pheug

Nikkei is a price index. Total return with dividends reinvested probably surpassed the 1989 peak a while ago, but it's hard to find that data on it.

The bulk of that profit came from volatility spike, rather than simple directional price difference between strike and underlying ("delta").

VIX (https://finance.yahoo.com/quote/%5EVIX) was at 14-15 just a month ago and now at 65 - one of the highest ever! What a time to be alive after years of suppressed volatility!

Both calls and puts increase in value when volatility increases.

> Who is the counterparty to these trades?

Typically large trading firms, market makers.

> When the market sentiment is negative, the number of these counterparties are narrow and dwindling

That's why his $2600 initial investment made in a much calmer market is now worth $310k :)

For the right price you'll find somebody willing to sell you the insurance you seek. But in a volatile market like today, that price may be way too much for most people.

When market crashes, volatility shoots up through the roof and significantly increases value even for such deep out-of-the-money options, by far more than the difference between strike and underlying price. Essentially he made most of that money on volatility spike.

AQR disagrees with you: https://www.aqr.com/Insights/Research/White-Papers/Pathetic-...

It's the latest HR innovation from Googleplex: doing your job is no longer enough, every employee is required to show off some "citizenship contributions" at the threat of getting their performance review docked one rating lower.

Interviewing is apparently the most popular choice to fulfill this requirement, so get ready for a glut of disinterested people interviewing you at google.

> My (limited) understanding of LambdaRank is that their model empirically minimizes NDCG, but does not have a strong theoretical backing for why it should work.

That's akin to saying that minimizing cross entropy empirically maximizes accuracy but there's no strong theoretical backing for that either

LambdaRank is one way of getting a smooth differentiable approximation to NDCG by slapping a sigmoid somewhere. The paper we're discussing now offers another way. Hard to say which way would turn out to be empirically better on problems of practical significance without actually experimenting.

They could be using some third-party code/services with closed source or under proprietary/non-AGPL-compatible license, and thus they can't open-source those parts under AGPL, as AGPL demands, even if they wanted to.

I needed to constantly move and tilt my head so much it hurt.

My specimen also had some problems with blurriness, I couldn't quite put a finger on what it was exactly, just it was definitely harder reading things on it compared to my previous setup, but this could be just an isolated manufacturing defect.

Got rid of it and went back to two 27" monitors: one primary straight in front of me where all the work happens, I don't have to move head while working on it. One secondary for less important side things. Both are 4K and so in total I have way more resolution this way than with a single ultrawide. Very happy with this setup.

Because Google pays them to do it and essentially to a large extent sponsors Firefox's development. Would you rather have Firefox die due to lack of funding?

DDG doesn't track what you click on, so unfortunately you are not contributing much to improving its and Bing's search quality by using it instead of Bing.

It's just you pretending search engines are simpler then they are. Noone's just ranking most clicked sites first - that's too naive and encourages clickbait. Clicks are an ingredient in the overall ranking system, which is generally also trained on relevancy labels by human annotators/raters/assessors. These human labels are far more reliable and can penalize all the bad clickbaity stuff which clicks can't, but unfortunately also very expensive to obtain and thus low volume.

We may not know exactly why you clicked back, or why Joe wasted 15 minutes on a site, but from all clicks in aggregate and from the human-labelled data, we can tell how that correlates with the clicked site's relevancy and quality and utilize your clicks accordingly to improve the ranking.

Finding such correlations is still by far and large what all search engines do, rather than trying to truly understand the query. It's only now starting to change with recent advances in deep learning and NLP.

Clicks are noisy and have their problems (especially clickbait), but as far as ranking goes they still blow out of water everything else you can use. Analemma_ is right on the money. We need more Lemon Pledge^W^W clicks!

Speaking as someone who worked for another big web search engine in the past (not G, not B)

And counting good/bad clicks for a given (query, url) pair is just the tip of the iceberg. There's a lot of other interesting stuff you can do with them _if_ you have the data. Deep learning/NLP with clicks as training signal is probably the most exciting area to me, to name one. Unfortunately almost all of the data currently gets nabbed by google, other search engines are just getting by with the scraps. And it's very hard to bootstrap a competitor from scratch - for example Cliqz had to resort to some shady deals with Mozilla just to get any data to start with.

