Monday, October 30, 2017

Why I like intelligent machines spying on me

Last week on Reddit people repeatedly freaked out about artificial intelligence machines (Facebook, Google, Amazon) spying on us, humans. Listening to keywords we utter through phone apps that run in the background; inferring who our friends are based on WiFi networks we connect to, and so on.

And I have to admit that in principle I actually like the idea of advertisers spying on me. For two very different reasons.

One, I am a big fan of targeted advertisement. The day when Instagram finally figured that I'm not buying a new luxury car, but instead started advertising books, I celebrated. I actually tried to pat the AI on the back, to trick it into showing even more books-related ads (I'm not sure it worked, but I kept promoted posts on the screen for a bit longer, as I think it tracks it, and then also clicked on some every now and then). Because I'd really much rather stare at book covers or fancy musical instruments than at cars and fashion items. I'll never buy either, but hey books are so much more enjoyable! And as long as ads are unavoidable, at least let's pick the ones we care about.

Second, I think the privacy is dead, but the society is in denial and does not realize that. It's a huge topic, and I have 5-6 draft posts saved that I never have time to finish, but basically the concept of privacy as we knew it is gone quite some time ago. We shed parts of our identity all the time, and it only takes some time and effort to figure out everything. Have you heard of this artist who collected hairs on the floor of Grand Central station in NYC, ran DNA tests on them, and then reconstructed faces of their owners? It was more of an art project, because we are not yet that good in facial reconstruction from DNA, but in principle it's quite doable. In 2-3 years if not now. Or have you heard about how it's possible to ID the driver based on how they turn, accelerate and break? The information that is recorded by any GPS device with a built-in accelerometer (aka smartphone). Privacy does not exist, yet there are no social or legal protections for the new world in which privacy does not exist. The sooner we realize it, the better. And in a crooked way, Facebook spying on people may accelerate changes in the society that would protect individuals from impacts of sudden exposure.

Again, that's a huge topic, but if you think of it, the main risk of living in post-privacy world is that some forces (your government, police, health insurance company) can access your life much easier than you can access theirs, and has much more to gain. It's the asymmetry of power that is dangerous, not the absence of privacy itself. Once we realize that everything that is hidden will be revealed, we have at least some chance of making sure it won't destroy us. It's like with Equifax breach: the breach is not the problem (some leak was bound to happen sooner or later), the problem is that our whole lives can be ruined by a single stupid number. Don't shoot the messenger, you know? And I think in this case Facebook and Google are, in a way, the messengers.

Friday, September 1, 2017

Books: "Lab Girl" by Hope Jahren

Everybody in the sciences have apparently read this book long ago; and I know of a professor who made all their students read it. So I felt a bit ashamed not having read it, and had big expectations about it.

Well, it's a great book, and I'm happy that it happened to be about scientists.

It's not a good "book about scientists" though.

I think it's an important distinction (and a horrible wording). My "liking" or "not liking" this book, or "recommending" vs. not recommending it to anybody (say, a student) would really depend on this framing.

I mean, there's genre f iction, and every now and then there's a person in fiction who happens to be a cook, or a policemen. This fact on itself would not make it a book about cooking, or about police, would it? "Indiana Jones" franchise is not really about archaeologists (even though the protagonist happens to claim this profession the calling of medicine), and "The adventures of Sherlock Holmes" is not really a book about the calling of medicine, even though one of the main characters is very much a doctor. And then there are books and movies about medical doctors specifically, that are created as such, to cater to the interest of the public, or maybe to ignite this interest. And you can discuss whether a certain portrayal of a certain profession is truthful, positive, negative, etc., which is one sort of a discussion; or you can discuss the plot, the soap opera, and whether the characters are psychologically plausible, which is a completely different discussion. These are two different dimensions, and a book or a movie can be great in one, and horrible in the other, or vice versa.

So my main trouble with "The Lab Girl" is that, while I totally loved the book, I feel a bit uncomfortable that it was picked by the scientific community, and transplanted from one category to another. From a moving memoir of a person who also happens to be a scientist, it was made into a book about scientists. I totally see the temptation: for one, there are not that many touching, human, vulnerable books that would truthfully describe scientific life. Science is often present in a cartoonish form in scientific fiction, apocalyptic thrillers, or books about political conspiracies, but psychological, literary books about sciences are rather rare. Second, books about women in science, written by women in science, are not that frequent, to put it mildly, and very much in need. Third, the topic of mental health and existential struggle, vulnerability, and success, are all extremely important ones, and ones traditionally shunned and downplayed by the scientific society. This book suddenly filled quite a few niches that were under-occupied, and it resonated with readers.

