00:00SPEAKERDo you have merch yet?
00:01SPEAKERNo merch.
00:02SPEAKEROh, come on.
00:03SPEAKERWe got to cold open some merch.
00:05SPEAKERWe have virtual merch.
00:06SPEAKERWe have virch.
00:07SPEAKERYeah.
00:08SPEAKERThere we go.
00:09SPEAKERBecause some people in our community
00:11SPEAKERmade a yellow text on purple background generator,
00:14SPEAKERand you can kind of gin up a t-shirt with that.
00:17SPEAKERWe were making Brad and Will NFTs.
00:20SPEAKERThey're trading cards.
00:21SPEAKERThere's $99 each.
00:24SPEAKERSome of the prizes include dinner with me.
00:26SPEAKERI don't know if that's a good prize, but it's what we have.
00:29SPEAKERPlay golf on one of our many golf courses.
00:32SPEAKERThere's a picture of me riding a purple and gold
00:35SPEAKERelephant with no shirt on.
00:36SPEAKERI don't know why.
00:37SPEAKERWhat?
00:38SPEAKERMan, how do you just get into NFTs now?
00:43SPEAKERI really thought the merch should
00:44SPEAKERbe just Brad's old monitors.
00:51SPEAKERThose monitors have gone to the great beyond, unfortunately,
00:54SPEAKERat this point.
00:55SPEAKERI tried.
00:55SPEAKERI actually went on the Discord and said,
00:57SPEAKERhey, I'm about to e-cycle these monitors.
00:59SPEAKERDoes anyone want them?
01:01SPEAKERNo takers.
01:01SPEAKERBy e-cycle them, you mean you put them
01:03SPEAKERon a chair with a sign that said free outside your front door,
01:05SPEAKERright?
01:06SPEAKEROh, I scheduled actual e-waste pickup.
01:09SPEAKERWow.
01:10SPEAKERWell, they come pick it up?
01:11SPEAKERI thought long and hard about it.
01:13SPEAKEROK, I started to donate them to a goodwill situation.
01:17SPEAKERAnd even they said no.
01:19SPEAKERThey were like, no, thank you.
01:21SPEAKERI didn't actually try.
01:22SPEAKERI mean, I considered doing that.
01:23SPEAKERI considered going on Craigslist and saying,
01:25SPEAKERlike, hey, free monitors, just come pick them up.
01:26SPEAKERAnd then I thought about what's in those monitors
01:29SPEAKERand how old they are.
01:30SPEAKERYeah, there's a lot of guinea pig dust in there.
01:32SPEAKERNo, I mean, like, the cold cathode backlights.
01:35SPEAKERLike, that's how old those monitors are,
01:36SPEAKERis that they're way, way pre-LED.
01:39SPEAKERAnd I thought, like, the amount of electricity these things use
01:42SPEAKERand the closeness to their death they probably are
01:46SPEAKERis actually just--
01:46SPEAKERthat's like that's doing someone a disservice
01:49SPEAKERto make them take these monitors and use them,
01:51SPEAKERbecause they're going to suck up way too much power,
01:53SPEAKERand they're going to die at any moment.
01:54SPEAKERTaking them out of circulation was the right choice, I think.
01:56SPEAKERYeah.
01:57SPEAKERI think you should have slapped a Content@Town sticker on them,
02:00SPEAKERand they should have been at a Patreon rewards here.
02:04SPEAKERLook, Kishore, I'm not going to lie.
02:06SPEAKERI looked up to see how much it would
02:08SPEAKERcost to get purple sweatpants with yellow Will and Brad
02:12SPEAKERslapped on the ass cheeks would cost.
02:14SPEAKERUnfortunately, it was too much.
02:16SPEAKERThat's our hockey-- that's the hockey jersey of this podcast.
02:19SPEAKERI don't know of any prices too much for that.
02:21SPEAKERThe real question is, who goes on which cheek?
02:24SPEAKERWell, Brad goes on left cheek, and I go on right cheek.
02:26SPEAKERClearly, it's Brad and Will.
02:27SPEAKERThe "and" is in the middle.
02:28SPEAKER[MUSIC PLAYING]
02:57SPEAKERWelcome to "Brad and Will Made a Tech Pod."
02:58SPEAKERI'm Will.
02:59SPEAKERI'm Brad.
03:00SPEAKERJoining us today is a very special guest
03:02SPEAKERand our dear friend, Kishore Hari.
03:04SPEAKERWelcome back to the show, Kishore.
03:06SPEAKERI'm really excited to be here.
03:08SPEAKEREither this means the world is ending yet again,
03:12SPEAKERor there is a sale on bidets somewhere.
03:15SPEAKERThose are the only two options for me
03:17SPEAKERappearing on this podcast.
03:18SPEAKERYour bidet-fluencer gig is starting to pick up, I've heard.
03:21SPEAKERSo luckily, the world's not ending again.
03:25SPEAKERI didn't introduce you as our apocalypse consultant.
03:28SPEAKERSo we're here to talk about science today, though.
03:31SPEAKERIt's going to be fun.
03:32SPEAKERBecause I feel like we introduced you to this before.
03:35SPEAKERBut in case people don't know, you
03:36SPEAKERare a professional science communicator.
03:38SPEAKERYou tell people-- you explain the science to the people
03:41SPEAKERin a way that is digestible and understandable and fair
03:43SPEAKERand just and conveys the important stuff
03:47SPEAKERwithout being sensational or--
03:49SPEAKERlook, I don't know.
03:51SPEAKERAm I doing a good job explaining?
03:52SPEAKERI'm not a communicator.
03:53SPEAKERYou're really setting me up for failure, because I really
03:57SPEAKERwanted to talk at people instead of talk with.
03:59SPEAKERNow I feel like I have to be reasonable in my descriptions
04:04SPEAKERof science this year.
04:07SPEAKEROh, no, no, no.
04:07SPEAKERThere's nothing reasonable about science.
04:09SPEAKERThat's well-established.
04:11SPEAKERSo we're doing something a little bit different this week.
04:13SPEAKERWe wanted to do kind of a, hey, here's
04:15SPEAKERthe stuff that we thought was interesting in science
04:17SPEAKERthis year.
04:18SPEAKERIt's not necessarily the biggest stories,
04:20SPEAKERbut it's the stuff that we thought
04:21SPEAKERwere kind of across the board.
04:23SPEAKERKind of a show and tell.
04:24SPEAKEREverybody picked a couple of stories from this year
04:27SPEAKERthat were, let's say, inspiring.
04:29SPEAKERYeah.
04:30SPEAKERI would say this is not the indie hits of science
04:32SPEAKERthis year, because there were so many big science stories that I
04:36SPEAKERthink caught our attention that we are playing the hits today.
04:40SPEAKERBut there's a reason we're playing the hits,
04:42SPEAKERbecause the hits were amazing this year.
04:45SPEAKERYeah, and there's also--
04:48SPEAKERI looked at mRNA impacting cancer vaccine.
04:53SPEAKERThe work that went into the mRNA vaccines
04:54SPEAKERseems like it's going to completely change
04:57SPEAKERa lot of infectious disease stuff in the next three
04:59SPEAKERto five years.
05:00SPEAKERBut there was so much happening there,
05:02SPEAKERI kind of couldn't find one thing
05:03SPEAKERto pick, because there were 50 little, I mean,
05:07SPEAKERlittle big things that came out this year about it.
05:09SPEAKERSo we're not going to talk about that just upfront.
05:11SPEAKERIt's always hard to pick out science stories,
05:13SPEAKERbecause I think we default to discovery
05:17SPEAKERas a way of selecting science stories.
05:19SPEAKERBut I think we can all relate to the fact
05:22SPEAKERthat science is incremental.
05:24SPEAKERThere are all these mini progresses that
05:26SPEAKERhappen that lead to this thing that we label as discovery,
05:30SPEAKERbut it's really the agglomeration
05:32SPEAKERof all of that work together.
05:33SPEAKERSo when you're talking about stuff like vaccine development
05:37SPEAKERor drug development or a myriad of other things in health
05:40SPEAKERor in numerous other scientific fields,
05:43SPEAKERpicking the point at which you say this is a huge story
05:48SPEAKERis a little bit arbitrary.
05:50SPEAKERAnd so with that qualification, let's
05:52SPEAKERtalk about some arbitrary stories
05:53SPEAKERthat we found pretty interesting.
05:56SPEAKERHonestly, my guiding principle here
05:59SPEAKERwas, what did I find inspiring?
06:00SPEAKERWhat fired the imagination for me is kind of how I picked.
06:05SPEAKERShould we start with the most beautiful story of the year?
06:07SPEAKERYeah, I think so.
06:09SPEAKERWhich one-- that could be multiple things.
06:12SPEAKERI think we'll start with the just wonderful telescope,
06:14SPEAKERand which is what I encourage everyone else to call it
06:17SPEAKERfor reasons.
06:18SPEAKERIf you Google his name, you'll discover
06:20SPEAKERreasons why you should just call it the just wonderful telescope.
06:23SPEAKERAnd I'm going to start with a non-science element of what
06:26SPEAKERJWST brought us this year, and that's just pure beauty.
06:33SPEAKERI don't think it can be understated
06:35SPEAKERwhen I rewind the clock to the mid '90s how impactful it was
06:40SPEAKERwhen I saw the Pillars of Creation photo from Hubble.
06:45SPEAKERLike we all know, if we say the word Pillars of Creation,
06:48SPEAKERan image pops up in our head.
06:50SPEAKERAnd the meaning of those images to feeling
06:55SPEAKERthe imagination and inspiration of people across the world,
06:59SPEAKERin addition to all the people that
07:00SPEAKERhave talked about how those images encouraged
07:03SPEAKERthem to become astronomers and astrophysicists and the like,
07:07SPEAKERcan't be understated.
07:08SPEAKERAnd there is so much that emerged from JWST
07:11SPEAKERthis year that was just beautiful.
07:12SPEAKERPut aside the science for a second.
07:14SPEAKERMy personal favorite is a spiral galaxy.
07:17SPEAKERIt's NGC 628.
07:19SPEAKERThere's image with the MIRI instrument,
07:21SPEAKERwhich is the mid-infrared instrument and the JWST.
07:25SPEAKERAnd it's a nearly perfectly symmetrical spiral galaxy.
07:29SPEAKERHow does our universe make something
07:32SPEAKERso symmetrical at the scales that we're talking about here?
07:36SPEAKERAnd then to me, the image that I have framed in my house
07:41SPEAKERis the zoom-in of the Carina Nebula,
07:43SPEAKERwhich is a stellar nursery.
07:46SPEAKERAnd I just found the overlay of the zoomed-in part
07:50SPEAKERthat JWST took of the Carina Nebula over the Hubble image.
07:56SPEAKERIt still brings a tear to my eye,
07:57SPEAKERthinking about the vastness that that represents
08:00SPEAKERand the birth of stars and what I hope is the birth of life.
08:06SPEAKERAnd then there's the actual kind of science part.
08:09SPEAKERAnd I think the most important science to me
08:12SPEAKERthat JWST brings about is its work on looking
08:16SPEAKERat atmospheric composition.
08:18SPEAKERAnd this year we saw the initial signs of what it can do.
08:22SPEAKERSo it looked at one planet it looked at.
08:26SPEAKERAnd it looks at this by looking at the light of the star
08:28SPEAKERshining through the atmosphere of this planet
08:31SPEAKERand kind of coming into the instrument.
08:34SPEAKERAnd from that, you can use a spectrometer
08:36SPEAKERto actually analyze the potential atomic composition.
08:40SPEAKERSo it looked at a planet in the WASP-96 area, WASP-96b.
08:48SPEAKERThis is a planet that's 0.48 the mass of Jupiter.
08:54SPEAKERBut it's so close to its sun that it orbits it every 3
08:59SPEAKERand 1/2 days.
09:00SPEAKERIs it a 3 and 1/2 day long year?
09:02SPEAKERHoly cow.
09:03SPEAKERAnd so just imagine a planet that's
09:05SPEAKERabout half the mass of Jupiter circling its star
09:09SPEAKERevery 3 and 1/2 days.
09:10SPEAKERIt's incredible.
09:11SPEAKERIt's an incredibly hot planet.
09:12SPEAKERIt's about 1,000 degrees.
09:14SPEAKERBut they were able to image water in its atmosphere.
