HNHacker News
TopNewBestAskShowJobs

nyan_sandwich

200 karma · joined August 24, 2012

Mechanical engineer, programmer, and aspiring entrepreneur based in Vancouver, BC.

I'm currently interested in making technical knowledge and thinking more accessible and convenient, and fixing Science in general.

submissionscomments
nyan_sandwich··on Ask HN: What are the New Computing Ideas for the next 50 years?
AGI. The Intelligence Explosion.

http://intelligenceexplosion.com/

https://en.wikipedia.org/wiki/Intelligence_explosion#Intelli...

nyan_sandwich··on Tell HN: Startup school acceptance emails are out
My email is wwtivy@gmail.com Mail me and we'll chat/meet up
nyan_sandwich··on Tell HN: Startup school acceptance emails are out
>Email's in profile

you sure about that? I don't see it. And yes I'd love to meet up.

nyan_sandwich··on Tell HN: Startup school acceptance emails are out
Yikes I'm in!

Coming from Vancouver. I'll probably be staying for a few days around the even to absorb the atmosphere.

I guess I'll need to share cars with someone for transport.

nyan_sandwich··on Procrastination should be solved by lighting fires, not filling buckets
As an ADHDer, that's what it's like for me; it's not a lack of focus, it's a lack of executive control over focus. I can go head down for sixteen hours on a really interesting problem, but if I'm not feeling it, it just doesn't happen.
nyan_sandwich··on Procrastination should be solved by lighting fires, not filling buckets
Brilliant. I feel it, man. In fact, as of two weeks ago I am building this right now (in a small slice of spare time of course).

Thanks for the inspiration for a new aspect of it.

nyan_sandwich··on Procrastination should be solved by lighting fires, not filling buckets
I have exactly this problem, and I actually have clinical ADHD, but I'm too akratic to get to a doctor to get anything done about it.
nyan_sandwich··on Day 180: Finished
>Like saying, "Harry Potter is in Ravenclaw."

Well actually... In a certain really good HP fanfiction, Harry Potter is in fact a Ravenclaw.

http://hpmor.com

nyan_sandwich··on Technology and the College Generation
>And if they think that college is boring, wait until they have to sit through their first meeting at work.

I was with you until this. Boredom means you are not engaged to your full potential. If the "real world" is even more boring than school, people should be complaining. (And someone should be making a killing on that inefficiency.)

nyan_sandwich··on The Banality of Systemic Evil
>But the thing is, those leaks weren't one-man attempts to re-architect the US

This is a good point.

The same applies to many other things in history; communism wasn't some kid who thought up a new ideology in his basement and pushed it on everyone else; it took the endorsement of the intellectual elite and a broad movement to make it happen. Still it went badly for billions of people.

Your implication is that this stuff is happening whether we like it or not, and that that somehow invalidates criticism of the naivety of it all. But then you side with it. If it is an inevitable march of history, surely it is as reasonable to criticize it as it is to endorse it, or as futile to endorse as it is to criticize?

The thing is that these broad movements are made up of people like Snowden and Swartz and so on who make things happen because they believe in it, and we do have some control over that. For example, if the international intellectual community had been less naively infatuated with democracy and independence movements, perhaps they would not have pushed so hard on Rhodesia and the Belgian Congo, Zimbabwe and the DRC would be first-world countries instead of hellholes.

If you make the inevitable march of progress cool, it will happen faster, whether it's a good thing or not. If we make reasoned caution and sanity cool, we might get a better outcome for the future of humanity.

nyan_sandwich··on Decoding Vibrations From Nearby Keyboards With Mobile Phone Accelerometers [pdf]
Right, and when someone made the first transistor, it was no smaller than a vacuum tube. Clearly impractical.

Give it time. A proof of concept does not represent the upper limit of what a technology can do.

nyan_sandwich··on Ask HN: What are the best technologies you have worked with in 2013?
LuaJIT. The FFI is really, really, cool (inline low level access to memory and C in a high-level garbage-collected dynamic language? Yes please.)
nyan_sandwich··on The Banality of Systemic Evil
I agree that "the system" has no intrinsic value, and is purely an engineering problem. This is obvious. But this sentiment is often erroneously taken as an argument for naively re-architecting the current system.

Us engineer/programmer types know that the last thing you want to do is a total rewrite of a working system, because its precise setup often contains domain knowledge that we don't have. "I don't know what this piece does, but it's ugly and seems problematic" does not necessarily imply "this piece should be removed or changed using my current level of expertise".

The analogy is not perfect, because civilizations are not really intelligently designed, so they can often contain crud that's just crud, but the same goes for you own body. Do you have the domain knowledge to determine that the appendix does nothing, the little finger is noncritical, and the liver is critical? Do Snowden or Swartz have the domain knowledge of the workings of civilization to make a decision like that for everybody?

It's that kind of naive wholesale re-architecting that dropped the ball on the Congo, Zimbabwe, etc, and caused a hundred million to die by communism.

