Calibre users with Dewey and/or LCC and tags: Relative usefulness?

[kite]

I'm looking for the experiences of calibre users whose libraries show both tags, and DDCand/orLCC. (Dewey Decimal Code, Library of Congress Code)

I'm currently wanting to have DDC (and LCC as it seems almost no extra work).
But if many people with say +500 non-fiction works don't find it DDC or LCC useful then I'll find something else to dream about.

If you've got both tags and DDC/LCC:

How much use do you make of this metadata both absolutely and relatively?
(If you use tags once a year and LCC twice then I'll be thinking that "twice as much use of LCC as use of tags" isn't worth putting much effort into.

How many non-fiction works are in your calibre library?
(I wonder if DDC/LCC become more useful as non-fiction libraries get larger.)

[BetterRed]

@kite - Most of my books are non fiction (2000+ books and 60,000 other). I don't use Dewey or US Library of Congress codes and I've never felt the need. I do search on Author and the Tags I assign (small set thereof and no imports).

However, I tend to use full text search more than I use metadata searching, I often add the search results to the library as a Book - so I can find it again

BR

[kite]

Thanks BetterRed. I've got quite a few image (as opposed to text) pdfs which won't text search. But have heard something similar to what you write from another user who does have DDC.

I am steeling myself to downgrade having DDC and LCC from "something I really need" to "something that is absolutely essential for library-geek decor".

When you say "full text search" you mean something like
[q]Chris Jensen, Entrepreneur, engineer, scientist
Search for a phrase within every epub file in the current directory.

*nix based systems including Linux and Mac OSX:
grep "some phrase" ./*.epub

Windows:
findstr /C:"some phrase" *.epub
[/q]
?

[BetterRed]

@kite - I mean full text search using an index, so things like Windows Indexer, OS/X Spotlight, Linux Recoll.

AFAIK things like grep and findstr plough through the text when the command is issued, so on large datasets they tend to be slow... or very slow. I can search all my texts (books, scripts, email's etc) which is upwards of 600K items in under 30 seconds, if I limit the search to my book folders then 2-5 seconds.

The Index is on an SSD drive, but that's fairly recent, and I'm not convinced that it's significantly faster than it was when it was on a regular spinning-disk moving-head drive (other things are a lot faster - eg unpacking and packing zips and rars).

BR

[WizkidTX]

I would like to agree with kite. I believe adding information found in the Library of Congress's catalog would be very helpful. Here is a link to the book Dragonspell by Donica K. Paul http://catalog.loc.gov/vwebv/holding...bibId=13438666.

The LoC has very good info: such as Number of Pages, Edition Info, descriptions of included items (such as maps), ISBN number, Book Subject (could be tags), Form/Genre Description (could be tags), and heck it even has links to the Author Biographical Info, Sample Text, and Publisher's Description.

I would love to be able to rely on consistent phrased tags that make sense for the book in question.

How about you?

WizKidTX
San Antonio, TX

[kite]

There are many ways of looking at a personal library.
Some of these are;
a storage facility that allows reading the reading of books,
a workspace where you can read,
a recreational space for experiencing ideas,
and,
a place to explore the categorisation of ideas, knowledge and works containing such.

Many personal libraries would be engaged with in multiple ways. While Calibre doesn't yet offer a plugin that produces a holographic projection of a Victorian upper class private library it does recognise the importance of aesthetics in it's pretty new cover generator, Golden Compass steampunk styled name device and bookcase logo.

To me Calibre would feel more like my personal library (even though lacking high-backed, leather upolstered armchairs and a butler) if more metadata was available. I acknowledge that it is unclear how much use would be made of the extra metadata - this would vary from individual to individual. And may be used intensively only by the rare exception. However unless many people actually have DDC & LCC and other metadata we'll never find how useful it might be for the general user.

I have some small appreciation that "simple is better" as far as metadata goes.
But feel attracted to exploring what my library would be like with lots of meta data. And I could always jettison metadata that was making life too complicated.

I am not arguing that Kovid, or anyone else, should write the necessary code. Rather that I think many Calibre users would use such code if it was readily available so they could experience what more metadata would be like. And some people would be more satisfied with their Calibre library with extra metadata even if they hardly ever used the extra data as their sense of "librariness" would be better met.