HNHacker News
TopNewBestAskShowJobs

daralthus

810 karma · joined February 11, 2010

twitter: @bonsaielectric
submissionscomments
daralthus··on The Cycle from ‘Zen and the Art of Motorcycle Maintenance’ Comes to Smithsonian
It treats the mind as a machine that needs regular maintenance. That's the appeal to hackers.
daralthus··on PPX: Probabilistic Programming EXecution Protocol and API Based on Flatbuffers
Alex Lew gave a great talk on using PP for data cleaning tasks: https://www.youtube.com/watch?v=MiiWzJE0fEA
daralthus··on Layered Programming (2013)
Interesting, although the features might leak into other codebases and services too, at which point compiler directives won't be enough.

I believe this is a lot more tractable with Unision's[0] content addressable hashing. It's kinda made for that although more with an eye towards updates to distributed systems.

[0] https://github.com/unisonweb/unison

daralthus··on DriverKit
Could this be used for better nvidia egpu drivers?
daralthus··on Muse: designing a studio for ideas
I am in completely favour of spatial solutions, but sooner or later you hit the problem that the graph of associations don't map well to 2D unless you denormalize it (like in databases). Which just means if you want to reference an item in multiple places you have to duplicate them and you loose the link between them.

The solution for Muse would be to not put the item in the drawer but the whole area around it as well. Then when you paste it back somewhere you create a wormhole in the space that shows the original place now in a different context. This way you can keep the intuitive zooming and panning and see duplicated areas that update properly.

To be fair there are already some ways they try to solve this with their point style wormholes or the "expand and edit content on the drawer" interaction, but I think they should fully embrace that thought space is not linear.

daralthus··on Ex-Google engineer describing the company's role in China censorship
There is no reason to take survival as the only metric here, no need to oversimplify the problem.

"The best way to predict the future is to invent it" - Alan Kay

daralthus··on Firefox removes core product support for RSS/Atom feeds
I am personally looking for the exact opposite. Read my newsletter subscriptions in my rss reader, so that I can choose the time I read them and keep email for important stuff.
daralthus··on Ask HN: What can be done to prevent a climate catastrophe?
Sorry for painting a negative picture, but when poor countries are destabilised they create conflict for otherwise stable western countries too. Simply because it creates an opportunity for bad actors to gain strategic power.

For example, without becoming too political, it seems a plausible strategy for Russia to try and control migration from the east to gain influence over the EU.

Depending on the hyperbole, similar conflicts could become war scenarios where your wealth would not be enough to protect you.

Considering this, getting rich is certainly a good advice, but might not be enough. We need to work on de escalating the situation even without any altruistic reasons, if you want to maximise your personal survival chances.

daralthus··on Ask HN: Please help me find a sci-fi novel
Thanks, that was a good idea! The story is Reasons To Be Cheerful by Greg Egan
daralthus··on New Theory Cracks Open the Black Box of Deep Learning
> certain very large deep neural networks don’t seem to need a drawn-out compression phase in order to generalize well. Instead, researchers program in something called early stopping, which cuts training short to prevent the network from encoding too many correlations in the first place.

This would explain the success of attention mechanisms quite well. If you are good in selecting the right input you don't need to discard a lot later.

daralthus··on Artificial Intelligence Uses Videogame Footage to Recreate Game Engine
They wrote a blog post a couple of month ago with more details: https://medium.com/@mark_riedl/automated-game-understanding-...
daralthus··on How Magic Leap Works – From Field of View to GPU
Can't you use the photonic lightfield chip for computation too? ;)
daralthus··on David Chang’s Unified Theory of Deliciousness
> Suddenly you aren’t just thinking about what’s happening inside the picture; you’re thinking about the system it represents and your response to it.

This is what true art is about.

His use of "unfamiliarity" as a tool to evoke a shift in your point of view is actually a classic artistic device called "Defamiliarization"[1]. That is: The artistic technique of presenting to audiences common things in an unfamiliar or strange way in order to enhance perception of the familiar.

Or as Viktor Shklovsky put it first: The purpose of art is to impart the sensation of things as they are perceived and not as they are known. The technique of art is to make objects ‘unfamiliar,’ to make forms difficult to increase the difficulty and length of perception because the process of perception is an aesthetic end in itself and must be prolonged.

On step further it can be described with the "Wundt" or "Hedonic curve"[2] that basically says, the most interesting experiences are those that are similar-yet-different to those that have been experienced previously.

[1]: https://en.wikipedia.org/wiki/Defamiliarization [2]: http://natcomp.liacs.nl/images/wundt.jpg

daralthus··on Yahoo Sale ‘Book’ Reveals Financial Meltdown and Big Bet on Mobile Voice Search
The question is simple:

You have deep learning, what would you build with Yahoo's data?

daralthus··on Computing the Uncomputable
This:

Perhaps it is fair to say that “finite” does not mean what we have always thought it to mean. What have we always thought it to mean? I used to think that I knew what I had always thought it to mean, but I no longer think so.

