Is there a text version of this? I keep seing it mentioned, probably really interesting, but I LOATHE videos or postcasts...
4,559 karma · joined November 3, 2011
What would help a lot would be smth. like a high-power-UV portable box that could fit in a backpack (or a chemical equivalent), to allow at least some imperfect but useful re-sterilizing of masks... single-use N95s are hugely wasteful...
I think our current focus on "perfect" and "100% correct" (whatever that even means...) hurts managing pandemics A TON! An approach where we'd recognize the probabilistic nature of contagion and fight for +1% probability of desired-event-X1, -0.1% prob. of undesired-event-X2 etc. would be much useful... at whole-population scales all measures will be sloppily implemented in practice and we need to accept than and just "tweak the odds" until "we're mostly winning"...
(For sure, use better proven and standardized methods in hospitals and other settings with well trained people paid to implement measures correctly. But don't expect most of the average population to bother doing things the optimal way even when they've been educated about it... there's opportunity costs to everything, even to "giving a f"!)
They help because:
1. you have a significant chance of being already infected with a mild form and contagious even if you don't know it - wear a damn mask to protect the really vulnerable people around from your viruses!
2. any masks reduces even by a tiny amount the chance of infection - you eat more veggie fibers to reduce your chance of colon cancer by a single digit percent ...wear a goddam mask to reduce even if by very little the risk of catching (and then spreading) COVID-19!
3. by wearing a mask you'll be imitated by a few others and make it socially more acceptable for others to wear too - some of the people who end up wearing makes because of you are probably already infected and don't know it, so you'll have limited spread by promoting the "fashion"! (also, a few are probably immunodepressed or otherwise sensitive and might not even know it so simply copying a "fashion" could save their lives)
PLEASE, wear a mask, any masks, even wear it badly, but do! Do it for others and to slow the spread of disease worldwide enough to allow healthcare systems to have a chance of coping with it, and to give vulnerable or unlucky people developing the rare severe cases a decent chance of survival!
And wash your hands!
(Yeah, 50-70% of the human population will get this, eventually. But it's incredibly stupid if we get it all at fucking once! Unless you're shorting some stocks or pursuing a dubious political agenda, you don't want this to happen.)
EDIT+: ...but don't go the other extreme, buying them at exorbitant prices despite being healthy and not at high risk, or hoarding them.
That's enough signal to make me not bother reading this. Writing this here to prevent other from wasting their time. Don't know the agenda behind, but can see the author(s) is/are trying to "pull" the Overton-window in a really really nasty direction concerning many things...
[1]: https://abe-winter.github.io/2020/02/11/books-facts-ip.html
Don't let it die until we have alternative suitable anonymous and un-blockable-by-cyberattacks (eg. "it must still work while the internet is down") currencies!
(Note: no, we don't have yet, all current crypto currencies are "barely working prototypes" at best, all of them have issues that still need ironing out! Bitcoin itself is like a proof-of-concept-pre-MVP that got "rushed into production". And the blockchain itself... we NEED to have a way of fully erasing history from it, like paper money do, we don't yet know how to do that while keeping the nice properties of blockchain... all these are open research questions, Satoshi only made the first baby step!)
Just accept that there will be a temporary extra increase in mortality, and move one, carry on all life and economic activity as if nothing much happened, because it didn't! Promptly silence the panic mongers. "Live and let die", so to say, until solutions (vaccine, drugs) come to market.
Paraphrasing what a wise American once said in a sad September day: go about you business! go shopping! (ok, maybe do more of it online, but whatever...)
In this particular case, even the orange guy has a better/healthier attitude towards this and more sense than VCs and the stock markets!
OSS is all about leveraging "hidden" advantages agains competitors that have clear and obvious financial advantages over you. If you get it right, you win!
Of course, if you don't like risks and don't like "uncomputable" prices, or "prices unique to each buyer", then sure, be conservative, and stay away from FOSS.
Please edit "white people" to "American and Chinese white people"... people from smaller countries are more prone to knowing more languages :P Also, people who migrate or travel a lot. Also, since race discrimination is still a thing in the US and not only, smart people of color do their best to appear at least as smart as they really are, and knowing foreign languages probably helps a lot.
Being white and of the right nationality in a developed country simply allows you to be lazier, and you can afford to appear "dumb" in casual life, or to focus on only what you care about, hence you'd probably not bother learning any foreign languages, the ROI would be very bad for you, others will learn your native language instead... We all have limited amounts of brainpower and we try to invest it as well as we can.
