furi

To add to this there are* serious issues with naming. I wrote an app that ran on IPFS last year and throughout testing there was a persistent problem with needing to issue the testers with a new hash/link for each new version because IPNS, IPFS' decentralized naming solution, effectively doesn't work. It's supposed to give each node a unique name which can be repeatedly mapped to different hashes but publishing a new name -> hash mapping can take several minutes and then lookup of that name can take several minutes even on the node that published it in the first place. Names also need to be republished every 24 hours or so or they disappear from the network, which in my opinion harms the claims of decentralization.

* = unless something has radically changed in the last few months and they just haven't closed the GitHub issue: https://github.com/ipfs/go-ipfs/issues/3860

I like the idea of biodiversity as much as the next person but the climate paradigm is we have to do something or we're all going to die. "Green" causes are not guaranteed to align with each other and when they come head to head I think we should favor the one that is about continuing to exist over the one that is fundamentally an aesthetic or philosophical concern, the fruits of which we will be unable to enjoy if we die out.

That's a double edged sword though. Humans being bad at telling the difference between cheating and skill is not good for training your AI to spot cheaters but at the same time it lowers the bar for how good your AI needs to be. The metric for whether or not your anti-cheat works is "whether or not humans can tell others are cheating". It doesn't matter so much if it can't identify the subtler ones if the players can't either.

There was a point a year or two ago where everyone was recounting these stories of how AI targeted advertising was terrifyingly accurate, it could tell you were pregnant before you found out for yourself, etc. At the time I was routinely opening YouTube with cookies disabled and getting suggested YouTube generated categories with a super low hit rate, stuff like "Recipes" that would be filled with nothing but general interest food videos (i.e. "you won't believe what we found at this market", "we made the largest burrito ever", never with any instructions), "Metal Music" containing a single acoustic cover a metal song and half a dozen EDM tracks or "Role Playing Games" stocked to the brim with Fortnite videos.

I just checked now and it's better than it was but there's still a "DIY" category containing one DIY video, a Japanese cooking tutorial and a video of somebody putting iPads in a bucket of slime. I'm not sure I have very much faith in their ability to deeply infer things about me from a limited dataset when they can't identify videos with ingredients in the description or sort music into genres given the set of all music (and millions of comments on it) to work with.

>> Artifact had in effect a perfectly reasonable economic model, but players did not see it that way.

I don't think it really did though. Artifact took Magic the Gathering's payment model verbatim (pay for a starter set, pay for booster packs, pay for entrance any activities where you might win more cards) but removed the thing that makes that work for Magic the Gathering: face to face interaction with other human beings at your Local Game Store. If we compare it vs. it's more direct competition (digital CCGs) it's a worse proposition for the average non/low spending player, but even if we do the kindest thing and compare it against what it's directly copying it's still a strictly worse deal.

I think Twitch would occupy the same category of "band-aid" services described in the article. It's a commercial simulation of hanging out with friends but scaled up so one person can provide the experience of a friend for a few hundred or thousands and the price scaled down to match ($0 for most participants, with some people paying for emojis or whatever it is spending money on Twitch gets you).

I'm a little lost here. The article doesn't mention video games anywhere that I can see, the words "video" and "game" don't even appear in it. There's a handful of passing mentions of "screens" but the article fairly clearly dismisses that convenient explanation and classifies it as a symptom rather than a cause.

> translations so often don't work in a one-to-one fashion, it'd be risky to expect them to do so.

That's hardly a point in What3Words' favor. They dragged words into it where previously there were only numbers so now they're responsible for finding a solution to the problem of words' cultural specificity. I don't think "our coordinates contain zero information that crosses linguistic boundaries" is a good solution to that problem.

These were useful for parameters (conceptually anyway, I haven't done any hard labwork with this yet):

Kirt R. Williams, Kishan Gupta, Matthew Wasilik - Etch Rates for Micromachining Processing—Part II

Mark R. Jackson - Effects of Radio Frequency Power and Sulfur Hexafluoride Flowrate on Etch Rate of Silicon Dioxide

S. A. Moshkalyov, C. Reyes-Betanzo, R.C. Teixeira, I. Doi, M.B. Zakia, J.A. Diniz, J. Swart - Etching of Polycrystalline Silicon in SF6 Containing Plasmas

I.J. Kima, H.K. Moona, J.H. Leea, N.E. Leea, J.W. Jungc, S.H. Cho - Silicon nitride etch characteristics in SF6/O2 and C3F6O/O2 plasmas and evaluation of their global warming effects

http://microchem.com/pdf/PMMA_Data_Sheet.pdf

These for further general process design/example values:

http://ww2.che.ufl.edu/unit-ops-lab/experiments/semiconducto...

These are by no means the be all and end all, they were just useful for the parts of the process I've looked at in detail so far and there are often several options at every step for chemicals and approaches, these were suited to what I was aiming for.

