I'm making a new tool for writers. With it, you'll be able to write your essays on "layers"
The problem? Tweets are easier to read than long-form essays, as they require less time commitment. If the content is not good on a long-form article, you'll find out way too late. With this tool I'm developing:
Layer 1 is the shortest version of your essay, the 1 min read — like a tweet. The idea boiled down to the shortest version
Layer 2 is the same text from layer 1, but with extras added here and there. What's already read by you is in black ink. What's new is in blue ink. This is the 2 min read version
Layer 3 shows everything from Layer 1 and 2 in black ink, but what's new is now in blue ink. and you keep doing that until you get to the full version.
I can post some screenshots here of my mockups, as I'm a designer. PM me if you find this intriguing!
The first thing that you see is the first layer (1 min version). Go right for 3 and 5 min version!
——
Edit 2: since I'm seeing the upvotes and the emails, I quickly made this sign-up form for the people who want to be updated when the product is done: https://layered-ink.webflow.io/
I would put up the https://layered.ink link but the domain hasn't been propagated yet.
@Admins — please do let me know if this is not permitted so I can take it down. Apologies if so.
Can anyone vertically scan 5 paragraphs of a long-form article in a 1-2 seconds and ambiently detect keywords that signal relevance? By ambient, I mean in your peripheral vision, without even knowing which word you're looking for or actively reading anything.
I acquired this "skill" about a year ago and now I use it when I suspect there may be filler text or introductions to concepts I already understand. I know it works because I uncannily land on interesting but otherwise nondescript passages. When I scroll back up I find that I did indeed skip the filler. Of course, this isn't voodoo or unprecedented. It's just funny that I've read so much that my brain basically has a regression model for various semantic characteristics that signal novelty by subject area. I suspect I also switch into different modes of scanning based on the writing style detected.
Because of all that, I actually prefer no layering at all. Reuters articles are Layer 1 compressed and I find them annoyingly curt.
Maybe, but I never did this before I tripled my reading volume while attending grad school. I thought skimming entailed some degree of horizontal scanning and actively reading clusters of words. In contrast, I don't even read clusters of 2 words when I do this and I don't aim to comprehend the skipped passages. I just automatically detect that they're irrelevant with basically zero reading.
Experiences may vary, but I've always thought of skimming as something that could be done to various degrees. Some examples:
1. going through pages looking for interesting textual structure, for example dialog, numbers, capitalized words, etc.
2. scanning vertically or in a zig-zag fashion, using peripheral vision to look for interesting words
3. reading the first few words of a paragraph before deciding if it's interesting or not.
I normally use 2, then try 3 on a paragraph that passes 2's test. Only when a paragraph passes test 2 and 3, do I decide to read it with my (temporary) full attention. 1's more for when I'm searching for something in particular or when what I'm reading follows a specific format and I'm interested in a particular section with distinctive structure.
Wow. That is next-level stuff. This makes me think of how varied the human experience is, even for the most routine activities. I've heard the word "skimming" hundreds or maybe thousands of times before, but clearly it is an inaccurate abstraction of the range of things someone can mean when they say "skimming."
Here is a dictionary definition: "The action of reading something quickly so as to note only the important points." If you showed me that definition before I started skimming effectively, I would have never extrapolated the behavior you described from that word. I didn't even know people did that.
-----------Idea ----------
I could see value in an open source "Verbose Dictionary" where people from all ages and walks of life would be able to add their own definition of a word. There would be two general rules when someone makes an entry:
1. They would need to write their definition in a verbose manner. I'm thinking at least three sentences, usually more, but also not as long as a Wikipedia entry.
2. They would need to be as open and honest as possible, so that there's minimal translation loss between what they're communicating and what we take away from it. Importantly, the word needs to be described with both intellectual and emotional cues. The author would also be able to let the reader know basic facts about themselves, and there would be some mitigations against trolling or fakers.
Over time, we would converge on a more universal language to maximize our mutual understanding of what each of us truly feel and think about words, concepts, current events, people, etc.... There would also be optional tags and slider scales (i.e. 1-10) that can attach to each definition so that you can correctly communicate the breadth, depth, and magnitude of your thoughts and feelings on the word. Critically, the goal would not be to achieve groupthink and converge on the same definition. That would be contrary to the premise of the Verbose Dictionary.
