--- Log opened Mon May 30 00:00:15 2016 | ||
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 00:44 | |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 00:47 | |
-!- ArturSha1 [~ArturShai@185.66.252.126] has quit [Ping timeout: 252 seconds] | 00:53 | |
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-lclfruqykddmlvgd] has joined ##hplusroadmap | 01:04 | |
-!- Gurkenglas [Gurkenglas@dslb-178-000-182-226.178.000.pools.vodafone-ip.de] has joined ##hplusroadmap | 01:11 | |
-!- ArturSha1 [~ArturShai@185.66.252.126] has joined ##hplusroadmap | 01:20 | |
-!- Gurkenglas [Gurkenglas@dslb-178-000-182-226.178.000.pools.vodafone-ip.de] has quit [Ping timeout: 244 seconds] | 01:23 | |
-!- sandeepkr [~sandeep@111.235.64.4] has quit [Ping timeout: 260 seconds] | 01:46 | |
-!- sandeepkr [~sandeep@111.235.64.4] has joined ##hplusroadmap | 02:03 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 240 seconds] | 02:05 | |
-!- esmerelda [~andares@172.58.41.16] has joined ##hplusroadmap | 03:27 | |
-!- esmerelda [~andares@172.58.41.16] has quit [Changing host] | 03:27 | |
-!- esmerelda [~andares@unaffiliated/jacco] has joined ##hplusroadmap | 03:27 | |
kanzure | hmph | 03:38 |
---|---|---|
-!- sandeepkr [~sandeep@111.235.64.4] has quit [Ping timeout: 244 seconds] | 03:44 | |
esmerelda | Up late I see | 03:52 |
-!- sandeepkr [~sandeep@111.235.64.4] has joined ##hplusroadmap | 03:53 | |
esmerelda | So apparently pelvic bone actually changes shape before and then reverses after menopause | 03:53 |
esmerelda | So apparently bone remodeling post puberty does happen | 03:53 |
esmerelda | I'd like to remodel some bones | 03:54 |
-!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has joined ##hplusroadmap | 03:56 | |
kanzure | http://www.nature.com/news/two-hundred-terabyte-maths-proof-is-largest-ever-1.19990 | 03:58 |
esmerelda | Why does this seem like an existential crisis to mathematicians? | 04:04 |
esmerelda | Computers are just another level of abstraction for them | 04:04 |
esmerelda | Programs are information-theoretically equivalent to expanded proofs | 04:05 |
-!- Vicarious [~diepfriet@CAcert/Vicarious] has joined ##hplusroadmap | 04:24 | |
-!- zadock [~outsider@81.180.223.50] has joined ##hplusroadmap | 04:54 | |
-!- zadock [~outsider@81.180.223.50] has quit [Ping timeout: 244 seconds] | 05:05 | |
-!- Urchin [~urchin@unaffiliated/urchin] has joined ##hplusroadmap | 05:20 | |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 05:27 | |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 05:28 | |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 05:29 | |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 244 seconds] | 05:30 | |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 05:30 | |
maaku | eh this one is super sensational | 05:30 |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 05:30 | |
maaku | basically someone else had already shown that if it held tue for numbers <=7,825 then it holds true for all numbers by induction | 05:31 |
maaku | so they just exhaustively checked each set of colorings for the numbers <=7,825. | 05:32 |
maaku | I'm guessing that the 200TB proof is just each of these checks appended together, although it's a asmall program that did it | 05:33 |
-!- augur [~augur@76-218-206-38.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 05:33 | |
-!- _hanhart [~hanhart@static.141.80.201.138.clients.your-server.de] has joined ##hplusroadmap | 05:34 | |
-!- augur [~augur@76-218-206-38.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 276 seconds] | 05:38 | |
kanzure | :frowns: so there's no 200 tb coq proof? | 05:49 |
maaku | well there is, but it's literally "Coq proof of possible coloring #1, Coq proof of possible coloring #2, ..." | 05:52 |
