HNHacker News
TopNewBestAskShowJobs

andrew-lucker

158 karma · joined May 10, 2015

Linguistics + PLT = ?
submissionscomments
andrew-lucker··on FRP in Rust, still a bit early
If Rust ever breaks out of just Servo, I'd be pleasantly amused. Somehow I could never imagine a Node programmer suddenly reaching enlightenment and choosing to take on the burdens of "ownership checking" willingly.

Rustaceans are called that for a reason, and that usually involves some level of PTSD from multithreaded C++ security audits or worse. Pick your poison.

andrew-lucker··on The Mind in the Whirlwind
There seems to be a general trend in literary communities now to talk about the nature of consciousness. Not much pretext is involved anymore apparently.
andrew-lucker··on Materialism alone cannot explain the riddle of consciousness
I feel like philosophers have an over-active desire to simplify. In their mind consciousness must be something elegant and having refined gem-like qualities.

As we see now, consciousness is probably a lot more fuzzy than we had predicted. We now have robot appendages for amputees that interface more and more closely with the natural brain. Even from this cyborg point of view, the patients don't seem to think so much about it other than simply "OMG, thank you".

The question of where "soul" exists is older than science, and not really something I could add anything to.

andrew-lucker··on Geohot’s new automated-driving device can only be redeemed by coughing up data
Is this literally everyone's business plan now? Collect data, ignore sales.
andrew-lucker··on Trump is tame compared to Indian prime minister Narendra Modi
Yeah, but not if Trump gets rid of our cash too and replaces it with higher denominations with his face all over it.
andrew-lucker··on Translating Unix Philosophy into Modern Environments
Same advice as "don't use eval" in Javascript, applies to system calls too.
andrew-lucker··on Does all AI need a learning component?
My personal passion is linguistics, and I find this whole LSTM or probablistic grammar stuff ridiculous. It would be nice to see some research on connecting knowledge models rather than just trying to mimick speech like a parrot.
andrew-lucker··on Does all AI need a learning component?
Sorry for the lack of substance. I don't like to get too specific on Medium, but rather just use it to dump general feelings and emotion. That seems to be the community standard too.

Reposting to HN is different though. What gave me the original feeling of FDD rather than full spectrum research is all of the things that don't get reposted or show up on news. I am sure that people are still doing that research, it just doesn't "have a hat on", as one of my professors would say.

andrew-lucker··on Show HN: Tripartisan – an echo-resistant political news aggregator
Also, OP could add an OCEAN score or similar using broad questions to get a sense of where a user is coming from. As soon as a site gets big enough for abuse, it seems simple voting has always broken down quickly.
andrew-lucker··on Ask HN: What are you working on and why is it awesome? 2017 Edition
I'm working on a port of OpenAI gym/universe to the Rust Programming language. It will be awesome because everyone is interested in AI now, and Rust is cool too.

https://github.com/andrew-lucker/rust-openai

andrew-lucker··on Neural Net Learns to Play Atari “Kangaroo”
I could, but the HN community is really diverse. I think there are genuinely more people interested in just watching stupid machines play video games. I could be wrong, but that is my working hypotheses.

On the other hand, if there is enough interest, I could start live-coding Rust or similar. I would love to do something to support building the Rust community, but I don't want to keep talking to an empty room again and again.

andrew-lucker··on Neural Net Plays Atari “Fishing Derby”
Yes, just a baseline. Trying to arrange a comparison with further strategies X vs Y etc.

Still working on the OpenAI plumbing though, so still a ways before I get to play with all the different research agents available.

andrew-lucker··on A professor’s response to an anonymous letter
Anathema indeed.
andrew-lucker··on Neural Networks for the Prediction of Organic Chemistry Reactions
What is the state of the art here? What models do we currently have to predict chemical/organic inputs/outcomes?
andrew-lucker··on AlphaGo shows its true strength in 3rd victory against Lee Sedol
First game, AlphaGo looked ambiguously weak until end of mid-game when it suddenly looked very strong.

Second game, AlphaGo looked obviously strong throughout.

So the probabilities are maybe not entirely wrong.

For reference I am 4k rating, somewhat below darkforest(1d amateur) and so very very far from 9dan pros that AlphaGo is beating. Usually, pro matches are hard to understand but the excellent commentary and abundant community analysis helps greatly.

andrew-lucker··on Ask HN: What are you working on and why is it awesome? Please include URL
Test all the things! https://jad.subarctic.org/services

Startups usually don't have much time or budget for testing, so we're improving the core technology to bring issue visibility to more businesses.

