DEVONthink for academic research - ADVICE

Readdle’s app is not going to be able to OCR handwriting, though. I imagine it will be a struggle to find an app that will do it. MyScript by Nebo on the iPad has a very good handwriting recognition engine, but as far as I know it is only for recognising text as you are writing it.

As to Liquid Text, I thought it was going to be wonderful, but ended up never using it because whatever you did was effectively trapped inside that application. Export was not good (though I suppose it might have improved since I last used it a coupld of years ago). The same applies to MarginNote. Nice in some ways, but difficult to export material to other programs.

1 Like

… just doing something very obvious here:

best! oliver

No need to tear them out if I don’t want to. I have a standing scanner like library would use, just really started playing around with it and trying to get through the massive heap of journals and books chocked full of annotations class notes, and any other paper item cluttering up my apartment. Less than stoked about the results of OCR, likely due to my chickenscratch, but I still need to double check the printed text as well to get a little better idea whether that’s all it is. I know DT had been having issues with OCR a while back too (basing this solely off of general forum chatter) , not sure if it’s been resolved.

Are you having OCR issues? I’m not noticing any OCR issues…

I am, though I wouldn’t say that they are necessarily related to DT, like I said probably has something to do with the quality of my chicken scratch handwriting. I’d want to go back and look at some of the stuff I’ve imported via my scanner too. The fact that it has it’s own OCR that it runs makes it difficult to parse out where the breakdown is coming from.

But the handwritten journals I compiled to archive for myself in Devonthink are completely unreadable, at least according to the OCR when I ran concordance using one of the previuos tips posted here to verify the accuracy of the OCR in DT, so I’m assuming that problems exsited at least on some level.

Oh. You you said “ I know DT had been having issues…” which didn’t seem to be the case.

??? I"m confused, thought I threw the caveat in there that observation was based solely off of general forum chatter since I hadn’t been able to figure out where the issue was originating from.

Maybe read/use the whole quote next time instead of cherry-picking it [shrug].

I’m fairly certain DT doesn’t do handwriting OCR. ABBYY (the OCR engine used by DT) does in fact seem to support it, but only in certain scenarios:

I might look into creating a script for handwriting OCR for DT if I can find some open source tools, but don’t hold your breath. :slight_smile:

This might be a possible script for taking snippet-notes. My fifth iteration of a note-taking script for reviewing literature.

This might be a possible script for tagging complex document. My fifth iteration of a tagging script for reviewing the literature.

+1 for you to indeed report back on this!!

Thanks for sharing! I have to say, I really like the idea of a monthly journal document. It sounds like a useful dumping ground for those sort-of-related-but-not-really-related thoughts and connections that pop up when writing. At the moment, for me, these either 1) get scribbled down on a Post-It and added to a pile of other Post-Its (which survives until it becomes overwhelming, at which point I scoop them into the bin!); or 2) take me down a rabbit hole so deep that whole days are lost reading about something that which may (or may not) be useful at some point in the future. Having a single document for these unfiltered thoughts, free of any particular structure seems so obvious. Of course, this only works if the document is reviewed regularly — which is why I like the idea of your monthly document. Do you schedule in a review at the end of each month to incorporate thoughts into a more permanent form/filing system?

This appears to have a correlate to GTD where you clear your head by writing down all that’s on your mind and review those items at a certain frequency. That leaves for more ‘space’ in your brain to analyze, as you’re not constantly keeping track of those ideas.

You might consider using a mind map structure to do that, and create one with groups in DT, Come to think of it, @BLUEFROG and @eboehnisch: would it be an idea to have the possibility to create a topology view of the groups in a DT database? You can use the classic tree view in the sidebar of course, but space might be limited to unfold them all and look at them at a glance. Like the related words, word cloud or frequency graph do with the data within a document, it would be a way to visualize the groups and their containing files.

In line with GTD I try and include a verb and ‘taskify’ those ideas, although of course the more abstract the idea gets, the more difficult this becomes. The concept is though, that those ideas become more actionable when reviewing them. You could include verbs like ‘assess’, ‘write’ or ‘search’ for example. In my experience jotting down some fuzzy concept is quick, but you might not understand after a month what you tried to tell your future self.

As to the method of the review you asked @spoiltvictorianchild about, I personally depend the frequency of it on the priority of the idea/task. You could for example tag an idea as ‘daily’, ‘weekly’, ‘monthly’, ‘quarterly’ or ‘yearly’ and have a smart group that displays those ideas in a corresponding group. So if an idea has high priority, you tag it ‘daily’. You then review the ‘daily’ smart group on a daily basis (and perhaps set a daily alarm for it), and you’ll run into that idea every day. Likewise, the ideas with a lower priority get a timeframe tag that is larger than ‘daily’ and you’ll not be bothered to assess those before the appropriate time. But of course you’re free to look over them at every moment you want to.

1 Like

So more like a mind map with the root in the middle?

1 Like

Yes, exactly.

Whether it’s feasible or necessary to display all containing individual files at the lowest level is something to research, but if your group structure grows over time it can be difficult to keep track of the structure just by looking at the tree view.

Definitely an interesting idea but hard to solve from a UI standpoint. We’ll keep it in mind.

Perhaps the ‘related words graph’ could be adapted to use groups and allow the spanning tree protocol to only use up/down relation of those groups. Clicking a bubble expands the groups attached like in the tree view.

You’d get a group view that might look something like this:

PPL              CPL
|                 |
Certification - Planned - ATPL - (many more groups)
Flying DB (root) — Navigation — Airport Charts
Flight plans - Archived - before 2020
|                 | 
Planned          2020

Oh, I’m glad to hear you found this useful! As I wrote above, I print it out at the end of each month and go through it, so that’s a built-in review. These days I’ve turned it into something of a “work journal,” where I am logging the hours I work, what I did, and so on. At least for this month, I’m actually not breaking off and creating a new monthly file, because I think that break is arbitrary and this way I can easily scan my work history in one place. I’m still transcribing voice notes and putting random things there, but these days I’m mostly writing things for my dissertation so my freewriting is happening in one of a couple of Scrivener documents related to that.

This is a good reminder to pull out ideas that don’t fit in that project and keep them somewhere, though. I didn’t really write this above but I’m trying to embrace redundancy as a principle, to just give me more chances to see something. As a result, some of the things that make it in to the journal are notes I’ve already written somewhere else – I’ve started to keep a collection of small standalone notes and quotes in DT, almost like a little blog for myself. There’s a lot of overlap between this stuff, and as I said I print out the journal once a month, lol. Like I said, embracing redundancy!

@anon6914418’s idea of tagging ideas with the frequency you should revisit them sounds helpful. I don’t think I can keep up with that level of organization, but I have a couple of tags like “possible epigraphs” or “would teach this.” I forget if I have one for “could be an article” or just “rabbit hole to chase down in the future” – I’ll definitely think of expanding my tagging system in that direction.

1 Like