4 ms·
Having recently picked up a Remarkable, I can empathize with some of these frustrations, but only some. Good things first: The writing and drawing experience i
by throwanem 5y ago
Having recently picked up a Remarkable, I can empathize with some of these frustrations, but only some.
Good things first: The writing and drawing experience is excellent. It's enough like paper and pencil that I've occasionally found myself trying to shade with my fingertip, as I would when drawing with graphite. The medium fineliner tool is perfect for notetaking, and the calligraphy pen tool does a solid enough job of feeling like a flex-nibbed pen that it's taught me some things about how to use one of those which I previously hadn't grasped. Battery life is ample, and palm rejection is if imperfect then at least tolerably okay.
The OCR is surprisingly good, if your handwriting doesn't suck. Mine doesn't, by dint of having spent a few years using fountain pens - we missed a trick dropping those in childhood instruction here in the US; they're a tool you just can't use long-term and not improve your hand. For fun yesterday I hand-wrote a trivial shell script, without really trying to write more neatly than usual, and fed it to the reMarkable's OCR to find it fell only two characters short of perfect accuracy. Especially given that, I can't really accept the claim that the OCR is bad - if you can't write clearly enough for humans to read, is it fair to expect a machine to do better?
Now the bad, of which there is regrettably quite a lot: The first-party software ecosystem is basically nonexistent, and the on-device capabilities are weirdly rudimentary. In paper notes I index by topic and page number, which is probably the simplest possible system of organization to maintain over such a corpus, and I expected thoughtfully built software on the reMarkable to do at least as well. But there isn't anything, so I'm faced with the need not only to come up with a new system that withstands freely rearrangeable pages whose numbers no longer can be relied upon, but also in so doing to work around the slowness and lack of tactility of the reMarkable as compared to paper. The page turning gestures are indeed not that reliable, and slow when they do work. Worse, the absence of even a basic bookmarking system forecloses any kind of rapid access to a specific page anywhere.
Even the few features that exist tend not to be completely implemented. For example, layers exist but aren't opaque; if you draw in white on a layer below lines in gray or black, the white appears on top. It's hard to even come up with a mental model for how this is implemented, and it violates the expectation created by every other graphical program that provides layers. They're still useful for protecting some content on a page from accidental erasure, but they're only useful for that. The lack of a "save as template" feature means that, to use eg the custom layout I've drawn up for a sleep diary, I have to either keep it as a blank page which I laboriously duplicate and move for each new page I need, or go through a convoluted flow of exporting, saving as PNG, and reuploading via the onboard SSH server. Too, if I want straight lines for that layout, I need a physical ruler to create them - which is fun for a little while, but would be preferably replaced by a straight line tool in the actual UI.
The hackability of the device is clearly excellent, and the community ecosystem quite rich. I can see where the folks at reMarkable could imagine this was enough, but it isn't. The device sells as being "like paper but better" right out of the box, but that's only true for people who have no idea how to get the most out of paper. But I do know how, and the device has at every turn frustrated my efforts to apply that knowledge to its use.
I have to say that I've found the reMarkable on balance mostly disappointing. I don't know whether they got so excited about the tech for itself that they forgot to put thought into making it pleasant to use, or if the tech cost so much to productize that they can't afford to build the software it deserves, but either way it comes to the same in my view.
I haven't yet decided to return it, but I also haven't decided not to. Maybe I can hack it into being the tool it really should have been out of the box, or turn it into a "newspaper" annotable RSS reader, or something.
In any case I expect before long I'll be back to taking regular notes with pen and paper, because the reMarkable has a very long way to go before it can provide a comparable level of utility.
- Syzygies 5y agoThank you. I have a career of scanned math research notes, in felt-tip color on 32 lb laser printer paper. Teaching through the pandemic forced me on to an iPad Pro and Apple Pencil. I quickly tired of the primitive Notability interface, and moved to Concepts. One rejects many art apps by testing whether handwriting is rendered naturally (or better), then most of the rest if one wants scalable vector art rather than bit maps. Concepts has the most advanced algorithmic support of those left, as far as I can tell. I certainly have complaints, but in some ways their layers are smarter than Adobe Illustrator layers. For instance one can group art without moving it all to one layer. This is as big a deal as lexical scope in programming languages. I yearn for the calm of a separate device just for pencil notes, but I fear this is like buying a wind surfing board. A week after learning to balance on this board, I'll want a trickier board. Algorithmic drawing is still at the Dartmouth BASIC stage, and the future is exciting. One imagines future drawing platforms with the self-referential power of Lisp. One doesn't imagine moving backwards.