jonluca

IAmA mod here - we have a bot that updates it every time a guest schedules a new AMA, so it should always be up to date. We also manually add events in as they come from other sources.

Bill likes to do his AmA's as "surprises", though, so we don't usually schedule those!

This is great!

Part 2 is going to be more analysis focused, checking the delta in how many experiments these companies are running, and how quickly they are moving.

I'll also mention how to properly disguise it - I'll use Discord as an example. Thanks!

It's a triple encoded payload that loads in a large HTML blob onto the page. The payload is 99% similar to Amex's actual page, it just submits the data to the attackers domain, and has a few extra fields like mothers maiden name, elementary school, etc.

The purpose of the obfuscation is 1) to prevent automated scanners and 2) prevent debugging of the script.

Since we did static analysis it did not impact the result.

I actually wanted to do that but thought the cost/benefit wasn't high enough - I'd pretty much need to manually place all the airports information on the map if I wanted to use the library I was using (Cytoscape).

Visualizing the DFS at the last step was good enough :) Maybe in the future!

Yeah you might be right. How do I flag this for deletion? Any mods seeing this, could you remove it?

Jane Street pays interns a little more than $60/hr. Pretty competitive. Definitely in line with unicorns and big tech ($50/hr @ FB and AirBnb, $45 @ Apple, etc)

I've tried to transition fully to python3, but yet it should mostly work in python2.7 as well!

I did not read the book on nltk. I've been into information/coding theory recently (just finished Information by James Gleick) and thought I'd try my hand at something closer to NLP/ML. Very little background in the topic - I've taken a few courses in college and chatted with a few friends that know ML fairly well, but besides that I relied on blog posts and papers found on arxiv!

That thought crossed my mind. I was thinking of trying to get a training set and teaching these models based on previous works by Jay, Madison, and Hamilton. However a lot of what I found for each of them was behind a paywall, or too hard to grep through to actually get the data. For instance all I could find in my (admittedly superficial) digging on John Jay was a book by UVA called "The Selected Papers of John Jay". It costs $90 and contains correspondence both too and from John Jay. It seemed like too much overhead for a weekend project so I just settled on unsupervised.

This is my first project in unsupervised NLP, so let me know if there's anything obviously wrong with the article or methodology.

If there's anything I've missed or if anyone has any tips on how to get it faster I'd love to hear them! I'm also considering switching to Prezto, but for now ~.4 seconds is good enough.

Anyone have experiences improving their shell?

There are 256 bits in ODM_RESERVED, and 8 ODM_RESERVED, so I think there are 32 bits. That's 32 major FW revisions (as it seems they're only blowing a fuse for every X.0.0 version)

Webpack does dead code elimination through UglifyJS, see here https://webpack.js.org/guides/tree-shaking/

The reason why there was such a small effort on optimization in the first place is that this isn't meant to be a revenue generating site, or one that focuses on "ultimate performance" - it's more of just a personal, online resume. Any time I needed something new, I'd just throw it in, because "it worked and looked good on my machine". If I found a cool js snippet that needed an entire library, I'd just throw it in. An argument could be made that that's part of what's wrong with modern webdev, and I wouldn't necessarily disagree, but this wasn't trying to be the fastest or most efficient site out there. I think there's a pretty distinct difference between making a scalable, well optimized webapp and just making a personal site as an online resume.

This was me documenting my updates - I just revised the article with a few more updates, and the total site size is 250kb, 99kb without images. This still isn't great, but total page speed is pretty quick (js execution takes ~70ms on my 2016 MBP).

Thanks for the feedback!

Thanks for the reply. I agree that 1.66mb is insane for what my site does, and I'm working on minifying it as much as possible. I got it down to ~90kb, which is still not amazing, but is significantly better than 1.6mb. Fonts were taking up more than 90% of that, which I just stripped.

I do disagree with the idea of including code from others being a bad practice/smell. The detriment and added load time of using a library like jQuery is well worth the functionality it provides, IMO.

Thanks for the comment and for pointing out a problem with the article, though - I've revised it!

Original author here - thanks for your feedback, just updated the article. I kind of settled at ~1mb because I thought it was good enough. I got it down to ~90kb (which still isn't great, but is fine for these purposes) by just stripping out font awesome.

Thanks!

Hopefully it'll fix some of the issues with getting redirected on desktops when people share links here/on reddit