awirth

I've noticed this pattern too.

For example, try to draw the character 口 on the 'draw' functionality on jisho.com -- the correct stroke order is (cardinal directions for 3 strokes): 1. NW -> SW 2. NW -> NE -> SE 3. SW -> SE. If you try to do it any other way it will fail.

If this is a problem for you (e.g. learning Japanese and don't know stroke orders and/or don't care to learn them exactly) -- the Google Translate handwriting input for Japanese does not have this behavior (or at least as much) and will effectively recognize your incredibly shittily reproduced kanji.

Well, respect for coming out and saying it directly.

This is called ScrollToTextFragment. See https://github.com/WICG/ScrollToTextFragment

I think they just started doing this a few hours ago. I only noticed it in the "Featured Snippet" box.

Edit:

> I was surprised because Quora can’t know what the search result looked like and so doesn’t know which part of the text to highlight.

Did Google at some point stop sending referer headers that include the query? The last time I checked (~10 years ago, admittedly), they did, which was often collected for analytics. Regardless, it's the browser doing the highlighting in this case, not the site itself.

From this scripts, it looks like this doesn't convert to black and white, so this might work with Hanko (Japanese stamp seal) too?

Akamai has a BGP based DDoS mitigation service via their prolexic acquisition.

As I was trying to understand this space coming from a 'docker user' background I was incredibly confused by the two different definitions of "privileged containers" ("what do you mean if I don't add --privileged it's still privileged?") -- so I figured others might appreciate the pointer as well.

Now that I understand what's going on better, I definitely agree that docker does not abstract the kernel APIs in a way that makes it easy to understand what's going on underneath the hood. Is that a good thing? I honestly don't know.

I'd encourage anyone interested in learning more to check out the codebase for JessFraz's contained.af

Note to anyone confused: The docker concept of '--privileged' is separate from what the LXD folks are refering to as 'privileged containers'. The LXD folks are talking about mapping UID 0 into the container, whereas (IIRC) the docker flag disables dropping capabilities and the seccomp syscall filters (and maybe some other things? I can't remember off the top of my head).

The equivalent docker functionality is userns-remap or sometimes just "user namespaces".

Yeah, agree in this specific case about local research (baring DMCA issues). Most H1 scopes seem to be remote targets as opposed to downloadables though.

"The worst H1 or a client can do is kick you off the platform."

As a hacker on hackerone, this is not my understanding of the relationship. Generally speaking the programs give you "authorized access" under the CFAA conditional on following the disclosure guidelines. I don't know about for other countries, but for the US I'm pretty sure this means that breaking the guidelines means you've retroactively committed a felony.

Now seems a little questionable about if any federal prosecutor would actually take the case, but it definitely doesn't seem like a strictly civil issue to me.

Strongly agree on all other points though.

Etymonline suggests it might be a combination of both definitions: https://www.etymonline.com/word/troll

This movie specifically has trolls under bridges, although no tolls of course.

Ah, sure. Down detector's map suggests it's global: https://downdetector.com/status/facebook/map/

Instagram also doesn't seem to be serving hot and cold content from different hostnames/IPs for me, so it couldn't be that either.

Also check out the official status update at https://developers.facebook.com/status/issues/34337305659485... and https://developers.facebook.com/support/bugs/610822019411772...

No, I checked if it was my own cache, and it's all HTTPS so it couldn't have been my ISP's cache either.

I compared a few celebrity's instagrams with accounts of friends with many fewer followers. Celeb's images loaded but less and less as you scroll back through time, friends images all 503'd. Still seems to be the case.

Some popular images seem to load for me, but longer tail stuff doesn't. Maybe they accidentally flushed their edge cache and then their origin couldn't handle the load.

Shameless self plug - I cleaned up my take on a very similar idea a bit today after seeing this post: https://github.com/allanlw/hgscn2

Basically, it uses a data URI and Webtorrent for a similar concept.

Note this uses golang.org/x/tools/go/packages and golang.org/x/tools/go/ssa to get to the go SSA AST and then generates some LLVM IR from that. It's not implementing a go parser or the transformation to SSA.

A few years back I was working on a WAM to LLVM compiler as a way to learn LLVM. Never fully finished my implementation, but this book really was an amazing reference. I particularly enjoyed the way the "tutorial style" that built up starting from a very simple language and explaining the design choices being made while extending the language to eventually become Prolog.

I don't know where the code is now, maybe I should make a point of digging it up and fixing it up to work with a recent versions of LLVM.

That's different than preserving:

  <pre>

  foo</pre>

"Case in point, your comment and the parent don't even have the same syntax because one of you couldn't, from the top of the head, come up with the proper one."

I'm not sure what you mean by this. They're both valid. I don't find the one enclosed in {} nearly as confusing, because once you're inside curly braces you can have arbitrary javascript expressions so all bets are off.

The thing that's strange is being able to have a JSX Element as an attribute value without enclosing curly braces. Perhaps the use case of passing arguments to custom elements is more common than I thought thought.

Specifically "<foo a=<bar/>/>" is weird.

When I was first introduced to JSX I found it very confusing because it was sold to me as "HTML in Javascript" but it's... really not. In particular these things threw me off:

  - You need to close tags you wouldn't need to close in HTML (breaks copy+pasted HTML)
  - Argument quoting isn't optional (breaks copy+pasted HTML)
  - HTML style Comments aren't allowed (breaks copy+pasted HTML)
  - Whitespace gets stripped for all elements (breaks PREs)
  - There's a JSX-specific syntax for "fragments" that seem to only exist for parser reasons?
  - Variable references work inside of text nodes but don't inside of attributes.

After I googled a bunch and read the spec it helped a bit, but it left a bit of a bad taste in my mouth. I'm still left scratching my head over why you should be able to use an element as the value for an attribute though, like this:

  <foo a=<bar> test </bar>/>

"Text pages of DLLs are shared between processes"

I thought this wasn't possible with ASLR'd relocations all over the place in the text?

Doesn't Chromium already lower their revenue from other lacks of integration that are included in Chrome?

Yes, Akamai offers QUIC support on some CDN products. (Note: I work at Akamai)

Blender has a full game engine included in it, although it looks like it's getting cut with the next release: https://www.blender.org/features/game-creation/

This reminds me of the documentation for how the game Liquid War did their pathfinding. The interesting relevance is that it does not use a uniform grid when calculating the gradients, but instead coalesces squares into power-of-two squares to "cheat"

https://ufoot.org/liquidwar/v5/techinfo/algorithm

<script async> does still block the document onload event from firing until it's been executed.

This is an interesting alphabet, even Base58 (Bitcoin addresses) have "o" and "1".

There's some information on how they generated it here in the spec[1].

The characters that are used in Open Location Codes were chosen by computing all possible 20 character combinations from 0-9A-Z and scoring them on how well they spell 10,000 words from over 30 languages. This was to avoid, as far as possible, Open Location Codes being generated that included recognisable words. The selected 20 character set is made up of "23456789CFGHJMPQRVWX".

[1]: https://github.com/google/open-location-code/blob/master/doc...

Is this something that is considered easy to build locally? When I last looked at the open source options a few years ago it was basically all researchware grade stuff. Amazon is offering this is a production service with SLAs.

This reminds me of the Miracle Jones short story "Kickstart My Bad Life" http://www.miraclejones.com/stories/kickstart-my-bad-life.ht...

Ah, I misremembered I guess. I could have sworn I had to type it in to authenticate the chromecast device to my phone for initial bootstrapping.