Tainnor

41 ms·

841.
by Tainnor 3y ago
For all of Canonical's faults (e.g. snap really seems like a bad technology), Ubuntu is still a Linux distribution and if you don't like a certain thing, you can always swap things out. If you don't like GNOME, use KDE, Xfce
842.
by Tainnor 3y ago
> Flying in a 737 Max is safer than getting vaccinated, I think the Astra Zenica clotting thing was about 50 per million and that was the one I got. That's a false equivalence because the choice wasn't between exposing yourself
843.
by Tainnor 3y ago
The reason why it's only such a "small" increase is precisely because planes are being grounded. Imagine that after the 737 Max incidents in 2018, they hadn't grounded the airplanes. How many people might have died then?
844.
by Tainnor 3y ago
Ultimately, there aren't that many people who have the technical knowledge, the ideology (that the internet should be for the good of the general public, not for corporations), the dedication and the time - and those who do exist are s
845.
by Tainnor 3y ago
I love that some 20-30 years after that famous chat somebody still fell for it.
846.
by Tainnor 3y ago
Yes, but not every sequence in R converges. That's of course also true of a compact space, the author was being imprecise there. What they meant was definitely sequential compactness, as mentioned by a sibling comment.
847.
by Tainnor 3y ago
Of course, people are people and react to certain stimuli. But I still remember the internet of old. Not only the small, semi-gated internet forums, but at also when Facebook just started out and you would just talk to your friends and not
848.
by Tainnor 3y ago
Social media is of course not responsible for all our problems, but I do blame specifically social media for making people (including myself) who are materially well off and live a comfortable life irrationally angry about things outside of
849.
by Tainnor 3y ago
Wow, that's wild. Not only are they writing down things everybody in that field should know, but they fail to even properly understand the very basics of regular languages. Literally what you learn in the first week of an intro to CS c
850.
by Tainnor 3y ago
> Any non-trivial architecture is not just Turing complete, it can perform hypercomputation (solving the halting problem), if it can use arbitrary precision numbers. Can you explain this part? (It doesn't seem like CS theory 101 to
851.
by Tainnor 3y ago
It appears that it's exactly the distinction between fixed and arbitrary precision that decides whether or not transformers are Turing Complete: https://openreview.net/forum?id=HyGBdo0qFm&noteId=r1lxKiAJlV I think
852.
by Tainnor 3y ago
> Yes. Anything with finite memory is not Turing-complete, and transformers are finite-state machines. So what? Anything with infinite memory is un-buildable. I think that's missing the point. A Turing Machine description is indepen
853.
by Tainnor 3y ago
This response by the authors to the reviews to me looks like they didn't really understand the objection: > First, it's worth noting that different reviewers sometimes gave opposite critiques of the paper, e.g. Reviewer erh8: T
854.
by Tainnor 3y ago
You're right that we can never find conclusive empirical evidence for something being truly random, but we can say with confidence that a PRNG is not random because we can look at the algorithm.
855.
by Tainnor 3y ago
That's not a uniform distribution. Probability doesn't remember previous state, so the chance of picking the number 241 out of 1 million numbers remains exactly the same after picking 241 once. In particular, the chance of getting
856.
by Tainnor 3y ago
Technically, it's not actually uniform, or "random" - it only has to be indistinguishable from random, e.g. if you had access to a true random source and a pseudo-random one, it should be computationally infeasible to disting
857.
by Tainnor 3y ago
Swiss public transport is IMHO special in three respects: a) density of the network, b) frequency (even remote areas get a bus an hour), c) punctuality. I'm not saying no other country can match this (I haven't been everywhere), b
858.
by Tainnor 3y ago
There is a sense in which the even integers can be considered "half" of all the integers: https://en.m.wikipedia.org/wiki/Natural_density "Counting stuff" can be generalised to infinite sets in diff
859.
by Tainnor 3y ago
You're constructing a paradox where none exists. The completeness and the incompleteness theorems don't contradict each other, they are about fundamentally different things (and it's really a shame we use the same terminology
860.
by Tainnor 3y ago
"A system encompasses all facts derivable within its framework" is a very vague sentence, but the way it's written I would consider it false. Second order logic is incomplete, for example. The fact that different set of axiom
861.
by Tainnor 3y ago
Not a contradiction, certain areas skew the statistics, but Cheddar isn't really all that popular in, say, Italy, France, Switzerland or Spain which are all known for their own local cheese production.
862.
by Tainnor 3y ago
I don't think that Kafka is bad per se, it's just that it's often used as a message queue which it doesn't seem optimised for. Half of the stuff you need for a proper MQ (e.g. DLT handling) isn't provided by Kafka p
863.
by Tainnor 3y ago
> insanely popular in certain areas of the world.
864.
by Tainnor 3y ago
What Grandi actually wrote was that Parmesan Cheese, which was imported into the US by Italians, continued to evolve in Italy but not in the US. I have no way of verifying whether this is true but it wouldn't be terribly surprising if
865.
by Tainnor 3y ago
The semantics of Java aren't really optimised for FP the way they are in Haskell (e.g. no TCO, no guaranteed immutability prevents certain optimisations), but if you write FP code in Java that leads to stack overflow, you're still
866.
by Tainnor 3y ago
The kind of anti-intellectualism that is espoused by this blog post is what sometimes makes me want to leave this profession. Certainly, there are many people who overcomplicate solutions - for a myriad of reasons. The idea that therefore i
867.
by Tainnor 3y ago
You can use a VIM plugin for many IDEs, including IntelliJ.
868.
by Tainnor 3y ago
You haven't answered my question: at which point do you say you "know" Gödel's first theorem? Or for an easier example: at which point do you "know" Java? Do you need to understand all the intricacies of the cl
869.
by Tainnor 3y ago
> Nothing particularly wrong with saying "outrageous" things if that's what you personally believe. You can have that opinion, but I'll just point out that there's enough people who feel alienated if straight up
870.
by Tainnor 3y ago
I hate this idea that discussion on HN is inherently "better" or "more civilised" than in other forums. Some HN users will say the most outrageous things but with a superficial layer of "civility". Parts of thi

More ›