You're overhyping the knowledge a bit. It can't have been gathered over that great a span of time because he was only in his late teens when he did this. Also I've been able to work out most of it myself over a period of about a year using the book "Introduction to Semiconductor Manufacturing Technology" by Hong Xiao for general process layout (what steps are necessary, in what order, with what chemicals, etc.) and a scattering of papers, chemical datasheets and ResearchGate discussions for the specific parameters to use.

If anybody would like to get into this kind of thing I highly recommend that book, it's very approachable for somebody coming from computing without a particular semiconductor/chemistry/material science background.

To add to this, disabling WebRTC when you're not using it (media.peerconnection.enabled in Firefox) is also a good idea as when it's enabled websites can use it to derive your IP address behind a proxy/NAT such as a VPN or a home router to better identify which machine you are. I also turn off WebGL (webgl.disabled in Firefox) because it seems to expose a large attack surface for fingerprinting* but I'm not too well versed on the exact risks of it.

* = According to PanOptiClick.

I can't help but feel that without also adding a right to life for insects or similar this swiftly leads to the conclusion that we should pave over the rainforests to minimize long-term suffering.

Humans talking to other humans in the local area and doing things together exists all over the world and has done since the dawn of time, what part of it isn't scalable?

Semiconductor manufacturing involves a lot of precisely controlled processes. You put the wafers into furnaces and pass a gas over them for X time and Y flow rate at Z temperature to impregnate the wafer with various chemicals. You put them in low pressure plasma environments to etch them, again for X time at Y flow rate. There are half a dozen more of these as well, like applying metal and implanting ions.

These values are experimentally tightened to get the highest possible accuracy to the desired effect and improve the number of working chips that leave the factory. If the power cuts out you don't know what conditions the wafer experienced while the system was winding down completely uncontrolled and your processes haven't been designed for the wafer going through the ramp up twice.

The reason why it's lost so much output is because modern semiconductor processes have hundreds of steps and (I believe) a lead time in the months, so the amount of material that's in flight at any one instant has to be huge to get any reasonable throughput.

>All you're saying is that DRM isn't perfect. Nothing is perfect, and it isn't exactly a revelation.

DRM and servers are fundamentally different in that securing a server is an achievable goal. There is nothing fundamental that stops you from exposing an interface without any holes in it, even if it's quite hard. DRM is the polar opposite. Where servers are physically isolated from attackers in a manner that allows for perfect security* DRM is physically colocated on the attacker's machine in a manner that explicitly denies perfect security.

Servers are also broken into fairly sporadically for short periods of time and many of them never at all. Data stolen from servers usually slowly goes stale as people change their passwords and so on. On the other side of the fence I cannot think of a DRM that wasn't compromised relatively quickly and excluding anti-cheats once DRM is compromised it stays that way forever.

You can even see the discrepancy in the availability of files. I can pirate basically any game almost immediately after launch but if I want background production files lifted from server, even for an ancient game, the Half-Life 2 beta is almost the only example. One of them is certainly more niche but not enough to explain the size of the gulf.

Both are examples of imperfect things but there's always going to be a line between "imperfect" and "too imperfect to bother with" and personally I feel DRM falls on the "too imperfect" side of that line.

* = Assuming breaking into the data center is outside of the threat model, which it usually is.

Do we know the problem is in the software? There's a lot of other things that go into securing a large system like this, training and testing staff to resist phishing attacks, apply security patches promptly, maintaining least privilege as requirements, hardware and staff change, etc. It seems to me that unless your software package encapsulates every use case and enforces the security protocols itself the only defense is an on-site security professional who is listened to.

Mass shootings contribute fewer than 1% of US gun homicides as far as I know so any policy based around them should raise serious question marks on that basis alone. Otherwise not to be pedantic but that doesn't lead to mutual exclusivity. If there was another amendment saying "the state must take every measure to prevent mass shootings" then you would end up in a "choose two out of three" scenario but as far as I'm aware there isn't one. Satisfying both the 2nd and 4th amendments at the same time is easy to do, you simply accept a certain level of carnage at the intersection of the two. It might be unpalatable for many but it's hardly impossible.

I've written a couple things in it, mostly little web servers that serve a questionnaire or something, more recently a Lisp and a rudimentary voice chat. I really really like it. Compared to the languages I'm coming from (C/C++ mostly, some Python & C#) the compile time checking is unparalleled. I frequently find that once I've fixed all the compile time errors the program works first time, something I've never experienced in any other imperative language. The standard library is nice too. It's still missing a couple things, for example hash functions better than SHA1 must be accessed by providing their string names to the OpenSSL module at the moment, but overall it's well designed and miles ahead of C/C++ in functionality. My main gripe is the state of Windows support, it works in the Linux subsystem but I spent three hours trying to work out how, for my voice chat program it involved:

1. Turning on something in PowerShell.

2. Installing something from the Windows store.

3. Adding a new repository to Apt.

4. Installing Crystal and all the dependencies of your application.

5. Compiling your application.

6. Installing PulseAudio for Windows.

7. Connecting PulseAudio for Windows to Linux somehow. (I gave up upon opening the guide for how to do this bit).

Hardly a process an end user can be expected to do.

