Metadata Modifications?

[TheMadGuy]

Well, I am a totally newbie to this forum but I use calibre for years meanwhile and have archivated several thousand of ebooks. Before I regsitered I read several threads in here that some of my wishes are impossible to be fullfilled (like one book being part of two seperate series), but maybe the other ones could be interesting for someone else, too!

First of all: Why is it always named "author"? That possibly sounds like a stupid question, but there are several books where there's the name of the so called "author" but then in brackets "Publisher"
Would it be an idea to place a checkbox before the author-field, standard un-checked, which I could check, if it's not the author but the publisher credited?

The field with the tags could be filled endless. But in my opinion it might become much easier with two additional fields in the mask:
1. Type (Novel, Short Stories, Textbook, Biography....)
2. Genre (Crime, Fantasy, Horror)
As those "tags" are - or at least should be - part of the metadata of every book, it would make handling the database much easier. Especially as I could see much easier what terms are allready part of that list.
I am sure that most of you might agree that so often the tags are filled with several different (mis-)spellings of the same term and sometimes in that field there is half of a novel!

I know it's a lot I am asking for, but the tags could also be modified at least to reduce their different "versions". Is it an idea to make a tag one word automatically? That means: If I enter a tag and press space, then automatically a semikolon appears and makes sure that the next word is a new tag?
This idea started when I tried to organize the tags for the first time - and comes up again whenever I add new books. Different tags have to be separated with semikolons, and if it's only one word for a tag (no change possible), things like this can't happen anymore. That's just an example of tags:
Book 1: medicine thriller; novel
Book 2: medicine-thriller; novel
Book 3: medicine, thriller, novel
Book 4: medince; thrilller; novel
Book 5: medicinethriller; novel
.....

Finally I'd appreciate a field I'd like to call "subtitle"
It could (not sure if it will) make it easier to avoid or at least find duplicates. There are so many books with a subtitle on the front page. (This title now is just fiction) Title: Summer Rain - Subtitle: The Dream of the Geese
So one enters that "Title - Subtitle", the next one "Title: Subtitle" and possibly the third one "Title (Subtitle)"
The versatility of that possibilities so often makes it anywhere between difficult and nearly impossible to care about the data as perfect as possible.

Tell me your opinion about one or more of my ideas for modifications or how you handle the problems I have caring about my data.

Thanks for support!

[BetterRed]

@TheMadGuy - some of what you want can be achieved with custom columns e.g. here's my custom column definition for Subtitle



The Author and Title column names are 'nominal names', in my Journals library I use them for Journal Name and Issue Identifier respectively - e.g London Review of Books and Volume 38 Number 19.

I tag via Shift+F2 in the Book List, which pops up the Tags Editor dialogue, it's a more deliberative user interface, so there's less chance of creating spurious entries or assigning the wrong ones.

BR

[TheMadGuy]

Partially that's a nice idea. I thought about those custom fields several times and used them for a while, but whenver I get something new, there is a lot to modify. That was the reason for me to write this in the hope to implement those custom fields as standards

[Terisa de morgan]

I'll give my ideas about some of your comments:

Originally Posted by TheMadGuy

The field with the tags could be filled endless. But in my opinion it might become much easier with two additional fields in the mask:
1. Type (Novel, Short Stories, Textbook, Biography....)
2. Genre (Crime, Fantasy, Horror)
As those "tags" are - or at least should be - part of the metadata of every book, it would make handling the database much easier. Especially as I could see much easier what terms are allready part of that list.

If you like in that way, you can create two additional columns which meet your needs (I have already create my genre column, with no relationship to your definition).

Originally Posted by TheMadGuy

I am sure that most of you might agree that so often the tags are filled with several different (mis-)spellings of the same term and sometimes in that field there is half of a novel!


