Prereading and reading tactics

We are strong proponents of pre-reading, yet the issue is only briefly mentioned in our course. There is a good reason for it. There is no one good prereading strategy, but several recommended strategies for various scenarios. Instead of discussing various pros and cons I will describe “a day in my life”. My days are not exactly like this, but I will emit children screaming, commuting, coffee breaks, lunch break (culmination of the day), multiple job assignments, occasional meetings and zombie collapse in front of TV with iPad in my hands. I take lots of smaller and longer breaks which are not mentioned here.

Prereading and reading news blog

News blog is the best media to train prereading. First I read title, if the title is remotely interesting I read the excerpt with ~5000wpm speed. If the excerpt is remotely interesting, I open the article as a new browser tab (this method is suited for PC not mobile) and forget about it until my tabs are full of articles. I use 24 inch 1920×1080 monitors, so this takes some time. Once I am happy with the result I preread the whole article with variable speed. I start at 5000 wpm and read the whole article. If the acticle was a waste I close the tab. Otherwise I reread the intersting paragraphs, usually around 1500 wpm. I allocate around 1 min per article. Very seldom the article is very long or complex or useful, in this case I move the tab to the left of my gmail tab for further integration (occasionally rereading, typically as a blog post idea in google docs/asana/trello or a mail to a friend). Rereading the article for the third time, my retention of the article is very good for very long time.

Prereading and reading tech articles

Then I am asked to solve something I have no idea about. I do some google search magic [subject for some other posts] and open several tabs with cool scientific articles. For each article I preread the whole article at 5000wpm, because there are potentially cool ideas hiding in different places of the articles. From ~20 articles I preread I select 3-5 and send them to my colleagues. For each of these articles I already have mental colouring of the interesting areas. I reread those areas at 2000 wpm, and then again at 200 wpm the definitions and the formulas. After reading all articles, I stop to think. Finally I reread the whole 3-5 article batch at 3000wpm to generate fuller understanding of the subject. Occasionally I understand that the articles are too fragmented and I need to read some fundamental book….

Prereading and reading tech book

Typically books are more boring than articles since they need to be systematic, so I try to avoid long books. If I do need to read long books, I read preread the introduction sections at constant speed of 5000wpm until I find the subjects I was looking for. I mentally colour the interesting sections and yet continue to read at 5000 wpm. When I finish the book or the section (I do not do more than 100 pages at 5000wpm without getting a serious pomodoro break), I revisit the interesting sections at constant speed of 1500wpm. If I do not understand something I revisit it after finishing a subsection (between 1/2 and 3 pages) at low speed of 200wpm. Occasionally I reread the whole interesting sections at 5000wpm to generate more unified understanding. The “unified understanding” means proper linking between various smaller structures of mental markers.

Prereading and reading computer code

Then I need to implement the subject in computer code. Computer code below 10,000 lines is small, between 10,000 and 100,000 lines is medium and above 100,000 lines is large. For large codes it usually takes several days until I generate their structure in my mental markers. For medium code [if the code is readable and not overly engineered] it takes a day or two. Orientation in small code base should take ~20 min. Then I am left with several markers for short code segments of 200-400 lines with mental marker p[er segment. Orientation within a short code segment is an issue of seconds. Typically I need to modify a given code in 3-5 segments to integrate my ideas within the code. I do spend some time testing that everything works as expected. If fact debugging takes a large part of my day.

Technical writing

Finally I need to document what I have done. Usually I just poor my mental markers into powerpoint presentation and add visualization of testing results before and after the modification of the code. Occasionally I reformulate the presentation into a patent or a specification document for my colleagues. Then I use speedwriting – I follow the structure of my mental markers but also add the details for each marker.

Fun reading

By the end of the day I am exhausted. I visit some funny site like http://www.fishki.net or http://www.omgfacts.com/. I do not preread them, since the articles are very short and fun. I read about 2000wpm and focus on images 🙂

I hope this answers some questions regarding my personal reading style. I preread at 5000 wpm, select interesting objects and reread at 2000 wpm, reread some hard places at 200 wpm, and reread the interesting objects again at 5000 wpm. And I take a lot of breaks between reading. What is your reading tactics?