09:19SPEAKERWater that close-- I mean, this star is not as hot as our sun.
09:22SPEAKERBut it was able to image that.
09:25SPEAKERAnd then you can imagine what this unlocks.
09:27SPEAKERAnd just this weekend, as we're recording this,
09:31SPEAKERthere was an astronomy conference
09:33SPEAKERwhere they started to unveil how they studied for about a week
09:37SPEAKERand 1/2 planets in the TRAPPIST-1 system.
09:41SPEAKERThis is a system that the Kepler telescope observed in 2017
09:47SPEAKERand identified seven planets around a star that's
09:50SPEAKERstill cooler than the sun.
09:51SPEAKERIt's about 39 light years away.
09:54SPEAKERAll the planets in this system are closer in its orbit
09:58SPEAKERthan Mercury is to our sun.
10:00SPEAKERSo it's a really tight orbital structure.
10:03SPEAKERAnd they're all rocky planets.
10:05SPEAKERAnd because these planets are much smaller
10:08SPEAKERthan the Jupiter-like planet that I talked about earlier,
10:11SPEAKERit's hard to tease out the signal.
10:13SPEAKERBut over the next few months, they're
10:15SPEAKERgoing to be deeply analyzing that data.
10:17SPEAKERAnd they're already getting some atmospheric composition.
10:20SPEAKERSo they looked at one of the planet that
10:23SPEAKERis closest to that star.
10:25SPEAKERAnd they found a significant amount
10:27SPEAKERof hydrogen in its atmosphere.
10:29SPEAKERWe're really going to be looking at this over the next year
10:33SPEAKERand seeing what do we know about the atmospheric composition.
10:37SPEAKERAnd when we start to see water in a planet that's
10:40SPEAKERin the Goldilocks zone, what could that possibly mean?
10:44SPEAKERThe imagery, the long-distance imagery
10:47SPEAKERthat's been dominating the dews has been so prevalent.
10:49SPEAKERI had no idea that they were even using the telescope
10:51SPEAKERfor planetary science.
10:52SPEAKERThat's wild.
10:53SPEAKERIs-- I don't know a ton about like xenobiology.
10:57SPEAKERIs the presence of water still considered
10:59SPEAKERlike the necessary precondition for life to emerge?
11:02SPEAKEROr is it just considered like one of many?
11:05SPEAKERI mean, obviously, there are like other liquids,
11:07SPEAKERyou know, other conditions.
11:08SPEAKEROr is that still the holy grail?
11:10SPEAKERIt is one of many.
11:12SPEAKERBut without water, we can't even have a starting conversation.
11:16SPEAKERIt's amazing.
11:17SPEAKERThe images that strike me out of JWST,
11:22SPEAKERlike when they've gone through and recreated or redone
11:24SPEAKERthe deep field surveys where 30 years ago, I guess,
11:29SPEAKERthey aimed Hubble at a small patch of what
11:31SPEAKERwe thought was dark sky and just kept taking pictures there,
11:34SPEAKERkept taking pictures there, until there was a big composite
11:37SPEAKERof what looks like 1,000 galaxies.
11:41SPEAKERAnd the deep fields of JWST are even more breathtaking
11:46SPEAKERthan those images after 30 years of doing these.
11:49SPEAKERYou'd think we wouldn't be surprised that there's just
11:51SPEAKERalways more stars, always more galaxies.
11:53SPEAKERBut here we are.
11:55SPEAKERDo you think there's potential for the big number
11:58SPEAKERto go up again?
11:58SPEAKERI feel like it was like, what, 20 years ago that it was like,
12:01SPEAKERoh, there are roughly, I don't know, 100 billion known
12:03SPEAKERgalaxies.
12:04SPEAKERAnd now I think that number's up to like 2 trillion.
12:06SPEAKERI'm not sure the number of galaxies matter.
12:09SPEAKERThe thing that woke up is I think
12:10SPEAKERwe have a very good handle on the mass of the universe.
12:14SPEAKERSo these are all kind of rounding errors
12:18SPEAKERin the larger scheme of things.
12:20SPEAKERSo yeah, maybe there's a little bit more visible light out
12:23SPEAKERthere than what we think.
12:24SPEAKERBut it still does not explain the giant gap
12:28SPEAKERwe have in the mass of the universe versus what
12:31SPEAKERwe are able to see and what is emitting light.
12:35SPEAKERSo even if there is a change like you're talking about,
12:38SPEAKERit's still an overall rounding error.
12:40SPEAKERI think it matters more to our psyche knowing
12:43SPEAKERthat the vastness of space is just incredible to behold.
12:48SPEAKERCan we talk about dark energy now?
12:50SPEAKEROh.
12:51SPEAKERSorry, I know you have a great segue here.
12:53SPEAKERThat was an amazing segue.
12:54SPEAKERI'm sorry.
12:55SPEAKERJust table it for two seconds.
12:56SPEAKERI'm just going to ask one more question here.
12:59SPEAKERYou're talking about JWST being used to study known exoplanets.
13:03SPEAKERDo you know if it's also going to be employed in detecting more
13:06SPEAKERunknown exoplanets?
13:07SPEAKERBecause I know the existing methods like, what is it,
13:09SPEAKERkind of planetary transit across the star
13:12SPEAKERand studying wobble of star under gravitational influence,
13:16SPEAKERall that stuff?
13:17SPEAKERDo you think new methods of finding more planets
13:19SPEAKERare going to emerge from this thing?
13:20SPEAKERYeah, I don't know.
13:23SPEAKERMy hunch from talking to some astrophysics colleagues
13:27SPEAKERis because time on the telescope is so precious,
13:31SPEAKERbecause there's so many demands on it, that in the near term,
13:34SPEAKERmost of the proposals they're accepting
13:36SPEAKERand areas they're going to point at
13:37SPEAKERare places that we've already imaged before.
13:40SPEAKERSo don't expect JWST to be pointing at just unknown areas
13:45SPEAKERwhere it's just going to randomly find a transit,
13:48SPEAKERjust because we have so many targets already
13:50SPEAKERto look at more deeply.
13:52SPEAKERNow, four or five years from now, 10 years from now,
13:56SPEAKERis that going to be the case?
13:57SPEAKERI don't know.
13:58SPEAKERThat's a good question.
14:00SPEAKEROK, Will, we can move on.
14:01SPEAKERI would love to talk to whoever it is that
14:03SPEAKERadministrates time on that thing.
14:05SPEAKERWho is the gatekeeper for who gets to use that thing
14:07SPEAKERand for how long?
14:08SPEAKERIt can't be a fun job.
14:10SPEAKERI mean, typically they have a committee on stuff like that.
14:13SPEAKERSo people review papers and judge
14:17SPEAKERbased on the relative merits and what
14:19SPEAKERthey think is going to be the outcome or the take, the haul.
14:23SPEAKERBut yeah, it seems like a lot.
14:24SPEAKERAll right, dark energy, there's your segue.
14:27SPEAKERYeah, dark energy, perfect segue.
14:29SPEAKERI learned about quintessence this year, which
14:31SPEAKERI didn't know about before.
14:34SPEAKERFor people who aren't familiar, when Einstein came up
14:37SPEAKERwith general relativity 100 years ago,
14:40SPEAKERhe added this thing called the cosmological constant
14:42SPEAKERbecause they felt like the universe should
14:44SPEAKERbe static at that time.
14:45SPEAKERThey're neither expanding nor contracting as it is,
14:48SPEAKERas it was, as it always will be.
14:50SPEAKERAnd then Hubble, a few years after that,
14:53SPEAKERdiscovered redshift and discovered
14:55SPEAKERthat that implied that the universe was expanding.
14:57SPEAKERAnd Einstein famously said that his biggest mistake was putting
15:00SPEAKERthe cosmological constant.
15:01SPEAKERAnd the biggest mistake of his life
15:03SPEAKERwas adding the cosmological constant.
15:05SPEAKERSo he pulled it back out of general relativity,
15:07SPEAKERthrew it in the trash.
15:08SPEAKERAnd then you fast forward 60, 70 years,
15:12SPEAKERand we were able to measure the distance to extremely distant
15:17SPEAKERsupernova very precisely.
15:19SPEAKERAnd we found, again, that the universe is accelerating,
15:21SPEAKERnot just expanding.
15:22SPEAKERSo it previously went from static
15:26SPEAKERto constantly getting bigger at the same rate
15:28SPEAKERto constantly getting bigger at a changing rate.
15:32SPEAKERAnd that implied the existence of dark energy,
15:36SPEAKERwhich is a force--
15:37SPEAKERin a world where the universe is made up only of matter
15:40SPEAKERand nothing else, then the universe
15:42SPEAKERwould eventually crunch down into a single point
15:45SPEAKERwhere an infinitely dense black hole, yada, yada.
15:48SPEAKERThe dark energy is the thing that we
15:49SPEAKERuse to balance out the sheet to make it seem as though--
15:54SPEAKERso there's something pushing out from the force of gravity
15:56SPEAKERthat would be pulling all things together.
15:58SPEAKERI'm just watching Kishore and hoping
15:59SPEAKERthat he continues nodding here, because as long as he continues
16:02SPEAKERnodding here, I know I'm doing OK.
16:05SPEAKERBut the idea that the power, the force of the dark energy,
16:10SPEAKERwhich still, frankly, seems kind of bullshit to me,
16:15SPEAKERis constantly expanding, is changing strength over time,
16:20SPEAKERis really interesting.
16:21SPEAKERAnd these three folks, Kazma and Andre, Ana Iyas, I think,
16:26SPEAKERand Paul Steinhardt did the math this year
16:31SPEAKERon their contestants' series.
16:32SPEAKERAnd if they're correct, the upshot
16:34SPEAKERis that there's a little less time than we expected before
16:37SPEAKERthe universe begins to contract.
16:39SPEAKERWell, I think there's been--
16:41SPEAKERmy understanding is for the last several years,
16:42SPEAKERthe thought was that it would never start to contract,
16:44SPEAKERthat the acceleration would only continue,
16:46SPEAKERand the universe would just expand forever.
16:48SPEAKERAnd I feel like that's gone back and forth
16:49SPEAKERat least once in our lifetime.
16:51SPEAKERSo we're arguing at a chalkboard right now.
16:53SPEAKERI want everyone to kind of put aside the idea
16:57SPEAKERthat physicists or astrophysicists
17:00SPEAKERare having these heated debates about, like, no,
17:03SPEAKERthe universe will contract.
17:05SPEAKERAnd no, they're having math arguments at boards,
17:08SPEAKERis what's going on.
17:09SPEAKERAnd Will's right.
17:10SPEAKERDark energy is basically a huge unknown
17:13SPEAKERin a number of mathematical equations
17:15SPEAKERthat explain an enormous part of the universe.
17:18SPEAKERBut it's something we've never observed, right?
17:20SPEAKERWe haven't observed it.
17:21SPEAKERWe don't have experimental dark that exists.
17:23SPEAKERIt's just an explanation, right, or a fudge.
17:26SPEAKERAnd the thing that we should acknowledge
17:28SPEAKERis there is this thing called inflationary theory that
17:31SPEAKERexplains, like, femtoseconds after the beginning
17:34SPEAKERof the universe.
17:35SPEAKERThere's this rapid expansion of the universe
17:39SPEAKERthat leads to all of these long-term consequences,
17:42SPEAKERincluding the continued acceleration,
17:46SPEAKERa part of the universe that, like, Sal Permoer and others
17:49SPEAKERreceived a Nobel Prize for from their work in the mid-'90s,
17:52SPEAKERestablishing that.
17:53SPEAKERNow, inflationary theory is just a theory.
17:55SPEAKERThere's a number of experimental projects
17:58SPEAKERlike BICEP2 in the Antarctic that
18:00SPEAKERhave tried to really confirm that in aspects
18:03SPEAKERof inflationary theory.
18:04SPEAKERAnd the three scientists you mentioned,
18:08SPEAKERthey study inflationary theory and are
18:10SPEAKERtrying to come up with explanations
18:12SPEAKERfor pieces of that theory that don't kind of work.
18:15SPEAKERAnd this idea of quintessence is really
18:17SPEAKERsaying, like, this cosmological constant--
18:21SPEAKERlike, there's no reason.
18:22SPEAKERThere's just this random constant for the universe.
18:25SPEAKERLike, there isn't, like, a pie for the universe.
