Help w/ Metadata

[mmobes]

I’m still relatively new to Calibre and I’m struggling with the metadata for my books. Not sure if there are some best practices or if anyone can suggest what I am doing wrong?

I just imported the entire Expanse collection. None of them were labeled as a series and most of them were missing descriptions. So, I went in and edited metadata, went to download and then found that depending on the book, it gave me a couple options. Sometimes they included what book in the series, sometimes they included new covers, and sometimes it included none of that.

Is there a way to make sure it always pulls the metadata from, say, Amazon itself? I tried pasting the ASIN from Amazon and then seeing what it pulls but it didn’t improve anything.

Is there a consensus best practice for getting accurate, complete Metadata information or is this just going to be a struggle?

[Karellen]

It's not answering your question, but just throwing this out there.

I never use the metadata scrapers. I just type the info in myself manually. Why? A few reasons...
1. I don't always agree with the series numbering
2. I don't like 25 tags/genres per book.
3. I want first published date, not some random publication date of that particular release.

For the Series issue, install the Manage Series plugin and you can quickly organise the series in bulk. Go To Preferences>Plugins and search Manage Series (or maybe it is installed by default, I don't remember)

[Sirtel]

Originally Posted by Karellen

It's not answering your question, but just throwing this out there.

I never use the metadata scrapers. I just type the info in myself manually. Why? A few reasons...

Same here. I often change overlong title and series names, so no point in downloading.

[davidfor]

Originally Posted by mmobes

I’m still relatively new to Calibre and I’m struggling with the metadata for my books. Not sure if there are some best practices or if anyone can suggest what I am doing wrong?

I just imported the entire Expanse collection. None of them were labeled as a series and most of them were missing descriptions. So, I went in and edited metadata, went to download and then found that depending on the book, it gave me a couple options. Sometimes they included what book in the series, sometimes they included new covers, and sometimes it included none of that.

Is there a way to make sure it always pulls the metadata from, say, Amazon itself? I tried pasting the ASIN from Amazon and then seeing what it pulls but it didn’t improve anything.

Is there a consensus best practice for getting accurate, complete Metadata information or is this just going to be a struggle?

The metadata download uses Metadata Source plugins to scrape various web sites looking for the book. When you use the metadata download, it uses the sources you have installed and enabled to search each site and parse the results. Each plugin does this and returns it to calibre. Calibre then merges the results it they fit together, such as having the same ISBN. Then the results are displayed for you to view and select.

The issue with this is that the quality of the metadata you get depends on the quality of the metadata available. If you don't like the results from a particular site, then disable it. If you just want the results from Amazon, disable everything except for Amazon. But, if Amazon does not sell the book, it won't be found. Also, while calibre includes several good Metadata Source plugins, there are other available to install. These might give better results. That will depend on the site and the books you have.

As you mention series specifically, that is a pain. Not all sources show series info separately. Some include the series in the title and it has to be parsed out. And for some book record, it is just not there. For myself, I do the search, look at the results and choose the one I like the most. Sometimes, if a single source does not give me what I want (one has a good description, another has the rest of the details), I select one then repeat the search and choose a different result. That effectively merges them for me, but, I might have to use undo on one of the fields.

What @Karellen and @Sirtel does sounds like to much hard work to me, all that typing. The metadata sources have most of the information in reasonable format so I use it. The quality of the metadata has improved over the years. I think my main issue is where they include press claims in the comments or what is really a subtitle in the title. For tags, I just accept whatever comes, but, I've realised recently that I don't actually use them.

[theducks]

Automatic Metadata is a Cr*p shoot.
Configure the tool to your tastes:
In the MDE window, there is a Configure button at the bottom to Start with.
This is you 'In General" settings ((look at the top-right side for fields (AKA columns) )
The left side has POSSIBLE sources. remove ticks from less productive sites to save time.

Note the button directly blow this area. That is used to fine tune the selection even more. (what it allows, depends upon the metadata source PI involved. (this is just a shortcut to Preferences:Plugins:metadata source: <select pi>: Configure PI (button)