_0w8t

12 ms·

91.
by _0w8t 4y ago
Please do not bring kids. It is irrelevant here and can be even factually wrong. In a poor country kinds are forced to do all kind of dangerous work. If not for cobalt they can be sent to fields and be subject for all chemicals used in mode
92.
by _0w8t 4y ago
The bug in question would not happen if a reviewer demanded to split the complicated if. Yes, it would add a few lines depending on the overall control flow, but code in question is hard to read and must be written in a less dense way.
93.
by _0w8t 4y ago
I also think that the estimated growth was too big to be caused by such systematic errors. On the other hand we do not have models of a galaxy evolution based on solving equations of General Relativity. Typically the assumption is that Newt
94.
by _0w8t 4y ago
I do not see how it follows. A smart developer will choose a database, not a file system, if data integrity is important. So it can be simply the case that the transactional API with a lot of its limitations were simply not flexible enough
95.
by _0w8t 4y ago
It was plausible for the first case. But when the second time it happened on the Russian section again one starts to wonder if there was more to the story. I.e. why have not micrometeorites hit equipment of other countries?
96.
by _0w8t 4y ago
This describes the current situation in Russia with Putin surrounded by yes men. But it is not true about China where officials are frequently fired loosing pretty much everything. I.e. if one always say yes in China, the consequences for n
97.
by _0w8t 4y ago
Now that US has learned that the big ballon was spy one, it would not be an embarrassment but a reasonable assumption that smaller ones could be spy ones as well. So they should be stopped.
98.
by _0w8t 4y ago
It accumulates on top, turns into ice, but the bottom layers of the ice very gradually flow (tenth of thousandth of years) toward the see. Modeling that precisely is hard and we do not know if Antarctic as the whole gains ice or looses it.
99.
by _0w8t 4y ago
Crash logs for fp-less builds are rather useless. And useful logs are very valuable feedback that the users can provide to developers.
100.
by _0w8t 4y ago
Sensibly configured compressed memory brings benefits of swap without much of its shortcomings. So use that, not swap.
101.
by _0w8t 4y ago
Methanol produced using green energy is much better alternative to hydrogen. It is vastly easier to handle and store.
102.
by _0w8t 4y ago
This fixes a security vulnerability introduced in OpenSSH 9.1, https://lwn.net/ml/oss-security/20230202130212.GA15689@local... .
103.
by _0w8t 4y ago
Heat to electricity conversion ratio with a realistic Tokamak design will have 30% efficiency at best, accounting for other losses will require Q at least 5. But even that may be too low so to play safe the latest Tokamaks target at least 1
104.
by _0w8t 4y ago
While I do agree that the extended video went into hand waving arguments on few occasions, that is not because Helion may know something that general public do not, but rather because nobody knows! Helion own admission is that they use 2-d
105.
by _0w8t 4y ago
The new video by the same guy https://youtube.com/live/wK5v1Nw98xs from 32 min describes how Helion will get a lot of D+T neutrons from the secondary reactions despite all their efforts. Even without that the neutron f
106.
by _0w8t 4y ago
The new video by the same guy who created the original criticism https://youtube.com/live/wK5v1Nw98xs starting from 2:03:00 explains that 95% recovery efficiency is not the whole story.
107.
by _0w8t 4y ago
With Helion pulse design peak neuron flux luminosity will still be much higher than that was observed in nuclear reactors. Compared with tokamaks their magnets will be directly exposed to it. Plus their idea to convert fusion energy directl
108.
by _0w8t 4y ago
VSCode has vim extensions emulating keybindings similarly to Emacs Evil mode. Although I am not a fun of vim I ended up using one of those when I needed to switch between VScode on Mac and Windows on a daily basis due to uniformity of keybi
109.
by _0w8t 4y ago
I am very skeptical about Helion, https://www.youtube.com/watch?v=3vUPhsFoniw
110.
by _0w8t 4y ago
Tokamac cost depends on its volume which is like forth power of its magnetic field strength. If some engineering breakthrough allows to go to 40 Tesla in a tokamac compared with 20 of the designs like SPARK, then a tokamak that can be trans
111.
by _0w8t 4y ago
Typically an Intel-based NUC does not have particularly good low-power management for mostly idle workloads. So running rarely accessed network software will be like 15 Wt or 15 USD or 20-30 Euro per year. RPi will be 10 times lower.
112.
by _0w8t 4y ago
The cost does not include the power consumption. For Intel-based Mac Mini running a network filter I suppose it will be at least its idle number 20 Wt or about 200 KWt/year or 20 USD/year (using 10 cent per KWt as an average price
113.
by _0w8t 4y ago
Idiomatic Erlang and Go are fundamentally different. In Erlang there is one mailbox per process and multiplexing is archived via polymorphic messages. In Go there can be multiple channels and one multiplexes them using the select statement.
114.
by _0w8t 4y ago
It was germanium glass if I remember correctly. It was very essential that all thermal radiation passed through it.
115.
by _0w8t 4y ago
A simple rule of thumb is that for samples with N < 100 be very skeptical for the results as those can be archived simply by randomness on top of small systematic errors. Proper statistics helps to rule out randomness, but not systematic
116.
by _0w8t 4y ago
Those guns are not cheap. I once worked in a startup that wanted to put thermal cameras on commercial planes to measure volcanic ash concentrations. The cameras had to be behind a special glass that is transparent for thermal radiation. But
117.
by _0w8t 4y ago
As was pointed out in the article a late async review can ensure that somebody else understands what happens in the code. In a few cases this alone justifies such review even if it is too late to change the architecture. Plus besides proble
118.
by _0w8t 4y ago
As explicitly stated in the proposal the new rules about super/this calls are exactly the same as for a blank final field initialization. So the change simplifies some aspects of the language as it removes number of rules. Moreover, gi
119.
by _0w8t 4y ago
The hash can be uploaded somewhere. Alternatively a seed to randomize the executable can be derived from an immutable token like serial number. Yet another solution is to re-sort the executable into a stable order and compare the hash of th
120.
by _0w8t 4y ago
A local package manager that performs the linking can save the hash of the result.

More ›