18:27SPEAKERLike, what is this constant?
18:28SPEAKERSomething's constant where there's a number without which
18:30SPEAKERthe atoms wouldn't congeal, right?
18:32SPEAKERWhat is this constant represent?
18:34SPEAKERAnd quintessence is, like, an attempt
18:36SPEAKERto try to explain what that cosmological constant is.
18:41SPEAKERAnd if quintessence aligns with the theories
18:44SPEAKERthat these three have put forth, then there's
18:47SPEAKERthe possibility of the ramification of that
18:50SPEAKERcould be that the universe contracts.
18:55SPEAKERDo you all know Katie Mack?
18:57SPEAKERShe's a great astrophysicist.
18:59SPEAKERShe was actually Paul Steinhardt's student.
19:04SPEAKERHe was her thesis advisor.
19:05SPEAKERAnd she wrote a book called The End of Everything
19:08SPEAKERa couple of years ago that goes into all of these theories.
19:11SPEAKERLike, whether we just have a continual expansion
19:14SPEAKERand the heat death of the universe where just all heat
19:17SPEAKERjust sort of evaporates and then we're
19:19SPEAKERleft with the black void of nothingness,
19:21SPEAKERsome of these contraction theories,
19:25SPEAKERwhether the Galra from Voltron come in and take quintessence
19:29SPEAKERand transform it into energy, as I believe is canon.
19:34SPEAKERI missed that one in her book.
19:35SPEAKERYeah, that one got cut by the editors.
19:38SPEAKERI don't understand.
19:39SPEAKERIt must have been in the pre-reads, the proofs.
19:42SPEAKERBut I think this argument is--
19:44SPEAKERit's a decades-long argument that we're
19:47SPEAKERseeing continual incremental progress on.
19:50SPEAKERAnd the quintessence one is really--
19:54SPEAKERit's an interesting one.
19:55SPEAKERIt's a little mind-bending-y.
19:57SPEAKERAnd I think what becomes fascinating
20:00SPEAKERis as experiments come online to potentially explore
20:05SPEAKERelements that could give us hints and insights into this.
20:09SPEAKERYeah, and so it seems like--
20:11SPEAKERI don't know what those experiments are.
20:14SPEAKERAre we looking-- is this like a Nancy Roman telescope situation?
20:16SPEAKERIs this, you know--
20:18SPEAKERIt could be like BICEP2 is one that
20:20SPEAKERis looking at elements from early universe
20:25SPEAKERas well that would give us insight
20:27SPEAKERinto inflationary theory.
20:29SPEAKERSo yeah, there's a number of experiments.
20:31SPEAKERThere will never be the experiment,
20:33SPEAKERfrom what I understand, to confirm this.
20:35SPEAKERBut there are a series of experiments
20:38SPEAKERthat might give us insight into early inflationary time that
20:42SPEAKERcould give us some hype.
20:44SPEAKERBut I had a conversation with Brian Greene years ago.
20:48SPEAKERAnd he's like, we're reaching the edge where
20:51SPEAKERexperimental physics, like the amount of energy
20:53SPEAKERthat's required to explore some of these spaces with accuracy
20:58SPEAKERand precision, is just beyond our capability.
21:00SPEAKERWe'd have to start building things
21:02SPEAKERthe size of solar systems or galaxies
21:04SPEAKERto explore this to achieve the energy we'd need to look back.
21:08SPEAKERAnd that's just not possible.
21:10SPEAKERSo in this way, it's an exploration of the mind
21:15SPEAKERas much as anything else.
21:16SPEAKERMaybe this is how that perfectly spiraled galaxy got built.
21:21SPEAKERDo I-- sorry, do I understand right
21:23SPEAKERabout inflationary theory that the thought
21:25SPEAKERis that very minor features of that very early expansion,
21:30SPEAKERlike little nuances in the way matter was distributed,
21:33SPEAKERare what led to a lot of the current structures we see,
21:35SPEAKERlike the distribution of galaxies directly followed
21:38SPEAKERfrom very minor variations?
21:40SPEAKERYeah, I am upset on the universe's behalf
21:42SPEAKERfor what you called minor.
21:44SPEAKERBecause we're talking about like-- we're talking about things
21:47SPEAKERlike where it's like the energy to break apart
21:50SPEAKERthe bonds of atoms.
21:52SPEAKERLike we're talking about like-- you're talking about minor
21:54SPEAKERchanges in that kind of fundamental force.
21:58SPEAKERLike it's-- we're talking about energy soups of like magnitude
22:01SPEAKERthat we can't really conceive of.
22:05SPEAKERIt's just over a timescale that is so small
22:10SPEAKERthat it's hard to conceive, is what I think
22:12SPEAKERis really challenging about it for how I picture it.
22:16SPEAKERFemtoseconds, fractions of femtoseconds.
22:19SPEAKERAnd I'm sure I'm off by orders of magnitude.
22:21SPEAKERI think the tech pod science channel is probably
22:24SPEAKERgoing to correct me because it's probably
22:26SPEAKERlike more like 10 to the minus 35 seconds or something
22:29SPEAKERlike that.
22:31SPEAKERThe other thing I wanted to mention here
22:32SPEAKERis that I got very strong-- when I pull up this Google Docs
22:37SPEAKERdocument, I got very strong luminiferous ether vibes
22:40SPEAKERfrom the term quintessence.
22:42SPEAKERAnd I started thinking like maybe some
22:44SPEAKERof these teams of researchers need some like branding
22:47SPEAKERexercises.
22:48SPEAKERI guess you could say the same thing about dark energy though.
22:50SPEAKERWell, look, naming stuff is hard.
22:52SPEAKERWe all know that.
22:53SPEAKERHere at Brad and Will made a tech pod a name
22:56SPEAKERthat we came up with as a placeholder name
22:58SPEAKERand then accidentally published the podcast.
23:00SPEAKERYeah, oops.
23:03SPEAKERThe big question I have about all this stuff--
23:05SPEAKERso there were two interesting takeaways from this.
23:07SPEAKEROne is that the lower bound on when universal shrinking could
23:12SPEAKERstart is like 100 million years from now,
23:15SPEAKERwhich, while a long time on a human scale
23:18SPEAKERis relatively short on a cosmological scale,
23:22SPEAKERthat means we could see a universal contraction
23:25SPEAKERbefore the sun runs through its main course of fuel.
23:31SPEAKERAnd this is a novel--
23:33SPEAKERthe idea that the universe could fundamentally
23:35SPEAKERchange in the lifespan of our solar system
23:36SPEAKERis kind of novel to me.
23:38SPEAKERAnd I wonder, does universal contraction change things
23:44SPEAKERon a human scale?
23:45SPEAKERDo thermodynamics behave differently
23:47SPEAKERbecause things are getting closer to get--
23:50SPEAKERdoes anything change in the world 100 million years from now,
23:53SPEAKERassuming we make it that far?
23:54SPEAKERSo this is beyond my depth.
23:57SPEAKERBut from what I've read from Katie's work and some others,
24:01SPEAKERthe short answer is no.
24:02SPEAKEREven if the wildest kind of theories about this come true,
24:05SPEAKERit's still operating at a time scale where we'd all be dead.
24:09SPEAKERThe human race would be gone, just
24:11SPEAKERbecause even in the time scale you're talking about--
24:14SPEAKERSo like six, seven months from now?
24:16SPEAKERYeah, we have to expand well beyond our solar system
24:22SPEAKERand to really even consider that a possibility.
24:27SPEAKERSo I'm not willing to let climate change
24:30SPEAKERgo to the next generation.
24:31SPEAKERBut this one, I am really willing to let go.
24:33SPEAKEROK, fair.
24:35SPEAKERThis might be wildly oversimplified,
24:36SPEAKERbut I mean, you're kind of talking
24:38SPEAKERabout basic tenets of physics changing
24:40SPEAKERas the nature and scale of space changes, right?
24:42SPEAKERBut I mean, space is expanding rapidly right now,
24:44SPEAKERand we don't see--
24:45SPEAKERI wouldn't think about it that way.
24:47SPEAKERI really think about it as there's
24:51SPEAKERelements of our explanation of physics
24:54SPEAKERthat we have put placeholders in because we don't have
24:57SPEAKERa full definition of what that is.
25:00SPEAKERAnd so we can call that physics is changing.
25:02SPEAKERBut I think of it as more it's filling out.
25:05SPEAKERIt's a more complete physics that's
25:07SPEAKERcoming online as opposed to it's fundamentally
25:10SPEAKERchanging in some way.
25:12SPEAKERI don't have a good segue for the next one.
25:14SPEAKERWhat is the next one?
25:15SPEAKERI thought we'd talk about our friend--
25:19SPEAKERhuman endurance is the topic of the day, I think, here.
25:23SPEAKERIt's making lots of small incremental changes,
25:26SPEAKERadvancements in science over a long period of time.
25:30SPEAKERYeah.
25:30SPEAKERWhy don't we-- hey, science happened in the past, too.
25:33SPEAKERWhy don't we look backwards for a bit instead of forwards?
25:36SPEAKERSo this year, the wreck of the endurance,
25:38SPEAKERwhich was the ship of the final major expedition
25:42SPEAKERto the Antarctic in the kind of--
25:44SPEAKERwhat is the formal term for that period?
25:46SPEAKERIn the kind of late 19th, early 20th century,
25:48SPEAKERthe kind of heroic age of Antarctic exploration.
25:52SPEAKERLarge men with dogs traversing impossible terrain.
25:56SPEAKERA lot of European countries were funding expeditions
25:59SPEAKERto the Antarctic, probably for commercial and geopolitical
26:03SPEAKERreasons, if I had to guess.
26:05SPEAKERBecause, hey, it was a whole continent just sitting there
26:08SPEAKERwaiting to be dominated.
26:09SPEAKERIt was a 4x game back then, Brad.
26:11SPEAKERYes, yes.
26:13SPEAKERAnyway, are you guys familiar with that expedition?
26:16SPEAKERAnd Shackleton, he's a really--
26:18SPEAKERhe's kind of an amusing, larger than life figure.
26:21SPEAKERHe was like a huge drinker, apparently
26:23SPEAKERlike an enormously charismatic kind of womanizer.
26:26SPEAKERBut loved by all--
26:27SPEAKERSounds charming.
26:28SPEAKERLoved by all and sundry, though, apparently
26:30SPEAKERlike a very inspiring leader.
26:32SPEAKERWell, in the final analysis, he led this expedition down there
26:36SPEAKERthat never even made it to Antarctica
26:37SPEAKERand got stuck for two years, and nobody died.
26:40SPEAKERHe managed to, oh, engineer a rescue that saved all--
26:44SPEAKERI think it was 27 or maybe 28 men.
26:47SPEAKERThere are these incredible maps of the journey of where
26:51SPEAKERwe think the endurance went down.
26:53SPEAKERAnd then the rescue boat where they had to sail away
26:57SPEAKERand how they went onto land.
27:00SPEAKERAnd those maps are amazing because even--
27:03SPEAKERyou give me modern technology, I'm
27:04SPEAKERnot sure I survived that trip.
27:06SPEAKERYeah, it's crazy.
27:07SPEAKERSo essentially, they sailed to Antarctica in the endurance.
27:10SPEAKERAnd it got mired in pack ice and was trapped for,
27:14SPEAKERI think it was 10 months with various periods of drift
27:17SPEAKERhappening.
27:19SPEAKERIt got relatively close to Antarctica,
27:20SPEAKERbut never actually made landfall.
27:23SPEAKEREventually was so kind of crushed
27:25SPEAKERin the movements of the ice that it first
27:28SPEAKERstarted taking on a lot of water,
27:29SPEAKERand they had plenty of time to get all the supplies off,
27:32SPEAKERincluding, I'm sure, plenty of booze.
27:34SPEAKERAll the something like 70 sled dogs they had with them,
27:37SPEAKERet cetera, et cetera.
27:38SPEAKEREventually, it did sink after it had taken on enough water.
27:42SPEAKERYeah, this is not really germane to the discovery of the wreck,
27:45SPEAKERbut he and a small number-- so they made it to Elephant Island,
27:48SPEAKERestablished a base camp to try to survive.
27:51SPEAKERHe and a small number of men took a whaling boat
27:54SPEAKERand sailed 800 miles to the closest human civilization.
27:57SPEAKERIsn't a whaling boat like 30 feet--
28:00SPEAKERlike a 30 foot long rowboat in that period?
