"Treat infrastructure as possible failure points and prepare accordingly" holds just as well for the payment infrastructure. Interesting and fresh perspective, thank you for sharing.
5,397 karma · joined April 6, 2009
8½
return -ENOCOMPREHEND;
in fields as crowded as public health and medicine, experts voice all sorts of advice, citing all kinds of research, and often contradicting each other. At the end of the day, we regular citizens usually hear from those experts that got their voices amplified by operators of platforms (journalism, political institutions, etc.). In Robin's terms, the elites choose from the crowd of experts those that they like the advice the best.
Cf. >Lycoming provides a list of engines and fuels that are compatible with them. According to their August 2017 chart, a number of their engines are compatible with unleaded fuel.
>However, all of their engines require that an oil additive be used when unleaded fuel is used: "When using the unleaded fuels identified in Table 1, Lycoming oil additive P/N LW-16702, or an equivalent finished product such as Aeroshell 15W-50, must be used."*
I'm with you here - especially including the observation that it was the non-free market force ("It couldn’t be patented") that skewed the choice in favor of the inferior, poisonous option.
Side note, besides its anti-knock properties, the lead also had protective effect on the valves - with early metallurgy, the high temperature gasses wore out valves, in particular the exhaust ones; lead partly ameliorated that. It is a concern with older vehicles (aircraft and cars) and they may require leaded gasoline for that particular reason - or at least replacement of relevant engine parts.
A compounding concern is that of a propeller tip speed - when it approaches speed of sound, the physics of airflow become complex and it's very hard to maintain efficiency - and near impossible to maintain reasonable noise levels. This necessitates keeping blades of high speed propellers rather short. The Tu-114 was exceptionally loud for a plane, both due to propeller speed concerns, and also due to contra-rotating propellers inherently causing increased turbulence.
There is a good four-part series on propeller theory and practice:
https://www.youtube.com/watch?v=SgoPPg8oVt8&list=PLD2EcpzcvT...
I take umbrage with those characterizations.
The practices described (pre-arranged release of information, voicing mutual support in coordinated manner, agreed-upon language and form) have for decades been the hallmark of professional marketing and journalism. Back when print and broadcast media were the top game, those methods were used by the small groups of legitimate journalists and marketers.
Twitter correctly recognizes coordinated release of information as signal of particularly important and valuable content. People organically coordinate release of information for it to get its full due impact and attention. People also organically ask their friends and business contacts to chip in with an upvote or reblog (or whatever is the equivalent on Twitter). Calling Twitter's or users' behaviors "callous" or "inauthentic" when it's the regular people - that is way off the mark.
My uncharitable read of it is - this whole venture reeks of gatekeeping for the old-guard legitimate journalism.
Nothing in the article as posted indicates that the "disinformation influencers" were nefarious actors. For all the description given, it might have been grassroots citizens action, only labeled "disinformation" by officials or government-aligned sources. The end result is Mozilla making arbitrary choice between two opposing camps of political activists - and reports on Twitter users along those lines with clear hopes of getting them banned.
I'd understand the point if the activism was directly related to open internet, to freedom of expression, interoperability of services, ease of access and so forth - if there were concerns closely related to Mozilla's core mission. However nothing in the article nor in the linked PDF seem to allude to any of such concerns. It feels like a small group of Mozilla employees[1] ran this research and reported on users for their own private reasons.
[1] "in-house activists" might be a more charitable characterization
Curiously enough, we've seen anonymous services succeed for results unrelated to privacy-from-government: places like Omegle & ChatRoulette use anonymity for fun; places like 4chan use anonymity both for fun and also to avoid certain problems common to name-posting. Anonymity has long been a viable alternative put to good use in literature, arts, and entertainment. Perhaps the proper marketing would be along those lines?
Not if Elon succeeds with Starlink:
>>@thesheetztweetz: How does transmitting into a country without a local downlink work on the regulatory side?
>@elonmusk: They can shake their fist at the sky [0]
For context, certain countries like India have quite strict regulation of satellite comms, requiring special permission[1] even to use plain consumer tech like Iridium. I presume EU would also try to tightly regulate consumer satellite comms, just like it requires real (government issued) ID to use cell phones - specifically to register locally purchased SIM cards, again for national security reasons.
[0] https://twitter.com/elonmusk/status/1433123220643717120
[1] https://www.osac.gov/Content/Report/9db45731-1eec-477a-a7af-... >There are multiple instances of authorities confiscating undeclared satellite phones from foreign travelers upon arrival in India. The official notice states: "All foreigners travelling to India are hereby informed that it is illegal to use/carry Thuraya or other such satellite phones in India. Custom authorities in India may seize such phones and legal action may be taken against the passenger concerned."
Four words: The Intercept, Secure Drop. A one-way mail (content submission) system that runs exclusively on Tor, and thus can't be supboenated for users' IP address.