Yeah, sure, you can philosophise in the most general terms but that's not how you get people and companies stop doing wrong things. You get them when the people/entities who want to re-bring the despicable things back into discussion can be fined / closed / put-in-prison etc.
I don't argue for eco-fascism. But with some things we need to be ruthless, especially in order to be able to carry on our peaceful, understanding and democratic discussions concerning all the other things!
> The problem of climate change has been cause precisely by a lack of democracy
Well, yeah, but ya know, sometimes you treat cancer with radiation... Just like you sometime need to apply more force and violence to end a war sooner and minimize overall destruction. Expecting for the Marshall island's vote to be considered in the US or the EU is not practical problem solving.
In all the civilized world we've taken certain really BAD options off-the-political-democratic-choice menu, because we don't want them to even be considered!
Maybe we can do the same with things that are are really bad for the environment, like not having eco-externalities priced into things! This can be done without killing democracy as a whole. You cut a finger to save the whole organism! Maybe we should start seeing the environment as an extension of our bodies and treat violence against it similarly to how we treat physical violence - take it off the f menu!
Not doing this would only make things worse and, yeah, sooner or later some "(pseudo)eco-fascist" group would spring up and find groups to blame (for the f-up resulting from people not acting because they couldn't find a democratically-acceptable solution) and manage to sell hate well enough to get in power. And at that point everyone will be double-screwed: you'll have wrecked environment + "eco"-fascists!
Then pull it out of political discussion! We don't have political discussion about "how to hang people of color Q better" anywhere in the world (mostly) anymore because we've wrote it in the text of our f laws that violence and racial discrimination are forbidden! Not on the political menu anymore.
Same needs to happen with lots of stuff concerning climate, ecology, sustainability etc. - we need to restrict the political menu to only the right choices.
And use the states' monopoly on violence to enforce these choices.
Stop pending the argument that "it's politically impossible / toxic / whatever / unenforceable etc.", that's a dumb argument.
Just take politics out of the equation! Or more precisely, take "democratic choice" out of this particular policy decision!
If countries amend their constitutions to disallow any party or individual not in support of this to even be part of any election, then this particular aspect would simply not appear in political campaigns. Have "dark room" deals in most countries of the world put this into place and move on.
And once the major international players are on board, for any country not playing the same game, just kindly point the missiles at them!
In the end, most national governments would approve of such taxation because they'll have the opportunity to formulate it in their own terms and slightly increase national budgets (maybe investing a bit more in the military and gaining an advantage over their neighbors that don't do it... ugly but any incentive in the right direction helps). And multinational corporations would also not mind it much if it's done across the board, without favoring anyone... and actually corporations could be "hired to enforce the tax" in more advanced schemas, offering opportunities to profit while punishing misbehaving countries.
Probably makes sense, not what Go was designed for, but I really don't get big-G's choice of "one different language for each niche"...
I mean, ffs, Go and Dart are both garbage-collected and compiled, even their lists of pros and cons look similar. Couldn't they just blend their features into one language (like, eg., add generics + some syntactic sugar to Go, to make it more usable for app and GUI code too?) instead of fragmenting the mindspace even more? Why don't people see the advantage of "universal" languages? It's obvious that developer love them and they are empowered by them, hence the success of languages like Javascript/Typescript and Kotlin despite their obvious flaws and limitations!
Chill. You don't want to go that way.
Talk to anyone in a country that experienced that... I can tell you that the way actual communism subtly broke the minds of people and twisted their thinking patterns is horrible. I takes more than a couple generations for a society to recover from that.
Try to fix what you have, don't go "burn the system down" and throw the baby with the bathwater!
Now, US politics... that circus looks weird from the outside, but whatever works for them ¯\_(ツ)_/¯
Finally more breathing room. And hopefully less competition for capital, so that the people actually really innovating and inventing new stuff have more chances of getting it.
Overall there's only reason for optimism!
Yey, can we get to this future faster?! Joking aside, all the current eco-friendly tech diminishes comfort by a loooot... we need low-energy habitats that are actually comfortable ffs.
For example, traveling around Europe, I see that the new trend in the developed/western part is to shiver you ass out during the cold months and to melt your brains out during the hot ones in almost all public spaces! Like we're back in the pre-AC era! You need to get to Eastern Europe's bigger cities to enjoy comfy heating/cooling habits like proper heating in the winter (yes, I want my >25 C in winter!) and proper cooling in the summer (<18 C please!)... It's wasteful, but very much enjoyable! Snow fighting after/before coming out/in of a 30+ C heated house is bliss :D Same a blasting through an enjoyable heatwave after jumping out of a 15 C office and then back in. Life's little pleasures.
