I’d like to be able to use an ID that is in the format of %Y%M%D%h%m%s so for example, it would look like 20210217085902. Maybe this is already possible and I’m just not understanding how?
I see that it’s easy to insert dates and times but the format is pre-determined in a way that I can’t use for this. Having an option to make the ID for a page or block this way would enable me to more easily combine my usage of Logseq with other applications I like, such as Zettlr. It also makes the ID meaningful for a human to read.
Even if the IDs couldn’t be set to use a format like this, having a simple way to insert that somewhere in a page would be really useful.
Thanks for considering (or suggesting alternate methods).
Thanks for that link–I had not found that before. It’s a bit different but perhaps similar enough that it would solve what I wanted to do. I’ll keep my eyes open for updates.
I just realized something nice that is sort of related to this.
I can make Logseq and Zettlr interoperate a bit by including id: YYMMDDhhmmss as a YAML property.
Although I need to enter it manually in Logseq, if I move my Logseq folder so that it sits within a Zettlr workspace, then Zettlr will automatically recognize those IDs allowing me to link nicely to a note made in Logseq. Zettlr also recognizes the tags from Logseq and includes them in its cloud and searches this way.
The reverse is not true of course. If I create something in Zettlr it won’t automatically be findable in Logseq but at the moment this works well for the way I want to work. I like using Logseq for an unstructured flow of ideas, note-taking, etc. I like using Zettlr to expand on ideas more and for its hierarchical functionality. So normally, I’d begin things in Logseq but might then develop them more in Zettlr.
I would like to also endorse this option. I explained my reasoning to @tienson in another channel, and was asked to reproduce it here:
UID in file name is future proof. If you move away from org, you still have a unique identifier to search for. Also a UID gives you a time stamp that is human readable (an org ID is a hash that tells you nothing useful), and has the other advantage that sorting alphabetical in the file system gives you an order of creation even if the file system time stamp changes (like moving between file systems sometimes does).
As an example, I have a bunch of links that are broken in org because I moved a lot between org and md. But I can always do a search in deft for a UID and find the link target.
To that, I will also add: I suggest that if this is implemented, there be the option to either use YYYYMMDDmmss as @owlyph suggests, or YYYYMMDDmm, which is an equally (if not slightly more) popular method amongst ZK users.