--- Log opened Thu Jun 23 00:00:00 2022
00:23 < nmz787> TMA: sorry I was talking to this friend while with family on the side of the highway (noisy) so I didn't totally get what his reasons were for feeling uneasy. But he's like nearly retirement age and I consider him a pretty reliable person and sensible skeptic. I'm just here putting a feeler out because I don't have time to dig in more (nor have I ever really considered ocular evolution before)
00:25 < nmz787> he said something about one organism having their light-sensitive cells facing the light, while we have our light sensing cells obscured by a layer of nerves or capillaries or something
00:25 < nmz787> it made be think of the transition to backside illuminated CMOS chips
00:28 < fenn> it's a commonly referenced "design flaw" of the human eye, supposedly proving that it evolved because no intelligent designer would ever do something so dumb
00:28 < fenn> you can see the blood flow as purple sine waves or moving spots if you look at a blue sky
00:35 < nmz787> it will probably turn out to be something akin to saccades or something for resolution enhancement on a finer scale
00:37 < nmz787> well, not like the understood utility/reason for saccades. More like how some super-resolution camera imaging techniques vibrate/move the sensor by sub-pixel scales to increase effective resolution
00:38 < nmz787> .wik optical image stabilization
00:38 < saxo> "lens-based optical stabilisation sensor-shift optical stabilisation digital or electronic stabilisation" - https://en.wikipedia.org/wiki/Optical_image_stabilization
00:40 < nmz787> .wik pixel shift
00:40 < saxo> "Pixel shift is a method in digital cameras for producing a super resolution image. The method works by taking several images, after each such capture moving ('shifting') the sensor to a new position." - https://en.wikipedia.org/wiki/Pixel_shift
00:41 < fenn> (fwiw i believe the eye evolved, as did all life, but i don't have to support bad arguments in favor of that belief)
00:51 < nmz787> not sure what you mean there
00:51 < nmz787> does anyone here question evolution?
02:05 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
02:07 < adlai|alphanum> that is a difficult question to answer without further qualifying the word.
02:08 < adlai|alphanum> I don't think anyone denies that several different kinds of variation take place, and have taken place in the past.
02:09 < adlai|alphanum> does the anthropic principle suggest that variation is not perfectly random, because any regions of the total design space that did not result in sufficient intelligence to ponder evolution, would not get counted ?
02:10 < adlai|alphanum> if you stretch far enough, this perspective leads to a godless teleological cosmology quite similar to intelligent design, although it still does 'evolve' from nothingness towards the designer.
02:11 < adlai|alphanum> however, I don't think any of the channel regulars are conventional young-earth creationists, visiting the museum where you see dioramas of humans hunting dinosaurs, etc.
02:30 < TMA> nmz787: ah, the inverted retinas. the inversion seems to be from the fact, that there are two lineages of organisms, one of which (us) performs inversion of their body very early in the embryonal development (the cells that were first on the outside will go inside and vice versa) it is thought that the blood vessels come from the cells ending up on the outside, while the light sensitive ones are those that were on the outside initially and end up in the layer 
02:46 < lkcl> i know someone who ended up with order-rotated cone-rod connectivity. their father was the world's first case-study, and it was inherited to them
02:47 < lkcl> red was swapped to green, green to blue, blue to *rods* (black/white), rods to red
02:47 < lkcl> (!!)
03:32 -!- spaceangel [~spaceange@ip-78-102-216-202.net.upcbroadband.cz] has joined #hplusroadmap
03:40 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
03:40 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
03:49 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
03:49 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
04:19 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
04:19 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
06:02 -!- flooded is now known as _flood
07:02 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
07:02 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
08:02 < jrayhawk> fenn: i did george church's $300 package. they theoretically gave me access to the raw sequences, but i only ever used the assemblies in variant-call files.
08:02 < kanzure> veritas? they mailed me my money back saying they were shutting down.
08:04 < jrayhawk> nebula
08:05 < jrayhawk> also they will heavily try to railroad you onto a dumb subscription service
08:05 < jrayhawk> that somehow manages to be less useful than promethease
08:07 < jrayhawk> at the $300 level, phasing data was very hit-or-miss, but it was otherwise pretty good
08:45 < kanzure> do they do any methylome stuff?