28:02SPEAKERThat is my understanding.
28:04SPEAKERSo I know absolutely nothing about kind of maritime
28:06SPEAKERmechanics or whatever.
28:07SPEAKERBut what I read about this is that trip in that whaling boat
28:11SPEAKERis considered one of the most heroic episodes of naval
28:14SPEAKERor marine travel in history, essentially.
28:18SPEAKERAnyway, that's all fun.
28:20SPEAKERAlso, another fun little side note,
28:22SPEAKERone of his previous expeditions down there--
28:25SPEAKERthe campsite was discovered about 12 years ago.
28:28SPEAKERI think it was around 2010.
28:29SPEAKERThey found one of his previous camps.
28:31SPEAKERAnd there were multiple intact cases
28:34SPEAKERof whiskey and brandy still there
28:35SPEAKERthat were over 100 years old buried in the ice.
28:39SPEAKERAnd there is actually-- it's not amazing.
28:42SPEAKERI've had it.
28:43SPEAKERThere is a recreation called Shackleton
28:45SPEAKERof the blended malt scotch whiskey
28:49SPEAKERthat was found at that base camp that is on the market.
28:52SPEAKERI don't know that I recommend it.
28:53SPEAKERBut anyway, it's a fun side story.
28:56SPEAKERAnyway, let's get to the shipwreck.
28:58SPEAKERSo yes, there have been multiple attempts
29:00SPEAKERto find the shipwreck over the last, say, 20 years, I think.
29:03SPEAKERI didn't find a lot of detail about why the one this year
29:05SPEAKERwas finally the one that was successful,
29:07SPEAKERalthough I'm sure you pointed out
29:09SPEAKERthat Saab, the car company, made the submersibles that
29:13SPEAKERwere used in finding the wreck.
29:14SPEAKERI love it.
29:15SPEAKERIt's not the James Cameron sitting a little Steve Zizzou
29:21SPEAKERlooking submersible and going down to the wreck.
29:24SPEAKERIt is-- Will, you remember going and seeing open ROV
29:28SPEAKERat Maker Faire all those years?
29:29SPEAKERYeah, of course.
29:30SPEAKERIt's a scaled-up version of that is what this thing looks like.
29:33SPEAKERIt is kind of a flat ROV that has a camera at the front.
29:39SPEAKERIt's much larger than that open ROV
29:41SPEAKERthat Dave Lang and Eric Stackmull designed.
29:44SPEAKERBut it's this kind of flat submersible
29:46SPEAKERthat is just kind of maneuverable
29:48SPEAKERand has a big-ass camera on it.
29:49SPEAKERBox with propellers and instruments attached.
29:51SPEAKERI mean, that was Dave's whole thing
29:53SPEAKERwhen he pitched that to us, was like, hey,
29:55SPEAKERI wanted to make a cheap way to get instruments
29:57SPEAKERto the bottom of the underwater places.
29:58SPEAKERAnd it turns out it's a smart way to do it.
30:01SPEAKERWhere did they find the--
30:02SPEAKERso I mean, I assume the luck plays into this a fair amount.
30:05SPEAKERYou just go down and start looking, right,
30:07SPEAKERwhere you think it's likely?
30:08SPEAKERThey had predictions about where it had drifted to.
30:11SPEAKERAnd they were not exactly accurate,
30:13SPEAKERbut it gave them, obviously, a search area.
30:15SPEAKERSo it was found at the bottom of the Weddell Sea
30:18SPEAKERat a depth of almost 10,000 feet, which I don't
30:21SPEAKERknow a ton of shipwreck deaths.
30:23SPEAKERI know the Titanic is, I think, around 6,000.
30:25SPEAKER10,000 sounds extremely deep to me.
30:27SPEAKERThere may be deeper ones out there, but that's very deep.
30:31SPEAKERThe thing-- well, there are a couple
30:33SPEAKERthings that sort of fire the imagination about the story.
30:35SPEAKERThe other one is the kind of tale of survival.
30:37SPEAKERBut the thing about the shipwreck,
30:38SPEAKERto me, that is so kind of marvelous
30:41SPEAKERis how remarkably well-preserved it is.
30:45SPEAKERAnd there are a couple of thoughts.
30:47SPEAKERI mean, I'm sure this is still under study,
30:49SPEAKERbut there are a couple thoughts about why that is.
30:51SPEAKERNumber one, it seems that cold water wrecks just
30:53SPEAKERtend to deteriorate more slowly than wrecks in warm water.
30:57SPEAKERBut the bigger one here is that the scientists on the expedition
31:00SPEAKERhave inferred that there are no wood-eating microorganisms
31:03SPEAKERat this depth and this low temperature
31:07SPEAKERat the bottom of the Antarctic like there are in warmer waters.
31:10SPEAKERSo that thing, I mean, there's plenty
31:13SPEAKERof very nice 4K footage out there of this thing.
31:15SPEAKERAnd it's kind of chilling almost to me, like ha-ha pun intended.
31:22SPEAKERBut I get a little--
31:26SPEAKERyou get chills looking at it because it is just
31:28SPEAKERsitting there intact on the bottom of the Antarctic.
31:31SPEAKERAnd the lettering, I don't know if that's brass or whatever,
31:33SPEAKERbut the endurance lettering on the whole
31:35SPEAKERis just there and in excellent shape.
31:38SPEAKERI mean, it's really just kind of remarkable to watch.
31:41SPEAKERYeah, I encourage people to watch the footage.
31:43SPEAKEROn the life question, so scientists
31:46SPEAKERhave been analyzing the footage because it was shot in 4K,
31:50SPEAKERas you mentioned.
31:51SPEAKERAnd there is a lot of life down there.
31:54SPEAKERAnd most of it is stuff that we would ascribe
31:58SPEAKERto as like filter feeders.
31:59SPEAKERAnd they're feeding on something we call marine snow.
32:03SPEAKERIt is basically algae that die and fall towards the bottom.
32:08SPEAKERAnd so there's a lot of stuff like starfish.
32:10SPEAKERThey found a couple of lobsters that
32:13SPEAKERkind of look like hermit crabs that kind of show up
32:17SPEAKERunder a UV light.
32:19SPEAKERSo it's teeming with life.
32:22SPEAKERBut there isn't the kind of bacterial organisms
32:25SPEAKERthat would digest wood.
32:28SPEAKERAnd there aren't the kind of currents
32:29SPEAKERthat we'd expect at the bottom of the sea that would just
32:32SPEAKERphysically rip it apart as well.
32:34SPEAKERThe other thing is that the wreck is in a protected area.
32:39SPEAKERSo you can't touch it.
32:41SPEAKERLike we can never--
32:43SPEAKERaccording to current law, we will never
32:45SPEAKERremove anything from it.
32:47SPEAKERSo I know like people at the British Museum are sad.
32:50SPEAKERThey can't steal another thing to add to their museum.
32:53SPEAKERBut it is just going to stay down there.
32:56SPEAKERAnd there's something about that that is lovely to me.
32:59SPEAKERAnd there's something about this that is also
33:01SPEAKERsad because we should have never been able to find this.
33:04SPEAKERThis is a story of climate change
33:06SPEAKERallowing us to find this ship.
33:09SPEAKERThis should have been under ice that we never
33:12SPEAKERcould have gotten through, right?
33:13SPEAKERThat is a depressing aspect of this that I--
33:16SPEAKERyeah, the fact that it is a protected historic site.
33:19SPEAKERIt's under treaty.
33:20SPEAKERI mean, it's actual international agreement
33:22SPEAKERthat has designated it as such.
33:25SPEAKERBut if you compare it again to something like Titanic
33:28SPEAKERthat has been repeatedly damaged by people exploring it
33:31SPEAKERand like Cameron literally landed a submersible
33:34SPEAKERon the deck of the Titanic at one point
33:36SPEAKERwhile he was down there shooting it.
33:38SPEAKERAnd so that wreck has just been abused excessively
33:40SPEAKERby explorers.
33:41SPEAKERSo like for this to sit there as a monument
33:45SPEAKERto this historical event and not be messed with
33:47SPEAKERis kind of beautiful in a way to me.
33:50SPEAKERWell, and this is a wood hull ship too, right?
33:53SPEAKERYes.
33:54SPEAKERSo like the metal hull of the Titanic
33:57SPEAKEReventually was going to rust and go away.
34:01SPEAKERThe fact that the wood has nothing eating it there
34:03SPEAKERmeans that this conceivably could be there
34:05SPEAKERfor a really, really long time, which I think is kind of lovely.
34:09SPEAKERAlso, nobody died on it, which is a really refreshing change
34:11SPEAKERfor most shipwrecks, it turns out.
34:15SPEAKERSo we've talked a lot about science.
34:18SPEAKERI kind of think it's time to do a little science.
34:20SPEAKERSo--
34:21SPEAKEROh.
34:23SPEAKERWow, that is a hell of a segue there.
34:25SPEAKERBrad, in the show notes, there's a link to a Google Colab site.
34:28SPEAKERAnd there's a sequence that I want you to put into that site.
34:31SPEAKERAnd we'll kind of come back to it in a couple of minutes.
34:34SPEAKERBecause I want to talk about AlphaFold and EMSFold,
34:40SPEAKERwhich I think are some of the most important developments
34:44SPEAKERin structural biology in a generation.
34:48SPEAKERAlphaFold is a product of DeepMind, Google offshoot.
34:52SPEAKEREMSFold is a project of a meta.
34:55SPEAKERI'll get into the differences here in a second.
34:59SPEAKERFirst, a slight kind of oversimplification
35:01SPEAKERto talk about how important proteins are.
35:03SPEAKERSo I'm on a tech pod.
35:06SPEAKERSo I'm going to kind of make a strained analogy here.
35:10SPEAKERSo we think of DNA as physical memory.
35:13SPEAKERRNA kind of works as sort of the RAM of the system.
35:18SPEAKERAnd proteins are the software.
35:19SPEAKERThey're the thing that does all the work that
35:23SPEAKERresults in the interfaces that actually meaningfully do stuff.
35:27SPEAKERSo protein functions, they make up antibodies, enzymes,
35:34SPEAKERstructural report in cells, messengers.
35:36SPEAKERThey transport molecules around our body.
35:41SPEAKERAnd the sequence of those proteins,
35:43SPEAKERwhich are made up of amino acids.
35:44SPEAKERAnd in humans, there's 20 common amino acids
35:47SPEAKERthat make up those proteins.
35:48SPEAKERTheir sequence determines their structure,
35:50SPEAKERand their structure determines their function.
35:53SPEAKERNow, when we talk about a system that has 20 amino acids,
35:59SPEAKERthe length of a protein is basically--
36:02SPEAKERit could be infinite.
36:03SPEAKERBut with 20 amino acids, so you just chain them together,
36:07SPEAKERand that forms these proteins, the possible proteins
36:10SPEAKERyou would consider would be like 20 to the n power,
36:13SPEAKERdepending on the length.
36:15SPEAKEREssentially infinite.
36:16SPEAKERBut because of constrained systems we live in,
36:18SPEAKERit's not infinite.
36:19SPEAKERThere's finite.
36:20SPEAKERBut it is a massive, massive number
36:22SPEAKERthat we think is possible out there.
36:25SPEAKERAnd so determining the structure of these proteins, especially
36:28SPEAKERthe ones that are known, are essential for us
36:30SPEAKERto unlock all sorts of information
36:32SPEAKERon how things operate in these systems.
36:37SPEAKERAnd right now, the way a lot of this is discovered
36:39SPEAKERis we will get a protein.
36:41SPEAKERWe'll actually image it.
36:43SPEAKERWe'll make it like X-ray crystallography or other methods.
36:46SPEAKERAnd basically say, this is like the sequence of a protein
36:49SPEAKERand kind of work backwards.
36:51SPEAKERWell, these AI systems came in and essentially said,
36:54SPEAKERwhat if we can actually predict the structure of the proteins
36:58SPEAKERbasically based off the sequences of proteins
37:01SPEAKERthat are known and the shapes of proteins that are known?
37:04SPEAKERSo AlphaFOD, a few years ago, took
37:06SPEAKERon this idea of we're going to train on this shared
37:09SPEAKERdatabase called a protein, the PDP.
37:12SPEAKERIt's a protein structure database.
