nbush

This is one of the dangers. AI boosters would say that humans already do this compression and it was accelerated by mass media and then the internet, and that model memory + context can be broad enough that compared to human capabilities the opportunities for depth and variability are even greater. But I think we know which way this optimization usually goes. Even the notion of a "fine-tune for subtlety" is a contradiction.

From the "About the author" page, regarding the author of this article:

> Aaron Horwath is the director of AI operations at a creative technology company where he is focused on implementing AI in a way that supports both humans and business.

This post triggered a wave of bittersweet remembrance: when Watsi was announced I signed up and was genuinely excited for what I thought a well-positioned incubator like YC could accomplish in general redistribution. I was very wrong! But to see this not being a shutdown or acquisition announcement is heartwarming. Keep linearly existing!

This post finally got me to dig back up the ultimate word trivia website, valiantly hosted on Tripod and still maintained: https://jeff560.tripod.com/words1.html

I remember doing this as a CSS copy-protection joke back in 2013. [1] Seems like a similarly petty thing for a company like FB to try and like many have said an extremely low-level arms race.

[1] https://github.com/nbush/headache

Reading your parent comment and the replies has me kind of dumbfounded at the wasted potential. Did they ever address these ideas directly? Would opening up the format in even the most limited ways (like focus) have exposed the IP?

Anyone remotely interested in this idea should check out Stanislaw Lem's His Master's Voice [1], which is an entire (short) novel about how and why an extraterrestrial message would be encrypted. It's worth mentioning how why turns out to be the most compelling question.

1. https://en.m.wikipedia.org/wiki/His_Master%27s_Voice_(novel)

I think this might be something that Netflix is overlooking. The community-like aspect of watching something simultaneously with the rest of the word is extremely powerful. The Twitch generation could prove that live broadcasts or some kind of time-locked release schedule is much more likely to build momentum.

Granted, Netflix's data is probably much more convincing than any vague feeling of community, but I wonder if the "indulgent viewing" they mention is the side-effect of our brains failing to cope with sudden overflow of stuff to consume.

Not really versioning, but Ormr [1] is built around completely non-destructive editing with history baked into each project. Still in beta but worth a look.

[1] https://www.getormr.com/

The strangest part is that they don't even show the bulge in the product renders on the site: http://images.apple.com/iphone-6/overview/images/design_deta...

It's as if they're in denial.

Almost 11% in 2011 [1], about the same as it is today.

1. http://www.reuters.com/article/2011/08/16/us-walmart-idUSTRE...

I'm sorry, but I have to challenge such a broad statement. The classification of literature is an awful game but Dune is hardly a default candidate for serious study, and I feel a view of literature with it near the top is unfortunately stunted. Just speaking as someone who has taken many of the courses you mention, although not an expert.

Nice work! Here's another solution posted in an earlier thread: javascript:d=document.createElement("div");d.innerHTML="<style>span:nth-child(odd) {display:none;}</style>";document.body.appendChild(d);

I don't think there's any easy way for the obfuscation to stay ahead of JS reversal. Thanks for taking a look!

Thanks! Yeah, there are many JS ways around this, but it would be more than enough to stop most people... including myself in most cases.

Cool, I love counterintuitive (and even counterproductive) solutions just for their own sake. I tried to do a similar thing with text a while ago: http://nbush.github.io/headache/

If you want to take the effect to its hilarious conclusion, you can even do depth of field using SVG filters: http://nbush.github.io/depth/

Microsoft Research took the ambient background idea a little further with some cool IllumiRoom demos. [0]

[0] http://www.youtube.com/watch?v=L2w-XqW7bF4

The examples became more and more complex while the code remained evolutionary and understandable. Amazing work and a huge resource for canvas drawing and those unfamiliar with javascript. Very impressive.

To be honest, I'm most impressed by the directional lighting simulation in the header. Any hints on how it was done? Mouse position and image opacity?

Nice! That's probably the easiest way. I guess you could scatter the spans a bit more randomly, but I couldn't find a way around JS completely.

I agree. Therefore, the need for obscurity begets user frustration begets obscurity. A closed karmic circle.

Very nice. Different but distinctive and useful design.

Row selection highlighting would help with navigation; I find myself getting a bit lost when parsing such long rows. I would also like to see a bit more info about how the site works.