jc4p

Bah did the link in my comment not work? I tried updating my original comment to link to http://www.politico.com/story/2015/03/state-department-email... but now there's two versions of my comment, one pre-edit in this chain and one post-edit in another chain I can't find. This article is heavily biased, I think, but it mentions the post Powell rulings. In the Judicial Watch testimony of Cheryl Mills she mentioned the post Powell decisions.

I tried updating my post but because this post was flagged away it seems to have duplicated somewhere else: https://news.ycombinator.com/item?id=12037507 -- This article explains it a bit, it's also light on the details. If you read the Judicial Watch testimony of her staffers, a lot of them touch on this.

I hate this argument because Colin's actions were so overt that extra regulations were passed after he left to flat out make what he and and others did against standard operating procedure. Stop using this as your argument.

> you have to have a cable subscription which offers the local broadcast

I watch all NFL local broadcast games over a cheap digital OTA. Can the same thing be done here for people without cable?

I think one of the points Adam repeatedly mentions in the article is that a contract in itself isn't really transparent. I get exactly what you are saying, it makes logical sense from the perspective of someone who knows what's going on, but in all of these discussions my thoughts are on the young naive person signing up for a deal they don't fully understand.

I think stock options should be valued at zero. In an a-will employment state you, and with a typical contract, you can get fired at any random day and if you don't have enough money to exercise your vested options within 90 days, they're gone. I don't consider that reliable.

In regards to the parent comment of yours, I think if it was stated as "proactively answer" vs "answer" it'd be less arguable.

I don't think you can do anything here once a culture's already been set. I also think that repeatedly pestering execs to ask about their efforts on this (as I've done at my current company every few weeks for the last 6 months) makes them think you're about to jump ship and want "free money", so watch out :)

I know it _just_ launched but oh my god how do you launch such a slow site. I'm trying to load the "System.Drawing" API docs and Chrome inspector just shows a GET pending alongside a _lot_ of analytics and marketing calls being made. I waited to see how long it'd take, 30s to load the page: http://i.imgur.com/Z6y1ulSr.png

Anyway, ignoring the launch issues: My #1 annoyance with MSFT documentation is sub-classes being on a different page. If you want to call a DirectX method which takes 4 inputs, you need to open up a new tab for each of those inputs to see the possible values / etc. I hate it, very very much. I hope their new documentation doesn't have this issue, I'll check back again in a few weeks to see if I can actually load pages on this site.

I agree with you on the IT mindset, but there's a huge difference between getting users in a small company to run updates or whatever the IT need there is vs overriding settings on something someone actually owns.

My company's IT has accidentally done things like this before and even when it's on a corporate machine a popup coming up and saying "gonna restart in 5 to do updates" SUCKS while you're working.

Are we just talking about API documentation? Stripe's the obvious go-to for that, but for big business stuff I was pleasantly surprised with HubSpot's API and documentation. Their own UI uses the API, which means debugging a lot easier I think: http://developers.hubspot.com/docs/overview

This is really good news. Plaid has an amazing API, it makes it very easy to get your own financial data. I'm trying to analyze my own spending habits / make a budget-allocator using my own patterns, so it's been insanely helpful. My big fear with all small SaaS's if they just suddenly shutter, so a new round of fundraising is always good news :)

An option isn't equity, but I would wager that a lot of people who subscribe to the "if I have equity I'm family" model are naive enough to not know that.

I definitely agree with your attitude on them, I really wish startup culture didn't frown upon electing to receive higher salary and no equity.

What about if your equity is in stock-options that haven't been exercised, and after you leave you only have 90 days to exercise them? :)

This is so 100% on point. I think the Netflix NYT article from yesterday had a more realistic tone regarding the "family" aspect: http://www.nytimes.com/2016/06/19/magazine/can-netflix-survi...

> The key concept is summed up in the 23rd slide. “We’re a team, not a family,” it reads. “Netflix leaders hire, develop and cut smartly, so we have stars in every position.”

Your last sentence here is really important. If that's the mindset a person is in, when a business-centric action happens they won't feel as betrayed. Trust me, feeling betrayed while also fully logically knowing you shouldn't feel betrayed because it's a corporation in a capitalist ecoystem is such a bad feeling.

The community around Tensorflow is great (lots of people that try to recreate results from new papers using TF), but if you're worried about putting all your eggs in one basket (or want to be a higher level up) you should checkout Keras if you haven't yet. It lets you write generic nets that can run on Theano or TF.