37:14SPEAKERAnd we're going to enter these competitions called CAS.
37:16SPEAKERThese are annual novel protein prediction competitions
37:22SPEAKERthat are hosted by academics every year.
37:25SPEAKERAnd they're coming up with models
37:27SPEAKERto predict the proteins for these competitions?
37:28SPEAKERYeah, so AlphaFOD came in with an AI model
37:31SPEAKERto essentially say, we're going to predict the protein.
37:33SPEAKERSo what the competition would do is
37:35SPEAKERthey would actually go through and develop and do
37:38SPEAKERan X-ray crystallography on a protein,
37:41SPEAKERunderstand its structure, have the model
37:43SPEAKERtry to predict its structure without telling them
37:45SPEAKERthe information, and then compare the results at the end.
37:48SPEAKERIt's a little more complicated than that,
37:50SPEAKERbut that's the basic idea.
37:51SPEAKERAnd so AlphaFOD goes through a series of these competitions
37:54SPEAKERover the years and gets better and better
37:56SPEAKERand starts improving on the scale of these models.
37:59SPEAKERLast year, it essentially--
38:02SPEAKERthis is CAS 14, so it might be a year and a half ago.
38:05SPEAKERAnd essentially, there was no difference
38:07SPEAKERbetween the proteins in real life that we image
38:11SPEAKERversus with the predicted structure from AlphaFOD.
38:14SPEAKERAnd that was sort of heralded as like, holy crap,
38:17SPEAKERwe got something here.
38:19SPEAKERAnd then this year, AlphaFOD released
38:23SPEAKERa database of structures of their structure,
38:26SPEAKERtheir predicted structures of 200 million proteins
38:30SPEAKERin the human body.
38:31SPEAKERAnd their kind of theory of the case
38:35SPEAKERhere is that those 200 million proteins, those
38:39SPEAKERare the known proteins, and those
38:40SPEAKERwill be the limit of human proteins in the human body.
38:44SPEAKEROh, they think that's a complete set.
38:46SPEAKERThat is close to the complete set.
38:49SPEAKERAnd so I'm not a machine learning expert,
38:52SPEAKERso I'm going to kind of qualify this.
38:55SPEAKERBut they use this system called a--
39:00SPEAKERwhat's it called-- a multi-fold machine learning
39:02SPEAKERmodel versus meta EMS fold uses a transformer
39:06SPEAKERmodel, which is a natural language processing model.
39:09SPEAKERBut anyway, so they essentially use this model
39:13SPEAKERto essentially map how the chemical bonds actually
39:16SPEAKERshape and form so that they're able to predict
39:19SPEAKERthese structures.
39:20SPEAKERSo Brad, any chance that finished running?
39:25SPEAKERIt is, in fact, on.
39:25SPEAKERSo what I gave Brad, I gave him an input for a protein
39:29SPEAKERsequence, and he put it into AlphaFold and actually ran it.
39:35SPEAKERAnd the output of that Google Colab
39:37SPEAKERis that it shows you the structure of that protein
39:41SPEAKERand gives you a 3D rendering of it if you scroll down.
39:45SPEAKERYes.
39:46SPEAKEROh, it's actually literal 3D.
39:47SPEAKERYou can manipulate it.
39:48SPEAKERHand it around and zoom it.
39:50SPEAKERSo that sequence was hemoglobin, which I gave you.
39:55SPEAKERAnd so what you see on your screen
39:57SPEAKERis you just image the predicted sequence--
40:00SPEAKERthe predicted structure from AlphaFold of hemoglobin.
40:03SPEAKERAnd you can-- hemoglobin's a well-imaged protein,
40:07SPEAKERas you can imagine.
40:08SPEAKERYou can Google hemoglobin and compare it.
40:13SPEAKERAnd what you'll find is they're almost identical.
40:18SPEAKERThe place that they're not identical
40:20SPEAKERis that proteins also don't just exist upon themselves.
40:24SPEAKERThey interact with other cofactors,
40:26SPEAKERwhether it be with other proteins
40:28SPEAKERor with other confounding elements.
40:31SPEAKERSo in case of oxygen, there's--
40:33SPEAKERin case of hemoglobin, there's oxygen and other cofactors
40:35SPEAKERthat it's interacting with.
40:38SPEAKERAnd so the shape of hemoglobin that AlphaFold produces
40:42SPEAKERis super close.
40:43SPEAKERIt's like it's great for science,
40:45SPEAKERbut it's not perfect because it doesn't model that interaction.
40:49SPEAKERAnd so beyond the AlphaFold releasing this database
40:53SPEAKERso that you can now query for a structure of protein
40:55SPEAKERand then go potentially make it, which accelerates
40:58SPEAKERthe scientific process, scientists
41:00SPEAKERhave been hacking AlphaFold-- hacking AlphaFold.
41:04SPEAKERThat was air quotes.
41:05SPEAKERI realize that doesn't come through on podcasts.
41:07SPEAKERYeah, it's an audio podcast.
41:09SPEAKERSorry.
41:09SPEAKERThey've been hacking at AlphaFold
41:11SPEAKERto actually start to mimic how proteins interact as well.
41:16SPEAKERAnd so it's been fascinating how they're now taking AlphaFold
41:19SPEAKERand taking it to the next level.
41:21SPEAKERAnd so CASP 15 was just a few weeks ago.
41:25SPEAKERAnd AlphaFold actually didn't enter this year.
41:28SPEAKERBut scientists who took AlphaFold and built upon it
41:33SPEAKERdid enter.
41:33SPEAKERAnd they started to model how proteins interact.
41:36SPEAKERAnd it's showing just--
41:38SPEAKERthe reason I think there's a huge story is we now
41:41SPEAKERhave a system of predicting structural biology in a way
41:46SPEAKERthat we hadn't for years.
41:48SPEAKERAnd the level of acceleration that we're
41:50SPEAKERseeing in terms of how scientists are using this
41:53SPEAKERis accelerating really, really quickly
41:55SPEAKERto the point now where scientists don't even
41:58SPEAKERhave to guess a protein structure.
42:00SPEAKERThey can go.
42:01SPEAKERThey can query it.
42:02SPEAKERThey can, in a matter of minutes,
42:03SPEAKERimage it and create a predicted structure
42:06SPEAKERand then go to the lab to actually do
42:08SPEAKERsome of the work on the confounding
42:11SPEAKERfactors and some of the hard work about putting this
42:13SPEAKERinto practice.
42:15SPEAKERMeta's work is very similar.
42:17SPEAKEREMSFold predicted 600 million proteins
42:20SPEAKERbecause it was across multiple organisms.
42:23SPEAKERBut they used a natural language model,
42:26SPEAKERwhich isn't as accurate as AlphaFold because it isn't
42:29SPEAKERlooking at the geometry of the potential shapes
42:32SPEAKERof the actual compounds themselves.
42:37SPEAKERBut from this transformer-based machine learning model,
42:41SPEAKERyou also get a much quicker response.
42:44SPEAKERSo what takes 10 minutes from AlphaFold
42:46SPEAKERtakes only a couple minutes from EMSFold.
42:50SPEAKERSo the amount of information that you
42:52SPEAKERcan generate so that you can get to some
42:54SPEAKERof the experimental work is so fast now
42:58SPEAKERthat it's just reshaping the entire landscape of how
43:01SPEAKERproteins work.
43:02SPEAKERThis is the baseline work that we
43:04SPEAKERneed to do to think about any development of drug
43:07SPEAKERtargets, of disease modeling, mechanisms of how cells function
43:13SPEAKERand work.
43:13SPEAKERIt's an incredible tool.
43:15SPEAKERThe interesting-- I was going to say the interesting thing
43:18SPEAKERabout this is that understanding these connections
43:21SPEAKERand understanding how different things interact with it
43:23SPEAKERlets us derive better drugs and things
43:25SPEAKERlike that that can bind directly to protein sites.
43:28SPEAKERBecause a lot-- so for people who
43:32SPEAKERaren't biologists or don't remember their high school
43:34SPEAKERbiology, proteins basically let your cells
43:37SPEAKERdo chemical reactions at lower temperatures than they would--
43:41SPEAKERtemperatures won't kill you, basically, for all--
43:44SPEAKERis the easiest way to describe it.
43:45SPEAKERIt does a bunch of other stuff as well.
43:47SPEAKERBut it does that by grabbing onto something,
43:49SPEAKERand then the protein moves in some way that
43:51SPEAKERwas ineffable and understandable until very recently.
43:54SPEAKERAnd now we have a model for that that we can get in and look at.
43:57SPEAKERAnd it's really, really cool stuff.
44:00SPEAKERIt's fascinating.
44:01SPEAKERThe people behind AlphaFold are on their way to Nobel Prize.
44:04SPEAKERThey won a breakthrough prize this year,
44:07SPEAKERwhich is definitely a good sign that things
44:09SPEAKERare on the horizon for them.
44:11SPEAKERI will say one thing about EMS Fold
44:13SPEAKERthat I think is really important is EMS Fold, very different
44:16SPEAKERmodel, because it's a natural language model,
44:20SPEAKERit really looks at sequences really well.
44:23SPEAKERIt's going to help us look at mutations in proteins
44:27SPEAKERa little more efficiently than AlphaFold will.
44:29SPEAKERSo while AlphaFold can really do incredible work on just
44:33SPEAKERthe fundamental proteins that exist,
44:35SPEAKERyou can go to a database called Uniprot,
44:38SPEAKERwhich is like a database of known protein sequences,
44:41SPEAKERpull out a sequence, plug it into AlphaFold.
44:43SPEAKEREMS is going to start to predict mutations
44:46SPEAKERin those proteins.
44:47SPEAKERIt'll help us attack a whole host
44:49SPEAKERof other biological diseases and issues.
44:56SPEAKER- Who owns this work, Kishore?
44:58SPEAKERIs this owned by the companies that do the models?
45:00SPEAKER- Who owns what?
45:02SPEAKERSo the code is owned by AlphaFold, which is part of Alphabet.
45:07SPEAKERBut the output is not, I mean,
45:09SPEAKERyou're basically running a piece of software
45:12SPEAKERand they don't claim proprietary ownership
45:15SPEAKERof the sequence that you put into it,
45:17SPEAKERnor the output from what I understand.
45:20SPEAKER- Fascinating.
45:20SPEAKER- The site you sent me is interesting.
45:22SPEAKERI don't know what the name is for this particular
45:25SPEAKERGoogle product that allows you to run code
45:27SPEAKERon some kind of cloud instance.
45:28SPEAKER- Oh, that's a Google Colab.
45:30SPEAKERSo you can basically create an instance
45:32SPEAKERand run all sorts of, you know,
45:35SPEAKERthere's all sorts of intersections
45:36SPEAKERto Google products and stuff.
45:38SPEAKERI've always used it for like cloud-based queries,
45:42SPEAKERbut the fact that AlphaFold is built into Colab
45:46SPEAKERso that like anyone can do it,
45:48SPEAKERyou can just go to this website and input a sequence.
45:51SPEAKER- Yes, that was gonna be my question
45:53SPEAKER'cause you can go step by step down here
45:54SPEAKERand expand each step of the processing
45:57SPEAKERand look at, you can literally look at the Python
45:59SPEAKERthat it's running to process this stuff.
46:01SPEAKERSo, okay, so knowing that they set that up
46:03SPEAKERahead of time is cool.
46:04SPEAKERI'm curious, I don't know that any of us could answer this,
46:07SPEAKERbut like, I would love to know like what the hardware,
46:09SPEAKERthe machine learning acceleration situation looks like
46:12SPEAKERfor a lot of these research teams.
46:13SPEAKERLike, are they running like a lot of like custom high-end
46:15SPEAKERASICs for this stuff?
46:16SPEAKEROr do they just have like racks of off the shelf 40 90s
46:20SPEAKERor like, I don't know.
46:21SPEAKER- It depends on their budget, Brad.
46:23SPEAKER- Yeah, I figured that was the case.
46:24SPEAKERBut I guess the real question for me is,
46:26SPEAKERare they kind of limited for budgetary reasons,
46:31SPEAKERprimarily probably to consumer hardware
46:33SPEAKERor are there like much higher end,
46:35SPEAKERlike high-performance computing kind of solutions
46:37SPEAKERout there for stuff like that?
46:38SPEAKERI mean, obviously Google has deep pockets,
46:39SPEAKERso I'm sure this effort is fairly well-funded.
