timkpaine

A lot of these libraries also suffer from really ugly syntax, vs e.g. my library which overloads operators and attempts to "transparently" plug in to python's native syntactic sugar https://github.com/timkpaine/tributary

Having worked on "Dagger", you may be interested in https://github.com/timkpaine/tributary

You can see how it's tied together in this presentation https://youtu.be/M9o9SF5-Pzw

There is a great talk on this as well:

https://youtu.be/M9o9SF5-Pzw

You may also be interested in: https://github.com/timkpaine/tributary/blob/main/docs/exampl...

Just an observation, the male/female ratio on this support letter is dramatically higher than on the letter pushing for resignation https://rms-open-letter.github.io/

The recruitment angle has always been pretty successful, and now with JP's extensive open source work exposure to collaborators and emerging projects has become valuable for participants at these conferences.

With the pandemic, I moved us from being just a pycon sponsor to sponsoring the psf directly (https://www.python.org/psf/sponsorship/sponsors/), though without a senior MD championing these types of initiatives it's always up to lower level employees being vocal and committed enough to push.

Is it that different? https://github.com/timkpaine/tributary/blob/main/docs/exampl...

Idk about using these libraries, but its almost impossible to find generic graph libraries that aren't designed around either ML or alternatively scheduling batches. One such example is my own, https://github.com/timkpaine/tributary

If you just use perspective (which this library uses for grid views), support for dataframes is native.

Really exciting to see the development of an open source conda server, getting locked into "Anaconda Enterprise" is a huge barrier to leveraging the environment at scale in a medium to large size organization.

For those interested, with sympy's lambdify you can do some cool stuff. Here I create a lazy-evaluated version of pricing a third order greek: https://github.com/timkpaine/tributary/blob/master/examples/...

A (now deleted) tweet from a presentation by Chef directly to ICE would suggest otherwise.

https://twitter.com/ccnim/status/968863798827372544?s=21

It is a bit slow, but there's some exploration of keeping a hot pool of prexecuted kernels available. At the end of the day it's still 1 user 1 kernel though, and this is important for security/isolation reasons.

Jesus....https://www.ncbi.nlm.nih.gov/pubmed/24785997

I also recommend looking at ipyannotate if most of your workflow is in jupyter

I am a very sweaty person, and I always get destroyed by mosquitos more than those around me. It's funny how sometimes meaningful research merely confirms things people already believe to be "obvious".

What's more scary than fake sex tapes is the potential for easy plausible deniability for real criminality.

Bar-style licensing* typo

It's funny how no major CS curriculum at a major University includes a legitimate ethics course. In the modern era, programmers are akin to lawyers or politicians; we wield immense, and often implicitly trusted power over hundreds of millions of opinions. It's long past due that programmers be required to adhere to ethical codes, maybe even to the extent of bad-style licensing.

Surprisingly we don't

https://www.smithsonianmag.com/science-nature/world-facing-g...

There's a few options in between, namely the "industry professor" and the "industrial researcher". Many universities will hire someone with suitable industry experience who has previously adjuncted to teach classes and occasionally do research as a "Visiting professor" (but you have to be good at both sides, think https://en.wikipedia.org/wiki/Bjarne_Stroustrup).

There are also numerous roles in industry which do research (even Oracle has a research lab!). This is probably the easier route, especially if you're not as keen on teaching, and more and more companies are open to people doing speculative work and publishing papers (note that i'm making the assumption we're generally talking about software development, ML, hardware development, etc).

you forget how seriously lazy developers are, remember left-pad?

its like watching people learn about the warranty clause of software licenses in real-time!

I recommend the pynq from digilent, it'll help ease your way into hardware design.

Similar to a library I've been working on as well: https://github.com/timkpaine/tributary

Without addressing any of the actual arguments..

>Any webserver will be happy to serve up .mp4 content. There are free players (JWPlayer for instance) and you are entirely free to serve up whatever you want from your own hosted service.

This echos the classic dropbox comment, "you can just do it yourself". A layperson isn't going to.

>That capricious decision can be appealed, you can take Youtube to court if you feel they have wronged you.

You can sue whoever you want, but you will lose due to lack of resources.

> Because it never was theirs in the first place

As the facebook controversies have show, people have a remarkably bad time using common sense and making rational decisions on the internet, so while you are totally correct, its silly to assume people think that way.

that sweet, sweet koolaid

1. Snapchat was listing, not raising capital. Until we see a similarly high-profile company try to raise equity (and I doubt we will), there's no reason to worry for IB.

2. If a client wants to buy/sell snapchat, the investment bank is still going to make broker fees on the transactions.

Does anyone have some good statistics on school-shootings vs workplace shootings? I have to imagine the latter is far more prevalent than the former