08:46 < jrayhawk> Nope
11:10 < L29Ah> .t https://arxiv.org/abs/2206.11090
11:10 < saxo> [2206.11090] Microscopic 3D printed optical tweezers for atomic quantum technology
11:23 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
11:23 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
11:32 -!- mirage335 [~mirage335@64.79.52.86] has quit [Quit: Client closed]
11:32 -!- mirage335 [~mirage335@64.79.52.86] has joined #hplusroadmap
12:36 < L29Ah> .t https://phys.org/news/2022-06-artificial-photosynthesis-food-sunshine.html
12:36 < saxo> Artificial photosynthesis can produce food without sunshine
12:44 < kanzure> yes we call it nonphotononsynthesis
12:46 < fenn> https://lacker.io/math/2022/02/24/godels-incompleteness-in-bash.html
12:46 < muurkha> sweet
12:47 < muurkha> fenn: was it you that told me about the security problem with sharing git repos?  I can't remember
12:47 < fenn> no
12:48 < fenn> jrayhawk was complaining about initializing a git repo in shared mode, or something like that
12:48 < muurkha> oh yeah
12:48 < fenn> escalation vulnerability
12:48 < muurkha> I mentioned it just now in https://news.ycombinator.com/item?id=31852847 but couldn't remember the details
12:49 < fenn> https://gnusha.org/logs/2022-04-28.log
12:50 < fenn> jrayhawk> man git-init, /shared
12:51 < jrayhawk> https://www.spinics.net/lists/git/msg298544.html
12:52 < muurkha> thanks!  link added
12:54 < muurkha> https://www.spinics.net/lists/git/msg298572.html makes it sound like they were going to fix it?
12:57 < muurkha> so maybe if it's not fixed it would be worth sending a patch
13:03 < fenn> muurkha: i prefer your directory listing to some web 2.0 div infested javascript nightmare, fwiw
13:06 < fenn> "If you want them to be able to push, you'll need to give them an account on the same server and either set umasks and group ownerships and permissions appropriately or set a POSIX ACL. Alternatively they can do the same thing on their server and you can pull from it. There are reportedly permission bugs in recent versions of Git (the last five years) that prevent this from being safe with people
13:06 < fenn> you don't trust"
13:06 < fenn> this makes it sound like pulling from someone else's server is unsafe
13:07 < fenn> i'd suggest moving "Alternatively..." to the end
13:07 < muurkha> it can be, depending on how you set it up
13:07 < muurkha> too late, edit window expired
13:07 < fenn> pff
13:07 < muurkha> sorry.  it was only about 10 minutes ago
13:08 < fenn> how long is the edit window? and is there some sort of "you must edit only once" rule?
13:08 < muurkha> 2 hours
13:08 < muurkha> no
13:09 < muurkha> I think a reasonable middle ground would include a header with a repository title, a link to the cloning URL, some nice colors and gradients and rounding and dropshadows and fonts, header lines of hashes and commits that last modified each file, rendering README.md, and most importantly, the ability to navigate commit history
13:09 < fenn> yes you can do that in most web servers that have directory listings
13:09 < fenn> not commit nav links
13:10 < muurkha> hmm, I didn't know Apache's fancy indexing was fancy enough to query git for commit messages
13:10 < muurkha> https://git.libre-soc.org/?p=binutils-gdb.git;a=tree falls a bit short but it's partway there
13:10 < muurkha> but that requires running special server software
13:10 < fenn> readme and css and cat'ing a header
13:12 < jrayhawk> muurkha: the patch would have to generalize lockfile handling of stuff in the root of GIT_DIR into a subdir somehow, which is actually already special-cased specifically for HEAD and packed-refs for legacy support reasons, but does not other stuff, like shallow_XXXXXX, which is presently broken
13:16 < jrayhawk> judging from junio et al.'s level of investment in the obviously broken repository format, there's probably a number of other files involved that i haven't run into yet
13:16 < muurkha> fenn: yeah, it could do those things
13:16 < muurkha> jrayhawk: maybe, yeah
13:19 < jrayhawk> https://secure.fairlystable.org/cgit/piny-code/commit/?id=e405f85ab352f90802a00d26c110693d884cc01d https://secure.fairlystable.org/cgit/piny-code/commit/?id=69965d3c022b04dd26df5f5e2b1217c438df0b4c is the minimal workaround (that still leaves shallow pushes broken)
13:20 < muurkha> heh, secure.fairlystable
13:22 < muurkha> seems like a reasonable thing to propose, though it's not the fix Michael proposed and Junio approved.  did they never do it?