Not only would I find such a dictionary highly captivating, but I also think that it can serve as a useful tool for conceptual mapping or things like artificial neural networks. What are your thoughts? Would you use the Verbose Dictionary?
It would have a different aim than Wikipedia. Wikipedia, is, above all, a source of knowledge. That's where I go when I want to chain-read about the German Revolution of 1918. It doesn’t have the emotional and sociological theme I’m describing.
VD on the other hand would be a real-time, K-clustered map of the human experience. If you go to the entry for Love, you’d see people from all over the world sharing what loves mean to them, anecdotes and all. Crowdsourced feedback would surface the best entries to the top. Political topics would solicit good-faithed micro-blogs with the specific aim of humanizing each other and learning why we believe in what we do. You'd be able to go back in time to see where people's minds and hearts were at on a certain day. Algorithms and strict moderation would ensure that diverse viewpoints are shared and treated in good faith. Maybe I’m wrong, but I think if there’s a dictionary for swearwords, there’s certainly room for a dictionary of the human experience.
Check out the below study published in Nature in 2019. It makes the claim that cultural values of openness are what give rise to democracy, rather than the other way around. We need more of that openness right now. https://www.nature.com/articles/s41562-019-0769-1
The purpose of a dictionary is to define words. A word with a thousand definitions is not defined. It is the opposite of defined. If a word has several diverging meanings, perhaps it's time to create more words.
Super valid point, especially if you're doing what keenmaster does (wish I could tag him). Frankly, I've got FOMO when reading long stuff that interests me
Fun. Quick user experience report: expanded "atoms"; the effort to find the boundaries of the insert was a thing ("where does it start? ah, next sentence.", "oh, I read this bit already, so I'm past the end"); with an unmoved mouse, I tried clicking "atoms" again to toggle the insert away (didn't). Perhaps color the insert?
Thanks for the feedback, there should be a fade-in when the text expands, and there's an option to click again to close, but IIRC it's not enabled on the demo. I think it's unnecessary, but it bothers some people to not be able to close them back up.
Fantastic Stavros, I love it. Really close concepts we've got here. I thought of exactly what you've made a few weeks ago as... some words may or may not be jargon on a certain "layer" of my essay format. So what I barely-envisioned was what you put into code here. Will bookmark it!
Yep! I was thinking of your layers idea while making this, but I wanted something that would allow people to drill down in a more targeted way (rather than generally). You're right that it usually ends up being used to define terms, but you can also use it well for elaboration.
I think the ideal solution will be some sort of middle ground, where much of the drilling down will be done by clicking on specific words or sentences, but a slider would allow you to generally expand more text in the article, for things that aren't exactly related to other sentences.
It does take a bit of doing, and drilling down to another layer should also be able to delete text, so the wording flows better.
If you want to talk about this more, feel free to send me an email, I find this problem very interesting.
I would love if it was possible to further drill down into topics (expand further in already expanded parts, up to many levels deep). I long ago hoped to one day try and write an explanation of Shor's algorithm in such a format, with many layers of drill-down available, to cater to readers of different level of knowledge. And also with possibility to collapse back the areas, and bookmark a specific expanded state of the text.
Is collapsing back also allowed, as I asked above? :) didn't see any clear UX indication to that effect :( Not 100% essential (I guess could try adding it as a contribution if the others were there), but how about the bookmarking aspect I also mentioned?
Neat. This layered approach reminds me of the Youtube series where they explain concepts such as Quantum Physics at 5 different levels of difficulty (children, highschool/uni students, phds, field experts).
- Maybe limit the layers to 2 or 3. The example you've got in Invision has eight layers which gives me decision paralysis. Do I want the "7 minute read" or the "9 minute read"?
- Consider an onramp from Twitter. People have big audiences there, and linking out to Layered Ink might work better than those 50 tweet "tweet streams".
1. Hmm, the concept is: the more layers, the better. This way, you can easily "pick your pace" — if you're short on time and you want easily-consumable content, you'd skim through 1-min versions. Ideally if at a certain point you say "holy shit, I love this", you'd jump to the full version. The same way you don't just skim your favourite author's new article — you just go to the full version because he's proven time and time again that his style fits you
2. Oh? Tell me more? Did I get you right: showing it to ppl who bypass the 280-char-limit by having tweet storms/threads?