But at the same time, I feel deeply uncomfortable with the idea of normalizing some of the messages of this book: that science requires special sacrifices, that it demands from its followers not just monastic existence (which would be bad enough), but sort of transcendental, esoteric transformation, incompatible with free time, with family, with life, with pretty much everything. Science as a calling, a flame that consumes you, burning from within, the insatiable quest for knowledge, and so on and so forth. A field that you can enter only after being hazed by your elders, because if they don't torture you now, you won't be ready for the tortures of real life (sorry for the spoiler, I hope it's a minor one). I hate it. I mean, for every extreme feeling there exists a person who can live this hype and be happy, or normal, or functioning, so I gladly accept that there are people in the world that feel like that. But I would hate to live this kind of life, and I don't want any student ever think that to be a scientist they need to experience these extremes. Ideally, I want students to be mistaken that science can be a normal routine profession, maybe a bit demanding, maybe a bit under-payed and under-appreciated, all things considering, but still just one profession out of many, that just happens to be also be fun. Because I suspect that one of the ways to make scientific word less torturous is to raise a generation of people who would expect it to be normal, and demand this normality from those around them.

To sum up, I really liked this book, but I regret presenting it to a student last year without finishing it first. I hope we'll have more books written by scientist that would mention science casually, between love affairs and problems with teen-aged kids. Books that would use scientific metaphors, and describe real scientific anecdotes as a backdrop for the main story. So that this lovely book could become one of many, but not necessarily the one.

Thursday, August 31, 2017

New personal page, and a review of new Google Sites

Google has being slowly promoting their new Google Sites for about a year now, and recently I've moved most of my website to this new platform. Behold:

https://sites.google.com/view/khakhalin

Pros:

  • It looks really pretty, with these large sliding images on top and everything
  • Looks perfect both on a wide screen and a mobile phone (old google sites couldn't cope with a phone screen at all). I was critical of this feature at first, as I thought they were pushing the phone layout a bit too aggressively, but it turns out that I was just formatting it wrong. If you just add one block of text below another, sure, it will look poor, kind of like mobile wikipedia, when opened on a computer. But what you should do is add columns, or put images on the side. Create a matrix. In a mobile format it will nicely reshape into a column-vector, so that's the best of both worlds.
  • Intuitive interface: it totally feels like Google was inspired by the recent development of Paper by Dropbox (which is awesome by the way), as the interface is clean, clear, and easy to use. It may be a bit harder to move large blocks around, but still possible.
  • If you know html, you'll appreciate that the blocks you add follow <div> mentality, and there's a logic behind div-embedding. While it is not shown to the user explicitly, if you ever worked with divs, you'll immediately recognize the structure (and beauty) of it.
  • I like the little magic thing they do when you put images behind text (they adjust the color of the font and the lightness of the image)
Cons:
  • For now they don't support tables, and I need tables to publish protocols. But I think the Google team promised to eventually introduce them.
  • For now they only allow 1 level of subpages, but this thing they explicitly promised in one of their blog posts, so it should, theoretically, come live within about a year.
  • Very few styles for now, and it's impossible to create your own styles, but then again I think it will be changed in the future.
  • Impossible to attach files to the page, which was a very nice feature in old sites, both for hosting large-resolution images, and for uploading pdfs. I'm not sure whether they plan to implement it; I hope they would, otherwise I'll have to use external ftp storage for files.
But overall it definitely wins over Wordpress and Weebly, in my opinion, and hey it's free!

Thursday, August 17, 2017

On race and teaching props

In one of my classes, I run a short lab on the rubber hand illusion, so naturally when I ran it for the first time I was in need of some fake rubber hands. I figured that for the illusion to work well, I'll need hands that are at least somewhat similar to real hands real people have. So my intention was to buy a bunch of small hands, a bunch of large hands, some light-skinned hands, and some dark-skinned hands. About 20% of our students are dark-skinned, and I guessed it would be silly to make them work with pale pink rubber hands.

(I think there is actually a study that showed that the illusion still works if the hand is of a different skin tone, or of a different size and shape than your own, but that it does not work that effectively. But I wanted to just demonstrate the illusion, and so needed the most robust effect.)

I went online to order some hands, and lo and behold... There are no fake brown hands on sale.

Look for yourself, here is the google images search for "Fake hand". Here's amazon search. All hands are light pink, not even tanned. Isn't it weird?

Why is it so? About third of all fake hands are visibly zombified, so maybe it's considered a faux pas to manufacture dark-skinned zombie parts? That would be weird, but who knows, people are weird. Another third of hands are non-zombified Halloween props. Don't African American celebrate Halloween? Again, I don't know, I'm a foreigner, I have no idea. Maybe they do. Maybe they don't. I'm sure some do. Still no Halloween hands for dark-skinned people.

