No lesson link!

I'm late to the introduction party. I've been here since mid-June when I discovered and signed up for Tana One. I first learned of Tana from the Ness Labs newsletter written by its founder, Anne-Laure Le Cunff, in late 2022. I got Tana access at the end of that December by signing up for @Santi Younger's first Tana class. Prior to Tana I used hand-written notes and DEVONthink part-time. I eventually tried DEVONthink full-time, but found it too compartmentalized for my needs. I then went through Obsidian and Logseq to end up with Tana. Tana's model comes closer to my own mental models of information organization than any other tool I've used or that I've looked at.

@Nick Silhacek There must be one or two things that I'm missing here. You have a node (or a field) as the target of Transcribe audio. I'm not familiar with the command's behavior, but it seems to me that the process you've shown still creates two independent transcripts, since you have two independent executions of the command. So how does editing the transcript in either the node or the field get reflected *anywhere* else? Also, with two independent Transcribe audio commands are you even guaranteed that the two transcripts are identical? Finally, why wouldn't making one transcript and then issuing the command 'Move and leave reference' work? Right now I can't take the time to try to duplicate your work and test it myself, but does any of what I said make sense?

@Nick Silhacek I found this FAQ answer at the bottom of the "AI command nodes" in Tana Documentation. Might be useful, but I didn't try to play around with it yet.

After posing my questions, I tried to answer Q1 by testing. My results seem to indicate that, as is too often the case with GPT, it gets its result from wherever the hell it feels like. Some appear to come directly from the URL; some appear to come from the <meta name="description" ... /> field in the webpage header; and I got one good hallucination.

@Nick Silhacek Good to know about the tooltip. I was not aware of that. Thanks.

@Nick Silhacek I confess I had already thought of a couple of these, but I was curious what you'd have to say. I'm not sure yet which approach I want to use. Build title from fields is ugly, because probably 99+% of the time the extra field is useless. To help show when this is actually needed, I wish Tana would give a warning whenever you tag something and the name already exists associated with another unrelated tag (not part of the same inheritance hierarchy).

@Tana Central It turns out that I will be able to attend Q&A on July 1st. Again, happy to discuss my findings if people are interested.