keredson

this. i was involved in a lot of interviewing / hiring at one of the top quant firms. i was shocked at how much this factored in.

two sigma, de shaw, aqr, etc.

not true, although i'm sure there are one-off instances where there's no encryption. but in general there's lots of encryption on each of the different levels of secure government networks.

> Aren't all orders eventually public anyway?

no. most orders are not executed on the public exchanges, but by internal matching via your broker and a market maker. (also most exchanges only support trades in multiples of 100 shares.)

it's the quality of the libraries above all else that keeps me coming back to python. truly world class.

There are definitely people who create frameworks who shouldn't, out of wanting to be "the guy", or not wanting to learn existing tooling. But frameworks/libraries are code too. SOMEONE has to write them. And sometimes there are genuinely good/unique ideas. It's good that strong coders w/ genuine knowledge of the problem domain tackle them. Your statement is equivalent to "I've read bad books, therefore I'd never hire an author."

I haven't. But when my rent is < just the interest on the loan for the house I'm in, it's hard to beat.

In the Seattle far-suburbs I'm renting at 3.36% by your metric. And people keep asking me "why don't you buy?!?" THIS. :)

agreed. to be clear i like this project a lot. we desperately need innovation in this area. this current github+jira duopoly on issue tracking is annoying.

ah, bummer. yeah, "closing" a bug in a branch / PR, and then having that status automatically tracked across all merges into master / other branches is i think a huge win for this kind of idea. (vs. manually tracking it per branch, which is a good feature but no one ever seems to use, i believe because it's just to labor intensive.)

huge fan of this model. i started my own proof of concept of it a while ago, tho didn't take it this far. i assume you can track the status of the bug independently per branch?

You have to get it linked on HN for some shameless self promotion. For instance, check out my project https://github.com/keredson/gnomecast, a linux native Chromecast broadcaster for local files, which supports MKV files, subtitles and 4K! I'm only a few stars shy of 1000! Help me out here guys! =P

yeah, that's a sticky problem unfortunately. on the plus side real transcoding is pretty rare given h264 is pretty much everywhere now. most of the "transcoding" gnomecast does is repackaging an unmodified h264 stream from one container format to another, which FFMPEG can do at 30-100x (rather than 1-2x for real transcoding).

a GUI that does exactly that (and the occasional transcoding / reboxing when necessary, to handle mkv files, etc.): https://github.com/keredson/gnomecast

in my experience someone claiming you're being defensive is a last-ditch (but effective) effort to shut down a conversation they feel they're losing. it's impossible to prove you're not being defensive (and if you try you only prove them right), or you give up making your (prob. legitimate) point.

either way it's a chicken-shit ad hominem attack.

I did not. But nor did I write the conversation code. That was an existing python lib.

nice! starred.

yes. the whole thing's only like 500 lines of code. literally wrote it in about 2 evenings after work. should be easy to port to pyqt.

i actually wrote this to transcode on the fly (if you look at the code there's "wait_for_byte" etc), but there are certain limitations i discovered. mp4 files have what's called a "MOOV atom" which is a descriptor stored at either the beginning or the end of the file. but you can only write it once you're completed transcoding. so that pretty much kills it for mp4. vlc uses webm which doesn't have this limit, but you can't put a h264 stream in a webm container, so you HAVE to transcode it to V8, which is slow as balls. (0.1x in my testing, tho you can force 1x if you accept lower quality.) considering almost no one has media in V8, i decided to opt for mp4. but it works if you switch it back to v8+webm.

also they're transcoding to V8 and webm. so pretty much transcoding everything.

thanks!

yeah, but they don't support subtitles as far as i can tell. my frustration w/ VLC's casting is what drove me to write this.

that said, not opposed to PRs that add support for it.

i haven't tried, but you're right in theory very possible. i'm really a fan of OS specific GUIs tho - i'd advise someone use the same underlying python casting API to write a native windows frontend.

thanks! i'm with you on disappointment w/ VLC's latest release. normally they do great work and i love VLC overall, but i waited so long for the casting and it's buggy as heck, and still didn't do subtitles! (my wife's EAASL so it's important for us :) of course all i did was just wrap a GUI around an existing python lib that already worked great, so i really just did all the easy stuff.

casting a video's audio track to an audio device works. adding support for the audio container should be trivial. can you open an issue on GH for it so i can ping you when done?

yep

Regarding the article's point about managing your schema, there are only two real options IMHO:

1) "The database schema is the official definition."

Programmatically generate what ever ORM objects (at build time) in a 1-to-1 fashion from a schema dump. This is the approach DKOs use: https://github.com/keredson/DKO As long as the code generation step is done as part of the build process, you'll have none of the normal code generation headaches, and your build will fail if you've made a code incompatible schema change.

2) "Your ORM objects are are the official definition."

And generate the schema definition automatically. The common process of this is that most "generate schema" functions are stupidly lazy, and drop the work of calculating the diff from an existing schema on the developer (forcing them to write migrations). This is unacceptable in my eyes, just as it would be if my version control software wanted me to write my own diffs by hand in order to make a commit. I strongly prefer automatically generated diffs, like in https://github.com/keredson/peewee-db-evolve. So you can do non-destructive schema changes. It's a model I've re-implemented for any new ORM I wind up using.

also https://github.com/keredson/DKO

yep! it's the standard now for new dev boxes where i work. the samsung mu6300 40" is the best option from our experience. how to config:

https://medium.com/@keredson/using-the-samsung-mu6300-40-4k-...

stay away from the LG TVs. they don't have true 4k RGB.