If we go down this road, Hamkins’ result takes on a different significance. It says that any subjectivity in the notion of ‘natural number’ may also infect what it means for a Turing machine to halt, and what function a Turing machine computes when it does halt.

Is beautiful. It seems like another proof for a leaky abstraction and incomepleteness.

We have to forget finite or admit that it has bounds on what we can explore with it, as it already defines our model.

daralthus··on Idea Debt
While 'debt' might be inaccurate, keeping lots of ideas in head definitely cripples you in my experience. They force you to always consider them as an option every time something new comes up and that quickly leads to decision fatigue.

In the meantime 'idea investment' is spot on.

daralthus··on Ask HN: I'm a dull person since software doesn't excite me anymore. What to do?
I highly recommend the book called Impro from Keith Johnestone. It could help with the uninteresting feeling a lot.
daralthus··on A Good Night’s Sleep Is Tied to Interruptions, Not Just Hours
In conclusion, earplugs are the best investment you can get.
daralthus··on A Note on Two Problems in Connexion with Graphs (1959) [pdf]
"One of the bad things about pictures is that they are almost always overspecific. One cannot make a picture of "an arbitrary triangle": as soon as one has made it, it has either an obtuse angle or not, whereas for "the arbitrary triangle" the property of having an obtuse angle is explicitly undefined."

Still one can make a picture of all the arbitrary triangles at the same time and in an interactive visualization explore them to get the complex behaviour and constrains the notion "arbitrary triangle" actually represent. For a good example have a look at: http://worrydream.com/LadderOfAbstraction/

daralthus··on Beyond Light Table
Thinking of programming as a dialogue with the computer is a really strong metaphor that I gladly recognize as something that inspired you [1][2].

Short: You provide the goals and the constraints, then through a back and forth dialogue with the computer you describe and clarify the problem (and also your understanding). You keep on negotiating and providing more constraints until it simply finds a possible solution (the problem domain is now computable) or the best fitting solution now satisfies your needs (in time).

The hardest parts are probably: 1. Collecting, developing the constraint solvers. Should we imagine that collection more as platform of user contributed (programmed/trained) packages? 2. Developing the dialouge that is not just textual but visual (in both ways) (notepad vs IDE vs Theater(?!) don't forget macros/acting out!)). Showing is always faster then writing and gives one better cues on the closures/framework of the actual discussion.

[1] http://scattered-thoughts.net/blog/2014/07/21/imperative-thi... [2] https://gist.github.com/jamii/8957881c8eaa4035f4ae

daralthus··on Show HN: Beyondpad – New kind of note taking and data management solution
you might be interested in http://opp.io/
daralthus··on Ask HN: I feel like an 'expert beginner' and I don't know how to get better
Get an internship at a company you think you will like to work at, something with 10-15 people I think is ideal. If you can't find one yet make some small projects help some friends or build something at a hacker space and try again.
daralthus··on Synthetic Muscle Made of Fishing Line is 100 Times Stronger Than the Real Thing
Somebody who read the paper, can you tell us how much was the achieved length change? It doesn't seem that much in the video.
daralthus··on Ask HN: Which language/web framework to learn for employability?
"skate where the puck is going to be"
daralthus··on Let me tell you about a game I made with the sole intention of watching it die
Reminds me of another game "4 Minutes and 33 Seconds of Uniqueness", which could be the antipole for this.

http://www.kloonigames.com/blog/games/4mins33secs

daralthus··on Don't reinvent the scrollbar
"But Ben!" - Not to be discouraging, the "natural flow" is a fair point in your article, but genereally it feels highly arrogant.

Some experiments may go wrong, and it is ok that you point that out, but telling them not to reinvent anything is another thing.

daralthus··on This is a web page
Damn, you have this medium full of potential that could be anything and you just want to mimic paper...
daralthus··on Flickr gets new UI, new Android app, 1 TB free space
First it looks good, but only because the old one. When you try out the new look, it is immediate that it lacks any consistency in its design or style whatsoever... (maybe because it is not rolled out fully yet. I guess riding the tumblr hype is more important now.)

On the ux side bringing the pictures to the front is pretty reasonable. Unfortunately it stops with a masonry (which is questionable in itself) and a profile header. Everywhere else it is just too noisy, smells like marketing and distractions that stop you every minute from enjoying the pictures. Just look at the home page with the "sign up" popup.

Yahoo needs designers and style as badly as acquiring the next thing every year...

daralthus··on Pirates playing a game dev simulator complain about piracy
Clearly, this is marketing at it's best.
daralthus··on Programmatic Art
Also check out Nature of Code[1]. You could read it online.

[1]:http://natureofcode.com/book/

← PreviousPage 5 of 9Next →