But the last third of fake hands are actually props for different kinds of beauty industries. There are "nail mannequin hands" (google it) that are used to showcase nail art, then there are training mannequins that nail polishers and such use to practice, and finally hand mannequins to show jewelry in window stores. And if you google them, all of them are pink - with the exception of jewelry hand mannequins that also come in pitch-black (that looks cool and artsy, but totally unrealistic). And I'm pretty sure dark-skinned people use all sorts of nail beauty products, and rings, and manicure.

But there are no naturally dark-skinned fake hands on the market, period.

I actually thought I found one, which was ridiculously expensive compared to "white hands", and ordered it, but it turned out that the photo was bad, and then hand was barely tanned. Maybe a bit brownish, but definitely not what I was looking for.

The practical outcome for the students that day was that the dark-skinned students had to work with weaker effects than light-skinned ones, and were also reminded of a lot of interesting social aspects of race in science that I totally did not want them to be reminded of on that particular day. Of course I told them the whole story of hand-hunting before the lab, and we all laughed about it in a sad wise laugh of a person who've seen worse things in their lives, but I was still annoyed and disappointed by the whole situation.

Wednesday, May 24, 2017

Books: A God that could be real by Nancy Abrams

About a year ago I entered my e-mail in a giveaway for a book on theLibraryThing web-site. The idea of these give-aways is that you kind of promise to read and review the book in exchange for a free copy. And then I suddenly won a free copy! So I found myself reading it.

The book is called "A god that could be real" by Nancy Ellen Abrams. The topic of it, to put it simply, is to reinvent and re-envision the concept of God that would not contradict modern cosmology, yet would also be an active personally engage-able God, a type one could conceivably pray to, as opposed to a cold rational impotent deistic construct. So Abrams takes a chisel and starts working on the clumsy block of our standard God-related preconceptions (such as omniscience and omnipotence), trying to get to the believable core, while not destroying the piece entirely. To summarize her message in one sentence, the author revisits post-Hegelian dialectics and applies it to religion, defining God as a truly (but dialectically) existing teleological representation of meta-humanity.

My biggest problem with this book is that Abrams rejects, ignores, and sometimes plain hides from the very idea of Platonism, be it the original Greek version, the Jewish interpretation of it (something in the style of the Gospel of John), or Hellenistic and Medieval versions. She chooses to understand all key theological words, such as "to create" (as applied to the Universe) in a very practical, matter-of-fact way, which then obviously leads to contradictions. But she pretends that these purely materialistic (as opposed to idealistic) interpretations are the only possibly ones. Which is of course not true.

In a similar fashion, she ignores, or politely dismisses, the mystical tradition. Her book is dedicated to the active process of defining and redefining God, which contradicts the very central message of most mystical traditions (from St. Symeon the New Theologican, to St. John of the Cross, and early Hassidic rabbis). While not all authors spell it out explicitly, the core message of Judeo-Christian mysticism is that not only one cannot define God, but even more strongly, it is not possible to formulate a logical predicate, a true/false statement that would contain God as an object, as God is not an object of our reality. In mystical monotheism, God is fundamentally a subject, staying outside of our world (and in most extreme versions - outside of our logic as well). God is the subject in the sense that s/he makes our world exist; God exists us. But we cannot ensnare God in our statements or trap God in our definitions, however clever they might seem (see apophatic theology).

Abrams ignores, or rather dismisses with a half-a-paragraph-long hand-wave this entire stratum, and entire school of thought. Which is annoying, as it means that in order to test and taste the validity of any of her statements I need to first translate them into their weaker versions, and only then think about them. Because of course these weaker versions are very possible: the whole concept of angels (messengeres) in Judaism, and then Christianity, as well as related concepts of Metatron, Son of Man, Messiah, prophets, Wisdom of God, etc. - all grow from this very same tension that made Abrams write her book: the tension between the intuition of God as infinitely "another", and the intuition of God's full presence in one's existence here and now. The impossible leap through this void was again and again reinterpreted by religious thinkers as different kinds of messengers, projections, incarnations and deputies, from the Angel wrestling with Jacob, and to the concept of Church as the body of God. Abrams is not the first person to fight this very fight.

