https://www.ics.uci.edu/~eppstein/ca/
http://www.math.cornell.edu/~lipa/mec/lesson6.html
http://kaytdek.trevorshp.com/projects/computer/neuralNetwork...
http://psoup.math.wisc.edu/mcell/rullex_life.html
I love it!
The clock has me thinking I need to get out my FPGA based simulator again and to hook it up to a nice 720P display :-).
For a hypothetical device you'd "buy in an instant" how are you imagining the output (and controls). Like, would you want a box with its own display? Or what?
Just googled that, it's a whole archive (yay!). Which demo was it? I obviously want to see something with GoL and "every Intuition window call in the book on purpose" :)
Cell is currently:
Empty Type 1 Type 2
T 7 | . 7 | . 7 | .
y 6 | . . 6 | . . 6 | . .
p 5 | . . . 5 | . . . 5 | . . .
e 4 | . . . . 4 | . . . . 4 | . . . .
3 | . . . . . 3 | . . . . . 3 | . . . . .
1 2 | . . . . . . 2 | . . . . . . 2 | . . . . . .
1 | . . . . . . . 1 | . . . . . . . 1 | . . . . . . .
0 | . . . . . . . . 0 | . . . . . . . . 0 | . . . . . . . .
+----------------- +----------------- +-----------------
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Type 2
Empty Type 1 Type 2
T 7 | E . 7 | E . 7 | .
y 6 | E . 6 | E . 6 | . .
p 5 | E . . 5 | E . . 5 | . . .
e 4 | E . . . 4 | 1 . . . 4 | . . . .
3 | 1 . . . . 3 | 1 . . . . 3 | . . . . .
1 2 | E . . . . . 2 | E . . . . . 2 | . . . . . .
1 | E . . . . . . 1 | E . . . . . . 1 | . . . . . . .
0 | E . . . . . . . 0 | E . . . . . . . 0 | . . . . . . . .
+----------------- +----------------- +-----------------
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Type 2
If the cell is empty, matrix 1 applies. If it has 3 Type 1 neighbors exactly, then in the next iteration it will have a Type 1 value. If it has more than 3 Type 1 neighbors or fewer than 3 Type 1 neighbors it will be Empty in the next iteration.
If the cell is type 1 then matrix 2 applies. If it has 3 or 4 type 1 neighbors it stays type 1, if it has fewer than 3 or more than 4 neighbors it becomes empty on the next iteration.
There are a number of ways to expand this, you can create two 'co-species' where they follow the original rules for their species and ignore the other except for conflict (3 neighbors of type 1 and 3 neighbors of type 2). You can create 'evolving' species where an over population of type 1 neighbors generates a type 2 cell which behaves using existing rules. Etc.
You change the rules by placing E, 1, or 2 in each matrix spot. Any spot that has more than a total of 7 neighbors is invalid (so the top right diagonal half of each matrix).
Also, your matrices for empty and type 1 each have an extra dot after the E on the first row (I think).
I think IPFS [1] or something similar might have the potential to resurrect the geek net. IPFS with a distributed mesh network infrastructure seems like it would be slow, inconvenient for commerce, and probably attractive only to the kind of people that made the early net interesting, especially with the current net as competition.
There are plenty of other times when a more open, informal setting is appropriate, and beginners and journeymen are welcome. But not every time.
> Otherwise I don't see it as more than a toy.
I was a kid so I could well be wrong, but my memory of the early net was that most 'regular' people kind of saw it as a toy. It wasn't until the start of the .com boom that it really became something different. Being seen as a toy might actually be good in this case.
I like the idea of using a different tech stack, though it would have to be very simple to keep up with (and surpass) current levels of innovation. Hence, the terminal?
> my memory of the early net was that most 'regular' people kind of saw it as a toy I think you're a few years ahead of me, then. I don't remember much of that time. But the difference might be (and I hope not) that the general web was a much more unique tool back then, and people knew about it from sheer novelty. I think the chicken/egg problem of users is harder to solve in this case. Not saying it can't be done, just saying it has to be solved intentionally and with effort.
> people knew about it from sheer novelty
This is a really good point. The essence of the question probably boils down to how many people like us there are out there, who want an alternative badly enough to work on it (and in it) while it's small and sparse. The only real feature I can think of that might seriously set it apart from today's net for everyone else is privacy protection, which could be built in from the ground up.
> the other net would be more interesting to just browse around and share stuff on.
What if that was the focus of the system? Make it super easy to publish and find Information. Webpages are honestly kind of hard to work with, and the tendency is to make them way too flashy. Remake the incentives of the system based on what was learned from the first version, and something really cool, and super useful, might be the result. I imagine that a terminal-interface focus might be instrumental.
You complain about ads, but think about what they enable. How many times do you use google per day? It lives off ads. Do you use any social network? They live off ads.
Ads are certainly not the best way for us to obtain those kind of commodities, but between having them and not having anything I still choose them.
Saying that the ends justify the means certainly simplifies things. It's not a compelling argument to me; I suppose that I'm too much of an idealist.
> Do you use any social network? They live off ads.
Indeed, they do. And I use the benefit of my education to block them every way that I can, pushing the cost onto others. Without ads, or at least some kind of financial support, networks like that would fall apart. Necessity being the mother of invention, and interpersonal connections being necessary, we'd invent another solution. I like the idea of small, independent, federated social networks, but I don't know that they'd actually be practical.
It doesn't get rid of silly css and "too clever by half" javascript for the "content site" but dumping the ad networks makes a difference.
Every time I suggest this some ... person... says "but the ads pay for the Internet!" I just look at them as if they're wearing a suit made of dead kittens.
The new web is no longer restricted to the ivory tower; you can get a lot better answers on StackOverflow than you could have ever gotten from the old academic internet. But it does have to feed its new master, capitalism.
That's simply not true. By the mid-90s academics were already a small minority on the web. The OP's sample links are biased because it's a somewhat academic topic.
There was plenty of free and cheap hosting around back then, and for those types of pages bare-bones shared hosting works just fine. For the longest time I paid about $60 a year to host a static site, and it worked great. Hardly the type of money that I needed sell out to advertisers to recoup my costs.
Even GeoCities, while mostly crap, hosted its fair share of useful content.
> The new web is no longer restricted to the ivory tower; you can get a lot better answers on StackOverflow than you could have ever gotten from the old academic internet. But it does have to feed its new master, capitalism.
Nowadays there is technically a lot more information available, but there's significantly more crap to dig through to find it.
And FWIW, I disagree that StackOverflow is "better" than the old style internet. Most StackOverflow answers are shallow - they answer the question, but there's no researching or learning, it's usually just an answer handed to you. It's like the old "teach a man to fish..." thing.
(Not that I'm discounting things like the excellent low-level questions on things like cache misses/alignment/"why are sorted lists faster to process", but I think they're rarer.)
If anyone here knows their way around the StackOverflow API and has some time, I think it could lead to an interesting couple of hours.