kanzure | why are you awake | 06:10 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 06:17 | |
-!- Orpheon [~Orpheon@7.152.62.81.dynamic.wline.res.cust.swisscom.ch] has joined ##hplusroadmap | 06:18 | |
maaku | good question | 06:38 |
* maaku sleep | 06:38 | |
-!- augur [~augur@76-218-206-38.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 07:24 | |
kanzure | does anyone have that old pic of a shotgun or rifle used to blast pellets at petri dishes for transformation reasons | 07:34 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 07:38 | |
-!- eudoxia [~eudoxia@r167-57-148-105.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 07:58 | |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-aqufgrmiimlzajqs] has joined ##hplusroadmap | 07:58 | |
-!- nildicit [~nildicit@unaffiliated/nildicit] has quit [Ping timeout: 240 seconds] | 08:52 | |
-!- eudoxia [~eudoxia@r167-57-148-105.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 08:54 | |
-!- CuriousCat [~CuriousCa@unaffiliated/wye-naught/x-8734122] has joined ##hplusroadmap | 08:58 | |
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-lclfruqykddmlvgd] has quit [Quit: Connection closed for inactivity] | 09:20 | |
-!- tariq786 is now known as kaaliakahn | 09:38 | |
nmz787 | abetusk: this is some C code that is seg-faulting on me at the moment... but was offered as a good example of how to use the BRLCAD API (which is largely accessible via command-line tools) to generate rastered g-code: http://brlcad.org/~sean/tmp/g-gcode.patch | 09:55 |
nmz787 | compiled with: gcc -I ../../../brlcad-svn-trunk/src/other/tcl/generic/ -I /usr/brlcad/include/brlcad -L /usr/brlcad/lib/ gcode.c -lbu -lrt | 09:57 |
nmz787 | also see http://brlcad.org/w/images/6/66/Converting_Geometry.pdf section "2.3 Converters Currently Available in BRL-CAD" | 09:57 |
-!- redlegion [~x@unaffiliated/redlegion] has quit [Quit: Time for me to pass out.] | 10:08 | |
-!- redlegion [~x@unaffiliated/redlegion] has joined ##hplusroadmap | 10:09 | |
-!- CuriousCat [~CuriousCa@unaffiliated/wye-naught/x-8734122] has quit [Quit: * * * * *] | 10:09 | |
-!- redlegion [~x@unaffiliated/redlegion] has quit [Client Quit] | 10:13 | |
-!- augur [~augur@76-218-206-38.lightspeed.sntcca.sbcglobal.net] has quit [Remote host closed the connection] | 10:15 | |
-!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has joined ##hplusroadmap | 10:17 | |
-!- Orpheon [~Orpheon@7.152.62.81.dynamic.wline.res.cust.swisscom.ch] has quit [Remote host closed the connection] | 10:18 | |
abetusk | nmz787, need help? | 10:21 |
nmz787 | well I don't have much time before I have to leave and go do some heat-seal bagging | 10:24 |
nmz787 | My plan was to convert that C code into Python | 10:24 |
nmz787 | but if I could get the C working with little effort, then maybe I'd call it a day and forget about it | 10:25 |
abetusk | alright. Is the brlcad python stuff in a useable enough state for anyone to download and start playing with it? | 10:25 |
abetusk | yeah, seems pretty straight forward, assuming it works the way I think it does | 10:25 |
nmz787 | hmm, it should be pretty OK | 10:25 |
abetusk | set up some callbacks for 'hits' and 'misses', fill in an array that represents the bitmap | 10:25 |
nmz787 | it basically just appends strings to a list, saves that list to a text file, then shoves that file to BRLCAD as a TCL script | 10:26 |
abetusk | compile it with debugging on, run it with valgrind and it should tell you exactly where it crashed | 10:26 |
nmz787 | mmm | 10:26 |
nmz787 | as far as that C code, I simply installed BRLCAD packages in ubuntu, and was able to compile with that one-liner I pasted earlier | 10:27 |
nmz787 | oh, I guess I have included a brlcad-svn repo header... not sure if I would be able to use only the stuff that the ubuntu package installed | 10:27 |