Overall, it took me about half a year to finish this book. Every now and then, feeling guilty, I would try to pick it up, but it just would not go. I do hope that it could make a more pleasant read to a non-scientist (as it talks a lot about science), or a non-believer (as it talks a lot about God), but if you happen to be interested in both religion and science, then this book may feel too slow for you. At the same time, the book is definitely thought-provoking. It contains several interesting, poetic metaphors that were a pleasure to read, and a nice discussion of physical scales that could be welcome in any classroom. It has some really nice material, but it did not quite work for me as a book, and I definitely cannot subscribe to its main argument.

Monday, May 22, 2017

Against economic growth

One thing that really annoys, or maybe even frightens me, is economic growth. Because when growth becomes a goal, and not a means towards a goal, you are dealing with cancer. Not an organism, not a healthy tissue, but a cancerous outgrowth.

I'm fine with change. It may be Buddhist, or it may be nervous, anxious desire to change everything again and again, but while it may be empty, at lest it is not self-destructive.

Stasis, stability, conservatism, tradition are also fine. Perhaps dead a little, but fine. I like it.

Even improvement is fine, as improvement, efficiency, can still be self-contained. If I learn to do my job better and better, I can end up freeing some time to walk around the block, or play a banjo.

But growth is self-destructive.

And unfortunately it seems that our society is addicted to growth.

But then maybe I'm wrong, as I know positively nothing about economics.

Friday, May 12, 2017

Small linguistic victories

I was talking at the board, describing some imaginary scientist that does some imaginary experiments, and I referred to this scientist as "she".

This sure does not sound like an achievement: of course there are women scientists! My science classes are typically between 80% and 100% female after all. But for me every time it feels like a small breakthrough, for a very different reason. The thing is, while I teach only in English, inside my head I still think a little bit in Russian, at least to some extent. And in Russian all nouns are gendered (they are either masculine, feminine, or neutral). So for me, at a very intuitive, subconscious level, all cells are "she", while all neurons are "he". Neurons are dudes, cells are lasses, while nuclei don't care - nuclei are "it" (neutral). For me, mitochondria are always maternal, and not just because they carry mDNA and can prove maternal lineage, but because they are feminine, while receptors and channels are clearly masculine. It is probably ridiculous, but it's true, and follows the patterns described in the literature (there are some famous studies, one on bridges, and one on kitchen utensils)

And so, "of course" (sarcasm here) "a scientist" (учёный, ucheny) is masculine in Russian. But it's worse than even that: "a scientist" in Russian, while grammatically a noun, does not read or sound like a typical noun, but has all signs of an verbal adjective that is used as a noun. It's somewhat similar to English nouns like "grown-up", "initiate" or "trainee": essentially a "grown one", "initiated one", or "trained one" respectively. So in Russian the word for "scientist" is literally "the learned one"; and this "one" is grammatically male. In Russian, "A scientist" etymologically means "the learned male". Only think of it. "The learned male!". It's bizarre!

It is a very curious case, and seems to be almost an exception. Most words for professions and occupations in Russian are not like that: "a teacher", "a professor", "a student" are all normal nouns; some are masculine but can be used for females without a grammatical collapse; some have both masculine and feminine forms. All of them are more or less OK. "A scientist", however, totally hurts my brain. On several occasions, when I tried to make my writing in Russian more gender-inclusive, I was attacked by Russian speakers (women and men alike!) for ruining the grammar. Because if you write in Russian something like "When a scientist is tired, she takes a break", it short-circuits the grammar. It reads like "When a well-learned male man is tired, she takes a break", which is seriously weird.

And no, there's no "learned female" form for this word (like for German der Professor / die Professorin). If you try to forge a "feminine form", it sounds very unusual, and is completely unheard of. I gather with some effort it should be possible to introduce a "feminine version" by force, kind of how notation "they" was re-introduced in English for the purpose of gender inclusiveness, but so far nobody bothered to even start playing with these possibilities. And probably the society will be very resistant to changes like that, for quite a while. And even then, the controversy would probably remain, as it remains in German, as having two different words for male and female people in a profession is a bit weird.

But in my private life, for my private brain, this whole strange situation means that while I work mostly with female scientists, it is still curiously hard for me to refer to an imaginary "placeholder" scientist as "she" in speech. Because grammar representation in late life bilinguals is weird. Even after years of training, it feels that I have to consciously override some circuit in the Broca area to make it happen.

If you are a native English speaker and cannot relate to this story at all, try to imagine that you learned a language where the words for "scissors" or "pants" are singular, and not plural, as they are in English, and you now need to say something to the effect "my scissors is red", or "give me a pants". Would not it feel weird? I bet it would! Even though you would know perfectly well what you are talking about, and how to properly say it, you would still have to consciously override something in your brain to use "trousers" as a singular.

So, here you go: small linguistic victories.