/tech/ - technology

>>1753
got some more advanced reading material senpai?
some kind of fundamentals in designing these, or more complex examples with the actual models attached/illustrated?

I'm always stuck with just piling in more and more nodes, but I understand a big part here is connecting multiple simple NNs together, but nobody online seems to care about actually implementing these, instead they rant about how cool they are.

That's really cool. I wish my library had a 3d printer, or they were economical enought that I could justify purchasing one. The biggest obstacle to me experimenting with embedded hardware is the fact that I'd have no case for what I design / build. 3d printing seems to be more niche than 3d printing enthusiasts make it out to be though.

>>1735
maybe they are economical, but you just seem like a person that has asked for too many handouts and they felt obliged to deny you. try going to the library person next time in a firm, but non-needy attitude

>>1737
what?

>>1524
>noncommercial license
That's nonfree

>>1674
lmao what

>>1331
of course, that's called capitalism, and they are obviously smart enough to still have a big market. the thing is, can they make enough money to change the cellular ownership so we can start moving towards VoiP calling.

the people in this thread are crazy. $599 is a good price for a smartphone of this caliber. I'm going to save for one and make it totally custom.

anyway, does anyone know how I can preorder from these guys without divulging information? like if I wanted to secretly mail it to (i should go to /paranoid/ shouldnt i?) an undisclosed address and put fake names and just pick up the package locally?
I dont currently have an address and it would really be convenient, expecially if purism is one of those fbi comprimised companies

>>1693
You are not free, your mind is licensed to bullsoykaf.

Statistical mechanics and cellular automata

There is a good discussion to be had on point of philosophical implications. We know from history that all advancements in science have followed with philosophical systems that act as sort of meta-proof or build on methodology of that particular science - i.e. Hobbs builds hes political philosophy on newtonian mechanics which can be seen from his fundamental principle of inertia which he adopted and applied it to social science. Who would have thought that we would have political philosophy (that is considered basis for modern liberal society) founded on new laws discovered in physics?

In quantum physics we have physicist fight over philosophical systems that quantum theory implicates. Bohr and Heisenberg have idealist philosophical outlook rejecting positivist, deterministic explanation of nature, for them everything is phenomena going that far that they claim that object of observation doesn't exist until we observe it.

Most interesting in my opinion dialectical materialism. Marx builds it on Hegels dialectics which reject concepts of isolated causality in classical mechanics. Bohm expresses similar criticism: Thus, there is no real case known of a set of perfect one-to-one causal relationships that could in principle make possible predictions of unlimited precision, without the need to take into account qualitatively new sets of causal factors existing outside the system of interest or at other levels.

Problems of isolated cause and effect and determinism that challenged classical mechanics in its contradiction to quantum mechanics give legitimate consideration in Hegels philosophical system. Therefor there are plenty of marxist philosophers that make effort to merge chaos theory with dialectical materialism. I guess that could be said for classical hegelians.

i.e. these are some accounts on hegel and chaos theory

It is easy to identify cause and effect in isolated cases, as when one hits a ball with a bat. But in a wider sense, the notion of causality becomes far more complicated. Individual causes and effects become lost in a vast ocean of interaction, where cause becomes transformed into effect and vice versa. Just try tracing back even the simplest event to its "ultimate causes" and you will see that eternity will not be long enough to do it.

The impossibility of establishing a "final cPost too long. Click here to view the full text.

>>538
Thank you for contributing, I thought this thread would die of neglect.

What you say is very interesting. Indeed, certain advances in maths, for example category theory, have had great rebound on philosophy, in lots of different areas (logic, but also ontology!). But I'd develop this on a more maths-specific thread.

I would like to stress on the fact that chaos is not only quantum. Actually, since quantum physics has a hard foundations problem, more mathematically-flavored accounts of chaos are realized in the realm of classical mechanics.

But the determinism hard coded in classical mechanics is nothing to be against. In fact, chaotic systems are, in a certain sense, ultradeterministic.

What I mean with this is that you may chose a property you want a solution to realize, and a chaotic system always has at least one such solution. So the dynamics in the chaotic realm is so rich it is akin to random processes: you may flip a coin and decide from the outcome that some solution should have some property, and the chaotic system will have such solution. This is a twisted version of determinism, where the possibilities, whose evolution along their "worldlines" is perfectly causal, are so many that the result is non-causal, entropic.

