has quit [Max SendQ exceeded]
14:55 -!- Storyteller [~Storytell@unaffiliated/storyteller] has joined ##hplusroadmap
14:57 < kanzure> .title https://news.ycombinator.com/item?id=15834006
14:57 < yoleaux> Woman with Transplanted Uterus Gives Birth, the First in the U.S | Hacker News
14:58 < Storyteller> pretty cool stuff
15:07 -!- rpifan [~rpifan@207-244-191-189-dhcp.mia.fl.atlanticbb.net] has joined ##hplusroadmap
15:15 -!- Storyteller [~Storytell@unaffiliated/storyteller] has left ##hplusroadmap ["Leaving"]
15:33 -!- darsie [~username@84-114-73-160.cable.dynamic.surfer.at] has quit [Ping timeout: 248 seconds]
16:18 < nmz787> Pattern Projection with Focus Detection - Typically, software used on optical instruments can have difficulties determining what is in focus and what is out of focus. The accuracy of the optics may limit how effectively the focal plane can be resolved. By using advanced optics to project a sharp pattern that will only be in focus at a precise distance, the software and optics are able to determine
16:18 < nmz787> what is in the focal plane more easily and with greater accuracy.This also allows for more flexibility in measuring transparent samples as the projection will only be in focus on a real surface.
16:20 < nmz787> hmm, the zdot machine seems like it wouldn't be terribly hard to reproduce http://nanoscience.com/products/optical-profilometry/optical-metrology/
16:21 < nmz787> seems like some implementation of https://en.wikipedia.org/wiki/Bahtinov_mask
16:24 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
16:25 < nmz787> hmm https://www.cloudynights.com/topic/547413-where-to-find-maskulator/
16:27 < nmz787> I grabbed the two programs there, bahtinov grabber and maskulator
16:27 < nmz787> they are just zipped exes
16:28 < nmz787> I wonder if the source can be dumped, maybe they're C#
16:29 < nmz787> "Microflash has more "stopping power" than a hollowpoint. Measured at 329 m/s or 1184 km/h using a chronometer. It rotates at 80120 RPM, once ever 9.7 inch."
16:29 < nmz787> so we can use bullets for ultracentrifugation needs
16:34 -!- JayDugger [~jwdugger@47.185.237.246] has quit [Quit: Leaving.]
17:01 < nmz787> fun read http://web.archive.org/web/20160330144621/http://www.njnoordhoek.com/
17:15 -!- Gurkenglas [~Gurkengla@dslb-094-223-133-187.094.223.pools.vodafone-ip.de] has joined ##hplusroadmap
17:53 -!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Quit: the neuronal action potential is an electrical manipulation of reversible abrupt phase changes in the lipid bilayer]
18:03 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 255 seconds]
18:47 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has joined ##hplusroadmap
19:05 -!- ebowden [~ebowden@2001:8003:1074:bc00:c55:aa1f:a318:537] has quit [Read error: Connection reset by peer]
19:05 -!- ebowden [~ebowden@2001:8003:1074:bc00:c55:aa1f:a318:537] has joined ##hplusroadmap
19:12 -!- mindsForge [~nak@75-171-23-2.phnx.qwest.net] has quit [Remote host closed the connection]
19:51 -!- Gurkenglas [~Gurkengla@dslb-094-223-133-187.094.223.pools.vodafone-ip.de] has quit [Ping timeout: 260 seconds]
19:57 -!- ebowden [~ebowden@2001:8003:1074:bc00:c55:aa1f:a318:537] has quit [Remote host closed the connection]
19:57 -!- ebowden [~ebowden@110.141.92.153] has joined ##hplusroadmap
21:39 -!- justan0theruser is now known as justanotheruser
21:43 -!- DataPacRat [~dan@vdsl-113.itcanada.com] has joined ##hplusroadmap
21:51 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has quit [Ping timeout: 255 seconds]
21:52 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has joined ##hplusroadmap
21:58 -!- mindsForge [~nak@75-171-23-2.phnx.qwest.net] has joined ##hplusroadmap
22:07 -!- yashgaroth_ [~yashgarot@2606:6000:cd4d:3300:f5e0:f867:a11d:8d52] has joined ##hplusroadmap
22:09 -!- aeiousomething [~aeiousome@gateway/vpn/privateinternetaccess/aeiousomething] has quit [Ping timeout: 248 seconds]
22:10 -!- yashgaroth [~yashgarot@2606:6000:cd4d:3300:f5e0:f867:a11d:8d52] has quit [Ping timeout: 240 seconds]
22:11 -!- jtimon [~quassel@164.31.134.37.dynamic.jazztel.es] has quit [Ping timeout: 248 seconds]
22:25 -!- yashgaroth_ [~yashgarot@2606:6000:cd4d:3300:f5e0:f867:a11d:8d52] has quit [Quit: Leaving]
22:25 < nmz787> https://hackaday.com/2017/12/02/solving-a-rubiks-cube-with-just-two-motors/
22:26 < nmz787> so cool
23:05 -!- aeiousomething [~aeiousome@183.82.170.54] has joined ##hplusroadmap
23:23 -!- aeiousomething [~aeiousome@183.82.170.54] has quit [Ping timeout: 248 seconds]
23:37 -!- aeiousomething [~aeiousome@183.82.170.54] has joined ##hplusroadmap
--- Log closed Sun Dec 03 00:00:32 2017