While I get your premise due to concerns of law, I think it is entirely feasible - and hinges on execution, marketing ability. We already have IP-hiding technology, whether Tor or Freenet or other such. The concern is "are we good enough yet to make it a sustainable business?"
For sake of example: if a hypothetical competitor to ProtonMail was to offer sign-up and email access only over Tor protocol, it would effectively be safe from police's demands to start logging IPs - thanks to technical measures. The actual difficulty is in the business side: getting enough paying customers to install & enable relevant browser or browser plugin.
Granted, the police could try to force the hypothetical competitor to install malicious JavaScript that would try to gather & leak users' IP address or other identifying information through other means, but that's solvable in the longer run just as well.
This assumption was tested extensively by several platforms, and it failed miserably all the way. You can still witness it yourself, on Facebook, where almost everybody posts under their own name and photo - and yet trolling, brigading, mass reporting, flamewars, and all other continue in force. Even worse, you'll find plenty of bot accounts on places like Twitter, with perfectly fine name and picture, while insightful and dissenting content is delivered by anonymous, or ironically branded, accounts.
The correct solution is to stop trying to shoehorn our preconceived notions from offline life to online activities - and to reminding individual users that power is in their hands. The solution to Eternal September is to teach and integrate newcomers - and sometimes to also gatekeep - rather than trying to become a campus police of sorts. The much different nature - the vastly larger scale - of the internet makes it entirely unsuitable to govern by offline intuition.
so far GitHub Copilot is more feasible as tool for humans doing code-coverage for its input code, "given enough eyeballs, all bugs are shallow" style. When a developer goes, "huh, Copilot generated insecure code, better report it to the original project it learned it from" - if only Copilot was able to link to the original project, it would all be great and useful.
Feedback - a small bug - somehow it misses / hides the "½" character in my bio (profile description).
For all the warts of private services, there's alternatives available, and the journalists & public opinion don't side with them by default - but rather with the user.
Government and regulated institutions have endless ways of denying service - whether by finding procedural errors or by stalling. Iron property of bureaucracy. For example: KYC and AML laws hurt small businesses at random, including countless PayPal drama stories.
Again the difference is in culture and in reporting: account "suspended" is often reported and perceived as unjust and raise understandable outrage (the egregious cases aside), while legal proceedings are usually reported and perceived as just and proper.
My point is, if we were to change the private platforms to strict regulation or outright government provided services, almost all suspensions would be reported and perceived as the user's own fault, and probably for the better of the society. Which puts the incentives of such supposed service-provider completely out of whack.
Case in point: a government-provided account would be suspended "for the duration of investigation into spreading misleading information about health/elections/finances/etc.", and reporters and people would largely shrug it off as "well that's for the better". Oh and trying to open an account with another government would raise all sorts of red flags, naturally.
Every single example you name happens just the same under regulation / government bureaucracy. Replacing private services with government services or tight regulation of industry doesn't improve error rate nor incentives. Consider the three key differences:
1) When the regulators / the bureaucracy causes you trouble, it is legally and culturally considered your own fault - "you filled the form badly" or any other such. The error rate gets worse, if anything.
2) When the regulators / the bureaucracy causes you trouble, the media has little incentive to hound them persistently. Private companies buy advertising, and the media can push them until they do; conversely the government doesn't buy advertising, but instead can launch endless "lawfare" against journalists, thus making it rather unprofitable to hound them over - what legally and culturally is "your own fault" anyway. That's mis-aligned incentives.
3) There's no competing government to come to, unless you're willing to uproot your whole family and move abroad. And in particularly egregious cases even that is either illegal or made difficult and lengthy - hey, we're having a pandemic right now.
That we know and obsess over the mistakes and problems caused by private service providers is prim and proper - as it is thanks to the interest of journalists and aligned incentives. The same would largely pass unnoticed and at larger scale, under regulation / bureaucracy. So let's not go backwards, shall we?
The third option, un-coordinated / de-centralized services are technically feasible, but quite hard to successfully build for organizational & financial reasons. Presumably we will get there eventually.
US scores poorly in those due to high sticker prices; the quality available for those who can afford is the best in US; both for specialist rare cases and for mundane cases.
The article paints a needlessly bleak picture.
The neutral reading of the practice is, "managers are able to take riskier hiring decisions, because they are given an allowed turnover rate".
Which surprisingly enough is a solution to the ever-growing worry of false negatives in hiring - i.e., overlooking good candidates whos resume or interview did not shine strongly enough, or who perhaps are from a shunned, misunderstood culture, or who otherwise did not fit the generic hiring practice prevalent in the society. This solution allows an organization to make riskier hiring decisions at a well understood rate - hopefully catching the false negatives that did slip through competing organizations' hiring process.
