4 ms·

One should understand that not all processes in life take 45 minutes. Learning meaningful content is like going to the gym or playing the piano - you cannot exp

by dangom 5y ago

One should understand that not all processes in life take 45 minutes. Learning meaningful content is like going to the gym or playing the piano - you cannot expect to see any meaningful results in a day. You'll have to put in the work for a very long time, but you'll get better at it as you go. Analogy aside, not all text is written to be read in a 45 min session. Set your mind that reading and absorbing content from a text book can take anywhere between 3 months to 3 years. Some concepts may take decades to really click in your mind. Luckily, the content in text books is usually broken down in chapters and sections for you, so consider each of those a "less text" content, and read "less text" every day. Set a goal of understanding a single section or chapter per day/week. At the end of a year you'll have read "long text" without ever reading much in a single reading session. Research papers can be read in anywhere from minutes to a couple weeks. But if you are new to the field, you may have to read them over and over, and that time can extend to months until you understand the implications of what is written. Don't rush it.

awild 5y ago
It should not take someone a week to read a research paper. Depending on the paper maybe a day but usually a few hours. Carefully studying it is something else and that might take a week, but not every paper needs that kind of attention and it's a valuable skill to learn skimming and differentiating between un-/valuable and un-/interesting papers.
- baby 5y ago
  > It should not take someone a week to read a research paper Either you have never read a research paper, or you are gifted, or you have been doing that for so long in your field that you forgot how hard it can be when you start.
  - spidersouris 5y ago
    Before arguing for or against this statement, perhaps we should precise what kind of research papers we're talking about. We're on HN, so I know most comments of this thread are dealing with scientific research papers; I'm a student in humanities myself, so I've never taken the time to analyze such papers and I am thus unable to defend or counter OP's claim on this particular question. However, their comment applies quite well to humanities. Except if you're totally unfamiliar with the subject, reading or understanding a humanitarian paper whether it be about linguistics, literature, history or whatnot should not take a week. Most papers are 20-page long and one day is far enough to analyze them. Some may take a bit longer, let's say a few days, if you want to catch and extract their very substance, but never much more than that. Counterexamples are welcomed.
    - hwbehrens 5y ago
      As a counterexample, this [0] was a paper that I read and summarized for some colleagues a while back. It's fairly related to my area of expertise, so I didn't find it too bad -- maybe a couple of hours to grok it and do a write-up. However, among my colleagues (all very talented in their own unrelated areas) it generated a fair amount of consternation, because although they had a week to read it, none of them felt like they had a good grasp of it until after I had presented it. The key takeaways that had jumped out at me immediately were lost in the noise. Whether or not a week is the right time, I can definitely empathize with taking more than a day to evaluate a paper, particularly in an area not very near to your own as there might be a lot of missing background to fill in. In contrast, my PhD advisor could flip through a paper in 30 seconds and pick up the key points it took me an hour to extract. In that sense, I really think that reading papers is a skill that can be trained. [0]: https://drops.dagstuhl.de/opus/volltexte/2020/13109/pdf/LIPIcs-DISC-2020-31.pdf https://drops.dagstuhl.de/opus/volltexte/2020/13109/pdf/LIPI...
  - waoush 5y ago
    I took a graduate-level reading course where the point was to critically read and discuss two research papers per week. These were research papers at top computing conferences in areas related to IoT. You definitely don't need a week for just one paper. I wasn't even a graduate student. Papers are usually around 16 pages. It would take me a good 4-5 hours to read and digest things, and even longer when it was my turn to present a paper, but never an entire week... Professors go over these things even faster.
- michael_nielsen 5y ago
  Right - if the paper is really good, it might take a few years to understand. Most papers can be read in < 10 seconds ("not interesting, next"). Most pretty good papers can be read in 5-10 minutes. Very good papers might take a few hours. And so on. I'm caricaturing, but only a tiny bit, this is how I read. And for good stuff, that I want to understand, a page an hour is (often) pretty typical for my first fairly thorough read. I've written two textbooks, and in both cases I thought of the book as the distilled outcome from reading about a dozen very important papers really thoroughly
  - LaserPineapple 5y ago
    Can you say more about this? What accounts for the difference in time spent on a pretty good vs a very good paper? Is it that you find the material in a very good paper more worthwhile, and thus are willing to expend more effort to understand it?
  - obua 5y ago
    I know you don't believe in formal logic, but maybe give this paper a chance: https://obua.com/publications/philosophy-of-abstraction-logic/2 https://obua.com/publications/philosophy-of-abstraction-logi...