abetusk | they updated their landing page | 10:28 |
-!- jtimon [~quassel@4.28.134.37.dynamic.jazztel.es] has quit [Ping timeout: 264 seconds] | 10:28 | |
nmz787 | I wonder if there is some intermediate format that I could pass to your gbl2ngc program | 10:28 |
nmz787 | since that really seems to do a nice job on the vector-perimeters and infill | 10:28 |
abetusk | "nice"...still pretty rough I think | 10:29 |
abetusk | but yeah, it takes in gerbers... | 10:29 |
nmz787 | apparently shooting rays is the stanadard method for interrogating 3D geometry | 10:30 |
nmz787 | in BRLCAD | 10:30 |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 10:30 | |
nmz787 | if we had IRC bitcoin based kudos... I would totally send you some abetusk ;) | 10:31 |
abetusk | I'm not that familiar with 3d modelers, especially at this level. I would have though it would be possible to intersect 2d or 3d geometry with your scene | 10:31 |
abetusk | ha, maybe I should set it up | 10:31 |
nmz787 | well you can do a boolean operation with a thin planar rectangle, doing a union will give you a slice of your model... then with that you can view it from top-down, and shoot rays in a raster fashion... finding when the rays 'hit' confirms a pixel | 10:33 |
abetusk | that's interesting. Can you get the underlying 3d geometry as a bag of triangles or something a list of polygons? | 10:34 |
nmz787 | yeah | 10:34 |
nmz787 | BOT is the bag of triangles export | 10:34 |
nmz787 | and I guess STL might count as something | 10:34 |
nmz787 | not sure of list of polygons otherwise | 10:35 |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 272 seconds] | 10:35 | |
abetusk | ok, so how about this: Intersect with a thin rectangle, get the bag of triangles and only consider lines that sit at the top of the thin triangle. Test for clockwise or counterclockwise line segments by shooting rays at an epsilon distance to the left and right of each line. Then stitch up the lines and you have your polygons with holes | 10:36 |
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:4c7c:85c5:666:6b88] has joined ##hplusroadmap | 10:38 | |
nmz787 | hmm, I have read a few times on the primitive of CCW but never really fully grasped it | 10:39 |
abetusk | if brlcad gives you other options, then great but I'm trying to construct it with the bag of triangles and ray shooting primitives | 10:40 |
-!- ArturSha1 [~ArturShai@185.66.252.126] has quit [Ping timeout: 272 seconds] | 10:40 | |
-!- redlegion [~x@unaffiliated/redlegion] has joined ##hplusroadmap | 10:40 | |
nmz787 | I mean using counterclockwise/clockwise as a primitive geometric operation | 10:40 |
abetusk | clipperlib represents geometry as polygons and polygons with holes, where the outer boundary is in one orientation (clockwise, say) and the holes are in the other (counter clockwise, say). | 10:40 |
nmz787 | this is what I've read and not fully internalized yet: https://www.toptal.com/python/computational-geometry-in-python-from-theory-to-implementation#introducing-ccw | 10:42 |
abetusk | that picture is horrible | 10:43 |
abetusk | http://www.angusj.com/delphi/clipper/documentation/Docs/Units/ClipperLib/Functions/Orientation.htm | 10:44 |
abetusk | http://www.angusj.com/delphi/clipper/documentation/Docs/Units/ClipperLib/Functions/SimplifyPolygon.htm | 10:45 |
abetusk | sorry, maybe I should say, what don't you understand about it? | 10:46 |
-!- augur [~augur@108-239-230-147.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 10:47 | |
nmz787 | mmm, I guess it just doesn't "click" I've read it a few times over the last year or year-and-a-half and still just doesn't really "click" (and I work on geometry calculations for PCB schematics and layout design... so it better sink in soon enough or I will be a sitting duck at work) | 10:49 |
