I sometimes get the sense that most PDF annotation products are built from the perspective of the PDF – rather than from the perspective of a user, annotating that PDF.
By way is example, 99% of them see too many taps to change highlight colours, or tools, or don’t treat a single highlight as a separate event from other highlights on the same page – or don’t easily present a way to remember most used colours or tools.
I suspect the point above is why so many of them look virtually the same. PDF front and centre, some form of toolbar on the side(s), or appearing at cursor point, with the standard fare of options in terms of interacting with the PDF. I’d love to see something examining the most typical requirements/approaches regarding annotating (doctor, researcher, teacher, academic, lawyer, student etc.), and having profiles and toolsets presented accordingly… in other words, switching up the paradigm, and starting with the user.
Very astute take on the purpose of these apps, in my opinion. Also, thanks for posting the link to that blog post — I always enjoy reading about reading.
Yes, thanks for that reference, @ygjose. It’s an excellent framework for this kind of “processing written ideas” work.
I agree with your take that LiquidText fits “level 4” the most. However, I think that’s because it does more than help you read. It’s about excerption, ordering, memo-writing, etc… Grounded Theory sorta stuff. In that sense, there are also a lot of other tools that fill in that gap, from index cards to Scrivener and Ulysses’s notes and reference material features to Curio, Tinderbox, all the way to Zettelkasten-type systems and Roam-like systems.
Since everything’s a stack, it might be a good idea for knowledge workers to conceptualize their writing tools as a stack, too. The framework summarized on Farnam Street is a decent template for what “full stack” reading looks like. So, we’re challenged to find the tools and processes that fit us the best on each of those levels.
Perhaps @BradG PDF annotation seems incomplete or “from the perspective of the PDF” because for many of us the annotation process is step on a road going elsewhere – leading outside of the annotation process to a book, or an article, or private essay consolidating excerpts and notes. PDFs are flat and linear, and annotations are, in essence, snippets or a reduction of the overall PDF. Personally, when I can get those snippets / notes out of the PDF and begin to work with them in relation to other things I’m reading, or other media, that the notes enter a broader context. Then, the annotation process broadens out and informs knowledge. Escapes the software, so to speak, to become something better.
I completely agree with you! And yes, I think LiquidText does more than comparisons. Although I haven’t tried it personally, from what I saw in the video tutorials provided by the website, it seems that LiquidText can do memo-writing and stuff like that.
I’m glad you mentioned Grounded Theory, which is one of the key theories in qualitative research. I don’t know if someone has ever talked about the similar thing before: I think when we do reading, especially scientific reading, we are actually doing something pretty similar to Grounded Theory. That is, at the first step, we collect relevant papers/data; then we constantly compare concepts from the papers/data; we sensitize concepts, write memos, reflect, etc. until our own provisional theory is saturated; finally, the ultimate theory is established.
Since doing scientific reading and notetaking is a qualitative process, I think a computer-assisted qualitative data analysis software (CAQDAS) would reach at its finest. If you remember what I posted in another thread, I recommended a QDA software called MAXQDA. This software certainly fits the fourth level reading. That’s why I’m using it for my whole literature reading, quotation, ordering, memo-writing, etc.
I consider LiquidText and MarginNote to be comparable in that they both offer the ability to play with the annotations directly within the app itself. Highlights is not in this same league. In my testing on an iPad Pro with an Apple Pencil 2, I found that LiquidText does not consistently import annotations from other PDF annotation apps and MarginNote does not reliably sync. I preferred MarginNote because I could tag annotations, I could write hand notes to annotations, and hand-written annotations were kept as one group.
In truth, I never took the time to play much with the annotations in LiquidText or MarginNote (e.g. mind mapping). But from my first tests, LiquidText seems to have the cleaner UI setup to start the process while MarginNote seems to have the deeper toolset to analyze and compare annotations (e.g. flashcards).