From my experience the scrobbler only work with music I own in Apple Music, not any music I'm listening to from Beats Radio or any of the recommended playlists.

I tried doing this with my own Last.fm data, if this slow badly coded Heroku app is still live you can see mine: http://lastfm-seasonally.herokuapp.com/user/pvam

Basically, I listen to what I like to listen to no matter the season. If you have last.fm change the last slash in that URL with your username and after a couple minutes (it's single-threaded pulling API requests) it'll show your analysis.

This is really cool! I love playing with my own Last.fm data. Some suggestions:

- Manually (or using your Facebook timeline?) place out big events in your life and see if you can see what kind of music you listened to before/after them that you don't normally. I listen to depressing shit post break-ups.

- Use your Twitter or any other feed to estimate your "mood" over time and see if you can find any music you listened to while you're angry

- Use The Echo Nest to get metrics for how happy/sad/energetic the music you listen to is.

- Come up with your own definition of albums played! I hate Last.fm's because, as you mentioned, it heavily biases albums with multiple tracks. In my personal analysis I always start off with a RLE for tracks played on the same album, and my metric for "albums I listen to the most" is "which albums do I listen to at least 3 continuous songs from the most".

I spent a few months this year trying to combine my Last.fm listening history with my geolocation history (I've mostly just stayed in the same place for the last 8 years, 3-4 years) and Forecast.io's historical API to be able to answer the question "am I really only happy when it rains?" -- Data warehousing is a hard problem, though.

I've been working on something similar for a long time. You can get far by using Last.fm to track your own music listening history and analyzing against notes (break-ups, rainy days, etc).

I love these blog posts. The timing is perfect for me too, I committed some code a week ago that said "I'm doing this manually because I can't figure out how to use `fit_generator`", thanks for the write-up Francois!

In an environment like ours where the split between genders is so drastic small bits of humor like this do less to make people laugh and more to help people who already feel excluded feel even more excluded.

This is rad! I have a bunch of LED matrices I currently use https://github.com/hzeller/rpi-rgb-led-matrix for, I need to spend some quality time with Anypixel to see how it compares / if I can combine the two.

I wish that "in browser previewer" link actually led to a usable in browser thing rather than just a folder on GitHub.

Neither does a lot of Android! Luckily the product I'm using uses https://crosswalk-project.org/ -- that sounded like an ad!

That's wild, I didn't expect any security-centric website in 2016 to be HTTP-only! The reasoning for not doing it is weird too, they could at the very least move the update logic over to a separate SSL endpoint.

Anyway, I'm not quite sure on the differences between them but I've been using KeePassX for years and recommend it thoroughly (as long as you're not looking for a easily synced or multi-user product): https://www.keepassx.org/

yeah I use nginx-rtmp for local video capturing (basically the functionality of a mirror) and the lag that needing to re-encode + chunk + starting off at the beginning of the chunks in the playlist introduces is meh. I think I'm going to transition my project over to WebRTC though, since the mobile device I'm using to display the video can handle that now.

This is awesome! There's a lot of interesting things that can be done with just the level of sequence generation AI we currently have, transforming the real data into a format for it is the most difficult part, I think.

This similar project from a few months ago does Mario level generation and discusses the training data sequence creation in greater detail: https://medium.com/@ageitgey/machine-learning-is-fun-part-2-...

Have you seen this demo of Jack McCauley's new laser based tracking? http://www.roadtovr.com/mts-virtual-reality-vr-tracking-syst...

I have an Oculus CV1, I had the DK2, and I've been making VR side-projects using their SDK for a couple years now. I'm really disillusioned about Oculus as a platform, though. I didn't even consider buying the HTC Vive because I've been riding the Oculus train for a while and had faith in them as the future of VR, but come on.

I really wish I could know how Carmack feels about this.

Sidenote: for what it's worth, I haven't used my Oculus since the first week I received it. It was supposed to work with glasses (the DK2 does) by shipping with different foam faceplates that can change how far off my face it is (my glasses fit in the Oculus, it's just that the lenses are so close to my eyes that my glasses scratch them) but they silently took that off the "What's in the box" months before shipping.

From what I can tell if you exercise a single option this still applies

Two didn't

I do warm-up problems like this even when starting hour long code interviews. I don't like FizzBuzz though, I like to use same difficulty questions that can be expanded onto other tasks, a point-in-rect sample question can turn into a real bounding box question.