[FRIAM] the slow red-pill

[uǝlƃ ☤>$]

Yeah, the gun analogy is also an unfortunate trawl. Massively asymmetric things like guns, nuclear bombs, and celebrity influencers are *distractions* from the main point.

What the 3 links I posted target are machines/system(ic)s with very SMALL effects ... nudges, incremental trends that bias over time. Again in that Ezra Klein podcast, Suzman outlines a radical sharing economy of some tribe where if a hunter is *too good* at huting, when everyone's sitting around eating the meat, they trash the meat. "This meat tastes like garbage!" and whatnot so as to symmetrize that hunter's status in the tribe. They don't want to tamp the hunter down too much, because, well, Yay Meat!!!

This same would be true of, say, a machine learning algorithm that *tends* to put more black people in jail than white people. It's nothing like a gun or Kim Kardashian's butt.

On 7/15/21 6:34 PM, Steve Smith wrote:
> uǝlƃ ☤>$
> 
>> I wouldn't call it "subliminal". I went through a similar process as I moved from TX to NM to CA to OR, gradually becoming more liberal the whole while, but quite *aware* of my transformation. 
> I can't claim the same except in hindsight.   Each insight I feel (in
> retrospect) that I gained along the way was somewhat
> cumulative/attritional.  I can name a few denouments along the way, but
> for the most part, the revelations were post-hoc fabrications to
> simplify a much more subtle/slow/complex process.   Maybe this is what
> you describe as one of our differences:  episodic vs diachronic?
>> I suspect something similar happens with the consumers of these progressive(ly radicalizing) campaigns. Sure, grandpa may not be that conscious of *how* radicalized he's becoming, but to assert that grandpa is totally unaware of his radicalization would be overly simplistic ... just like most of us are very aware of the increase in the acceptability of porn, or nerds, or violence in the media.
> I do think we all recognize and enjoy catching a wave and surfing it
> when we are lucky or good enough to anticipate it well.   What I'm
> thinking of feels more like riding a tide, which can either be very
> intentional or entirely otherwise.
>> It would be more appropriate, I think, to assert that whatever progression we're led through is *accepted* by us. Grandpa *wants* to be radicalized by Fox News. The existence of Fox News simply makes his desire somewhat acceptable ... like Pieter constantly trying to get us to watch Brett Weinstein videos. Pieter *wants* to be radicalized. Brett simply helps him do so. >8^D
> There is definitely a confirmation bias "hit" that I observe in myself
> and others whilst listening to/reading opinions which support those I
> already hold.  I perhaps even enjoy getting pushed a bit further by
> them.  Maybe part of my advancing senescence stage (can) include(s)
> losing my juice for being radicalized?  I'm not exactly seeking "center"
> or "moderate" but somehow there seems to be a complement to seeking
> self-radicalization?  It just ain't as much fun as it used to be.
>> Re: Volokh's hypothesis: I agree with you. He's simply trying to treat the *legal* situation without committing to the full fauna-flora in which it lives, which is why the Matthias article is relevant.
>>
>> I have a rebuttal to the Matthias argument that I purposefully did not link to, just in case my trawl led to an interesting catch. (Know 10 things, say 9.) And although Frankenstein's monster is a good start, it pales in comparison to the modern questions of explainable/interpretable AI and *ethical* AI. Can we *read through* an algorithm to effectively blame an algorithm's author? Or, if not the author, the post-authorship *user* of the author's product? Or, as a postmodernist might argue, should we take the author's product as a stigmergic naturfact and treat the author and her artifact as *excused* ... they were only being creative ... as with gun manufacturers ... the blame lies with the user of the artifact. Algorithms don't kill people. People kill people!
> 
> Gun's don't kill people, bullets do.  
> 
> I think the question is more complex than the general problem of
> ballistics and firearms.   When my niece discovered a small caliber
> round stopped between the panes of her patio door window last January 1
> it was most likely that the round was a stray intended exclusively for
> expressing the exuberance of the wielder of the weapon it came from
> rather than something intended to cause her (or her property) harm.  
> Few other devices would likely yield the same risk as the firearm/live
> ammo used to "celebrate" that night.
> 
> Even If we ignore the primary intention of firearms (obtaining
> significant asymmetric advantage for killing humans and animals) we
> might consider the "landscape" along with the artifex and artifact.  It
> is ever so much harder to kill someone with a stray arrow or dart than a
