kaivi

I'm just curious - why do I primarily see Chinese researchers publishing deep learning stuff on arXiv? Is it subsidized over there? Just look at the publications linked in this thread so far, 2/3 are from Beijing.

Well shit, I was just writing up a huge post exposing the guy on Reddit, and I've noticed that the scenes and timings in the video did not match with what I have downloaded.

Turns out, Mark himself has already admitted that 2 scenes were staged, supposedly all because of a dishonest "friend" whom he paid to place the box on his porch.

I've had the same thought when I first saw it. Some suspicious things: thieves talking to themselves, that lady throwing the box into her own garbage bin, zero attempt at disassembly or closer inspection of the box, no police involved.

Also GPS is just not that good for locating anything of that size in given circumstances, and it would not have worked in the parking garage.

Too many things could have gone wrong here, but they did not. The design is subpar in my opinion, for somebody who worked on a Mars rover. Custom printed board plus a bunch of smartphones, seriously?

Same here. Looks to me like it can even indicate 2 different states.

Same for me, it's "Catching up" and appears to connect to some ipv4 nodes, but nothing so far. Edit: it werks now!

Deep work.

Not a huge time investment, but an indispensable skill or revelation I wish I had before starting college. Think I've found the book here on HN: https://www.goodreads.com/book/show/25744928-deep-work

Hey, I've been using Rancher for a year, and I love it.

Just wanted to bring your attention to a YouTube channel from Rancher Labs, where "Learning Rancher" videos are posted regularly. But the content is all the same: boring 10min intro to every video, after which developers go through the same stuff, painfully slowly. Is it meaningful to produce this kind of content over and over again? IMO there are already way better Rancher tutorials by indian dudes on YouTube.

I've started with k8s thanks to Rancher, and I presume that the vast majority of Rancher users are doing the same. Rancher without k8s does not make sense, and it seems to me that k8s stuff is actively avoided in docs and tutorials. Maybe it would be better to make a series of tutorials on Rancher+k8s, and go through concrete topics like networking and CI?

I love Stripe as well, but one thing keeps me wondering - why do their bank payouts take so long?

My company is based in Norway, and I see funds arriving a week after they are marked "on the way to your bank". Money get sent like that every day, but arrive only after a week from a local bank. I know that local bank transfers are reconciled multiple times a day, and I can't see why there is a delay.

The money sending frequency setting seems just misleading to me.

Woah, did not know that. Thank you sir! Time to migrate my personal mail I guess.

I second this. Their service was exceptional for 3+ years I've been with FastMail. Got many small businesses I've worked with to migrate.

Only thing which annoys me is that their push-enabled iOS app does not support multiple accounts. It has been like that for years, I've heard that a new app was in the making, but nothing came out yet.

Woah, that would be $80k to process my dataset. I'll try renting hardware instead.

I've tinkered with face recognition in the past days, and am now waiting for a response to my GCE quota request for a V100 GPU. $8/h rate for a preemptible instance seems really cheap, yet I am not sure if I will ever be able to process my dataset.

Does anyone have a clue on how much it will cost to detect faces and extract 128d encodings from ~100M of 200x200 photos?

I've recently paid for a code signing cert from Comodo, and I'm still stuck in the process.

In order to verify your company phone number, it has be shown in any of the links like : (www.dnb.com) or (www.hoovers.com) including local/national registration agencies and reputable third party databases.. So please update the Company name,address and Phone number in any one of the above web site.

My company is registered in Norway, and having the company's email and domain listed in the national company registry does not help. I'm currently in SE Asia, and I have to go back to to this:

[...] you can send an attestation letter signed by your attorney, Certified Public Accountant or Latin Notary (where legally recognized) verifying the telephone number. You can download sample text for the letter [...]

We need Let's Encrypt for code signing. But how can we automate identity validation? Verify the e-mail address or phone number with a national registry, where possible?

There is a chat group in Telegram, called "BitGrail Trollbox". It doesn't have a direct link, but one can join when searching the group name through client application. The Bomber dude is there, and it seemed like there is a discussion about what car to buy with all that money. I was removed from it the instant I joined, maybe someone can join and quickly dump the chat log?

Thanks, I've enjoyed it a lot, strange how I haven't heard about Adam Curtis earlier. When I watched it, I've had a similar feeling as that during the first semester of philosophy class - things I've noticed and ideas I've had before were not novel, and here is a whole movie made to organize these things in my head. Thought the narrative of isolated bubbles in social networks would resonate with HN crowd.

Love stuff like that! I've quickly wrapped a prettifier for S3 xml listings in a userscript, so you can use it with Tampermonkey Beta. Tested on Chrome under OS X.

https://gist.github.com/kaivi/8114cbc2080da78d67c94238af6421...

Edit: Okay, the userscript won't run on larger XML files, gotta figure it out later.

I'd like to set it up as a subscription service like Domaintools. I'm also crawling and storing user's comments, friends, posts and links to media published in other groups.

It will make a great tool for profiling a person, but I'm not sure on how (il)legal it will be.

I'm hoarding profile data diffs from a well known social network. Been crawling every single user for the past 2 years and saving the changes. Had to stop doing it last month, after storage costs became too much.

Looks like it does all kinds of tricks at once -- histograms, SIFT/SURF features and edge patterns. I was looking for a simpler solution, like an effective MVP-tree implementation for perceptual hashes. Wouldn't that be faster and overall better?

This is all good, but how does one store, index and efficiently search for near matches? I have a side project with terabytes of photo URLs, and hashing&indexing them has been an itch I could not scratch.

This makes me wonder: is it possible to submerge the antenna in a special fluid or powder and send some modulated current through it, so that said fluid hardens in form of a radiation pattern? We could then MR scan it and see the real pattern in 3D. It would be super interesting to compare the result to a simulation. Any ideas on what that fluid should be?

180662, and I am still using it!

  There are 11.5m documents and 2.6 terabytes of information drawn from Mossack Fonseca’s internal database.

Is there a torrent for this or an online catalog?

I believe this works in the same way, just that there is no technical explanation: http://www.solargreencolor.com/

What does the pos parameter in image URL specify? Looks like some kind of a filter or historical juxtaposition: http://alice-logbook.cern.ch/aliceOnline/get_image.php?conta...

Yeah, I've stumbled upon this Monopoly image about 2 years ago, when I was developing a small JS-based website. It dynamically embedded random pics from image hosting sites like Imgur, and this Monopoly guy was everywhere.

I guessed that it must have been some sort of massive Facebook spam at some point in time, or something like this. After all, the guy waves dollars in your face in exchange for credentials.

I wish there were more articles about determining the frequency at which one page should be crawled. Some pages never change, some change multiple times per minute, and we do not want to crawl them all equally often.

Priceonomics blog: http://blog.priceonomics.com/

.edu domains are still popular targets in Western Europe, and access to webservers in .gov and .edu TLDs is the most priced there.

The cause for this is the belief that Google gives more weight to content and backlinks from these TLDs.

It feels and looks wrong: their website is a mess and the devices look no better.

Their domain has whois privacy enabled, and it was just registered this year. News of this product are spreading quickly, but I can not find anything else about this company on the web. Also, their Wikipedia page was removed today.

Has anybody heard of Linshof before?