How do modern researchers and readers manage their bookmarks?

[Novosedoff]

Hi there all,

I thought it might be a good idea to raise this practical issue because the data is becoming extremely abundant in the modern world, but the main problem is that the data is often unstructured and difficult to search.

It all starts from the very basic habits that everyone of us has acquired. Some old school folks still prefer to use printed books, they often make notes right on the book pages, and it takes them quite a bit of time when they need to find a paper they read 10 years ago. So for them it's all analog, in very slow motion.

Modern folks read from the screen, on tablet or computer. The notes they make are easy to share, but still hard to search when the notes are all about plots, tablets and graphics. Lets say in 5 years you wanna find a page from a 500-pages book you have read over the last week. What would you rather be doing now in order to make it happen fast in 5 years when both the name of the author and the title of the book are gonna be completely forgotten?

For a researcher reading over 20-50 new publications per week, the problem of recording and searching becomes even more persistent. So I am curious how you folks manage your notes and memories of the read stuff? What programs are you using?

 Thanks.

[caesar novus]

Here goes an answer, whose very feebleness may inspire better ones from others. First of all, I especially have a problem (re)finding things in an audio file, like an interview or an audiobook. Am I missing some player feature that can mark points or find things inside? I guess videos have a similar problem, but there is some structuring like chapters becoming popular.

I do have a trick for structured reference material I return to again and again. I create tables on a web page which may have headings for rows and columns. They are filled with data cells using dl and dt tags and furthermore may have structure and color codes inside. I tend to use words with perhaps every other letter being a live link. I edit it in a simple code editor that mainly indents and checks syntax, but still allows raw html input. It is labor intensive, but wonderful to switch to and take links out of the soup of bookmark lists. I throw such pages on both local files and free web hosting sites.

For the rest  I have that soup of bookmark lists. You can keep a number of them. I used to have near photographic memory, but that faded out midlife. Also when young I learned to speed read by just reading the first 5, 4, or 3 letters of every word depending how predictable the text - this useful for localized searching. I find it surprisingly easy to do blind searches in e-books like within kindle. Obvious search strings may be too numerous, but I try unusual strings that are probably near the section that I am interested.

Fun fact: this post is my first under Win11 which took countless failed attempts to install over months. Umpteen hours were and will be spent to maintain it. Mean while my Mac discreetly updates itself over and over with little fuss.

Edited February 5, 2022 by caesar novus

[Novosedoff]

3 hours ago, caesar novus said:

I do have a trick for structured reference material I return to again and again. I create tables on a web page which may have headings for rows and columns. They are filled with data cells using dl and dt tags and furthermore may have structure and color codes inside. I tend to use words with perhaps every other letter being a live link. I edit it in a simple code editor that mainly indents and checks syntax, but still allows raw html input. It is labor intensive, but wonderful to switch to and take links out of the soup of bookmark lists. I throw such pages on both local files and free web hosting sites.

For the rest  I have that soup of bookmark lists.

This is good stuff. I used to screenshot anything that I read and found useful or just curious. Over a couple of years I collected a batch of such screenshots all sorted in various folders.  Then I realized that it just made more sense to start tagging the images. So here it goes

https://en.wikipedia.org/wiki/Comparison_of_digital_image_metadata_editors

For audio one can also choose from https://en.wikipedia.org/wiki/Tag_editor

But I've chosen the one which is not even listed in the above wiki article:

http://fastphototagger.sourceforge.net/about.html

 It has both the Windows and Android versions which is quite nice, considering that I am accustomed to reading from tablets. It also supports the hierarchical keywords which took me a while to understand. The author is American bloke called Dennis, who is always ready to respond and to help. 

I mean once I started using the app it takes me literally seconds to find a page from a particular publication I screenshotted a year ago, while I would normally swallow 20 to 50 publications per week, containing multiple figures and tables, which often results in over 100 images saved  in  a week. The only drawback is that, as you point it, the process of meta-tagging is labor-intensive, no matter how fast and handy the app may be. I tried to raise the issue on the exiftool discussion board, probing the very idea of automatic meta-tagging (which in my view is feasible, given all the OCR and machine learning technologies that we have at hand now), but it seems such technologies are not yet around for wide audiences. More important is sharing, because these days the information grows just too fast for one person to be able to read everything through (a known Russian scientist complained about it recently too). But the idea of sharing is not that simple because of the copyrights rights. If someone has published a book containing some interesting findings and I screenshot a few pages from it and share with anyone else, then the writer has the right to file a claim to stop me. The other issue relates to the fact that there are too many data sources one can find on the Internet. We've got arxiv.org for publications in natural sciences, academia.edu for historians, plenty of specialized journals, each of which runs its own archive.  But they all are disconnected from each other, so users ain't able to run one search query across multiple sources, which is a pity. While Web 3.0 should be all about making the semantic search easier.                   

Edited February 5, 2022 by Novosedoff