To expand, there always is a sort of "microdeterminism" in classical systems: the dynamics of one single point are totally determined by its equations of motion, well in contrast to the equations that govern quantum mechanics, which determine probability distributions.
So determinism is still there, but to really experience it, you must restrict yourself to an insignificant part of the system. Globally, or even locally (in the topology sense), determinism is not true: the evolution of a set of points is not guaranteed to mimic the evolution of a single point, and this is true also for non-chaotic systems (Gromov's non-squeezing theorem)!
For example we may take the single points of the set very close (small initial uncertainty), but the dynamics can separate them exponentially in time. As a result, the small initial uncertainty is propagated, deterninistically, to an enormous uncertainty at a further time.

All thePost too long. Click here to view the full text.

I just found this thread. It is a very interesting topic.
I'm going to study a course on ODEs next semester, so this is very good motivation.
I want to understand what the 'edge of chaos' means to physics.
Thank you for this intro Opie.

>>539
In regards to this post in particular: as above so below.

This all looks incredibly fascinating. I'll be reading the books posted here in the coming months for sure. Hopefully I'll have something to add when I've delved a little deeper into the supporting math and theory.

Never found the use of a wearable to very convenient. With my phone on me at all times, to have a other device just to control and view the basics seems very tedious. I could imagine getting one once technology allows them to be act on its own, without a phone only needed to link when it comes to texting, phone calls, etc.

>>117
I use a Casio F-91W but I have opened it up to put a small NFC chip capable of doing NDEF (storing soykaf for android, basically). I'm using it for my phone number right now, but not much else. Thinking of ideas that I could use it for.

Just a $50 casio solar runner's watch. Not hackable, not homemade, not cool. But it's a totally standalone zero maintenance gadget. No apps, no bluetooth, no online accounts or step tracking or any nonsense. It tells time and counts down and counts up, and I never need to charge it because six hours in the sun keeps it running for a year. I literally never need to think about it. I love that thing.

>>1667
EMPs near you silently
so watcha gonna do now, big boy?

>>1669
Not that anon but…

Literally nothing because minute-level precision of time is mostly useful to interact with societal constructs or for industrial processes. Survivalism doesn't have time to waste (heh) with such timing constraints.

I miss Kowloon, if only for the aesthetics. Ignore the fuck up blank post above….finger slipped.

>>1657
does missing it for only the aesthetic, imply that you had experience with it in other ways?

I find it interesting, perhaps. Sad to have gone away. but as I never was there, I do not think I miss it.

>>1642
Kowloon is a town in Hong Kong which used to exist in a jurisdictional grey area, and because their was essentially no laws or policing so it was super dense unregistered buildings with no building code.

>>1655
The point is, all the prep they need, you might as well just built something out of normal materials.

>>1670
They need galvanized and paint. Something that robots do to millions of cars every year on production lines around the world. Something the robots could do to prefab container structures and cladding prior to assembly, if they were churned out on a production line.

Wood rots. Brick and mortar needs re-pointed. Everything needs maintenance. The point is, cheap affordable housing does not come from currently existing materials and currently existing building practices and techniques.

Prefabricated houses have existed in the UK since the early 1900s when we needed to re-home large parts of the nation after the slum clearances and after the Blitz and other bombing raids during the second world war. Same for many European cities.

Today, prefab houses are a similar price to traditionally built homes because of low production numbers and bespoke high end finishes. When churned out in bulk and to one standard, such as for student or social housing, they're much much cheaper to produce, maintain and replace than any brick and mortar, permanent structure. It's already been proven in Europe.

If you can put together a container home production line to address the housing crisis, it will be the cheapest means of house construction we've seen since building with wattle and daub and as revolutionary as when Ford moved automotive production from bespoke coach built, to the production line.

The swastika next to it makes me feel worse.

>>1619
Yeah same, but don't worry, I fixed it

initially the text was pink with a black background

here we go, now with additional message

Kill your LEgo??
Swastika now is a emoji?

Check out Luke Smith's video walkthrough on this very subject, alice
https://youtu.be/Q4I_Ft-VLAg

I'm just learning to program, so I wouldn't know if it's the best way, but I configured my vim for Python according to this guide here:
https://realpython.com/blog/python/vim-and-python-a-match-made-in-heaven/#auto-complete
I'm still having a little trouble with the auto-indentation, but besides that it's working great as far as my capacities can judge.

>>1611
I've used the same guide to configure vim and it's working good so far.

For other languages I usually look up a 'styleguide' on how to write the code and then edit my config according to that, vim is a lot of fun and I should have used it a lot sooner, I'm glad you're starting with it