I know it's a lot I am asking for, but the tags could also be modified at least to reduce their different "versions". Is it an idea to make a tag one word automatically? That means: If I enter a tag and press space, then automatically a semikolon appears and makes sure that the next word is a new tag?
This idea started when I tried to organize the tags for the first time - and comes up again whenever I add new books. Different tags have to be separated with semikolons, and if it's only one word for a tag (no change possible), things like this can't happen anymore. That's just an example of tags:
Book 1: medicine thriller; novel
Book 2: medicine-thriller; novel
Book 3: medicine, thriller, novel
Book 4: medince; thrilller; novel
Book 5: medicinethriller; novel
.....

I have all my standard tags associated to a "fake" book, and I choose between them. So, I begin to type the tag and I choose between the offered words.

Originally Posted by TheMadGuy

Finally I'd appreciate a field I'd like to call "subtitle"
It could (not sure if it will) make it easier to avoid or at least find duplicates. There are so many books with a subtitle on the front page. (This title now is just fiction) Title: Summer Rain - Subtitle: The Dream of the Geese
So one enters that "Title - Subtitle", the next one "Title: Subtitle" and possibly the third one "Title (Subtitle)"
The versatility of that possibilities so often makes it anywhere between difficult and nearly impossible to care about the data as perfect as possible.

Tell me your opinion about one or more of my ideas for modifications or how you handle the problems I have caring about my data.

Thanks for support!

You can create a text custom column, named "subtitle", to include that information. It's quite easy, really. You won't be able to populate it from the metadata download, but I it's the only limitation I can think of.

[Terisa de morgan]

Originally Posted by TheMadGuy

Partially that's a nice idea. I thought about those custom fields several times and used them for a while, but whenver I get something new, there is a lot to modify. That was the reason for me to write this in the hope to implement those custom fields as standards

What's the benefit of being standard (subtitle, I'm thinking of) against custom?

[BetterRed]

Originally Posted by TheMadGuy

Partially that's a nice idea. I thought about those custom fields several times and used them for a while, but whenver I get something new, there is a lot to modify. That was the reason for me to write this in the hope to implement those custom fields as standards

A snowflake's chance, the database hasn't been changed for several years, if it were to change millions of users would be affected. A lot of thought went into making it extensible with custom columns

"...but whenever I get something new, there is a lot to modify."

Do you mean a new book or a new column? It's easy to go overboard with adding extra columns.

I prefer to edit through the book list, where I can order the fields to suit what I'm doing. If you prefer the dialogue boxes - try the different layouts in Preferences->Behaviour->Metadata Edit (single) layout options and have a look at Preferences->Tweaks->Order of custom column(s) in edit metadata

BR

[fidvo]

Originally Posted by BetterRed

I prefer to edit through the book list, where I can order the fields to suit what I'm doing. If you prefer the dialogue boxes - try the different layouts in Preferences->Behaviour->Metadata Edit (single) layout options and have a look at Preferences->Tweaks->Order of custom column(s) in edit metadata

BR

I can't believe I never knew this was there. File this one under "you learn something new every day."

This is going to save me a lot of time with my work flow.

[TheMadGuy]

Originally Posted by Terisa de morgan

What's the benefit of being standard (subtitle, I'm thinking of) against custom?

Well, my other big database is DVDProfiler.... Every user fills profiles and they are all contributed to a database. With creating standards (not to call it "rules") everyone fills the fields with the same system and everyone could profit from.

Just imagine if all the data you enter is sent to a database where everybody else could use what you have allready done and doesn't need to fill it again. I've been teached that it's not necessary to invent the wheel twice

[Terisa de morgan]

Originally Posted by TheMadGuy

Well, my other big database is DVDProfiler.... Every user fills profiles and they are all contributed to a database. With creating standards (not to call it "rules") everyone fills the fields with the same system and everyone could profit from.