> the people most vocal against Article 13 are those who want to leech other people's creations for free.

Why would such people be at all concerned about Article 13? Torrent sites and swarms have been in flagrant disregard of the law since forever, one more law isn't going to change much. Likewise the other popular pirate mechanism, the old "upload an archive with the content to Mega or similar" technique, already frequently uses encrypted archives to make sure only people coming through the right channel (often an ad laden website + several even more ad laden URL shorteners) get access to the content, making a filter requirement ineffectual there as well.

Seems like the problem here is that the various safety nets don't extend to such people in the first place. It would also probably go a long ways to curbing the "employ ten people for 4 hours a week each" strategy we see used to avoid similar obligations.

I think that's true in a historical sense, it was women who joined the workforce as the second household income and started the vicious cycle I proposed (if such a thing exists, what do I know about economics).

But you can still construe the problem both ways, is it women increasingly working outside the home that society takes as an unhealthy unquestionable good or men almost exclusively working outside the home which has been taken as good (or at least normal) since time immemorial.

And it's the same thing with the solution, if we come to the conclusion that somebody should take a smaller part in the workforce to help correct the birth rate and other problems two income households produce, the historical role of women in "creating" the problem doesn't imply they should be the ones to go.

If you take it as a given that all the men work and will continue to work then yes, women entering workforce means the birth rate declines. But that isn't a given, men can leave the workforce just as women can enter it. My point is that I think it has less to do with women working and more to do with both partners working simultaneously.

I don't think it's particularly a women-centric issue and you're never going to have a real discussion about it while it's posed as one. The majority of the burden of children obviously falls on women but time wise it's a small component, mostly concentrated in the first year or two, versus the 20 years of needing to maintain a household, take the kids to school, etc. that follow. The woman is always going to need take a half a year or more off (ideally) for each child, that's fixed, but who takes the following 20 off is up to the couple.

The question is really whether or not two income households are unquestionably a good thing, to which the answer seems like a strong "No". Even without the birth rate issue I struggle to conceive of mechanism by which it isn't a vicious circle of more people entering the workforce, driving incomes down (and driving the cost of scarce resources like certain housing up), forcing more households to become two income to keep up with no obvious benefit for anyone but the people doing the employing.

This is just a hastily assembled excuse for not living up to the expectation created by their name. OpenAI have also failed to release their Dota 2 model which has absolutely no security implications, despite the fact that it cannot be properly tested without public release. OpenAI isn't.

Maybe I missed something, but my experience was that Typescript was capable of making useful conclusions (i.e. MyLibrary.MyType.width exists, but MyLibrary.MyType.dog does not) about other libraries, delivered to it via NPM, that were also written in Typescript. If your libraries were written in JavaScript to be run in the browser you were stuck doing `declare var mylibrary: any;` and writing "unsafe" code. This is also true of any global variables/functions/etc. that exist in <script> tags in your HTML.

To be clear I don't expect it to be able to do this, reading JavaScript that was written without types and deciding what the types are sounds more or less impossible, it's just a demonstration of how you tend to end up with an imperfect solution as a result of the decisions made early on in the web's development.

The web environment was never intentionally designed to be what it has become (a system for drawing fully featured user interfaces and even games) and so isn't really equipped for it, you could argue it's missing a lot of widget primitives, the view-updates-as-string-manipulation is archaic to put it kindly and the whole thing is untyped and shackled to a weak language without so much as a proper this keyword.

This environment is simultaneously so restrictive (in that you can't really introduce new languages except by having them transpile to JavaScript) and so bad that people end up trying to tie it into something more cohesive (React's virtual DOM, Typescript) and then only partly succeeding (React's dependence on setState and things often being much easier to do with document.getElementById than in the React way, Typescript's inability to handle libraries intended for the web), leading other people to try do it better.

I think something's a little off here. I pasted in a paragraph from the PCI Express 3.0 specification document and it told me it was "average" for software developers, ineffective for retirees, college students and men but fairly effective for accountants and women. It then recommended a bunch of rather strange changes such as use "embraces" or "empowers" instead of "supports". I also couldn't get anything less niche than fairly heavy technical writing to produce a result outside of the "average" range.

I don't mean they've heard of it, indeed what most of these platforms support isn't Markdown but a similar simplified ruleset. I mean that more people than one might think are using the general concepts of Markdown on a daily basis.

Normally I would agree but Markdown-esque text annotation has been a feature of many commonly used platforms for years now. A quick search plus my own knowledge tells me the following platforms support it: Facebook Messenger, Skype, Discord, Slack, WhatsApp, Telegram and Reddit posts. It looks like LINE, Viber and Kik Messenger don't support it, but those are the only major ones that don't seem to.

Not for torrenting, for accessing/delivering ZeroNet sites over. (i.e. connecting to the torrent tracker to get the magnet in the first place).