Robin Sloan

Aiee, my guess was so close and yet so far. I had no idea you could *do* that. That’s amaz­ing! I want to learn how!

Won­der­ful.

(Can you tell if they alt-​​tabbed away to some­thing else?)

I was won­der­ing about that code. Saw snip­pets of it in the source, but didn’t piece it all together. That last graph may have been me jump­ing back to the top after read­ing a bit to look over changes you made between it and the draft.

HA! I totally bet that crazy one is mine. Mid­way through my read­ing, I real­ized YOU COULD CLICK ON “EURUSSSSS” for sound effects. Also I scrolled back to see the pic­ture of the gar­den to look for the four shal­low graves since, you know, I’m visual like that.

BTW, this is not unre­lated to the research that was done at my under­grad­u­ate lab ( http://​alice​.berke​ley​.edu/ ) , where they would (and still do) “watch” sin­gle bio­mol­e­cules go through some process (fold, unfold, bind, cleave, what­not), some­times over and over again, but one process at a time. By watch, I mean they would mea­sure the force required to hold on to the mol­e­cules and/​or the exten­sion of the mol­e­cules as a func­tion of time, using laser tweez­ers. The idea was that the kind of bio­chem­i­cal mea­sure­ments sci­en­tists nor­mally took–with a ther­mome­ter, a color-​​indicator, some­thing like that–were always bulk mea­sure­ments, and were just aggre­gate aver­ages of these very indi­vid­ual sto­ries that smeared every molecule’s tra­jec­tory together. By look­ing at one mol­e­cule at a time, the indi­vid­ual sto­ries came across more. Sev­eral com­mon tra­jec­tory fam­i­lies would emerge, and with enough exper­i­ments, you could quan­tify the rel­a­tive prob­a­bil­ity of each fam­ily and see how they would, in fact, sum up to cre­ate the aver­age bulk pic­ture. The graphs would look like these tracks, and there’d be these sharp spikes that would indi­cate some­thing had hap­pened: a ribozyme’s bonds had popped, allow­ing it to unfold, or an enzyme had bound to some DNA, chang­ing its ten­sion. It was really lovely stuff. 

So yes, quite won­der­ful to have a lit­er­ary analog.

Ooh, see, already you guys are onto some­thing. Inter­act­ing with (or telling sto­ries about) your own reading-​​graph… maybe that’s part of the fun. Not just hav­ing it drip silently away into the cloud.

I’m not the “And this one? No idea:” reader, but I eas­ily could be.

I rou­tinely go to pages, scroll up and down to get a lay of the land then then skim or read as the mood strikes. Some­times straight to com­ments, some­times savour­ing every­thing, some­times just scrolling down to find out that there is a pag­i­na­tion link and then fig­ur­ing out how to get the whole text on one page so I can hit the Instapa­per book­marklet and read it all later.

How do I know this? Because I found myself rapidly scrolling up and down before set­tling in to read THIS post.

Tim: This was at least a lit­tle bit dif­fer­ent, in that the story page is so stripped-​​down. But even so, I did see a fair amount of the behav­ior you describe (as evi­denced by the chaos of the aggre­gate graph). A few graphs in par­tic­u­lar seemed to whiz down the page, scop­ing it all out, and then return to the top to set­tle in for a more relaxed read-​​through.

yes­ter­day i was going through my (scant) blog hits and saw an IP address from a now ancient rela­tion­ship. he’d gone through many of the posts, enlarged some pic­tures, clicked on com­ment links here and there but ulti­mately exited with­out writ­ing any­thing. i could almost see him toy­ing with the idea then decid­ing it was just bet­ter he left. and i got all *affected* with these lit­tle lines of hits and links. it was amaz­ing to me!

so yeah, your lit­tle lines can tell some cool sto­ries of their own about so much more than just read­ing habits.

Have you tried label­ing the y-​​axis with the images and major plot points?

So… why are there a hand­ful of extra long threads at the top? Do you make more progress thru the story by find­ing more “easter eggs” or something? 

Maybe you should to a “choose your own adven­ture” like this… where the “most pop­u­lar” choices are com­piled into a flat story that gets printed?

Mikeho: That’s an amaz­ing idea.

re mikeho, yes, it should become like this:

http://​samiz​dat​.cc/​c​y​oa/

@sha — awe­some link… 

@Tim Maly — thanks, I kinda won­der if Annabel Scheme could be con­verted into a CYOA story now… hmm.

mikeho, that makes per­fect sense since now Annabel is in one or pos­si­bly many of sev­eral dif­fer­ent worlds…we should do it.

that reminds me, does any­one with more inter­net skills than I cur­rently have want to set up a bul­letin board for actu­ally dis­cussing the book itself? By now most of us must have read it…

Very inter­est­ing but of not much prac­ti­cal use I’m afraid. If you could some­how track eye move­ment across the page that would be a dif­fer­ent story. In any case the results would be highly depen­dent on the content/​layout of the page being read, so they could be poten­tially used to gauge the “browse­wor­thi­ness” of a par­tic­u­lar page.

[…] He is using some scroll snif­fers to track how read­ers nav­i­gate his texts.  Super inter­est­ing, right?   He is also using graph­ing util­i­ties to visu­al­ize the data.  Also inter­est­ing.  Go check it out: Robin Sloan’s code is watch­ing you read Robin Sloan’s code. […]