Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
thelastbender12
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
61.
▲
by
thelastbender12
6y ago
The distinction between natural language and programming languages is interesting, given the push towards machine learning and programming 2.0 that switches explicit instructions in code for training using examples.
62.
▲
by
thelastbender12
6y ago
Your comment is probably in jest but I'm curious - when the public employees' retirement funds are tied in to the equity market and a few companies like Apple/Google drive the price movement, does it introduce adverse incenti
63.
▲
by
thelastbender12
7y ago
Lowering the entry barrier for JS developers increases the plausible TAM if you make money from downstream developer tools/services, described as 'commoditizing your complement`. (ref https://www.joelonsoftware.com/
64.
▲
by
thelastbender12
7y ago
> it's the ability to reference any fragment of the text so that it lives in multiple places at the same time. I've seen Roam being mentioned as a graph database for notes before but your comment helps me see it! Would this be
65.
▲
by
thelastbender12
7y ago
thank you for the link! Do you know if browser extensions can access the webcam too?
66.
▲
by
thelastbender12
7y ago
Ha! The parent might be right about the young people though I'd think the reason for reinventing things is a lot more people writing code and preferring different styles. That is a positive! I don't think YAML/JSON are being
67.
▲
by
thelastbender12
7y ago
The cortex tool mentioned looks really useful to get a service running out of a trained model. Though I didn't really understand what the article is trying to get at. Storing your deployment configuration in yaml and json files is pret
68.
▲
by
thelastbender12
7y ago
Building a component one time and using CDN as a cache seems really efficient. The article also mentions plausibly extending this to support an application like Spectrum. Really curious how would that work since Spectrum has real time chat
69.
▲
by
thelastbender12
7y ago
Hmm, could you please expand? What I'm trying to get at is that the Bus at D and the bus at A expect to reach stop F at different times. So, the ETA between F and H is never quite the same for them, as the traffic conditions could chan
70.
▲
by
thelastbender12
7y ago
This is a pretty nifty solution! To nitpick, I'm curious how much the ETA between successive stops fluctuates. For example, if you have 2 buses on the route `A-B-C-D-E-F-G-H`, using the same value of T(F-H) for the bus starting off at
71.
▲
by
thelastbender12
7y ago
I used it sometime back to bootstrap type annotations for a web api, with pretty reasonable results. Though the codebase was admittedly small.
72.
▲
by
thelastbender12
7y ago
Ha! It really comes down to personal preferences. Distinctive style of presenting material is what really makes a lecture engaging to me. Also, as a selection filter given the abundance of content online.
73.
▲
by
thelastbender12
7y ago
To a reasonable scale, a redis instance behind a small web server should do this for you.
74.
▲
by
thelastbender12
7y ago
strongly agree. I love the extra clarity type annotations bring to the code, though going back to the start of the file every time I want to add an import is slightly dissuading.
75.
▲
by
thelastbender12
7y ago
Very much this! For numerical computing, Numba + llvmlite attempts to do it. I don't know however if this approach could be extended to other domains - say making a web framework. Given, python classes let you do so much tinkering, any
76.
▲
by
thelastbender12
7y ago
thank you, this looks very readable. I was very much looking for a treatment like this that spends more time on how the primitives - perceptron, backprop, etc. came about.
77.
▲
by
thelastbender12
7y ago
second the Mackay book recommendation! It is a lovely text with a ton of examples, good reference for a senior course on applied probability.
78.
▲
by
thelastbender12
7y ago
Thanks for asking this. I had a related question - have you found any other textbooks/courses which take a chronological view on the subject of neural networks, sort of a tour-de-force of their history? A lot of MOOCs and open courses
79.
▲
by
thelastbender12
7y ago
could you expand on why you see an aggregator like Paytm a closed financial ecosystem?
80.
▲
by
thelastbender12
7y ago
Don't you think it'd be the other way around - the older a language gets, the standard libraries degrade since you can't introduce newer patterns without breaking compatibility.
81.
▲
by
thelastbender12
7y ago
MIT OCW used to host their undergrad introduction to Physics lectures by Prof Walter Lewin which are pretty great. It was pulled down in 2014 but you'd definitely be able to find copies on Youtube.
82.
▲
by
thelastbender12
7y ago
Wikipedia :) - https://en.wikipedia.org/wiki/Niki_Lauda His grit, came back from a near-death burn injury in _six_ weeks to finish off the F1 season. The movie Rush (2013) made it more known in the popular culture.
83.
▲
by
thelastbender12
7y ago
To that point, most third-party libraries used also delegate the number crunching to C/C++ code, so most time is still spent outside the GIL.
84.
▲
by
thelastbender12
7y ago
curious, does Swift potentially fill this gap for you?
85.
▲
by
thelastbender12
7y ago
Not quite timeseries related but what are some good query processing solutions that let you make uni-temporal queries without too much application code? For example: assuming a simple application that employs event-stream based storage. * N
86.
▲
by
thelastbender12
7y ago
This sounds fun. Do you mind giving a longer description?
87.
▲
by
thelastbender12
8y ago
thank you, this looks interesting. The parent comment caught my eye since I'm pretty comfortable visualizing linear algebra with 2D arrays but struggle beyond that.
88.
▲
by
thelastbender12
8y ago
could you recommend a good source to pick it?
89.
▲
by
thelastbender12
8y ago
I use Github stars to bookmark a repo I might need later. 62k is not a very big number that way I'd think.
90.
▲
by
thelastbender12
8y ago
That is a good point. But presumably most of Uber's revenue comes from the residents of a city who are in town most of the time. Opening up in a new city doesn't benefit Uber as much as it does Airbnb where all properties are acce
More ›