Just imagine if all the data you enter is sent to a database where everybody else could use what you have allready done and doesn't need to fill it again. I've been teached that it's not necessary to invent the wheel twice

But your wheel is square for me, as I don't use that field. And I have that mentality... in my day job. For my personal use, I don't care about standards, mainly when it gives a false sense of security, because a template itself without clear rules for filling it (and everybody should agree to stick to that criteria) is of no meaning.

[Divingduck]

Originally Posted by TheMadGuy

With creating standards (not to call it "rules") everyone fills the fields with the same system and everyone could profit from.

A nice thought, but good beware me for standards in metadata like Amazon use to have. I spend a lot of time in editing metadata and my trust in swarm intelligence metadata isn't really the best - especially if I need to spend more time in cleaning up the "swarm mess" as it takes to do the job manually only once. There are some good sources available but none of them are in general good sources for all kind of publications and provided languages.
For the range of essential metadata I think, calibre already have more as of what most publisher want to publish in their books. Sure, I wonder too about things like not provided standard information for series and so on, but in calibre I can do everything I want to do. If I want to implement the complete Dublin Core Standard, there is nothing what prevent me to do it. (Well, maybe only my own mental fantasy in imagine of all possible potentialities what the program provides. )

[fidvo]

Originally Posted by TheMadGuy

Well, my other big database is DVDProfiler.... Every user fills profiles and they are all contributed to a database. With creating standards (not to call it "rules") everyone fills the fields with the same system and everyone could profit from.

Just imagine if all the data you enter is sent to a database where everybody else could use what you have allready done and doesn't need to fill it again. I've been teached that it's not necessary to invent the wheel twice

Fine, but including that field as standard in Calibre won't fill it in automatically. Calibre doesn't have an online repository of metadata. It downloads metadata from other sites that have no direct connection to Calibre. So even if it included the Subtitle field by default, that still wouldn't get you what you want. You would have to convince all the major ebook metadata sites to add that field to their databases as well.

And keep in mind that it's impossible to create a comprehensive set of fields that cover every possibility. You think a Subtitle field should be standard, but I have no need for that field. I think a Subseries field should be standard, but you may not have a need for that one. Ask everyone who uses Calibre what fields they think should be standard, and you'll probably have a list in the thousands, many of which overlap and duplicate functionality because people organize their books in different ways and always think their way is the "right" way.

Better to go with a minimum set of standard fields and let the users add extras as needed. Exactly the way Calibre already does it. Yes, the drawback is that you have to fill in these fields by hand instead of downloading them from an online source. But better that than the mess of trying to make everyone's favorite fields standard.

[BetterRed]

Apart from Series (which I don't use), calibre's other standard columns are a subset of those defined by IETF RFC 5013.

BR

[Fiat_Lux]

Originally Posted by Divingduck

If I want to implement the complete Dublin Core Standard, there is nothing what prevent me to do it.

For my Calibre library, I'd Ilike to have both CIP and Dublin Core Metadata Standard fields completely filled out. It just takes to long to fill out all of those fields, though.

I originally added CIP columns, when I found an extension that claimed to fill them in. Dublin Core Metadata was added, when I got tired of looking at tags that were irrelevant to the book.

Amber

[TheMadGuy]

Well, first of all I have to say "thank you" for all your feedback.

The reason why I thought about a "standard" (and I know that it IS possible) is the software I used to archive my DVD-Collection. It's named DVD-Profiler, and it's also used by thousand of people.

But as there is no database for calibre, filled by users, it's more than only difficult. Maybe anytime this will come where you can upload YOUR data when there is nothing so far. I hope that these days I am done with authors/titles/series, then the next step to handle is short descriptions of several thousand books - and bringing down the publishers to a shorter list (once again because of the several different spellings or ways they are written).

I am sure within the next 50 years a lot of things will change and might be "standardized" but I am also sure that I wont survive until that date ;-)

[Fiat_Lux]

Originally Posted by TheMadGuy

But as there is no database for calibre, filled by users,

There are services such as http://www.oclc.org/cataloging-subsc...dering.en.html that you can subscribe to.

Amber