robo editors?

[Alicia Henn]

http://www.nytimes.com/2011/10/16/books/review/do-androids-dream-of-electric-authors.html?scp=3&sq=androids+electric&st=nyt

Here's an amusing article about automated book writing, complete with
how essential human authors are to robo wiki-harvesting.

Alicia

[Jason Olshefsky]

On Oct 20, 2011, at 9:24 AM, Alicia Henn wrote:

> http://www.nytimes.com/2011/10/16/books/review/do-androids-dream-of-electric-authors.html?scp=3&sq=androids+electric&st=nyt
>
> Here's an amusing article about automated book writing, complete with how essential human authors are to robo wiki-harvesting.

The first thing I was going to do was to copy and paste your e-mail and resend it. Or make a script to do it for me.

Then I went and read the article and halfway through I didn't know what "diaspora" meant so I double-clicked, hit Command-C and was about to pop over to Dashboard and put it in the dictionary when a little text-bubble with a question-mark popped up. I clicked it and it loaded a pop-up with the American Heritage definition. Good robot.

Then I finished the article and thought, "wow, reCAPTCHA is going to get really really difficult soon." (Since, when you answer a reCAPTCHA you get two words: one that is known-human-readable, and one that is an OCR error, so when you answer it correctly, you are actually assisting automatic OCR. I figured it would soon be, "are these Wikipedia articles related?")

It's really too bad they make paper books out of drivel. [Insert joke about Congress.] I can see cranking out eBooks, but in a way, it would be not-quite-trivial to make something that makes a Wikipedia eBook (if there isn't one already.) "Hitchhiker's Guide" here we come.

---Jason Olshefsky
http://JayceLand.com/
http://JayceLand.com/blog/
585-789-1473

[Eric Scoles]

On Thu, Oct 20, 2011 at 10:22 AM, Jason Olshefsky <goog...@jayceland.com> wrote:

 I can see cranking out eBooks, but in a way, it would be not-quite-trivial to make something that makes a Wikipedia eBook (if there isn't one already.)

I used to have such a thing, in fact. It was an application that ran on my Archos PDA. It was text-mode only because the Archos didn't have a native web browser (had to install Opera), so it used Lynx. At that time (6+ years ago) I had to do an apt get to install it, but the install clearly could have been automated, and the documentation suggested that, at least with the tools Wikipedia was making available at that time and assuming you had a certain not unrealistically high (albeit higher than mine) level of skill with shell scripting and PERL, it pretty much bordered on trivial. 

It would probably be harder now because: Wikipedia is so very much larger now; and I think they stopped making the whole catalog available for download in XML by wget to whoever wants it. 

But since MediaWiki pages are very highly structured, someone with an intermediate level of JavaScript, Ruby, or PERL knowledge should in principle be able to script a scraper. It would take a hell of a long time to crank, though. 

--
--

eric scoles | erics...@gmail.com

[Dana Paxson]

The more I think about this thread and the article, the more interesting
this whole thing is getting. I'm trying to recall a section of one of
Neal Stephenson's novels, maybe Snow Crash (too tired and cold-ridden to
go digging right now), where some characters immerse themselves in a
kind of hive-semiconsciousness in a community built around that idea.
The robo editor notion and the unending streams of search-based
associative language and connection remind me for some reason of that
Stephenson idea.

The Web and its multimodal, multimedia, deeply-immersive engagement with
our minds may represent an embryonic form of that hive-mind gestating in
the human species, and all one needs to do is to run some searches to
tap into the stream. Once in, things become incomprehensible at the
more-conscious levels of thought but eerily resonant below them.

An example. I Googled three offhandedly-chosen words: "grandiloquence
tertiary quadrature" and got back 94,000 hits. Some were dictionaries
and lexicons, but then in came some strange things: a Hungarian blog, a
blog constructed of what appeared to be random sentences, a Polish blog
of random language, and I found myself having to resist the pull to
continue immersing myself in it all. So what's the pull, and why is
anyone pulling? This looks a bit to me like a tradeoff from conscious
rationality toward preconscious associativity, but I'm not any more sure
of this than I might be about anything that works on my mind where I
can't detect it well.

Maybe we're headed somewhere new - who knows?