abetusk | you have a list of x,y points, say [[0,0], [10,0], [5,5]]. Assuming the last point connects with the first, it makes a triangle (in 2d) | 10:50 |
nmz787 | sure | 10:51 |
abetusk | if you were sitting in the middle of the triangle and watched an ant traverse the boundary from the first point to the last (then back to the first point) you'd be spinning around | 10:51 |
nmz787 | seems this would be CCW | 10:51 |
abetusk | yeah | 10:51 |
nmz787 | if (x,y) | 10:51 |
abetusk | and if the points were laid out as [[0,0], [5,5], [10,0]] ? | 10:52 |
nmz787 | CW | 10:52 |
abetusk | so that's it. Nothing magical | 10:52 |
nmz787 | but now I've labelled them, but how does it help? | 10:52 |
abetusk | hm? They're labeled from their order. How does what help? | 10:53 |
nmz787 | i mean what does labelling them afford us? | 10:53 |
nmz787 | "A simple (but non-obvious) computational geometry algorithm for determining convexity is to check that every triplet of consecutive vertices is CCW." | 10:55 |
abetusk | I'm having a little trouble understanding where you're coming from. Usually polygons are defined as a list of ordered (x,y) points, indicating the edge boundary. The triangle above is really three edges, [[0,0], [5,5]], [[0,0],[10,0]] and [[5,5],[10,0]] but in order to make a polygon you need to know an ordering of those edges. You can see how it gets more complex for more edges. | 10:56 |
abetusk | http://debian.fmi.uni-sofia.bg/~sergei/cgsr/docs/clockwise.htm | 10:58 |
abetusk | please don't read that post anymore. It seems really misleading and bad | 10:59 |
nmz787 | hmm, but isn't a list already ordered? | 10:59 |
nmz787 | haha, ok | 10:59 |
nmz787 | I also won't send it to anyone else then | 10:59 |
abetusk | sure. I give you an ordered list of points making up the boundary of a polygon. Tell me whether the polygon is clockwise or counterclockwise | 10:59 |
abetusk | that post was talking about the cross product. Take the cross product of consecutive points and if they all have positive sign (say) it's one orientation, if they all have negative sign it's the other orientation. | 11:01 |
nmz787 | I see that, which gives you concave vs convex | 11:01 |
kanzure | http://www.theonion.com/article/voyager-probe-badly-damaged-after-smashing-end-uni-52996 | 11:01 |
abetusk | if it's mixed, then you can't tell. If it's concave then it will tell you the orientation | 11:02 |
abetusk | I think. I think his statement is at best misleading and at worst flat out wrong | 11:02 |
abetusk | the article he links to (which I linked to above) does a good job of explaining. | 11:03 |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has joined ##hplusroadmap | 11:05 | |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has quit [Quit: Textual IRC Client: www.textualapp.com] | 11:14 | |
-!- nildicit [~nildicit@unaffiliated/nildicit] has joined ##hplusroadmap | 11:21 | |
-!- Gurkenglas [Gurkenglas@dslb-178-000-182-226.178.000.pools.vodafone-ip.de] has joined ##hplusroadmap | 11:21 | |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has joined ##hplusroadmap | 11:33 | |
-!- Orpheon [~Orpheon@7.152.62.81.dynamic.wline.res.cust.swisscom.ch] has joined ##hplusroadmap | 11:47 | |
-!- maaku [~quassel@173-228-107-141.dsl.static.fusionbroadband.com] has quit [Ping timeout: 244 seconds] | 11:53 | |
-!- maaku [~quassel@173-228-107-141.dsl.static.fusionbroadband.com] has joined ##hplusroadmap | 11:54 | |
archels_ | kanzure: speaking of infinite torture, The Age of Em is finally coming out this week | 12:07 |
kanzure | i saw googe had indexed the book's contents the other day.. its chapters started showing up in search results. | 12:08 |