> bullet (based primarily on range).  We also mine/refine inestimable
> volumes of lead only to distribute most of it across the landscape from
> the barrel of a firearm.   There are now regulations in many wetlands,
> banning the use of lead shot for this very reason.   While the NRA
> shields firearm (and ammunition) producers from culpability in the (mis)
> use of their firearms against innocents, it seems inevitable that
> someone might successfully hold them accountable for providing a
> significant mechanism for poisoning the landscape with lead?
> 
>     https://undark.org/2017/01/30/lead-ammunition-bullets-hunting-copper/
> 
> I believe that the extravagant discussions held by Oppenheimer and other
> Manhattan Project scientists about the use of nuclear weapons are
> germane to this discussion?  It seems to me that there are significant
> differences between Edward Teller's role and Robert Oppenheimer's in the
> shaping of the cold-war landscape?  If Oppenheimer had been able to
> persuade Truman or Groves not to use Fat Man and LIttle Boy in Japan,
> perhaps the cold war would have been very different, and had Teller had
> his way, an early fusion weapon might have been used to "deter" Soviet
> nuclear weapons development.
> 
> I do think I am responsible for the consequences of my actions and that
> effects ARE transitive across causal networks.   Are hammer
> manufacturers culpable for those which were used for killing people? 
> Probably less so than battle-axe manufacturers.
> 
> I just read EricS's response and am now contemplating the implications
> of the differences between "who to punish" and "how to take
> responsibility".   Given our increasing awareness of small-worlds it
> seems that we become more and more culpable for not considering the
> implicative cascades of our actions/decisions including trying to
> understand the stigmergic field components of relevance.
> 
> - Steve
> 
>>
>> On 7/15/21 10:32 AM, Steve Smith wrote:
>>> glen -
>>>> There’s a new tactic for exposing you to radical content online: the ‘slow red-pill’
>>>> https://www.theguardian.com/commentisfree/2021/jul/15/theres-a-new-tactic-for-exposing-you-to-radical-content-online-the-slow-red-pill
>>> A modern application of "subliminal advertising" and "desensitization" I
>>> think?   As I moved in my life from Red/Libertarian circles to
>>> Blue/Green circles I noticed the subliminals and the desensitization in
>>> my *new* circles but had to be pretty far from the *old* circles to
>>> realize theirs were fraught in the same way.   I still cringe when my
>>> Blue/Green friends try to slip things past my filters, but I am more
>>> conscious by circumstance of the phenomena.   Red Rants these days seem
>>> to be *nothing but* attempts to normalize absurd things.   I suppose it
>>> might be part of a "great renormalization"?   Moderate Conservatives
>>> (and perhaps Liberals if they join forces) may find that simply steering
>>> clear of the radicalizing techniques will win them the attention, if not
>>> allegiance of a population who are under extreme political adrenal fatigue?
>>>> "There is no content moderation solution for a political problem."
>>> I think this is a central point.  
>>>> Related content:
>>>>
>>>> Common Carrier Status as Quid Pro Quo for § 230(c)(1) Immunity
>>>> https://reason.com/volokh/2021/07/15/common-carrier-status-as-quid-pro-quo-for-%c2%a7-230c1-immunity/
>>> I think this exposes the historical context/embedding well, but not sure
>>> if it fully covers the space that we are now exploring.  I would guess
>>> that media-studies researchers are all over this question in the
>>> abstract.   It seems as if the soft/social sciences have adopted a lot
>>> of quantitative methods whilst learning (or inviting in) mathematics.  
>>> What is the appropriate basis space for studying the dynamics of social
>>> media (companies, consumers, producers, exploiters/gamers)?
>>>> The responsibility gap: Ascribing responsibility for the actions of learning automata
>>>> https://link.springer.com/article/10.1007%2Fs10676-004-3422-1
>>> I believe Mary Shelley coined an interesting character and let him play
>>> out his nature in a somewhat allegorical setting some 200 years ago.  
>>> The genre that emerged in its  pattern is elaborate, but perhaps yet out
>>> of date.
>>>
>>> DaveW has lectured on the topic of Asimov's Robots and their ethics.
>>>
>>> Some interesting outliers in the genre include:
>>>
>>> Jack Williamson's Humanoids:
>>>
>>>     https://en.wikipedia.org/wiki/With_Folded_Hands
>>>
>>> and Stanislaus Lem's Medieval/Fairy-Tale highly anthropomorphic
>>> cybernetic creatures:
>>>
>>>     https://en.wikipedia.org/wiki/The_Cyberiad
>>>
>>> What might be the "new questions" that have emerged in the last 20-40
>>> years with the Global Internet and significant adoption?
>>>
>>>     https://en.wikipedia.org/wiki/Global_brain#History


-- 
☤>$ uǝlƃ