What do you prefer, tags or categories?
Today there is a weird mix of hierarchical tags, where you can both tag books,
but if the tagged word appears in several places the book will appear on all
relevant nodes.
I can keep this, or make more like categories, where each book appears only on
specific categories you chose. Still you can a word like "general" under both
science and hobbies, but it will be 2 different things (unlike now).
So, change it? or keep it?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
If I were limited to a single choice, I would vote for nested categories. Is
there a problem with having both categories and tags? When doing research,
dynamic tagging would be useful for managing reference sets?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm a proponent of faceted ontology. Instead of classify a book using a
category in a single massive tree, such as is used by the Dewey decimal system
or the library of congress, or a cloud of random tags, we use multiple small
trees to identify different characteristics of an item.
Instead of using a big tree that tries to capture everything about a book in
one tree, use several small trees, which each capture information about a
facet of the item.
So, we don't try to make a tree that lets you say something like
/English/Literature/Fiction or /Literature/Non-Fiction/French, we have a
language tree, with perhaps /language/european/germanic/english and
/language/european/romance/french and another tree that has
/prose/literature/fiction and /prose/literature/non-fiction and still another
for /genre/horror/zombie and /genre/speculative fiction/science fiction and
/future civilizations/known space/jinx or /future civilizaions/humanx
The current classification system makes it easy to create this kind of multi-
tree classification. It would be nice to have better support for structuring
these mini-trees. Especially (as mention by others elsewhere) distinguishing
between Zombie/Fiction/General and Science/Ecology/General where the two
General topics are completely unrelated, at least to my mind.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have 919 books in my eLibrary at this time. (I have a Symbol LS1203 bar code
reader which makes it very easy to scan the ISBN number from the cover of
newer books.) I almost always ask for the Google tags for a book.
When the Google tag for a book has multiple components, such as "Computers /
Computer Graphics / Game Programming & Design", I always create a tree
structure for the tag: Computers Computers / Computer Graphics Computers / Computer Graphics / Game Programming & Design
If I select the node for "Computers / Computer Graphics / Game Programming &
Design" I will only get the one book with that tag. If I select the "Computers
/ Computer Graphics" node, I will get that book, plus the 6 books with the
"Computers / Computer Graphics / General" tag.
As a result of the default tags, and the structured tags returned by Google, I
have a tag tree that includes both "tags" and categories. For me, that is the
best of both worlds.
I can locate books by single topic, for example, in the example image,
"Computer viruses" leads me to "Steal this computer book" by Wally Wang. This
book has tags for "Computer hackers", "Computer viruses", "Computers /
Internet / General", "Computers / Security / General", "Online information
services", "Social Science / Anthropology / Cultural", "Social Science /
Sociology / General", "Subculture". This lets me find this book using specific
tags, but also through categories that lead me from a general concept to a
more specific one.
I can also search for books related to topics. If I'm thinking about social
engineering, I might start with social science, and drill down through the
categories to find the book. Otherwise, using the tags, I might have to
associate social engineering with computer hacking (or one of the other tags)
to find this book.
So, I tend to like the current tag tree, but I would like to see some
enhancements, as mentioned in my feature request.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
What do you prefer, tags or categories?
Today there is a weird mix of hierarchical tags, where you can both tag books,
but if the tagged word appears in several places the book will appear on all
relevant nodes.
I can keep this, or make more like categories, where each book appears only on
specific categories you chose. Still you can a word like "general" under both
science and hobbies, but it will be 2 different things (unlike now).
So, change it? or keep it?
If I were limited to a single choice, I would vote for nested categories. Is
there a problem with having both categories and tags? When doing research,
dynamic tagging would be useful for managing reference sets?
I'm a proponent of faceted ontology. Instead of classify a book using a
category in a single massive tree, such as is used by the Dewey decimal system
or the library of congress, or a cloud of random tags, we use multiple small
trees to identify different characteristics of an item.
Instead of using a big tree that tries to capture everything about a book in
one tree, use several small trees, which each capture information about a
facet of the item.
So, we don't try to make a tree that lets you say something like
/English/Literature/Fiction or /Literature/Non-Fiction/French, we have a
language tree, with perhaps /language/european/germanic/english and
/language/european/romance/french and another tree that has
/prose/literature/fiction and /prose/literature/non-fiction and still another
for /genre/horror/zombie and /genre/speculative fiction/science fiction and
/future civilizations/known space/jinx or /future civilizaions/humanx
The current classification system makes it easy to create this kind of multi-
tree classification. It would be nice to have better support for structuring
these mini-trees. Especially (as mention by others elsewhere) distinguishing
between Zombie/Fiction/General and Science/Ecology/General where the two
General topics are completely unrelated, at least to my mind.
Well, I could just change the tags so it will be path specific, might be
trickier to manage but possible.
I have added an image to my feature request relating to the tree display. Here
is the link:
https://sourceforge.net/tracker/?func=detail&aid=2973872&group_id=245460&atid
=1126369
I have 919 books in my eLibrary at this time. (I have a Symbol LS1203 bar code
reader which makes it very easy to scan the ISBN number from the cover of
newer books.) I almost always ask for the Google tags for a book.
When the Google tag for a book has multiple components, such as "Computers /
Computer Graphics / Game Programming & Design", I always create a tree
structure for the tag:
Computers
Computers / Computer Graphics
Computers / Computer Graphics / Game Programming & Design
If I select the node for "Computers / Computer Graphics / Game Programming &
Design" I will only get the one book with that tag. If I select the "Computers
/ Computer Graphics" node, I will get that book, plus the 6 books with the
"Computers / Computer Graphics / General" tag.
As a result of the default tags, and the structured tags returned by Google, I
have a tag tree that includes both "tags" and categories. For me, that is the
best of both worlds.
I can locate books by single topic, for example, in the example image,
"Computer viruses" leads me to "Steal this computer book" by Wally Wang. This
book has tags for "Computer hackers", "Computer viruses", "Computers /
Internet / General", "Computers / Security / General", "Online information
services", "Social Science / Anthropology / Cultural", "Social Science /
Sociology / General", "Subculture". This lets me find this book using specific
tags, but also through categories that lead me from a general concept to a
more specific one.
I can also search for books related to topics. If I'm thinking about social
engineering, I might start with social science, and drill down through the
categories to find the book. Otherwise, using the tags, I might have to
associate social engineering with computer hacking (or one of the other tags)
to find this book.
So, I tend to like the current tag tree, but I would like to see some
enhancements, as mentioned in my feature request.
As for the barcode reader, a plugin I want to add is a barcode reader from
your videocam
Thanks for your answers. The chosen solution is using tags tree, yet with a
slight change from today.
Books will be tagged by words and their location in the tags tree.
So Computers -> General will be different from Physics -> General.
Also the google tags automaticaly breaks into trees.
You can check the behaviour in the test builds
Best regards,
Amir Shaked.