We manage writing tests/scenarios and customers get alerts, a dashboard, and optionally a public service-status page.

andrew-lucker··on Let Math Save Our Democracy
This article seems to assume that our current system is an imperfect attempt at populist democracy and that somehow we just got the numbers wrong. The history and current opposition is still that populist democracy is not good for rural politics. The argument is simple, most people live in dense cities, and if you counted votes evenly then nobody would ever care what country-folk care about at any level of government.

If the article is meant to be a persuasive essay then it may be good to at least acknowledge that dissent exists, particularly when that dissenting opinion designed or heavily influenced the current system.

"the commission says that its main objective was to give Latinos and Native Americans the ability to elect representatives of their choice"

even the article acknowledges the partisan nature of districting, but instead of recognizing that openly the author chooses to begin with the thesis "PARTISAN gerrymandering is an offense to democracy." Partisan gerrymandering was a founding principle of our democracy. By definition, as long as districts exist in elections, they will be partisan.

andrew-lucker··on The Moral Character of Cryptographic Work
Recognizing that other people have different opinions than you doesn't stop you from preferring or advocating your own. Objective morality is nice to the extent that you can have an intellectual discussion and come to a resolution. More often you just fight it out.
andrew-lucker··on P vs. NP: An Assumption That Runs the Internet
There are NP problems in network routing and analysis, but implying that shortest path is NP would be an oversimplification.
andrew-lucker··on [dead]
Economics as a science is just a backup major for people who failed the maths requirement for Operations Research. Economics isn't so much a science as a study of ethics regarding asset allocation. There is nothing ambiguous about a disagreement on what is ethical in any situation. That is fundamentally a non-sciency thing and just a disagreement of personal opinion.
andrew-lucker··on Java: Missing Features
So would this affect all flat memory allocation? Say if you wanted to allocate 100GB for a bloom filter, would you need to break it into chunks?
andrew-lucker··on Scientific Consensus Is Almost Never Wrong
Scientific Consensus Is Almost Never Wrong

on topics covered in Geology 101

except for that one time

andrew-lucker··on Leap-second decision delayed by eight years
http://physics.nist.gov/cuu/Units/second.html

Seconds have been independent of solar time for quite a while.

andrew-lucker··on Leap-second decision delayed by eight years
Why are these options mutually exclusive? Is there some technical limitation here preventing us from having one monotonic clock and a separate orbital clock?
andrew-lucker··on An Abrupt End to Debian Live
Debian community is a riddle, wrapped in a mystery, inside an enigma.
andrew-lucker··on Misunderstanding the Church-Turing Thesis
The author doesn't seem to be very familiar with the Church side of the Church-Turing thesis. That is where the "limits of computation" notion comes from.

"All total functions are recursively enumerable"

andrew-lucker··on Ask HN: How to improve spoken and written English skills rapidly?
1. Listen (a lot) to speakers/recordings of whatever accent you like

2. Read (a lot) of anything you find interesting

3. English grammar = memorization. We inherit grammar points from Old German, Old French, and Latin. The result is that native speakers mix and match grammar rules and it sounds weird when using a certain rule in a place where it logically should be OK.

It sounds like most of these problems can be solved with bulk exposure to the language. I know, easier said than done.

Also for the grammar, it would be good to have structured learning rather than just listening/reading to random things. https://learnenglish.britishcouncil.org/en/english-grammar/v... seems to be a decent resource, and British grammar is 99.99% same as American.

andrew-lucker··on Sabotage
Not related to overwork or sabotage, but the "talk it down" is expected in certain cultures and has a name: https://en.wikipedia.org/wiki/Self-deprecation
andrew-lucker··on The Hostile Email Landscape
It's only a matter of time before we start seeing peering issues between major email providers. This is not a system for sharing, it is competition.
andrew-lucker··on Programming and ?
For most development roles that I have encountered, the technology seems to be most important. However if you want to move into business, product, or management, then an understanding of how the software will be used becomes hugely important.

If you fundamentally understand what problem you are solving beyond simply the technology then you can answer questions like "what feature should be developed first?" or "how should we organize our data for the customer?"

You can definitely build a career out of being "just a programmer" and that is what I have done. However being a "clueless developer" is limiting and very successful people tend to have a mix of both code and industry knowledge.

← PreviousPage 2 of 3Next →