1,206 karma · joined June 6, 2019
[0] https://bitwarden.com/help/install-and-deploy-unified-beta/
Reader: https://www.kiwix.org/en/download/
Content: https://library.kiwix.org/?lang=eng
Kudos to the author for putting in the effort to make it so easily reproducible.
Years later, a line in Matthew Crawford’s best-selling manifesto of the manual arts, Shop Class as Soulcraft, echoed the feeling my mentors had instilled in me. There is a type of person, he writes, who “hates the feeling of dependence, especially when it is a direct result of his not understanding something. So he goes home and starts taking the valve covers off his engine to investigate for himself. Maybe he has no idea what he is doing, but he trusts that whatever the problem is, he ought to be able to figure it out by his own efforts. Then again, maybe not—he may never get his valve train back together again. But he intends to go down swinging.”I found an interesting counter perspective on the Mindscape podcast[0]:
And Ted Chiang, in that article, suggests that ChatGPT and language models generally can be thought of as a kind of blurry jpeg of the web, where they get trained to compress the web and then at inference time, when you're generating a text with these models, it's a form of lossy decompression that involves interpolation in the same way. And I think it's a very interesting test case for intuitions because I think this metaphor, this analogy, parts of it are pumping the right intuitions.
There is definitely a deep connection between machine learning and compression that has long been observed and studied. [...] I think comparing it to lossy image decompression is pumping the wrong intuitions because, again, that suggests that all it's doing is this kind of shallow interpolation that amounts to a form of approximate memorization where you have memorized some parts of the data and then you are loosely interpolating what's in between.
[...] the intuition is that there would be a way, presumably, to characterize what large language models and image generation models are doing when they generate images and texts as involving a form of interpolation but this form of interpolation would be very very different from what we might think of when we think of nearest neighbour pixel interpolation in lossy image decompression. So different, in fact, that this analogy is very unhelpful to understand what generative models are doing because, again, instead of being analogous to brute force memorization, there's something much more generally novel and generative about the process of inference in these models.
Someone already mentioned Ben Eater's Building an 8-bit computer. Nans2Tetris [1] is a similar course but completely in software, where you start from logic gates and end up with a fully functioning OS. Checkout the cool projects students built on top. [2]
[0] https://www.youtube.com/playlist?list=PLrxfgDEc2NxZJcWcrxH3j...
One can hope :)
The first principle is that you must not fool yourself—and you are the easiest person to fool. So you have to be very careful about that. After you’ve not fooled yourself, it’s easy not to fool other scientists. You just have to be honest in a conventional way after that.
All this to say, both the current state of our attempts to reach the next frontier of human knowledge (string theory, say) and Sabine's criticisms follow age old traditions of back and forth in science. Both are needed, the former to make actual progress and the later to keep the bad ideas in check.
Having said that, I personally detest the particular populist-with-lack-of-nuance approach Sabine (and others like her) take. I like Sean Carroll's books [2] and his Mindscape podcast [3] for a more nuanced view. For instance, his recent podcast with Raphael Bousso [4] is such a nice overview of the current state of the art Physics. On string theory:
String theory, I think, has given us from my perspective, as far as I can tell, the only consistent candidates for theories of quantum gravity that are in any sense complete. In particular, it has led to the discovery of the AdS/CFT correspondence. That is a lot of letters. But what it means is that there’s a certain class of universes, unfortunately, not including the one that we seem to be living in, but still a very rich set of space times worlds that you can imagine, for which we have a complete quantum theory of gravity. Or at least it could be wrong. We haven’t tested it experimentally. But it’s a very significant statement that you have a theory which is complete in itself, it’s not missing anything. You could put it on a computer and run it, you can calculate things with it. And so that I would say is, from the physics perspective, the number one thing that has come out of String theory, there’s a lot of beautiful mathematical results, and many other applications that the theory has had.
[1] https://www.goodreads.com/book/show/19395553-our-mathematica...
[2] https://www.preposterousuniverse.com/books-courses/
[3] https://www.preposterousuniverse.com/podcast/
[4] https://www.preposterousuniverse.com/podcast/2022/11/21/218-...
For example, I could be a brain in a vat, being fed sense signals through wires that is making me feel that I live on this blue green rocky planet revolving around a big ball of fire. If so, "I am a brain in a vat" is the objective reality, even though the "I" inside this simulated universe may never be able to perceive the true reality.
It's true that we can never be 100% sure that our sense signals are truly objective and not something emergent (spacetime of our universe) from a more fundamental reality (say, brain in a vat). Then what should we do? I believe the best thing we can do is to follow the data. As long as we continue to observe that our universe is consistently and precisely obeying the physical laws we have discovered, we can axiomatically assume that we do live in objective reality. As we dig deeper, we may make new observations indicating we exist in a Matrix after all. We'd then have to update our mental models accordingly.
Would you opinion change if everything in Elasticsearch was in-house?
I think there is a broader question about spirit of open source licensing here. If I'm building a new DB from scratch, should I choose a true open source license and risk giving up competitive advantage in my ability to monetize it, over Amazon-like companies who have the benefit of brand and scale? Or is it okay to just make the code source-available, which is still better than closed-source?
I like the balanced approach some companies are adopting: make the code source-available initially and automatically transition to a open source license after N years.
Logic is not constrained to human conceptions but is objectively valid. One has to keep in mind that logic is based on premises and inferences. Given the premises and inference rules, the conclusion automatically follow.
