§ Conversation with Alok about how I read
Alok Debnath, a friend of mine claims he understood "how I read" based on
reading infinite jest and setting me experiments that allowed him to observe
how I read.
In his words:
Alok: I have seen you spasm on your cursor trying to read text (TODO: get a longer quote form alok about what this means)
He said that he never understood what the fuck I was doing until he read
Infinite Jest by David Foster Wallace (a phenomenal book, I loved it and recommended it to him,
which were enough recommendations to get him to read it, it seems).
§ Infinite jest
Alok: It took me reading that book twice to understand how you process text. Let's deep dive. You don't read sentences, from what I remember. You have a different model of chunking text. For what it's worth, I tried to remember what you spasm between when skimming vs when reading. There does not seem to be much difference between those modes for you, which was interesting. So I tried to note why you would sometimes go BACK rather than read sentence to sentence. I figured it was one of three things:
The three things are:
1. You went from verb to verb or action to action, or event to event, and then determined the significance of that event. You would move backwards only if the verb was significant enough to warrant its arguments being understood 2. You went from topic to topic, and would only go BACK if you think you missed a timestep in the movement between topics. 3. You read from line to line, regardless of the sentence, phrase, clause or syntactic structure, and would only go back if an item caught your attention.
According to him, I do a combination of 1 and 2, in contrast to others who
might do (3).
§ Why this works for me
you are uniquely adaptive to reading style, based on very little information. This is a good thing when there is a unique, singular style to the entire article, it is easy to templatize and then retrofit into how you want to get that information.
§ His take on my take on why SEP is trash
I ran a series of experiments to figure this out. I'd ask you to read a paper, a textbook chapter or something in front of me, and after you'd read a paragraph, ask you to explain it. Some times you were already reading something (mostly philosophy related) and that's ambush. Lastly, I used SEP as bait .SEP has not been written for people like you. And I was thoroughly surprised at your vehement disapproval to some of the articles (for their content ofc), but your veiled stylistic inputs as well.
you mentioned that the text of the SEP article on Derrida was malformed, which is a stylistic input, rather than a content issue.
He figured out that the SEP entry has been written in two merged styles - one is
a list of topics that Derrida talks about The other is a list of events which
weave together how those topics became tenets of his philosophy. The style of
writing generally based on the modern notion of "set-inductive" introductions
to topics Which doesn't work well with you. Because I then noticed how you read
code And I figured that you need to have a trace of the topics talked about in
case they appear again in the code, and you parse blocks, retrieve need, GC,
and move on to the next block. So order of arguments and the state of these
topics remains in your mind, along with significant events.
debnatak: You read text in a similar way, which is why set inductive writing is
the worst way to write for you.
§ Inductive writing
Apparently, this is a common philosophy of teaching where one is exposed
to a topic by "forewarning" them of all the subtopics, then a narrative is
weaved exploring the relationship between each subtopic, explaining them as
they come. It's the style in which school textbooks are written. So it is
neither ordered by event, nor ordered by topic. It's the job of the teacher to
guide the student across the text.
Now, the Derrida SEP article is written in a very similar manner, albeit a bit more well formed in narrative The text is not written in a manner where you can parse things by topic(i.e. first deconstructionism, then universality, then sovereignty) or by event (publishing book1 , then 2, then 3 or whatever). Therefore, this writing style is completely adverserial to your reading style!
§ Why I enjoy DFW and infinite jest
Infinite Jest, and DFW in general, clearly does not write like this. It is the antithesis, almost Your insane experience in garbage collection works when there are a large number of interconnected stories, people and threads being referenced in the same sentence.Given that you chunk differently, parse differently, and organise mental notes on what you have read differently than I do, it is not surprising that you understood DFW fairly well. DFWs writing style almost wholeheartedly abandons prescriptivist notions of punctuation and syntactic structure beyond the meager subject verb agreement, Which I think is also abandoned in some monologues. That would not be a large issue for someone who does not use punctuation as a mechanism of parsing sentence information, or even as an anchor of "begin here" and "end here".
§ Inference from me reading code?
According to him, this is similar to how I read code:
I saw you read word2vec.c in front of me And I was mindfucked at how you abstract information on the go Like you read bracket to bracket (I think), and keep track of "important" variables, especially function arguments and return values, and just summarize the rest debnatak: Not every operation needs to be understood, of course. But it is noted regardless. Functions are skipped over if not called, variables are ignored if not used. debnatak: You were never confused about the program flow debnatak: Idk, it seemed clean for code, and math ofc because representations
§ eye tracking data