As much as I think Swartz was right to challenge the rent seekers in science publication, and Snowden was right to undermine the NSA, I think there's really something to be said for being careful with this stuff.

Sure there's a lot of apparent problems in our current civilization, but please consider that we might occupy a relatively optimized position in a sea of dysfunctional possible societies, and that the exact mechanics of that may be nontrivial and beyond the understanding of you or me or any of these other "Generation W" troublemakers (of which I am one). The stakes are high enough that we really ought to be careful.

>Another superstition trampled under the relentless feet of reason.

This doesn't seem like rational thought, it seems like endorsement of destructive mob behavior backed by quasi-religious ideals.

Would you recognize naive problematic restructuring when you saw it, or are you just using this sentiment as another soldier in the war against some fixed enemy?

nyan_sandwich··on Nginx image processing server with OpenResty and Lua
Actually, Openresty is both faster and more obscure.
nyan_sandwich··on Ask HN: Why does evil succeed so well in our world?
It is demonstrably not the case that it is easier to make money through evil than good, at least historically; the world we live in now is quite nice compared to the past, and that's the result of selfish people following incentives. If incentives were aligned with evil, we would expect to live in an evil world, and we do not.

There is evil around every corner for sure, but it is outweighed by the good. Sure there are incentives for scummy business practices that add a bit of evil on the margin, but do you really think most companies are a net negative? That that marginal evil isn't outweighed by the vast good and wealth that is created as well?

The question you should have asked is "Why do i think incentives align with evil?", and the answer is "Because you take the good for granted and thereby fail to notice it."

We're on the Internet. Show me some evil that outweighs the awesomeness of that.

nyan_sandwich··on Why Aren't Schools Grouping Kids By Intelligence Level?
There are a number of obstacles between where we are and a world where intelligence is taken as seriously as it should be.

See most other comments around here though; the reason IQ is not used more often is that intelligence denialism is rampant. One reason is that as some comments demonstrate, egalitarianism and democratic principles are incompatible with IQ realism. It's pretty clear that intelligence segregation is going to turn out in practice to be racial segregation, so it's simply not going to happen in a society that has defined itself against such practice.

nyan_sandwich··on Ask HN: What features do you want in a modern spreadsheet app?
What is a spreadsheet, really?

It's an interactive programming environment with an interface optimized for data entry and some other things. The difference between Excel and, say, Matlab or R, is excel's strength at data entry, visual/intuitive nature, and total lack of accessible higher-level programming features.

So the ideal solution in this problem space, to me, looks not much like excel at all, but more like a high-powered computation environment a-la mathematica, except with roughly the execution model and visual/accellerated/intuitive interface strength (and then some, as you suggest) of spreadsheets. For example, the whole grid thing might have to go, instead making tables first class objects in the language, and then have a strong visual representation with an accelerated (not plain text) interface for manipulating them.

But maybe you want to take it in a different direction than I do? In any case, excel can trace a pretty direct line to VisiCalc, with very little fundamental reevaluation applied since 1978 or so. So you might want to consider not being bound too closely to the "this is how spreadsheets have always been done" thing.

nyan_sandwich··on The consumer driven economy is for people who don't naturally automate things
>They will still be under the control/ownership of capitalists because they won't be persons - everything that isn't a person technically has an owner.

>When machines do cross the barrier of deserving rights, we should reconvene and have this discussion again.

You are assuming the rule of law holds through all of this.

If an autonomous machine intelligence is capable of becoming militarily sovereign (in the sense that it becomes nontrivial to simply "arrest" or otherwise overpower), it matters little whether the rest of us think it has rights, or who nominally "owns" it.

If that machine is also able to outcompete human capitalists at wealth accumulation (assuming market holds) or human generals at military power (assuming nothing)...

And let's face it, it's not that hard for even lone smart humans to be militarily untouchable. Think cybercriminals: If today we decided that cybercriminals weren't people and were owned by the government, would the they care? No; they are sovereign through anonymity, and their legal status is not an impediment to their work. Likewise with autonomous machine intelligence.

nyan_sandwich··on The consumer driven economy is for people who don't naturally automate things
>1,2,3

Yup, except why do the developers not get the income? Presumably they are selling the machines?

>4

Yes. The key with this kind of logic is to relate it to the current situation, though; people with more workers can afford even more workers. Why doesn't it all blow up now?

There is more to success than simply scale of manufacturing capability; firms need to keep ahead of competitors, keep up with technology, manage the whole stack, etc, and then they can end up controlling the whole market (at the thing they do) and having nowhere else to go because the firms in other fields have an advantage that takes something special to beat.

Also, the neat thing about capitalism is that money to buy more work is rarely the bottleneck, because you can get a capitalist to invest in your scheme (if it looks any good).