by dangom 5y ago

awild 5y ago

It should not take someone a week to read a research paper. Depending on the paper maybe a day but usually a few hours. Carefully studying it is something else and that might take a week, but not every paper needs that kind of attention and it's a valuable skill to learn skimming and differentiating between un-/valuable and un-/interesting papers.

baby 5y ago
> It should not take someone a week to read a research paper Either you have never read a research paper, or you are gifted, or you have been doing that for so long in your field that you forgot how hard it can be when you start.
- spidersouris 5y ago
  Before arguing for or against this statement, perhaps we should precise what kind of research papers we're talking about. We're on HN, so I know most comments of this thread are dealing with scientific research papers; I'm a student in humanities myself, so I've never taken the time to analyze such papers and I am thus unable to defend or counter OP's claim on this particular question. However, their comment applies quite well to humanities. Except if you're totally unfamiliar with the subject, reading or understanding a humanitarian paper whether it be about linguistics, literature, history or whatnot should not take a week. Most papers are 20-page long and one day is far enough to analyze them. Some may take a bit longer, let's say a few days, if you want to catch and extract their very substance, but never much more than that. Counterexamples are welcomed.
  - hwbehrens 5y ago
    As a counterexample, this [0] was a paper that I read and summarized for some colleagues a while back. It's fairly related to my area of expertise, so I didn't find it too bad -- maybe a couple of hours to grok it and do a write-up. However, among my colleagues (all very talented in their own unrelated areas) it generated a fair amount of consternation, because although they had a week to read it, none of them felt like they had a good grasp of it until after I had presented it. The key takeaways that had jumped out at me immediately were lost in the noise. Whether or not a week is the right time, I can definitely empathize with taking more than a day to evaluate a paper, particularly in an area not very near to your own as there might be a lot of missing background to fill in. In contrast, my PhD advisor could flip through a paper in 30 seconds and pick up the key points it took me an hour to extract. In that sense, I really think that reading papers is a skill that can be trained. [0]: https://drops.dagstuhl.de/opus/volltexte/2020/13109/pdf/LIPIcs-DISC-2020-31.pdf https://drops.dagstuhl.de/opus/volltexte/2020/13109/pdf/LIPI...
- waoush 5y ago
  I took a graduate-level reading course where the point was to critically read and discuss two research papers per week. These were research papers at top computing conferences in areas related to IoT. You definitely don't need a week for just one paper. I wasn't even a graduate student. Papers are usually around 16 pages. It would take me a good 4-5 hours to read and digest things, and even longer when it was my turn to present a paper, but never an entire week... Professors go over these things even faster.
michael_nielsen 5y ago
Right - if the paper is really good, it might take a few years to understand. Most papers can be read in < 10 seconds ("not interesting, next"). Most pretty good papers can be read in 5-10 minutes. Very good papers might take a few hours. And so on. I'm caricaturing, but only a tiny bit, this is how I read. And for good stuff, that I want to understand, a page an hour is (often) pretty typical for my first fairly thorough read. I've written two textbooks, and in both cases I thought of the book as the distilled outcome from reading about a dozen very important papers really thoroughly
- LaserPineapple 5y ago
  Can you say more about this? What accounts for the difference in time spent on a pretty good vs a very good paper? Is it that you find the material in a very good paper more worthwhile, and thus are willing to expend more effort to understand it?
- obua 5y ago
  I know you don't believe in formal logic, but maybe give this paper a chance: https://obua.com/publications/philosophy-of-abstraction-logic/2 https://obua.com/publications/philosophy-of-abstraction-logi...