So a four sided circle cannot exist because we have defined circle to be a particular shape. A valid four sided circle would need a change in the definition of circle itself. But then we're not talking about the same thing anymore. Similarly, the same universe cannot contain both an irresistible force and an immovable object, because of the conflicting definitions of those terms.
Even an omnipotent God is constrained by logic. In a reality without logic, it'd meaningless to try to understand any aspect of reality. God can both exist and not exist at the same time. He can be both Good and Evil. Conscious beings can be both alive and dead. It is hard to see how a illogical universe can retain any meaning that we attribute to God and Creation.
# tlp-stat
defaults.conf L0019: SATA_LINKPWR_ON_AC="med_power_with_dipm max_performance"
defaults.conf L0020: SATA_LINKPWR_ON_BAT="med_power_with_dipm min_power"The misinformation is that people's belief of the world are based on statistics that were only true decades ago and haven't been updated. For example, the distinction between the developing and the developed world. The world has actually gotten better and we have reasons to be more optimistic.
However, I shouldn't have stated my own interpretation that way.
> a down-playing of issues of serious concern
Unless we really know the state of the world, we won't come up with the right solutions. For things that have improved, we should be optimistic and celebrate the successes. We keep hearing about the hunger problem in Africa, but did years of initiatives work or not? Are education programs really working or not? Are our efforts actually making the world better? We never seem to hear the success stories in media.
That doesn't mean we rest on our laurels and stop fixing the remaining issues.
As to the other concerns, this is not a traditional survey whose results are to be used seriously. The questions are meant to educate that many of us live with outdated notions of the state of the world. If you watch the talks, they show how even "experts" at the universities and world organizations such as the World Health Summit [0] fail to be better at the questions than a group of chimpanzees answering randomly.
I should have been more careful with my enthusiasm here. So the faults are all mine. I encourage everyone to explore the link and the talks and form their own opinion.
Given their goal is to show that we have reasons to be more optimistic about the world, I think the quiz setup makes sense. It highlights that we can't even distinguish between optimistic 10% and pessimistic 50% for issues we regularly hear about in the news. I'm exaggerating here, but the 50% would have been obviously wrong but isn't because we didn't look at the data at all, instead made an assumption based on our perceptions. This hard hit is a good thing and fulfills the goal of the quiz.
But yes, choices of [1%, 10%, 20%] would likely make more people choose 10% than among [10%, 30%, 50%] when 10% is the right answer. Thanks for the insight.
But thanks for sharing this. Always good to learn the nuances.
Gapminder is an attempt to show how outdated our view of the world are. The first time I had come across Rosling's TED talk (see my other comment), it did strike me how little I had actively looked into the state of the world and how the state is more optimistic than one can conclude from doomscrolling [0].
This classic TED talk [0] by Hans Rosling presents the same idea. I recommend watching all his other talks as well [e.g., 1-2].
[0] https://www.youtube.com/watch?v=hVimVzgtD6w
Well, do you have a source that does this mathematical calculation?
That's a lot of real estate!
1) We don't know how life originated on Earth.
2) We don't have a full idea of what happened during the Cambrian explosion.
3) We don't know how human consciousness manifests itself in the brain.
4) Evolution by random natural processes is not possible without deliberate nudges.
Is the claim here that 1-3 is simply not possible without an intelligent entity intervening in the natural laws? If so, I'm personally fine with it. God (who is the most popular potential Intelligent Designer) has always been a God of the Gaps. In the future, I am pretty confident we'll figure out the explanations using just the physical laws of the universe.
But if the claim also includes 4, then I think it can be discarded with a high confidence. We can already observe how evolution works at the virus and bacteria level. We have found the transitional fossils [0]. The science of evolutionary development biology [1] is already giving us great insights into how organisms translate the genetic code to build our complex bodies. It also shows how the same genes have been reused across species (e.g., the genes for eyes in the house fly and humans are the same). We don't need the Intelligent Design hypothesis to explain evolution.
(EDIT: I see from another comment that you already agree to the above point.)
The claim Intelligent Design supporters can make at best is that God created the first cells, perhaps nudged them a little during the Cambrian explosion, and perhaps again intervened before the evolution of Homo Sapiens. But all the rest happened the boring way, following the natural laws.
> we have never seen life come from non life.
Does not imply we never will. Given the countless other times this argument has been used to justify an Intelligent Creator when we didn't know something, I give very low credence to the idea that /this/ time is truly it. Unless someone can show mathematically why life cannot possibly emerge from the natural laws, naturalism remains the best hypothesis.
[0] https://en.wikipedia.org/wiki/Transitional_fossil [1] https://en.wikipedia.org/wiki/Evolutionary_developmental_bio...
"Whether it’s Dragonfly going to Titan, Clipper going to Europa, hopefully a lander going to Europa, hopefully a mission that’ll fly through the plumes of Enceladus, exo-planets going gangbusters, SETI hopefully taking on a broader and broader search and surveying those exo-planets. Within the next few decades we could potentially answer this primordial, age old question of, “Are we alone?” And that’s gonna revolutionize biology. It’s gonna revolutionize how we think about our place in the universe. And so for all of the pains and agony of trying to operate on these time scales, we do live in a beautiful time where we might transform the universe in which we live into a biological universe."
Along with the commercialization of space travel, I'd say there is a sufficiently large part of this universe we are yet to "touch" but soon will be able to.
I for one am tremendously excited about the future.
[0] https://www.preposterousuniverse.com/podcast/2020/04/13/92-k...