An Attempt to Exhaust Memory to Simulate a Place to Attempt to Exhaust

[bcj]

I'm definitely doing this, I just have no idea what or how.

First two ideas off the top of my head:

• Something using The Library of Babel
• A procedurally generated version of: "An Attempt to Exhaust a Place in Paris" — Georges Perec

[WhiteFangs]

I can't wait to read (and see how you implement) the second one!

[bcj]

I've spent some time thinking about what would be the most Borgesian thing I could do with The Library of Babel, and I think the answer lies in Pierre Menard, Author of the Quixote. I might try to find Don Quixote hidden within the library of Babel.

The simplest implementation of this would be to sanitize the Quixote so that it only contained [a-z ,.] then break it into 80-character 40-line pages, and search for each page using the library's fantastic search function. There are ~2,000,000 characters in the novel, so that would only be ~1000 pages. In that case, my 'novel' would probably just be the list of hex locations of the pages.

I don't know if I'd like this as the NaNoGenMo submission. I may like it as a NaNoGenMo submission.

[bcj]

I've yet to come up with any real way of doing number two, although it might involve attempts to exhaust a place on twitter

[hugovk]

Second one -- find (or set up your own!) webcam pointing out the window, hook it up to some computer vision library that verbosely logs what it sees.

[bcj]

I guess if I was using twitter, I could use a bot, have it follow anyone that follows it, then use its feed as the events it witnesses. I don't think there's quite enough there as an idea though.

[bcj]

The webcam thing seems like a pretty cool idea. I don't know if there are free computer-vision libraries that would be good enough though.

I'm a little bit thinking of trying to make some sort of city simulation and then logging what happens at a specific area.

[ikarth]

The computer vision thing depends on how you want to use the data; katie rose pipkin's cloud ocr is one approach. There are open source image recognition libraries out there, of varying quality.

I like the simulation idea too. There's a lot of scope to explore more of that approach, and I think it's been somewhat underdeveloped historically.

[jerry-clayton]

Perhaps instead of attempting to exhaust a place on twitter, you could attempt to exhaust a moment? For example, the first few hours after the Snowden Papers were leaked. You could collect relevant tweets and manipulate or sort them in some way.

[bcj]

That is an interesting idea, and I think that Life: A User's manual by Georges Perec takes place at exactly one moment, so it is very fitting.

[hugovk]

Perhaps instead of attempting to exhaust a place on twitter, you could attempt to exhaust a moment? For example, the first few hours after the Snowden Papers were leaked. You could collect relevant tweets and manipulate or sort them in some way.

Here's an archive 13 million "tweets that mention Ferguson, Missouri between August 10th and August 27th, 2014 subsequent to the death of Michael Brown". It's actually just the IDs, but you can use twarc (the tool that collected them) to "rehydrate" them to full tweet data. More about this archive here which also mentions an Aaron Swartz tweet archive, and there's quite possibly more archives over here.

[bcj]

Wow, that's pretty awesome. Thanks

[bcj]

Alright, I've pretty much settled on the idea that my main focus for NaNoGenMo will be simulating a city that I can generate a description of. Although I am reserving the right to about face once I recognize the logistics of that goal. I also do plan to spend some time playing with that Twitter data set because that's just cool

[enkiv2]

Will you be simulating the people in the city, or will this be more of a SimCity kind of view?

On Tue, Oct 27, 2015 at 3:16 PM Brendan Curran-Johnson < notifications@github.com> wrote:

Alright, I've pretty much settled on the idea that my main focus for NaNoGenMo will be simulating a city that I can generate a description of. Although I am reserving the right to about face once I recognize the logistics of that goal. I also do plan to spend some time playing with that Twitter data set because that's just cool

— Reply to this email directly or view it on GitHub https://github.com/dariusk/NaNoGenMo-2015/issues/3#issuecomment-151614952 .

[jerry-clayton]

If it's alright with you, I might take on that idea of exhausting a moment. I looked into the novel you mentioned and it is truly fascinating.

[bcj]

My current idea is that I would like to make a top-down simulation a city, but for (at least some of) the people who cross into the narrator's view, simulate them on an individual level. I don't have any experience in this kind of stuff, so I have a good feel yet for what kinds of things are going to be possible though. I am excited about that

@s-knob: please do. Even if I end up doing stuff with a moment, I feel like any two implementations would be different and interesting in their own right

[bcj]

As a bit of a warmup, I've completed my Library of babel project #117