cf

14 ms·

31.
Statistics for Hackers (2016)(speakerdeck.com)
3 points by cf 6y ago|0 comments
32.
by cf 6y ago
While I'm not sure how much the modern methods are amendable to napkin math as stated in the article a lot more methods use simulation which if you can code are pretty straightforward to get working. Jake Vanderplas's presentation
33.
by cf 6y ago
dual booting is a kinda a lot of work just for some free games. You also could charge a dollar since you are often offering a strictly worse product for Linux users. That's what happens in many other markets when you are trying to expa
34.
by cf 6y ago
Thanks! And believe me I want to be proved wrong on this. If you find any developers that were motivated to create a native Linux port of their game due to strong numbers on Proton I want to know!
35.
by cf 6y ago
There were Linux ports before Proton and I'm making the case that Proton does not create the right incentives to grow a market for Linux games. Like imagine a restaurant that served food to people in suits and people in tshirts but the
36.
by cf 6y ago
Basic question, if someone buys a game to run it through Proton do they look like a Windows or Linux user on your end? How would you feel about games being made available for free to Linux users if it uses Proton. If the game contributes ba
37.
by cf 6y ago
This has always been true, but Linux gets treated worse than OS X even though those users also have tiny market share.
38.
by cf 6y ago
Sure thing. Supergiant has previously released all their games (Bastion, Transistor, Pyre) with Linux support. With Hades they explicitly said would not come to Linux, and in fact cite the presence of Proton as a reason [1] Nicalis games ha
39.
by cf 6y ago
The problem with Proton is it is leading many game developers to actually drop Linux support. They basically say players seem to be able to get the game to work on Proton and that's that. But with no official support there is no guaran
40.
by cf 6y ago
Hard to edit from a text editor and requires using something specialised
41.
by cf 6y ago
I don't think there should be any single entity responsible for archiving. If you want something preserved there needs to be multiple copies across different organisations.
42.
by cf 6y ago
No conversation about England and housing is complete without discussing Right to Buy where affordable public housing is privatised and then has its value rapidly rise to meet market prices.
43.
by cf 6y ago
You say quaint. To me it seems remarkably out of touch with the real problems Americans face.
44.
by cf 6y ago
Well there is a dedicated paper on it. The short answer is the English description helps a guided search https://arxiv.org/abs/2003.09040
45.
by cf 6y ago
This looks really cool! I gotta figure out a way to contribute to this. Is this what OsmAnd uses to pull in transit information?
46.
by cf 6y ago
What's the best way to help? Do I just add the feeds to another website somewhere?
47.
by cf 6y ago
I'm curious is there something like OSM but for directions using public transit? Some public repository of bus and train times.
48.
by cf 6y ago
Thanks for finding that detail. I went to that article and missed it somehow.
49.
by cf 6y ago
I'm fairly sure Courant was persecuted for being Jewish. Other biographies are pretty clear on this and I can find no reference to what political party Courant belonged to in Germany.
50.
by cf 6y ago
If you want something pithy. AD makes it easy to compute derivatives, PPLs make it easy to compute integrals. In particularly, the kind that come from taking the expectation of a function with respect to some probability distribution.
51.
by cf 6y ago
I thought with libraries like Zygote there is some really nice stuff already in Julia. I'd say it's still early days for good autodiff libraries in general and I think we still haven't really explored what they can do.
52.
by cf 6y ago
The reality is that most of a modelling task is preprocessing your data before it can be passed to a probabilistic model and postprocessing to make decisions using it. The code is usually written in R or Python so there is a strong pressure
53.
by cf 6y ago
I think systems built around effect handlers like pyro and edward2 form a nice corner in the design space by being less reliant on metaprogramming, being fairly composable, and give nicer UIs. The downside is a trickier to write and debug A
54.
by cf 6y ago
The reason Julia is a good fit is it has both good numerics support (arrays, matrices, autodiff, samplers, etc) and good metaprogramming support. Probabilistic programming systems really need both to feel natural and well-integrated while s
55.
by cf 6y ago
Why not? If you pick your model well incremental updates to the model can be made fairly cheap. An open source example that's relatively usable can be found at http://probcomp.csail.mit.edu/software/bayesdb/
56.
by cf 6y ago
The main simplify function is full of heuristics and is itself a basket of more specialised simplify functions. One thing to try one or several of these specialised functions. Another tip I don't think gets mentioned enough, try to pro
57.
by cf 6y ago
You should definitely raise an issue if you got the bandwidth. The project has generally been responsive on this sort of stuff.
58.
by cf 6y ago
Oh man this is like 90% of what I wanted to make! Thanks for making this open-source!
59.
by cf 6y ago
Yes I want to adjust remote audio streams but I don't see where in the API I can iterate over each user's audio feed.
60.
by cf 6y ago
One thing I've wanted to make is something like gather.town where I can remix the audio and video so that different users sounded louder or quieter. But, I never figured out where in the WebRTC API that is done. It seems like I need to

More ›