13:22 < jrayhawk> Nope.
13:22 < muurkha> maybe they'd be more willing to accept a patch doing the thing they already approved rather than a different patch fixing the problem in a different way
13:23 < muurkha> I mean, they're humans, I have no idea, I'm just guessing
13:23 -!- heath [~heath@user/heath] has quit [Quit: WeeChat 1.7]
13:23 -!- heath [~heath@user/heath] has joined #hplusroadmap
13:25 < jrayhawk> yeah, i don't think me or micheal comprehended the extent of what lockfiles had broken in that thread (though junio probably understood what a bandaid the proposal was, which was part of his incredulity)
13:33 < jrayhawk> i proposed throwing money at a couple different folks to get something, but all of them cited not wanting to work with junio
13:33 < jrayhawk> er, to get something working
13:34 < jrayhawk> and i don't know C myself
13:37 < muurkha> heh, Junio seems a bit abrasive even in that thread
13:40 < nmz787> jrayhawk: you don't know C!!!!
13:41 < jrayhawk> i am a disgrace to my uniform
13:42 < nmz787> I just spent an hour or more screwing with my mouse left-click switch, but seem to have revived it from automatically double-clicking thanks to the pics+explanation on this page https://www.gonnalearn.com/inside-omron-d2fc-f-7n-microswitch/
13:43 < muurkha> now you have a US$100 mouse
13:43 < muurkha> (plus whatever the retail price was)
13:43 < nmz787> (you know you have ADD when you can't easily justify spending $6 for 5 new switches on Amazon next day delivery, because you want 'your mouse' working NOW)
13:43 < jrayhawk> oh yeah, i had that exact same problem with my G500
13:43 < nmz787> muurkha: well, at least I got it free from work
13:43 < nmz787> and it was used, so, I guess the company may have got it's moneys worth
13:44 < jrayhawk> i think i still have one working and one broken G500 in a box if you want parts
13:44 < nmz787> the old "take it apart, blow, spray solvent, and put it back together" trick
13:44 < muurkha> you must like it a lot
13:45 < muurkha> that reminds me, my girlfriend likes her mouse a lot, I should buy some extras for when it breaks
13:45 < nmz787> jrayhawk: nah I guess not, if it fails again I'll just 'buy it now' and let USPS do the legwork of driving around
13:45 < nmz787> muurkha: this is an M570... it's pretty nice
13:46 < muurkha> not familiar
13:47 < muurkha> is there a mouse enthusiast community like the keyboard enthusiast thing?  a friend of mine has a mouse that has thumbwheels that adjust the angle and height of each of its surfaces
13:47 < muurkha> it looks vaguely like a small model aircraft
13:47 < jrayhawk> gamers are meganerds about mice
13:48 < muurkha> yet somehow razer still has sales
13:50 < muurkha> jrayhawk: C is fun.  relaxing.  you should try it sometime
13:50 < muurkha> the first one's free
13:51 < nmz787> LOL
13:51 < nmz787> relaxing when your shit works and you can stop coding, maybe
13:53 < muurkha> well, aside from its hidden semantic complexities, it's very explicit about almost everything
13:54 < muurkha> so it's a nice respite from systems where everything is magic and you write five lines of code and have an all-singing all-dancing web service
13:54 < muurkha> which crashes once a week and you don't know why
13:55 < muurkha> and runs ten times slower than it should for, also, no apparent reason
13:55 < muurkha> also though when you write code in C you can use it in any language
13:56 < muurkha> (except browser JS or GLSL I guess)
13:56 < muurkha> and on any OS
13:56 < muurkha> and even on microcontrollers usually
13:56 < muurkha> so, although getting it to work takes more effort, you only have to do it once instead of 300 times
13:57 < muurkha> ...in theory :)
14:35 -!- spaceangel [~spaceange@ip-78-102-216-202.net.upcbroadband.cz] has quit [Remote host closed the connection]