>5

The second thing (poor get poorer) does not necessarily follow. Only the people who can't do anything that machines can't do better get poor. The working poor have more buying power because the machines create more wealth for less money.

In practice, despite a lot of automation, we don't have many people unable to do anything productive; people are very much more versatile than machines (for now).

More on that: http://lesswrong.com/lw/hh4/the_robots_ai_and_unemployment_a...

>In the future (my crazy thoughts):

Yes, eventually the set of people who cannot do something that machines can't do better includes everyone, and everyone becomes unemployed. (Which looks pretty good, IMO; who likes working?). There are more serious problems than unemployment at this point, though.

At this point, why do you expect the machines to still be under the control of the capitalists? If the machines can do capitalism, etc better than capitalists, any wealth that ends up in the hands of an autonomous machine will eventually increase to eclipse the human capitalists as well. (And of course, meanwhile the machines are increasing in effectiveness because they are better at machine design as well).

This is called the "intelligence explosion" or "technological singularity" scenario, depending who you ask, and looks like bad news, unless we do some serious stuff to make sure the machines work to specifically benefit humanity instead of just continuing to find better and better ways to do whatever other task they were doing. (Better ways that eventually include overpowering humans militarily and melting us down for raw material).

If you want more info: http://intelligenceexplosion.com/

nyan_sandwich··on Ask HN: Why isn't anyone using Openresty
I plan to in some upcoming projects, because I'm a Lua nut, but here's why people might not use it:

* Lua syntax can get annoying and is pretty low level (no list comprehensions)(I switched to moonscript[1])

* Lua has no libraries or established support community compared to eg. python. You have to do a lot yourself.

* Performance isn't the most important concern for most people.

* Openresty isn't really a "framework", it's pretty barebones. (Lapis[2] looks like it should help)

[1]: http://moonscript.org/

[2]: http://leafo.net/lapis/

nyan_sandwich··on The Colors of Dribbble
That (quoted) statement is probably wrong (with the uncertainty focused on how the future goes).

The Mantis shrimp[1] can see colors in up to 12 dimensions, instead of the pathetic three that we can see.

Further, a study done with mice which I can't find now involved implanting mice with the genes for a third type of color receptor, and they grew up being able to distinguish colors in three dimensions instead of two. In other words, if you add a few genes for more color dimensions, the brain adapts.

To advance the human capability to perceive color, it is probably just a matter of adding a few genes. This kind of stuff is within a century for sure. Our descendants could grow up being able to see visual subtleties we can't even imagine, if we choose to go that way.

[1]: https://en.wikipedia.org/wiki/Mantis_shrimp

nyan_sandwich··on How not to loose motivation about a side project?
I find it hard to lose motivation with beeminder (https://www.beeminder.com/) kicking my ass.

I used to have a lot of trouble.

nyan_sandwich··on ShowHN: GetURL - A CLI tool to get a public link for any file
Another:

    cat file | curl -F 'sprunge=<-' sprunge.us
nyan_sandwich··on Self-Taught Developers: Are You Missing Your Foundation?
At first I found the article kind of condescending. I was expecting something I didn't know, instead:

>hey self-taught developers, here's all this totally elementary stuff that you probably don't know.

I'm self taught, but I have a very solid computer science foundation because I found that stuff interesting when I was learning. Looking at the other comments here, I guess that's not the default...

nyan_sandwich··on The last perambulation of an ancient arthropod
The description of the poor little critter slowly dieing as it tried to find breathable water made me really angry at the universe. Death is such a horrible thing.

Maybe I shouldn't feel empathy for a nonsentient bug, but similar things happen to thousands of humans every day...

We shall have to work faster.

nyan_sandwich··on Evolution could explain the placebo effect
In the same sense that Physics can explain it...

It might be better to refer directly to psychology and neuroscience to explain the placebo effect

nyan_sandwich··on Introducing the 4chan API
um... make a 4chan app that lets users up/downvote threads, then builds a naive bayesian model of what keywords (eg. "toasting epic bread") are correllated with the kind of threads you like. Netflix-like. A sort of automatic cream-extractor.

de-anonymizer based on posting times, writing style, what baits them to respond, etc. The Thread-Local unique ID's would help, giving you more stuff that you knew came from same user. Don't know if this one is practical. Kindof scraping the bottom here...

That's it for now.

nyan_sandwich··on Introducing the 4chan API
create a markov chain 4chan slang generator.

track usages of phrases over time. (thinking of the recent evolution of "rustled my jimmies" derivatives)

See what topics are trending

Fuck maybe I should build some of this...

nyan_sandwich··on Introducing the 4chan API
It's not like the html form does anything special. Forms don't have funny markup you have to scrape or produce. It's just a POST.

I guess there's mime/url encoding or whatever, but that's hardly an issue.

nyan_sandwich··on Free Book: An Introduction to Programming in Go
vs Java-like

float64 myfloat64array[] = new(float64[5], ...)

← PreviousPage 2 of 3Next →