After we invest so much of our lives in developing technology, we should at least enjoy the simple comforts and pleasures it freaking offers!
Everything is both!
Of course you'll have more and more proprietary-algos / blackbox-ML/AI-systems filtering content. And of course the content can't be manually reviewed (because SIZE) and you can't have the platform be responsible for it (what does it even mean to "be responsible for content"?!).
It's sad there's not much we can do to regulate away some of the toxic infos (like the "platform vs publisher" meme currently floating around...), but here it's actually a case where doing nothing and stomacking some of the bad consequences is 1000% better that doing something! Because that "something" whatever it may be will limit fundamental freedoms simply by having something exists in this space!
The point would be to limit/regulate targeting: either (a) they're a no-login and no-user-personalization place, and they do no targeting everyone gets a random sample from the same content (I'd really prefer this!), or (b) it needs to be very clear what kind of targeting is allowed... and the line gets very blurry here, amplifying hate speech for clicks and eyeballs can probably pay well and there need to be ways to solve this problem...
Anything that can metabolize lots of kinds of multicellular living matter could can do insane amounts of damage even if it's not a pathogen... for example agriculture is quite frail and we depend on it!
We're starting to be good at (1) using machine learning to "learn" designs for components much faster than naive-genetical-algorithms (aka "evolution"), and this will only get better as ML improves, and (2) we're getting better and better at simulating-in-software biological-like processes (and quantum computing will help with this). So we are pretty much in shape to vaaaaaastly outpower 3bn years of evolution.
A virtual + physical nanobots-evolution-gym where AI agents evolve and compete at creating better and better self-replicating nano-machines will give you the nightmare scenario easily...
Basically anything related to the combo of "ML + nanotech" should be carefully watched and monitored!
(And if you're willing to not start from scratch but instead use existing bio-building-blocks like viruses and bacteria, and don't have as high goals as "consuming all matter on earth" and instead you settle for "most multi-cellular living matter" then technically we already have the capabilities. You probably know what I mean.)
1. Use them if they take <5 parametets - no point making a function if you need to pass all the local context into it
2. Use them iff they return something that can be clearly named and have an explainable and understandable TYPE, even in dynamic langs where you don't explictly type things - if you'd end up returning a tuple of 3 arbitrary things that can't be "named as one single concept" it's a code smell
3. Kepp those functions local, maybe define them inside the calling function, at least private if they're methods, anything - if your single caller function becomes unwillingly part of an api, which can esily happen in Python with a _funk() method that users could ignore it's private, you've lost
But in fastai 1.0 it was all bundled together in one big yarn, with everything depending in the end on some data loading classes that depended on GPU driver etc.
Anyway, it was really bad architecture and dev practices in the codebase I was working though, the tested behavior would probably not have matched production one 100%... I don't blame fastai much for not helping with a broken workflow, but I prefer more barebones and less opinionated frameworks, aka using tf or pytorch directly, since some times you really need to get that "broken" thing running in production before you work on a refactored version of it :P Fastai seems very research-oriented and opinionated.
I'll definitely look into fastai 2.0 though :)
I get it that for any serious use you'd want a GPU, but for learning and toying around you might want to be able to run and debug code on your freakin macbook! Is that too much to asks? (Some of us do code in IDEs, not in notebooks + vim on server, and we'd want at least our test suite to be able to run locally ffs!)
(Also, hopefully they've got rid of the lovecraftian architecture with methods that can mutate an object's class [?!] - I understood the practical appeal and why they did it, but as a software engineer with sympathy for functional-programming that almost made me wanna barf :|)
Anyway, fastai is awesome for learning and experimenting, keep up the good work! I just hate it that it's so obnoxious to use and learn for anyone with a more traditional software engineering background...
Soldier on, till death and beyond! When did we become such suckers for short term results at the expense of the future...
Probably the production versions of those models are suboptimal in different ways but work better in practice...
How can you code know what data to expect without typed field? Your code size can grow up even 10x if you need to assume that any record can have any shape...
You could have more complex and user defined types in an ideal super-SQL, like "int or map:string->bool" etc., but you WANT types. They reduce complexity at all levels! You want them in you code, you want them in your data!
Sure, go ahead, just use MongoDB with no constraints on collections structure... You'll end up with 5x more code in you application and 5x more bugs.
SQL and the relational model help you think more, code less (and slower - but this is an advantage) and have less bugs with less effort! If you're lazy and prefer thinking deeper to "working harder", you'll always prefer properly typed SQL :P (Unless you're working with stuff that's naturally shaped as graphs or tensors.)