At the end of the day, I realized that consistency and reliability are more important that the ability to play with annotations inside a PDF annotation app. Since MarginNote could not export its annotations (the PDF is flattened), I went back to a stable PDF editor (PDF Expert) and hope to explore the options to sync with external apps some day soon.
That’s actually where I ended up, too. I went back to PDF expert because it worked well and consistently. It is easy to import and export files, which makes it easy to interoperate with other apps (and people).
Any reason for using MaxQDA and not Atlas.ti?
I am toying around with the two of them, but one thing that frustrates me is that I can’t simply index my PDFs folder (that I keep and organize in DEVONthink); instead, I have to move them elsewhere (and keep them in an unknown format, in the case of Atlas.ti).
I’m using MAXQDA just because it was the first QDA software I knew. After that, I knew a bunch of other QDA softwares. I would like to try all of them, but it’s going to take me some time and also money to be familiar with them. So I just stick to my “beautiful” first impression.
I know what you mean. You mean you can’t create subfolders under the parent folders? I encounter this problem as well. But it doesn’t bother me too much. As long as my code system is well organized, which doc is in which folder isn’t that important for me.
Actually, my files are all in a single folder as I like the flat structure and use tags instead of folders to organize them. I meant that I cannot control this one folder where they all are since theses softwares import the file to a database elsewhere and change their format. But I just noticed that Atlas.ti does this, but with MaxQDA I can actually have them all in one indexed folder that I can manage with DEVONthink as well. Nice
LiquidText is about to see an update, according to their Twitter page and they are working on a macOS and Windows app, which also soon to be released. I’m not sure how this will affect their pricing structure.
Personally, MarginNote 3 has the best feature set on paper, but the interface is obstructing me to use it.
I don’t disagree with that in the slightest – but at times, when spending several days reviewing PDFs, one can (depending on the task/environment/context) spend many continuous hours inside the PDF annotation process – and it is here that my comments were directed at. I still feel that most of them are too similar, simply since their point of departure is focusing on the PDF, rather than the process.
So, whilst it remains a mere link in the chain, I still wish that the apps within this space (or a new one – such as seemingly being designed by @msteffens) would give some thought to who would most frequently use their app, and what their workflows would entail/require.
I acknowledge that it is almost impossible to please everyone – but then, that’s what the they all(?) end up doing – targeting the middle of the road, and not catering for use-cases… Which is possibly another reason why so many look/function the same…
I cannot but help thinking that If the market is already rich with PDF annotators that seemingly cover the ‘general’ approach, and do it well – then, in the absence of changing the emphasis, why would another one bring anything different?
I actually tried to build my app around a typical research workflow. Before getting started with my app, I’ve talked to many researchers, and read many posts where researchers described their own workflows. And I could draw from my own experience during my time in academics.
As a result, I’ve come up with a list of principles that I consider crucial. I will have to write a longer post to go into more detail. This is just a short summary:
Free annotations from the PDF & work with them: As @anon41602260 mentioned, annotations get only useful when they are extracted from the PDF. So my app gets them out of the PDF immediately, and you then interact with the resulting plaintext notes.
Notes are plaintext: This ensures maximum longevity, avoids lock-in, and allows for easy integration with other apps. The plaintext notes also give you a good data exit strategy.
Notes are atomic: One note per thought. This allows you to tag to the bullet point level, and it enables you to freely filter, rearrange and reuse them.
Notes are self-contained: Notes should be as self-contained as possible. I.e., the note’s body text (or its metadata) should contain all relevant information (like its tags, creation date, linking & citation info, related file info, and of course the Zettel ID). This way, you can do lots of things with your notes but don’t loose anything. All relevant info always travels with the note. E.g., you could transfer your notes somewhere where there’s no file path info, like a database, a single text file, or an app that stores files internally using its own naming scheme. Your notes would still remain intact and can be retrieved again w/o loss.