46:42SPEAKER- I think much has been written about this,
46:44SPEAKERbut I have not perused this myself.
46:46SPEAKERSo I don't know, but I bet you we could find out.
46:49SPEAKER- So we, Kishore, we know some people who run labs
46:53SPEAKERthat do a little bit of this kind of work
46:54SPEAKERand they do a mixture of cloud and local compute,
46:57SPEAKERdepending on what they want.
46:59SPEAKERGenerally, my understanding is that if you have the budget
47:02SPEAKERto do cloud stuff and you want faster results,
47:04SPEAKERyou might see faster results.
47:05SPEAKERBut if you have enough data to keep a big workstation
47:11SPEAKERwith 20 GPUs in it constantly running,
47:13SPEAKERthen it behooves you to do both.
47:17SPEAKERKishore, how are you doing on time?
47:18SPEAKERDo we have time to do a couple more or are you, okay.
47:22SPEAKERDo we want to talk about, do we want to talk about Dart next?
47:24SPEAKER- Yeah, yeah, let's Dart to the next topic.
47:27SPEAKER- Oh boy.
47:28SPEAKER- NASA's double asteroid redirection test,
47:31SPEAKERwhich the culminating event of which took place,
47:35SPEAKERI believe, on September 26th of this year.
47:38SPEAKERDart was conducted by NASA
47:40SPEAKERas a planetary defense coordination office,
47:44SPEAKERwhich is about as sci-fi a division sounding name.
47:47SPEAKER- I assume that's like three people sitting in a closet
47:49SPEAKERsomewhere in like an off-brand NASA campus someplace
47:53SPEAKERthat they don't let them out very much.
47:56SPEAKER- No way.
47:57SPEAKERI believe it's at, I think headquarters
48:01SPEAKERfor the planetary office is in DC,
48:03SPEAKERbut a lot of the work is run out of JPL.
48:05SPEAKERSo if you consider Pasadena, low rent.
48:08SPEAKER- Okay, no, Pasadena's nice.
48:09SPEAKERWe like Pasadena.
48:10SPEAKERWe're friends with people from Pasadena,
48:12SPEAKERbut it's still three people, right?
48:13SPEAKER- No, it's actually quite a large project.
48:16SPEAKERThis is really out of the Near Earth Observation,
48:20SPEAKERNear Earth Object Observations Program.
48:23SPEAKERAnd I, you know, it's one of those programs at NASA
48:26SPEAKERwhere they grant out money.
48:28SPEAKERSo there's people that work in-house,
48:30SPEAKERbut also like associated people out in the world too
48:33SPEAKERthat are funded to work on stuff.
48:35SPEAKERAnd they are just looking for stuff that might hit us
48:38SPEAKERor come close to us.
48:39SPEAKER- I feel better knowing that they're there then.
48:42SPEAKER- So this was an attempt to test one method
48:45SPEAKERof diverting the trajectory of an asteroid
48:47SPEAKERthat might collide with the earth.
48:50SPEAKERAnd in this case, they did that by literally slamming
48:53SPEAKERa spacecraft head-on at the exact opposite direction
48:56SPEAKERof the travel of the asteroid.
48:59SPEAKERThey chose, yeah, what's that?
49:01SPEAKER- Did it work?
49:02SPEAKER- Yeah, it worked.
49:03SPEAKERSo they reported results, I think about a month
49:05SPEAKERafter the impact or even less and said, yes, it worked.
49:10SPEAKERTheir threshold for success was to,
49:12SPEAKERI'm looking for the figure.
49:14SPEAKERThey considered altering the asteroid's orbital period
49:18SPEAKERby at least, I believe it was 73 seconds
49:21SPEAKERto be the threshold of success.
49:23SPEAKERAnd they actually altered it by about 32 minutes,
49:25SPEAKERgive or take.
49:27SPEAKERSo, you know, by NASA's own reckoning,
49:29SPEAKERthis thing was a resounding success.
49:31SPEAKER- Long history of under-promising and over-delivering.
49:35SPEAKER- They chose an asteroid called Dimorphos,
49:37SPEAKERwhich is itself a satellite of a larger asteroid
49:41SPEAKERcalled Didymos, I assume is how this is pronounced.
49:44SPEAKEREither of those were considered an imminent threat,
49:46SPEAKERbut they were, that was a good test subject for this.
49:50SPEAKERThe craft slammed into the asteroid
49:52SPEAKERat about 14,000 miles per hour.
49:54SPEAKERThey didn't load it up with really
49:56SPEAKERany kind of scientific experiments or payload
49:58SPEAKERother than what was required to get it there and a camera.
50:01SPEAKERSo you can go on Wikipedia or on NASA site
50:03SPEAKERand there's like a great succession of images
50:06SPEAKERof just closer and closer to this giant rock.
50:08SPEAKERAnd then you can see the very final image
50:10SPEAKERbefore it slammed into it right above the ground,
50:13SPEAKERwhich is fun.
50:14SPEAKERIt did have a piggyback kind of secondary satellite
50:18SPEAKERthat detached a couple of weeks before the impact
50:20SPEAKERand acted as sort of a cameraman to record the impact
50:25SPEAKERand see what happened from a distance there.
50:29SPEAKERSo yeah, like that's kind of the gist of it
50:31SPEAKERis that they have determined this to be, I guess,
50:35SPEAKERa fairly effective strategy.
50:36SPEAKERThey just literally yesterday at the time of this recording
50:39SPEAKERput out another release about some analysis of the results.
50:44SPEAKERSo as you might expect, it ejected a massive amounts
50:48SPEAKERof material when it impacted.
50:51SPEAKERI think they said about 2 million pounds of rock
50:53SPEAKERwere ejected from the asteroid.
50:55SPEAKERAnd they have determined so far
50:58SPEAKERthat the force of the ejecta altered the travel
51:02SPEAKERof the asteroid far, far more than the actual force
51:05SPEAKERof the impact of the spacecraft, like three point.
51:08SPEAKERI think they said that it amplified the alteration
51:11SPEAKERby like 3.6 times.
51:13SPEAKER- So you're saying that the deep impact
51:14SPEAKERof the rocket stopped Armageddon?
51:18SPEAKER- Sure.
51:18SPEAKER- So the image did from everywhere.
51:23SPEAKERSo like JWST was pointing at this,
51:26SPEAKERHubble was pointing at this.
51:27SPEAKERLike they basically were like, what if we just point
51:29SPEAKERevery telescope we have, we can at this thing
51:32SPEAKERto image the actual impact and the result and stuff.
51:36SPEAKERAnd those are beautiful to look at too.
51:39SPEAKERThis isn't the only way we can move asteroids.
51:43SPEAKERWe can do things like gravity tugs where we put something
51:46SPEAKERright next to the asteroid and slightly tug on it
51:49SPEAKERusing the gravity of the other object.
51:51SPEAKERAnd you can just change the trajectory
51:53SPEAKERwithout having to smack it.
51:54SPEAKERBut this is the most American way
51:57SPEAKERto change the trajectory of an asteroid.
51:59SPEAKER(laughing)
52:00SPEAKER- No, no, no.
52:01SPEAKERThis is the second most American way.
52:02SPEAKERThe most American way is to put Bruce Willis up there
52:05SPEAKERand then give him a big drill and a nuclear bomb.
52:07SPEAKER- Yeah, until there are nukes involved,
52:09SPEAKERI'm gonna say it could get even more American.
52:13SPEAKER- Is this a rocky metallic asteroid
52:16SPEAKERor is it a dusty asteroid, do we know?
52:18SPEAKER- It's rocky.
52:19SPEAKERThere is like some definite like regolith on it.
52:22SPEAKERSo it has, I can't remember what the composition was,
52:28SPEAKERbut it is a rocky asteroid.
52:30SPEAKER- It's not like a hard iron nickel asteroid.
52:32SPEAKER- I don't know, I saw it was described as a rubble pile
52:36SPEAKERat one point colloquially, if that tells you anything.
52:40SPEAKER- The pictures, like watching this happen live
52:45SPEAKERis one of the wonders of the modern age moments
52:48SPEAKERthis year for me.
52:48SPEAKERLike you had a live feed of images coming in
52:53SPEAKERevery couple of seconds from the cameraman
52:56SPEAKERand then also the live feed of the images coming in
52:58SPEAKERfrom the probe as it was about to impact.
53:01SPEAKERAnd the last shot you see looks like it's from about
53:04SPEAKERfive feet off the ground.
53:05SPEAKERAnd it has incredible definition of the rocks
53:07SPEAKERand the shadows and the surface of this asteroid.
53:11SPEAKERAnd then you get like a partial frame on the next one.
53:14SPEAKER- Yeah, about the top 10% of the image filled in
53:17SPEAKERbefore it impacted.
53:18SPEAKER- Yeah. - It's fun, yeah.
53:19SPEAKER- It's pretty amazing.
53:21SPEAKER- I'm glad you mentioned the gravitational tug
53:22SPEAKERbecause I had always heard that was kind of the most
53:24SPEAKERplausible or promising method of doing this.
53:29SPEAKERAnd there's something kind of like beautiful
53:31SPEAKERand elegant about it to me.
53:35SPEAKERYou could do this without destroying things,
53:37SPEAKERbut yes, of course we just.
53:38SPEAKER- Well, but also the fact that the ejecta,
53:41SPEAKERI mean, of course the equal opposite reaction,
53:43SPEAKERall that energy has to go somewhere.
53:44SPEAKERSo it makes sense that hitting it really hard
53:47SPEAKERwith something is gonna make something come out
53:48SPEAKERthe other side and slow it down even more.
53:50SPEAKERSo that's cool.
53:51SPEAKER- Although, and also maybe not surprising to know
53:53SPEAKERthat the ejecta has now assumed a kind of comet-like trail
53:57SPEAKERor tail form subsequently.
54:00SPEAKERSo there's that.
54:01SPEAKER- I'm just excited 'cause it's the kind of thing
54:02SPEAKERthat seems like it would be really hard to get funding
54:05SPEAKERfor people who won't fund climate change work.
54:09SPEAKERWhen, you know, it's one in every 65 to 100 million years.
54:13SPEAKERSo we probably have a little bit of time left.
54:15SPEAKERWe don't need to spend money on this right now.
54:17SPEAKERWe could wait until later.
54:19SPEAKERIt'll probably be fine.
54:20SPEAKER- Yeah, real quick, I wanna jump in here.
54:22SPEAKERSpeaking of the last giant impact,
54:24SPEAKERdid you see the story from earlier in the year?
54:26SPEAKERI just came across it last night
54:27SPEAKERthat the Tanis dig site in,
54:29SPEAKERI wanna say it's like South Dakota,
54:32SPEAKERwhich state is, North Dakota.
54:35SPEAKERClaims, this has not really fully been vetted.
54:38SPEAKERThere are certainly people in different disciplines
54:40SPEAKERwho seem still a bit skeptical about this.
54:42SPEAKERClaims that they found fossils of animals
54:45SPEAKERfrom the day of the impact of the-
54:48SPEAKER- The big one?
54:49SPEAKER- Yeah, how do you pronounce the impact site?
54:51SPEAKER- Yeah, Chicxulub, or I'm not pronouncing that right.
54:54SPEAKER- I believe, I've never heard it said out loud,
54:58SPEAKERbut yes, the asteroid that wiped out the dinosaurs.
55:00SPEAKERLike this dig claims to have found parts of animals,
55:04SPEAKERfish fossils, with all kinds of telltale geologic evidence
55:08SPEAKERof having been exposed to the factors
55:11SPEAKERon the day of the impact.
55:12SPEAKERAgain, this is still under study.
55:14SPEAKER- How would they know down to the day?
55:16SPEAKERThat seems-
55:18SPEAKER- Well, they had a calendar and it's like,
55:20SPEAKERhey, there's a big X and then the day before is circled
55:23SPEAKERand then that day is not circled.
55:24SPEAKERSo, it's just a process of elimination.
55:27SPEAKER- Who knows if it's maybe the next day,
55:28SPEAKERbut it's mainly analysis of traces of geologic material.
55:32SPEAKERLike they found the kind of glass spheroids
55:34SPEAKERin the gills of the fish that contain traces of,
55:37SPEAKERI don't know if you're familiar with the boundary
55:39SPEAKERin the geologic record.
55:40SPEAKER- We're talking KT boundary time, yeah.