They could, but if the court says the guy did nothing wrong after all, it would look rather bad on Uber to retaliate by banning him at that point against court's opinion and potentially opens them up for another lawsuit. Personally I don't see any business sense for Uber in doing this, unlike banning for chargebacks.

Social credit scores as seen on black mirror

Uber would just ban his account for life if he pulls this, most likely automatically without any effort to investigate and definitely not going to any courts.

A google engineer in search PA?

Actually $157k salary sounds plausible, it's in the range of base salary for L4 band which is the most common level at google AFAIK. Of course, there should be like an extra $100k in stocks and bonuses on top of it, and noone from FAANG cares about base salary figure alone.

Yeah, these micro pcs are really nice, thinking of getting one like that myself. But rpi4 costs an order of magnitude less ($35), it's in a different price class I'd say.

With Wireguard you set DNS server IP directly in the config file, it is not negotiable over the connection. So you can edit the config to set it to your pihole's ip. Or remove the DNS line altogether and then it won't touch your DNS settings at all.

With OpenVPN, yes, you'd get around 200 mbit tops - the thing is OpenVPN is single-threaded and a 1.5 GHz ARM core just isn't enough to cut it. I have an 1.8Ghz ARMv8 in a router and it peaks at ~240mbit, pegging a single core at 100%. Similarly clocked x86 core would fare much better and that's what I'm thinking of upgrading to.

With Wireguard you should have a much better performance however as it's multithreaded and rpi4 offers you 4 cores.

The bad ones don't outperform and need marketing to find suckers. The good ones won't take your money even if you beg, don't need to advertise themselves and are rarely if ever in the news.

Correct. It's even more complicated than just trusting a "no logs" policy. You have to also trust that noone with hardware access (rogue VPN company employees, datacenter people, law enforcement and intelligence services) intercepts your traffic or messes with the servers, and that the servers themselves don't get hacked due to some vulnerabilities. Recall how NordVPN recently got hacked because datacenter operator left vulnerable remote management software exposed.

Tor might have its vulnerabilities as well, like all software. Not to forget that pretty much anyone can run an entry guard today and at least associate your IP with usage of Tor..

But don't you have to provide to them a cc card "just in case"? That's pretty strong link to your identity.

Also the bandwidth rates all big cloud providers are charging are just extortionate

But then you get no privacy benefits of VPNs: pretty much all cloud providers will be to tell to law enforcement, given an IP address, who and when rented it from them.

And since you're not, I suppose, reselling your traffic to other users, the liability for problematic traffic will land on you.

There's vidir in debian's moreutils package which does pretty much exactly that.

Check all also all the other tools from that package: https://manpages.debian.org/buster/moreutils/index.html

The package's tagline is "collection of the Unix tools that nobody thought to write long ago, when Unix was young", so pretty relevant to the whole topic.

But why go through so much trouble instead of just using Tor, which AFAIK gives you 3 hops right out of the box?

https://duckduckgo.com/?q=%22my+user+agent%22 - check what the snippets say.

>They're looking at how users use Google Search because the data's there

This. Having worked in past life for one of their competitors, can confirm - what users click on (in SERP) is one of the most powerful signals for ranking. And who got (almost) all the clicks in the world? Google!

That's why it's so damn hard to beat them. It's the unreasonable effectiveness of data: more data (which they have almost all of) usually beats a smarter algorithm, and with 20 years R&D, theirs is surely not dumb.

Do the clicks belong to users or Google, that's an interesting question, though.

Paid "Plus" plan and connecting to their dedicated P2P servers in Switzerland and Sweden. Definitely getting throttled by them to ~20Mbps while torrenting. Tried a competitor VPN - no throttling, torrenting at full speed. Glad I only paid for a month - they won't be seeing any more of my money once the month's up.

Google pretty much knows (or can accurately estimate) exactly when a new document appears on the web and how many people are visiting it, they don't even need to rely on second hand social signals for this. They control the web's dominant crawler (Googlebot), browser (Chrome, which sends everything you type in the address bar to them by default), ads (Adsense) and tracking (Google Analytics) platforms.