Notes keep their context: This means one should be able to easily see the context of one’s note. I.e., easily jump to the quoted text in the PDF source, and back again. Also, always keep the reference info in the note. Ideally, the reference info would be extracted automatically.
Notes can be inter-linked: Directly linking notes to other notes is crucial since this actually captures your thought trains (tagging does not). Also, by creating a network of semantically rich and linked notes, your note relationships can be visualized. The visualization graph also shows the note’s backlinks, and it can even be used to navigate your notes.
Notes are human- and machine-readable: Both are important. The structured plaintext note is your data format. Having it not only human readable but also machine readable allows to build a data model from it, which in turn allows for all kinds of advanced features.
Notes can be reused in other applications: It’s very important that the app (your notes) can integrate with other apps, via import/export but Really also directly via scripting/automation.
Each of these points could be a post by itself. I had to type this quickly, so apologies for being somewhat terse. But my point is that I tried to build the app around these principles. And these were deducted from analyzing other researcher’s workflows.
I started using Highlights recently because it’s app is everywhere, simple, and clean.
The macOS app is a little frustrating because you can’t move pages of a document around or pull a page out of the document and save it separately. When you try to pull a page out form the side bar it creates a copy of that page in the document, which you are unable to delete. I contacted them about a month ago and they said they would “look into it.”
I think your effort is comprehensive and interesting – looking forward to see it develop. You might want to ask the admins to move your post to a separate thread so it doesn’t get buried and forgotten in this thread?
Please make zettel-anything entirely optional and configurable. The whole Z-mania will die off in a few months (as it usually does) so optional will be nice.
Ambitious and difficult, so how you pull this off will be interesting.
I am not sure how you can accomplish plain text portability (assuming it is based on some commonly used markdown library) at the same time as accomplishing the two requirements I quoted above. This would indeed be very useful.
I mentioned my own, perhaps niche, use case of needing to have a set of notes that cover multiple PDF documents (many to many) – which MarginNote does quite nicely, and LiquidText somewhat less well. Looking forward to seeing how to accomplish this sort of meshing with your solution.
Keep us up to speed as your work develops, please.
To echo @Ben_Wah comment, editing (deleting, moving, extracting subsets of) document PDF pages is a great feature. When it gets down to business and reference, having a shorter or reorganized PDF can make it far more efficient, e.g., getting rid of cruft, illustrations, chapters, or just keeping a key page or pages from a doc. I.e., make the document one’s own for the task at hand.
Also, on the source citations (linking). Highlights does this well, but superficially. Highlights (pleasantly) surprised me with links in the Notes to the source documents by default. Awesome. Unfortunately, the links do not inherit to subnotes that I split. So the source reference gets lost. Ouch. I would suggest by default (?) having the source document inherit to subnotes, even after edits. That way the source material, inspiration, reference remains handy.
Finally, I am glad you are focused on workflows. I believe the PDF programmers don’t eat their own dog food, get lost in the weeds as they program. For the bigger companies, the feature list is probably all about marketing (new sales, laundry list), not real workload realities, and the programmers 1) don’t know any better (are not part of the target market), and 2) just do what they’re told. So if you can stay on target and raise the bar, that would be great!
@occam Yes to me “Killer App” akin to App Killer such as a specific app kills every other app with the same functionality. For example, I personally think Excel is a Killer Spreadsheet app.
In this category, say Note Taking apps, there is a lot of different UI’s being tried out that segments the market and still has different feature sets that align with different user requirements.
I am still waiting for decades for the Killer email app. that’s a game-changer
Thanks. To avoid hijacking this thread any further, I’ve opened a dedicated thread where I’ve tried to describe the aspects of this app in more detail:
Your app(Keypoints) sounds really promising. I am also using these apps only for academic reading. Automation of the edit of highlights and availability to export bullet by bullet would be crucial for me, which I prefer highlights app for the latter.
Interestingly, the creator of Highlights app is also a researcher like you. Which rises some concerns about longevity of the app. And the improvement and new features down the road. Good luck!!
