livueta
Joined 8 years ago
Karma
2,379
Submissions
506
Member since
Nov 16
Recent submissions
Comment · 2 years ago
Sibling comment has good suggestions. I personally like Dell Optiplex models, e.g. 9020. Looks like right now the best you can do is about 40-45 shipped for a complete one, but I've snagged about six in the last year or two for 20-35. They come and go. Beware of sellers posting the same thing for 100+; they're worth max 50.
Comment · 2 years ago
Nope, and that definitely would have helped, but I bet 99% of people looking to use an old phone as a server aren't going to do all that either.
Comment · 2 years ago
The biggest issue I've found while trying to do this is a very high rate of battery failure while plugged in 24/7 under load. 4/5 phones I've done this with have become spicy pillows in the 1-2 year timeframe. They were all Samsung or LG devices, so not complete bottom of the barrel garbage batteries either. IME if you need a cheap server, a used enterprise micro pc has a much lower chance of potentially burning your house down, can be found on ebay for literally $30, and can be easily stuffed full of ssds.
Comment · 2 years ago
While I don't think you're wrong to say that the trends/regime were the same then as today, that's very much more obvious in hindsight than it was at the time. You've got to recall the triumphal attitude in the West towards China circa the run-up to WTO admission in the late 90s. Read Clinton on the topic: https://www.iatp.org/sites/default/files/Full_Text_of_Clinto...
The trendy thing was to believe that exposure to global markets and the Internet would inevitably result in further liberalization, and anyone who disagreed was probably a car-burning anarchist. It was only in the 2010s that the fact that reality wasn't quite that beneficent started creeping into mainstream neoliberal perspectives.
To be fair, China really had seen massive changes under Deng, so the idea that continued reform would eventually result in major political liberalisation wasn't completely insane. It went from Mao to a semblance of rule of (in retrospect, obviously "by" rather than "of", but whatever) law, so was extending the trend line really so implausible?
In retrospect, obviously, yes, it was implausible. But that general vibe meant that many really did think that the Sino-British Joint Declaration would be respected.
Of course, the actual history of how that whole thing turned out w.r.t. tearing up the treaty and burning the bits in the fires of national-security laws, has altered perspectives on how well Taiwan would cope under rule from Beijing. Now that the concept of one country, two systems - that idea of a special status - is widely understood to be a meaningless platitude, the assumptions that were common at the time of HK reunification are totally invalid. You can see this in trends of polling of Taiwanese attitudes to reunification: as HK has been strangled, the Taiwanese public has come to understand the worthlesness of any of the sort of assurances that were offered to HK, which at the time assauged a lot of objections.
Finally, there's a matter of basic military calculus: HK is a lot harder to defend than Taiwan. It's not a natural polity grounded in geography, and not potentially self-sufficient in the same ways as Taiwan. Thatcher didn't agree in 1984 because she loved and trusted Deng, but because he could credibly have taken it by force in a day and told her so, so, in a very real sense, there wasn't much of an alternative.
Comment · 2 years ago
It's been kind of enlightening seeing leadership at $BIGCORP push AI coding solutions like they're guaranteed to be a 10x increase in velocity in every context. Feedback from ICs isn't wholly negative - there are definitely situations where it can be useful, like quickly grokking common applications of common tools, or semi-intelligently applying a diff pattern that is more then just a regex - but there's a complete unwillingness to hear any feedback that isn't "this tech is a total paradigm shift that allows us to finally get rid of all these pesky and expensive developers". Reports of, for instance, the introduction of subtle bugs that take extended amounts of time to understand and fix, are met with outright hostility and accusations of incompetence. When a complex defect or escalation drags on, a common question is "why haven't you asked AI to fix it yet", belying a total misunderstanding of the sorts of tasks the tool is applicable to. The kool-aid is not so much drunk as rectally infused. If valuations are based on this sort of outlook, whew, this market is totally fucked.
Comment · 2 years ago
You can trade range for bandwidth by playing with the spread factor, an encoding tunable: https://meshtastic.org/docs/configuration/radio/lora/#modem-...
But generally yes, latency is too high and bandwidth is too low for synchronous voice. The upside is that real-world performance often exceeds the range of analog UHF even at significantly lower frequencies, e.g. GMRS.
Comment · 2 years ago
Sounds like the plot of a Fukumoto Nobuyuki manga.
Comment · 2 years ago
Re ISM band usefulness: yeah. I grudgingly admit that the propagation characteristics of HF and probably even VHF aren't a great match for the ISM governance model, but at least everywhere I've ever used radios, UHF HAM bands are as dead as a doornail and don't (fine, generally) propagate far enough for small numbers of negligent users to create major problems for large amounts of users. Take some 400mhz UHF and make it ISM. 902-928mhz ISM has a ton of great stuff like LoRa going on, but access to 400mhz bands would be a huge shot in the arm to the practical usability of projects like Meshtastic without significant risks to the general usability of that chunk of spectrum.
Another interesting thing about the history of protocol development in ISM bands is how the need to cope with noise/overuse/other forms of spectrum degradation has spurred enormous advances in encoding and signal-processing techniques, quite contrary to the initial expectation of useless garbage dumping zone bands.
Comment · 2 years ago
Comment · 2 years ago
> and then I'll use passkeys in keepassxc.
If the auth cartel deigns to allow it:
https://github.com/keepassxreboot/keepassxc/issues/10407
https://news.ycombinator.com/item?id=39698502
https://news.ycombinator.com/item?id=39706876
Attestation makes passkeys inherently anti-user, full stop.
Comment · 2 years ago
Comment · 2 years ago
An interesting facet of this that makes me feel somewhat more charitable to cashless payments in that context is that it's very easy to perform cashless payments in Japan pseudoanonymously, in that you can buy an IC card from a vending machine, load it with cash, and use it for payments all without explicit links to your KYC'd banking identity. Sure, if you attracted individualized security state attention they could probably pull security camera footage and figure it out, but that approach doesn't scale into turnkey tyrrany; sort of how like $5 wrench cryptanalysis doesn't prevent widespread strong cryptography fundamentally altering the envelope of what kinds of automated, at-scale privacy abuses are possible.
This is a fairly stark contrast with the West, where pseudoanonymous cashless payments basically don't exist. The closest thing (stuff like xmr aside, where the chance you can use it to buy a coffee is very low) are prepaid visa gift cards, but the UX on these is terrible for a lot of reasons and so aren't anywhere close to the practical usability of cash-charged IC cards in Japan.
Comment · 2 years ago
You might enjoy it if you're after thoughtful story-driven shows from the last decade. It's on my personal mental shortlist of those right alongside ODDTAXI and Ousama. Knowledge of rakugo is definitely not required.
Comment · 2 years ago
Have you seen Shōwa Genroku Rakugo Shinjū?