--
In the days when Sussman was a novice Minsky once came to him as he sat
hacking at the PDP-6. "What are you doing?", asked Minsky. "I am
training a randomly wired neural net to play Tic-Tac-Toe." "Why is the
net wired randomly?", asked Minsky. "I do not want it to have any
preconceptions of how to play." Minsky shut his eyes. "Why do you
close your eyes?", Sussman asked his teacher. "So the room will be
empty." At that moment, Sussman was enlightened.>That's apparently become such a technological challenge
I suppose it's more of an organizational challenge. Various factors like the difficulty of quantifying product- or service-manager productivity, or mistaking casual mass audience for core, engaged audience, etc.
Our culture still perceives interactive services and electronic content way too heavily through the lens of factory-like production of more countable widgtes, rather than of culture-and-religion like managing by making opinionated decisions, integrating things into cohesive, fractal form, and building lasting epic stories.
>There's no reason i should subsidize evil capitalist shareholders with the little money i have to live
Since I disagree with your stance :-) I'll propose a middle ground: try to compare various available payment/contribution channels with regards to: how large % they forward to the creator vs the cut taken, and how small is the channel's undue influence on the host socio-political system and the creators. Notably the (traditional) banking system, including some "bridge" companies like PayPal, fail the later part very much; they extend undue influence both broad socio-political, and also over the creators. Conversely, a specialized "patronize your favorite creator" services like SubscribeStar seems to be quite neutral. Of course nothing beats a direct payment :-)
There are two more key differences: Substack has serious marketing effort behind it. Big scale, well thought ought, multi-channel and multi-platform. It's the proverbial tide that rises all boats - the ones that docked at the Substack harbor, that is.
The other substantive advantage is uniform and clear user (reader) interface. The value of which cannot be over-stated for ease and please of focusing on the content, for comfortable familiarity, and for fast, smooth performance. It stands in stark contrast with ad-laden, noisy and ad-hoc interface of most online publications, be it newspapers or blogs. Its structure is very clear "flat trees", giving an almost linear navigating experience - as opposed to the multiply linked graphs of typical publications. The constrained options of Substack allow it to be, and remain, great. It even passes the "works just as well without adblock" test.
At the risk of doing it disservice by the way of risky comparison, Substack is everything that Google Amp ever wanted to be - and without the warts.
[1] https://www.axios.com/media-trust-crisis-2bf0ec1c-00c0-4901-...
Still less than the carbon footprint of the people running our current fiat-based financial infrastructure.
It's the bankers who have jets, mansions and yachts, not the bitcoiners.
Edit to clarify:
Not that riches is somehow "wrong" - it's just that with major value comes major costs, which in practice mean major energy usage.
The neutral reading of the practice is, "managers are able to take riskier hiring decisions, because they are given an allowed turnover rate".
Which surprisingly enough is a solution to the ever-growing worry of false negatives in hiring - i.e., overlooking good candidates whos resume or interview did not shine strongly enough, or who perhaps are from a shunned, misunderstood culture, or who otherwise did not fit the generic hiring practice prevalent in the society. This solution allows an organization to make riskier hiring decisions at a well understood rate - hopefully catching the false negatives that did slip through competing organizations' hiring process.
A competing ad service, naturally much smaller at first, would be hard pressed to replicate[1] even part of Google's privacy-invasive targeting, due to the costs involved. Even more importantly, a competing ad service would be nigh unable to replicate even part of Google's privacy-invasive targeting, due to the legal protections and regulatory oversight of privacy. It is much easier for a large, well established business to "continue as it always did" and get either a nod of understanding, or at worst a slap on the wrist from the regulators - than for a newer, smaller player to start doing anything shady. In the later case, stiff penalties and "making an example of" can be expected for variety of reasons - the new player tends to not be well connected in the regulatory circles, and tends to be less influential on the local economy either, thus there's little downside for slapping the new player hard.
This is yet another case where regulatory framework with somewhat arbitrary enforcement (relevant fines have wide ranges; judicial injunctions are optional and discretionary) entrenches and unfairly protects from competition the existing large market players.
--
[1] sadly, ability to replicate minute features is key to provide "bullet-point engineering" and getting sales to casual buyers who are easily impressed by long feature list - and to advanced buyers who are trying to squeeze every last bit of advantage from the service.
>Whether a student passes or fails is dependent on a combination of two assessments; >a report from their junior high school marked out of 300 points >and a written test covering Japanese, math, English, science and social studies and scored out of 700.
The report is worth up to 300 points which is about 10x the typical differentials mentioned in article. Going by the condensed explanation in article, is created by the junior high teacher(s?) of the pupil. I suspect a scoring inflation, quite possibly gender-selective - specifically to prop up pupils the teachers perceive as worthy moving onward.