55:42SPEAKER- Yes, so I found out last night that apparently
55:45SPEAKERthe term KT boundary is no longer in fashion.
55:47SPEAKERNow it's the KPG boundary.
55:51SPEAKERIt's for paleogene rather than tertiary.
55:54SPEAKERSo, it's now the cretaceous paleogene boundary.
55:57SPEAKERAnyway, what is it?
55:58SPEAKERThere's like a tiny layer of iridium
55:59SPEAKERaround the entire world at the point
56:02SPEAKERwhere the impact occurred.
56:02SPEAKERAnyway, they look for traces like that,
56:05SPEAKERprimarily in these fossils and are surmising
56:08SPEAKERthat these animals were exposed
56:10SPEAKERto the direct effects of the impact.
56:11SPEAKERLook Brad, some fish just like
56:13SPEAKERto huff indium-lined glass beads.
56:15SPEAKERYou can't stop them.
56:16SPEAKERThey're just gonna get them in their gills.
56:19SPEAKERYou can't, nothing can stop that.
56:22SPEAKERShould we talk real quick about fusion?
56:24SPEAKER'Cause that was the news this week.
56:27SPEAKERThe NIF, which is I think that I foolishly
56:32SPEAKERdidn't write down that with the acronyms.
56:33SPEAKER- National Ignition Facility.
56:35SPEAKER- Thank you, I thought that's what it was,
56:36SPEAKERbut I wasn't sure.
56:38SPEAKERShot a couple of weeks ago, shot 192 laser beams
56:42SPEAKERat a tiny, tiny little target filled with fuel
56:44SPEAKERand achieved fusion ignition in a controlled setting
56:47SPEAKERfor the first time on earth, which is very exciting.
56:49SPEAKER- This is inertial confinement fusion,
56:51SPEAKERwhich is not plasma controlled fusion,
56:53SPEAKERwhich is bound in a tokamak, which is different, but.
56:56SPEAKER- The plasma confused fusion is the thing
56:58SPEAKERI always think of as the Taurus shaped reactor
57:01SPEAKERwith the plasma floating in and like it powers
57:03SPEAKERthe impulse engines on the enterprise is all I know.
57:06SPEAKER- Magnetic donut.
57:07SPEAKER- Yeah, yeah, exactly.
57:08SPEAKER- Tokamaks power the engine in the enterprise.
57:12SPEAKERNIF was the location for the Star Trek 2009 engineering room.
57:19SPEAKERThey actually shot, that's where they shot.
57:21SPEAKER- Really? - The engineer.
57:22SPEAKER- Oh, yeah, no kidding.
57:24SPEAKER- I thought that was a Budweiser brewery.
57:26SPEAKER- No, the pipes are different.
57:28SPEAKERThis is like, you can see elements of engineering though
57:32SPEAKERin that 2009 Star Trek.
57:33SPEAKER- I'm cooing pictures of this right now, that's cool.
57:36SPEAKERSo yeah, the fuel target was about the size of a peppercorn.
57:38SPEAKERIt's enclosed in a really cool diamond shell
57:42SPEAKERthat's almost as perfectly round as we can make it,
57:44SPEAKERit seems like, and has just a tiny little tube
57:46SPEAKERthrough that they can feed the fuel in.
57:49SPEAKERThe laser put in 2.05 megajoules of energy into the fuel.
57:54SPEAKERThe reaction produced about 3.15 megajoules of energy.
57:57SPEAKERSo this is the big deal.
57:58SPEAKERWe put some energy in and we got more energy out
58:01SPEAKERbecause of the fusion reaction happening.
58:03SPEAKERNow there's an asterisk, 'cause of course,
58:05SPEAKERthe lasers are pretty inefficient.
58:07SPEAKERSo getting the lasers to produce 2.05 megajoules of energy
58:11SPEAKERrequired about 300 megajoules from the grid.
58:14SPEAKERSo we have, look, there's work to do,
58:16SPEAKERis the takeaway here. - Some optimizations.
58:18SPEAKER- Yeah, some optimizations.
58:20SPEAKER- Do we need to very quickly define fusion?
58:23SPEAKERFusion is kind of the other nuclear process
58:25SPEAKERby which energy can be released.
58:27SPEAKER- Yeah, it's the one we're all most familiar with,
58:28SPEAKER'cause it happens in the sun
58:29SPEAKERevery single moment of every day.
58:31SPEAKER- So all nuclear power plants operate on fission,
58:33SPEAKERwhich is splitting atoms, which produces a lot of waste.
58:36SPEAKERFusion, less waste, much more efficient,
58:38SPEAKERis what happens in the sun. - Essentially no waste,
58:40SPEAKERbecause you're basically taking hydrogen,
58:41SPEAKERfusing it together, and forming helium.
58:43SPEAKERWe don't go beyond that.
58:45SPEAKERWe don't have the ability to have helium form
58:49SPEAKERhigher energy compounds, but that's the fusion
58:54SPEAKERthat we're talking about here.
58:56SPEAKERAnd it's not just about ignition, which is initial fusion,
59:00SPEAKERbut actually sustaining that fusion for some period of time.
59:03SPEAKERAnd so this fusion actually lasted close to a few seconds
59:08SPEAKERto produce that much energy.
59:11SPEAKER- Which is a long time in fusion land.
59:12SPEAKER- And the way it produces energy is like
59:14SPEAKERhow every kind of reactor at scale works,
59:17SPEAKERis basically it generates a ton of heat.
59:19SPEAKERIn this case, it emits a ton of neutrons.
59:21SPEAKERThose neutrons will go get absorbed into something,
59:25SPEAKERlike a heating jacket or something,
59:27SPEAKERand that thing gets hot, and then we use that
59:29SPEAKERto heat water to generate steam,
59:31SPEAKERand that'll spin a turbine or something,
59:33SPEAKERand that'll generate energy.
59:34SPEAKER- So, I mean, the upshot is this is a process,
59:38SPEAKERlike work on fusion has been going on now for 70 years.
59:40SPEAKERWe started in the '50s, I think.
59:44SPEAKERIt's been a bunch of tiny iterative steps,
59:48SPEAKERbut this is kind of a big milestone along the way,
59:49SPEAKERit feels like to me.
59:51SPEAKERI'm not a fusion expert.
59:55SPEAKERThe further journey from here into SimCity-style
59:59SPEAKERfusion power plants that let us all get free
1:00:03SPEAKERof fossil fuels forever and run clean hydrogen,
1:00:06SPEAKERhelium-producing power plants,
1:00:09SPEAKERit seems like there's still lots more work
1:00:11SPEAKERto come in the future, right?
1:00:13SPEAKER- Yeah, I would say my friend Kyle Hill,
1:00:16SPEAKERwho's a science YouTuber, kind of put it in this way
1:00:20SPEAKERthat I think is easy to understand.
1:00:22SPEAKERThere's like four milestones we really have to hit here.
1:00:26SPEAKEROne is just to achieve ignition at all,
1:00:28SPEAKERhave fusion happen at all,
1:00:29SPEAKEReven if it's energy, like a net energy negative.
1:00:33SPEAKERThen there's what happened yesterday,
1:00:34SPEAKERwhich is achieving net energy positive,
1:00:37SPEAKERwhere some of the energy we put in from the lasers,
1:00:42SPEAKERin this case, we produce more energy out.
1:00:45SPEAKERThen there's the next step,
1:00:47SPEAKERwhich is the energy of the system put in
1:00:51SPEAKERis less than the energy that comes out,
1:00:53SPEAKERwhich means like handling all of the 300 megajoules
1:00:57SPEAKERcoming that fed from the grid into these laser lamps.
1:01:01SPEAKER- Well, and presumably like making the fuel
1:01:04SPEAKERand stuff like that too, right?
1:01:05SPEAKER- Right, so that's all part of that milestone.
1:01:07SPEAKERAnd then the fourth milestone is actually building a system
1:01:09SPEAKERaround all of this to actually generate power.
1:01:13SPEAKERSo all of those steps, you need higher and higher energy.
1:01:15SPEAKERAnd for the most part from like the '50s through,
1:01:19SPEAKERI would say like the early 2000s,
1:01:21SPEAKERthe energy increase that we're seeing along the scale
1:01:26SPEAKERwas following kind of a Moore's law kind of axis of gain.
1:01:30SPEAKERAnd then it slowed down in the 2000s
1:01:32SPEAKERas we had to build like bigger and bigger systems
1:01:34SPEAKERto actually generate the kind of energy output.
1:01:38SPEAKERLike you go to NIF, like the 192 lasers,
1:01:41SPEAKERthey are the most old school lasers you've seen.
1:01:43SPEAKERThey're built with these flash lamps, these giant tubes.
1:01:46SPEAKERThey're like freaking two feet long tubes.
1:01:49SPEAKERAnd there are hundreds of them that flash.
1:01:52SPEAKERAnd then the light pulse that's sort of generated
1:01:57SPEAKERgoes through these concentrators to make a laser.
1:02:00SPEAKERA laser pen that we have in our like hip pocket
1:02:04SPEAKERover the side has somewhat more advanced technology
1:02:08SPEAKERthan those flash lamps do.
1:02:09SPEAKERThat's a bit of a extreme here,
1:02:12SPEAKERbut I just wanna acknowledge like these systems were built
1:02:15SPEAKERwith old technology.
1:02:16SPEAKERSo there are efficiencies that will come
1:02:18SPEAKERfrom building more modern systems here
1:02:21SPEAKERthat'll help balance that equation.
1:02:23SPEAKERBut we still have a long way to go.
1:02:24SPEAKERSo the ratio of energies is oftentimes referred to as Q.
1:02:28SPEAKERSo the energy in versus energy out.
1:02:30SPEAKERSo that we got to Q1 here.
1:02:32SPEAKERWe need to get to something close to Q10
1:02:36SPEAKERto have a viable ratio to build a power generation system.
1:02:43SPEAKER- And that's what the goal of the ITER project is
1:02:46SPEAKERin Southern France, that the US and the EU and China,
1:02:49SPEAKERand I can't remember the other two countries are.
1:02:51SPEAKER- It's 35 countries, it's a huge project.
1:02:53SPEAKERThere's actually like, I think there's 12 Tokamaks
1:02:56SPEAKERcoming online in the next like 15 to 20 years.
1:03:00SPEAKERSo this isn't a small scale situation
1:03:02SPEAKERwhere we're relying just on one project.
1:03:06SPEAKERBut ITER is the largest one.
1:03:08SPEAKERWe've never achieved ignition with a Tokamak.
1:03:11SPEAKERSo there's a lot to prove with that technology
1:03:15SPEAKERwhere we're using magnetic fields.
1:03:17SPEAKERMagnetic fields that are 300 times as powerful
1:03:20SPEAKERas the magnetic field of earth to confine plasma,
1:03:23SPEAKERto force the kind of pressure to achieve that fusion.
1:03:27SPEAKERBut there's a lot of hope based off of the science
1:03:30SPEAKERthat that is the more legitimate pathway to power generation
1:03:33SPEAKEReven though it's decades off.
1:03:35SPEAKERNIF, to be cynical just for a second,
1:03:38SPEAKERbasically allowed us to put a little bit of deuterium
1:03:40SPEAKERand tritium, isotopes of hydrogen in the middle of this
1:03:43SPEAKERand shoot a bunch of lasers on it and test the output.
1:03:46SPEAKERWell, what other radioactive materials could we put
1:03:49SPEAKERinto a target and test in a facility
1:03:52SPEAKERwithout actually exploding it?
1:03:54SPEAKERYou'd almost think a nuclear weapons facility
1:03:56SPEAKERwould be really interested in that kind of technology.
1:03:59SPEAKERAnd guess what Lawrence Livermore National Laboratory does?
1:04:03SPEAKERSo there is some military purposes to this
1:04:07SPEAKERabout testing fuel.
1:04:09SPEAKERSo I wouldn't think about NIF as like the project
1:04:11SPEAKERthat gets us to a power plant
1:04:14SPEAKERbecause of the system that it's designed,
1:04:16SPEAKERlike you shoot a target, this hall rom as it's called,
1:04:19SPEAKERto generate the energy.
1:04:21SPEAKERNow you have to, even if that proves viable,
1:04:23SPEAKERnow we have to design a system where like a hall rom
1:04:26SPEAKERlike perfectly falls into place
1:04:27SPEAKERand we shoot it with all these lasers and ignites
1:04:30SPEAKERand then another one falls into place and shoots and ignites.