archels_ | cute | 12:10 |
archels_ | I wonder how quickly it's going to turn up on libgen | 12:10 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 252 seconds] | 12:29 | |
maaku | kanzure: one more reason for draining the channel -- archeological artifacts more than a few cm under the surface of the seafloor are preserved in the anaerobic environment... | 12:31 |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…] | 12:32 | |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has joined ##hplusroadmap | 12:51 | |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…] | 13:01 | |
-!- Gurkenglas [Gurkenglas@dslb-178-000-182-226.178.000.pools.vodafone-ip.de] has quit [Ping timeout: 250 seconds] | 13:02 | |
-!- Gurkenglas [Gurkenglas@dslb-178-000-182-226.178.000.pools.vodafone-ip.de] has joined ##hplusroadmap | 13:03 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 13:05 | |
-!- CuriousCat [~CuriousCa@2602:30a:2c94:5460:89e8:d1c3:cc88:d910] has joined ##hplusroadmap | 13:09 | |
-!- CuriousCat [~CuriousCa@2602:30a:2c94:5460:89e8:d1c3:cc88:d910] has quit [Changing host] | 13:09 | |
-!- CuriousCat [~CuriousCa@unaffiliated/wye-naught/x-8734122] has joined ##hplusroadmap | 13:09 | |
kanzure | .title https://news.ycombinator.com/item?id=11802629 | 13:16 |
yoleaux | The New Napster: How Sci-Hub Is Blowing Up the Academic Publishing Industry | Hacker News | 13:16 |
-!- Darius [~quassel@cpe-158-222-160-123.nyc.res.rr.com] has joined ##hplusroadmap | 13:31 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 244 seconds] | 13:44 | |
-!- Quashie_ [~boingredd@CPEbc4dfb5bf733-CMbc4dfb5bf730.cpe.net.cable.rogers.com] has quit [Ping timeout: 244 seconds] | 13:45 | |
-!- augur [~augur@108-239-230-147.lightspeed.sntcca.sbcglobal.net] has quit [Read error: Connection reset by peer] | 14:02 | |
-!- esmerelda [~andares@unaffiliated/jacco] has quit [Ping timeout: 276 seconds] | 14:23 | |
-!- augur [~augur@2602:304:cdac:e260:f451:eb27:7e43:60a6] has joined ##hplusroadmap | 14:25 | |
-!- Orpheon [~Orpheon@7.152.62.81.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 276 seconds] | 14:30 | |
-!- Orpheon [~Orpheon@7.152.62.81.dynamic.wline.res.cust.swisscom.ch] has joined ##hplusroadmap | 14:32 | |
-!- esmerelda [~andares@unaffiliated/jacco] has joined ##hplusroadmap | 14:43 | |
-!- esmerelda [~andares@unaffiliated/jacco] has quit [Ping timeout: 276 seconds] | 14:48 | |
-!- eudoxia [~eudoxia@r167-57-165-221.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 15:07 | |
-!- esmerelda [~andares@unaffiliated/jacco] has joined ##hplusroadmap | 15:08 | |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving] | 15:11 | |
-!- Gurkenglas [Gurkenglas@dslb-178-000-182-226.178.000.pools.vodafone-ip.de] has quit [Ping timeout: 244 seconds] | 15:47 | |
-!- esmerelda [~andares@unaffiliated/jacco] has quit [Ping timeout: 276 seconds] | 15:56 | |
-!- esmerelda [~andares@172.58.41.122] has joined ##hplusroadmap | 15:57 | |
-!- esmerelda [~andares@172.58.41.122] has quit [Changing host] | 15:57 | |
-!- esmerelda [~andares@unaffiliated/jacco] has joined ##hplusroadmap | 15:57 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 16:20 | |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has joined ##hplusroadmap | 16:21 | |
-!- ninja-bob [~textual@c-73-36-60-43.hsd1.il.comcast.net] has quit [Client Quit] | 16:22 | |
-!- eudoxia [~eudoxia@r167-57-165-221.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 16:26 | |
kanzure | instead of using microwells for the inkjet dna synthesizer, peter clark suggests using stencil templates (also with micro cutouts or whatever). | 16:50 |