15:13 < nmz787> apparently emscriptem and webasm are changing that too
15:14 < nmz787> I have my fair share of C and C++
15:14 < nmz787> even helped on a webserver (uwsgi)
15:14 < nmz787> and plenty of micros, linux stuff
15:15 < nmz787> I am proud to say C++ was my both my first computer language class and first college class, at age 14
15:16 < nmz787> but am glad to use Python when I can, and break things out into a C extension for performance
15:16 < nmz787> micropython was pretty cool on an ARM MCU
15:17 < nmz787> that was nice because I could do all my register programming and other simple co-routine sort of stuff in Python, but then for a specific few functions jump to custom C functions for doing DMA and ADC and processing statistics on data blocks (i.e. light DSP)
15:20 < nmz787> err, maybe it wasn't uwsgi
15:24 < nmz787> uWebSockets
15:25 < nmz787> sort of a C++ version of Python Flask style coding
15:28 -!- Llamamoe [~asmag@178235178198.dynamic-4-waw-k-1-2-0.vectranet.pl] has quit [Quit: Leaving.]
15:52 < TMA> at the University, the teacher of the C/C++ course told us that "C is an assembler with a chassis"
16:10 < kanzure> nmz787: how confident are you that you have all the edge cases of C/C++ down?
16:10 < kanzure> all the relevant ones, at least
16:40 < kanzure> when jrayhawk says he doesn't know it's probably not because he can't figure out how to place curly braces
16:41 < kanzure> ( { if x>2) ++x }
16:43 < jrayhawk> improper curling got me kicked out of physical education, too
16:52 < kanzure> https://whyy.org/segments/startup-offers-genetic-testing-that-promises-to-predict-healthiest-embryo/
16:55 < nmz787> kanzure: not as confident as in Python
16:55 < nmz787> reusing code is for-the-win (usually)
16:56 < nmz787> usually it's a lot of goofy type casting stuff that gets me stuck in C/C++
17:03 < muurkha> TMA: yeah
17:03 < muurkha> nobody knows all the quirks of C
17:04 < muurkha> well, maybe half a dozen people know all the quirks of *standard* C
17:04 < muurkha> but it's a very small number compared to the people who use C
17:07 < nmz787> as evidenced by like, nearly all the CVEs out there
17:08 < nmz787> (I assume)
17:27 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 248 seconds]
17:54 < lkcl> TMA: that's an extremely common misconception unfortunately ("c is assembler") i'm really dismayed someone's trying to *teach* you that.
18:00 < lkcl> for people who've grown up with scripted programming languages that "hide" direct access to libc6, system calls, and assembler, there seems to be a lot of puzzlement
18:07 -!- wrldpc [~igloo@cpe-67-255-88-60.maine.res.rr.com] has joined #hplusroadmap
18:08 < wrldpc> Fuck the media lab
18:08 < wrldpc> The prodigal Ben has returned.
18:09 -!- wrldpc [~igloo@cpe-67-255-88-60.maine.res.rr.com] has quit [Remote host closed the connection]
18:16 -!- _flood [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Read error: Connection reset by peer]
18:17 -!- _flood [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
18:18 < fenn> he didn't last long
18:19 < kanzure> he mentioned you recently on facebook
18:19 < kanzure> i don't usually know those things but for some reason that event generated an email that got to me
22:38 < fenn> "Dolt is the first and only SQL database that you can fork, clone, branch, merge, push and pull just like a Git repository. Dolt is a version controlled database. Dolt is Git for Data."
22:45 < fenn> sounds great for cleaning up metadata
22:50 -!- faceface [~faceface@user/faceface] has quit [Quit: Lost terminal]
22:54 < fenn> "Bug in production? Create a copy of the database on your laptop, start your services, change the production data to speed debugging." sounds like there are going to be a lot more large scale data breaches in the future
--- Log closed Fri Jun 24 00:00:01 2022