1:04:33SPEAKERAnd then it has to happen like every less than a second.
1:04:36SPEAKERLike you can imagine all the failures
1:04:38SPEAKERin like dropping a pellet perfectly into a line of lasers.
1:04:41SPEAKER- It's just a burp gun of ping pong balls
1:04:45SPEAKERgoing through a giant laser cannon, yeah.
1:04:47SPEAKER- I remember first going to NIF,
1:04:49SPEAKERI went to the facility in like 2009
1:04:51SPEAKERand they had like an animation of that
1:04:53SPEAKERand I laughed out loud.
1:04:55SPEAKERI was like, come on, enough with that.
1:04:57SPEAKERThat like, I understand the point of this system
1:05:00SPEAKERbut stop joking around with that being a viable way
1:05:03SPEAKERof producing a power plant.
1:05:06SPEAKERThat being said, this is an incredible milestone
1:05:08SPEAKERjust to get to Q1.
1:05:11SPEAKERAnd so the fact that we're seeing progress
1:05:13SPEAKERand hopefully seeing progress in the Tokamaks
1:05:15SPEAKERover the next 10 years is something to celebrate.
1:05:18SPEAKER- Is it an oversimplification to suggest
1:05:21SPEAKERthat once fusion is stably and efficiently achieved
1:05:24SPEAKERthat it's the end of the world's energy problems?
1:05:26SPEAKERI mean, is there even a need for diversification,
1:05:30SPEAKERlike kind of solar, like orbital solar collectors
1:05:33SPEAKERand stuff like that?
1:05:34SPEAKEROr is that like, is that the Holy grail
1:05:36SPEAKERor is it more complicated than that?
1:05:37SPEAKER- I don't wanna say it's the end
1:05:39SPEAKERbecause like tritium costs like $30,000 a gram right now.
1:05:44SPEAKERSo it's not like a cheap fuel.
1:05:46SPEAKERSo we have to come up with ways to generate the fuel
1:05:48SPEAKERthat's actually appropriate for these facilities.
1:05:52SPEAKERBut if you could do that,
1:05:54SPEAKERthere's enough like hydrogen in like a cup of water
1:05:57SPEAKERto power your house for a year with fusion technology.
1:06:01SPEAKERSo we get into the place where the energy output
1:06:06SPEAKERisn't the issue anymore.
1:06:08SPEAKERIt is finding the fuel and it has no emission
1:06:11SPEAKERfor all intents and purposes.
1:06:14SPEAKER- Yeah, I mean, the only thing that's bad
1:06:16SPEAKERis the neutrons eventually make the cladding
1:06:18SPEAKERfor the reactor a little hot,
1:06:20SPEAKERbut like compared to what comes out of a fission reactor,
1:06:23SPEAKERit's basically nothing, right?
1:06:24SPEAKER- Yeah, you turn off the input
1:06:27SPEAKERto a fusion reactor, fusion stops.
1:06:30SPEAKER- Yeah, turns out that's nice.
1:06:31SPEAKERIt's a good thing.
1:06:33SPEAKERKishore, thank you so much for coming by.
1:06:35SPEAKERWe always, it's always nice to have,
1:06:36SPEAKERit's nice to have a visit that's not about
1:06:38SPEAKERthe impending apocalypse due to a virus.
1:06:40SPEAKERSo we'll have to do this again sometime.
1:06:43SPEAKERI have one more kind of general question for you
1:06:45SPEAKERand this may not, you may not be the person to answer this,
1:06:47SPEAKERso I apologize in advance.
1:06:48SPEAKERBut like all of the stuff we talked about today
1:06:50SPEAKERare all kind of application science.
1:06:54SPEAKERWe don't talk a lot about theoretical science
1:06:56SPEAKERin just in general, much less the popular,
1:07:00SPEAKERthe popular science conversation world.
1:07:02SPEAKERAnd like all of these, all of this work
1:07:06SPEAKERis based on decades of theoretical research
1:07:09SPEAKERthat it seems like as a country,
1:07:11SPEAKERwe don't really do very much anymore often.
1:07:14SPEAKERWe spend all of our money and time and energy
1:07:16SPEAKERon applications for things that we've already figured out.
1:07:18SPEAKERAnd there's a little bit less on the actually,
1:07:21SPEAKERhey, what happens if we do this
1:07:23SPEAKERand how does this part of the world work
1:07:25SPEAKERin a way that may not have any kind of long-term application?
1:07:27SPEAKER- My short answer is first and foremost,
1:07:29SPEAKERwe talked about quintessence for a while,
1:07:30SPEAKERwhich is not a word that...
1:07:33SPEAKERYou're basically asking like,
1:07:35SPEAKERare basic research and its value
1:07:39SPEAKERto the larger kind of ecosystem of people?
1:07:43SPEAKERAnd talking about that more.
1:07:45SPEAKERStorytelling about basic research is hard
1:07:47SPEAKERbecause we are naturally inclined to say like,
1:07:51SPEAKERwhat does this mean?
1:07:52SPEAKERWhy does it matter?
1:07:54SPEAKERHow does it relate to like my life?
1:07:56SPEAKERAnd basic research essentially is in its purest form
1:07:59SPEAKERis curiosity driven.
1:08:01SPEAKERAnd like, we don't have any connection
1:08:03SPEAKERoutside of like increasing the body of knowledge
1:08:06SPEAKERwithout any like overly prescribed notions
1:08:10SPEAKERof where that kind of goes.
1:08:12SPEAKERWe still invest a ton of money in basic research
1:08:15SPEAKERfrom just a pure money standpoint.
1:08:18SPEAKERIt's just that the amount of money
1:08:19SPEAKERas percentage of our GDP has fallen
1:08:22SPEAKERsince the '50s and '60s.
1:08:25SPEAKERSo we're not investing nearly as much as we used to.
1:08:28SPEAKERA number of other countries are investing a lot more.
1:08:31SPEAKERI think the success of basic research
1:08:33SPEAKERis in everything that we use.
1:08:35SPEAKEROur phones to on down the line,
1:08:38SPEAKERlike the nature of technology that's in Hubble
1:08:42SPEAKERis in each of our smartphones
1:08:44SPEAKERis kind of a ubiquitous concept now.
1:08:47SPEAKERSo we don't know what we can achieve from all of this.
1:08:50SPEAKERBut I think by telling the stories
1:08:53SPEAKERfar enough on the pipeline
1:08:54SPEAKERthat it can capture the imagination,
1:08:56SPEAKERgives people insight that there's all this work
1:08:59SPEAKERthat has to come before it to get to that point.
1:09:02SPEAKERAnd it's important to support that endeavor
1:09:05SPEAKERas opposed to just looking at the pure application.
1:09:08SPEAKERAnd I think talking about projects like AlphaFold,
1:09:11SPEAKERwhich actually doesn't do anything on its own,
1:09:13SPEAKERit's a tool to enable that kind of work,
1:09:16SPEAKERhelps tell that story more.
1:09:17SPEAKERAnd I think that's the opportunity space
1:09:21SPEAKERis to tell more stories
1:09:22SPEAKERwith the underpinning of all of that work.
1:09:24SPEAKERI mean, so much of when I listen to FOSPOD
1:09:28SPEAKERis like that's the story of FOSPOD to me.
1:09:30SPEAKERIt's like all of these technologies
1:09:32SPEAKERthat enable the technologies that we use every day.
1:09:35SPEAKER- Jeremy Allison described it
1:09:36SPEAKERas he builds the plumbing of the internet.
1:09:39SPEAKERAnd I think that's maybe the most apt.
1:09:41SPEAKER- And it's the same thing here with science.
1:09:43SPEAKER- And just so people have an idea of scale,
1:09:45SPEAKERour funding in the US for science in general
1:09:48SPEAKERis almost infinitesimal
1:09:51SPEAKERcompared to the money we spend on other things
1:09:52SPEAKERlike military budget and stuff like that, right?
1:09:55SPEAKERSo we're not talking about a ton of money.
1:09:57SPEAKERWe're talking about--
1:09:59SPEAKER- We are still talking about billions of dollars though.
1:10:01SPEAKERSo it's not nothing either.
1:10:02SPEAKER- Of course.
1:10:03SPEAKERBut yeah, billions of dollars compared to,
1:10:06SPEAKERit's not like we're buying a fighter jet
1:10:08SPEAKERthat costs a few billion dollars.
1:10:10SPEAKERAnyway, I guess that'll wrap it up for us this week.
1:10:12SPEAKERKishore, thank you so much for coming by.
1:10:13SPEAKERWhere can people find you and find out more about you
1:10:16SPEAKERin what I assume is a post-Twitter world by now?
1:10:19SPEAKER- One can hope that it's a post-Twitter world.
1:10:21SPEAKERI have stopped podcasting.
1:10:23SPEAKERThis is the first time in 10 years I haven't had a podcast.
1:10:26SPEAKERAnd my wife put it to me this way,
1:10:29SPEAKER"The people will be okay."
1:10:31SPEAKERAnd I was like, wow, that's rough but also accurate.
1:10:36SPEAKERYou can find me places.
1:10:38SPEAKER- But where will people find out
1:10:39SPEAKERwhat you think about the latest MCU movies?
1:10:41SPEAKERHow will they know what you think, Kishore?
1:10:43SPEAKER- You can find me on the TechPod Discord.
1:10:47SPEAKERI'm also on Wills Discord.
1:10:49SPEAKERYou can find me places if you need to find me.
1:10:51SPEAKERAnd I will espouse you with all of my thoughts
1:10:53SPEAKERabout where the MCU has lost its way
1:10:57SPEAKERand why Andor is the best "Star Wars" in a generation.
1:11:00SPEAKERBut I will say I am going to post
1:11:04SPEAKERsome of my top 10 science stories of the year,
1:11:07SPEAKERlike written narrative pieces,
1:11:09SPEAKERand my top five science books of the year
1:11:11SPEAKERin the Discord as well.
1:11:12SPEAKERSo check that out.
1:11:13SPEAKERIt should post by the time the episode drops.
1:11:16SPEAKERSo encourage people to check it out
1:11:17SPEAKERbecause there's great books and narrative pieces on science
1:11:21SPEAKERthat can help for this exploration.
1:11:24SPEAKERAnd I'll probably just continue talking
1:11:25SPEAKEReven after you guys hop off the line
1:11:27SPEAKERbecause that's the only way to podcast now for me.
1:11:30SPEAKER- Thanks, Kishore.
1:11:31SPEAKERThat'll do it for us this week.
1:11:32SPEAKERAs always, the TechPod is a 100% listener-supported podcast.
1:11:36SPEAKERSo if you would like to support the show,
1:11:37SPEAKERyou can find out how by going to patreon.com/techpod.
1:11:40SPEAKERThank you, supporters, patrons.
1:11:42SPEAKER- Yes, thank you very much.
1:11:43SPEAKERNice to see you all.
1:11:44SPEAKER- As always, a very special thank you
1:11:46SPEAKERto our executive producer tier patrons,
1:11:48SPEAKERincluding Nick Johnston,
1:11:49SPEAKERPaddle Creek Games' makers of "Fractured Veil,"
1:11:52SPEAKERAndrew Slosky, Buddy Thorp, Stay Healthy crimes.
1:11:56SPEAKERI like that one.
1:11:57SPEAKERJust Wedge, Joel Krauska, Teal, Not That One,
1:12:00SPEAKERand James Kamik.
1:12:01SPEAKERThank you all so much.
1:12:03SPEAKER- Thank God.
1:12:04SPEAKER- I hope everybody, hey,
1:12:06SPEAKERI don't know where you are, but where we are,
1:12:08SPEAKERCOVID cases are spiking again,
1:12:10SPEAKERRSV cases and flu cases are spiking,
1:12:11SPEAKERso go get your flu shots.
1:12:13SPEAKERWear a mask, be safe out there.
1:12:15SPEAKER- It's rough out there.
1:12:15SPEAKER- Don't get sick at Christmas 'cause that sucks
1:12:17SPEAKERor the holiday season in general.
1:12:20SPEAKERAnd we will be back next week
1:12:21SPEAKERwith another episode of the TechPod.
1:12:24SPEAKERSee you all then.
1:12:25SPEAKER(upbeat music)