kanzure | ... "if you design the template cleverly you can actually centrifuge dna in it. You need one with 2 sets of joined wells. One that forms over the glass slide and one set joined to the first that the solution can simply flow into when you tip the slide - template assembly on its side. That should help with automation." | 16:51 |
kanzure | i understand his flow idea (sounds like my flooding idea) but not the centrifugation part | 16:54 |
-!- Quashie [~boingredd@45.42.8.189] has joined ##hplusroadmap | 17:01 | |
yashgaroth | yeah you can't centrifuge dna or protein if it's in solution, I mean you can if you want but they're not gonna sediment | 17:02 |
-!- saurik [saurik@carrier.saurik.com] has quit [Ping timeout: 260 seconds] | 17:11 | |
-!- saurik [saurik@carrier.saurik.com] has joined ##hplusroadmap | 17:12 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 272 seconds] | 17:13 | |
-!- esmerelda [~andares@unaffiliated/jacco] has quit [Ping timeout: 276 seconds] | 17:21 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 17:24 | |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-aqufgrmiimlzajqs] has quit [Quit: Connection closed for inactivity] | 17:42 | |
-!- Jawmare [~Jawmare@unaffiliated/jawmare] has quit [Read error: Connection reset by peer] | 17:57 | |
-!- drewbot [~cinch@ec2-54-167-187-37.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 19:47 | |
-!- drewbot [~cinch@ec2-54-161-114-214.compute-1.amazonaws.com] has joined ##hplusroadmap | 19:48 | |
xentrac_ | I've never understood why not, yashgaroth; shouldn't solvated complexes have a slightly different density than the bulk solvent? | 20:02 |
xentrac_ | and if not, couldn't you just saturate the solution with barium iodide or something in order to change the density of the solvated complexes that aren't solvating your protein? | 20:03 |
xentrac_ | I know I'm pretty naïve about chemistry and thermodynamics | 20:04 |
yashgaroth | as am I; if they're soluble then they won't pellet at g forces achievable in the lab...sure if you use strong salts or solvents or a density gradient they'll separate | 20:05 |
xentrac_ | oh, yeah, I didn't mean they'd come out of solution | 20:08 |
yashgaroth | it's mostly for the prior discussion of centrifuging the template array, written by a mathematician | 20:10 |
yashgaroth | even with any additives you'd like, centrifuging/separating DNA that small would require g forces that would wreck whatever material the array was built from | 20:11 |
xentrac_ | I was thinking that maybe you could end up with concentrated dissolved DNA in one end of the well | 20:12 |
yashgaroth | yes, but good luck removing the upper layer without pulling out all the DNA; even if you're just trying to speed up reactions by doing so, the necessary solvents would wreck enzyme activity | 20:13 |
xentrac_ | heh | 20:18 |
xentrac_ | also, sorry I didn't see you said "sediment" earlier | 20:18 |
xentrac_ | obviously you're correct; you're not going to be able to overcome electrostatic attraction between polar moieties and adjacent water molecules with the masses of the molecules | 20:19 |
-!- ninja-bob [~textual@207-181-243-222.c3-0.lem-ubr1.chi-lem.il.cable.rcn.com] has joined ##hplusroadmap | 20:22 | |
-!- Quashie [~boingredd@45.42.8.189] has quit [Ping timeout: 250 seconds] | 20:26 | |
yashgaroth | I mean there may well be an engineering solution to it, but considering the goal is ultra-low-cost, at-home use, something like a custom ultracentrifuge rotor seems a little counter-productive | 20:31 |
jcorgan | what types of solvent are we talking about? | 20:33 |
yashgaroth | alcohol precipitation of DNA is traditional, I'm sure there are others but no one uses them | 20:36 |
jcorgan | so, i'm not familiar with DNA precipitation, but, if i read correctly above, you have DNA completely solvated, right? | 20:37 |
yashgaroth | in aqueous buffers, yes | 20:38 |
jcorgan | and your goal is to selectively precipitate different fractions of it, or is it all at once | 20:38 |
jcorgan | (sorry, have only been lightly skimming the conversation) | 20:39 |
yashgaroth | I suppose you could do it selectively, not sure if there's a relation between % alcohol and DNA size for precipitation though | 20:39 |
jcorgan | ok, so i'm speculating here, but | 20:40 |
yashgaroth | all at once, though | 20:40 |
jcorgan | oh, that's easier | 20:40 |
jcorgan | call the solvent A | 20:40 |
jcorgan | find a solvent B with two properties--1) it is miscible with A, and 2) the solute is completely insoluble in B | 20:41 |
jcorgan | then you can start adding B in incremental amounts until the solvent system can no longer sustain the solute and it precipitates | 20:41 |
jcorgan | there might even be a range of concentration that provides selective "crashing out" | 20:42 |
yashgaroth | yes that's how it usually works, B doesn't need to be a solvent necessarily but yes | 20:42 |
jcorgan | ah, ok, i'm really unfamiliar with the area you guys are discussing but it's a common technique elsewhere, so shouldn't be surprised it's also used there | 20:44 |
yashgaroth | yeah the selective precipitation is popular, best example is proteins in ammonium sulfate but dna/alcohol is the other one | 20:45 |
xentrac_ | yeah, microliters of ethanol are definitely practical for at-home use, unlike ultracentrifuges | 20:45 |
jcorgan | it will take me awhile to catch up to what is obvious in this group of people | 20:54 |
yashgaroth | I have my small domain of biology and very little else, I wouldn't worry | 20:56 |
-!- ArturSha1 [~ArturShai@185.66.252.126] has joined ##hplusroadmap | 20:58 | |
-!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has quit [Read error: Connection reset by peer] | 20:58 | |
-!- ninja-bob [~textual@207-181-243-222.c3-0.lem-ubr1.chi-lem.il.cable.rcn.com] has quit [Quit: My Mac has gone to sleep. ZZZzzz…] | 21:00 | |
maaku | .title https://arxiv.org/abs/1605.08695 | 21:21 |
yoleaux | [1605.08695] TensorFlow: A system for large-scale machine learning | 21:21 |
maaku | new tensorflow paper out of google brain | 21:21 |
jcorgan | data-flow based systems FTW | 21:22 |
jcorgan | but any indication of the delta between this paper and the earlier TF one? | 21:23 |
jcorgan | also, might be stating the obvious, but the Keras library provides a really usable abstraction layer atop TF | 21:27 |
nmz787 | you could just use electrophoresis and dialysis tubing/agrose/PAGE gel | 21:35 |
-!- Jawmare [~Jawmare@unaffiliated/jawmare] has joined ##hplusroadmap | 21:46 | |
yashgaroth | I assume that's to do with our prior discussion and not tensorflow, but applying those to 4000 wells is tricky | 21:54 |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 22:04 | |
-!- CuriousCat [~CuriousCa@unaffiliated/wye-naught/x-8734122] has quit [Quit: * * * * *] | 22:09 | |
-!- withanr [notauser@gateway/vpn/mullvad/x-tfdqexmduobcdnqy] has joined ##hplusroadmap | 22:17 | |
-!- Darius [~quassel@cpe-158-222-160-123.nyc.res.rr.com] has quit [Remote host closed the connection] | 22:44 | |
-!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has quit [Quit: Leaving] | 22:57 | |
-!- augur [~augur@2602:304:cdac:e260:f451:eb27:7e43:60a6] has quit [Remote host closed the connection] | 23:08 | |
-!- withanr [notauser@gateway/vpn/mullvad/x-tfdqexmduobcdnqy] has quit [Remote host closed the connection] | 23:33 | |
-!- Orpheon [~Orpheon@7.152.62.81.dynamic.wline.res.cust.swisscom.ch] has quit [Read error: Connection reset by peer] | 23:53 | |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 23:55 | |
--- Log closed Tue May 31 00:00:16 2016 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!