P.S: I love Dynasty's logo!!!
Regarding (1), I think the parent still has a point. Too many choices can be stressful to the user.
How do you expect users to interact or choose between "7 minute version" and "9 minute version"? Are they expected to keep going up the layers one-by-one until the full version is reached? Are they expected to intuitively know which one to pick right away?
Too many possible choices also delegates the responsibility of picking the number of layers to the writer, which can be an additional stress.
If possible, I'd try to figure out the most common use cases. The interactions and thought processes of users should be intuitive and dead simple to describe. Other than, loving the idea! Keep it up. :)
So you think people would open another website and then go through the "expand" rather than just do it on the website they are currently on? I don't think so to be honest.
What you propose as the system for writing makes sense and probably reflects what many people already do.
However, I am curious to know what the tool does. In other words, someone can write a “thesis statement”, and the bullet points and an outline and so on in any editor. What does the tool do?
For what it's worth, a lot of good software doesn't actually "do" many new things, other than support a specific use case with a better UX than existed previously :)
It'll be a writing tool. A text editor. There are two ways to approach it:
1. Let people brain dump — write everything they've got on their mind. Click "new layer". Cut down on words. Click "new layer". Cut down on words. Repeat ad nauseam.
Every time you write a new layer, you're "editing" your article so as to make it shorter. In other words, you're purifying the idea
2. Let people write the shortest version, then write a bit more, then more.
I'm doing way #1 because I feel like road #2 prevents people from having a brain dump... I'd lose all my ideas if I'd start with the short verison
Mine as well. And the process is often lossy, so I'll keep old variants easily at hand. Eg, "this old variant had nice property X, or dealt with Y well, or suggests a direction worth exploring".
Brainstorming, well, "everything is a graph", but perhaps picture versions as columns, with each column showing all three levels, with transclusional editing between them. So you do a pass in whichever direction, create a new column, and easily grab material from previous variants, and repeat.
Very np, sorry... I was picturing a screen divided into columns. Each column is a stack of the various distillation levels, say most brief on top, then less brief, and so on.
Within a column, editing text in any level, updates that text in any other level it appears in. That's the transclusion. Since all levels are said to have a copy of the most-brief text, if you edit that text at any level, it's updated at all levels of that column.
When you start on your next column/version, you can easily grab text from previous version(s). And you might do that in different styles. If you copy the entire column, then the old column serves as a checkpoint, and the new for continued editing. Or copy just the most-distilled level, and then work downward, to "reconstitute in a more crystalline form". Or copy just the least-distilled level, and work upward, attempting a new distillation.[1] Or do these from a previous column instead of the most recent one. And the old columns/versions are easily accessible, to browse for inspiration, or to grab stuff from.
[1] Hmm... I guess copying just the least-distilled text "forgets" what the embedded more-distilled text regions of it were previously? So one can try playing with the "brain dump" text again, without the distraction of the previous distillation choices. Then when things settle down again, one can highlight "this region is now distillation level n".
For UI, perhaps highlight-drag-n-drop to more distilled levels? And if each level say had an "X" - delete and forget this level, then the different styles of column copying unify as "fully copy a column, and then X-away the levels/region-choices you want to discard"?
So... All levels simultaneously visible, and easily edited together. And multiple versions of them easily accessible, in support of doing multiple exploratory passes, with each pass able to easily draw from past attempts. And working in either direction: towards distillation, or building out from a more distilled seed.
That was the brainstormy vision anyway. I've no idea if it would work out. And for whom - I work in both directions (back and forth), and like a visual record as external memory (so I don't have to remember things or take notes), and like to just see it all (without view switching). Someone who doesn't want those, might find multiple columns to be distracting clutter.
Thanks. The bottom-up journaling style of use [1] looks tempting, even as someone who generally prefer graphs to trees.
PayWhatYouWant[2], but it seems a pity one can't easily play with it non-persistently, without giving them an email? I wonder if that's the right choice, funnel wise, for something with a "I can't imagine using it... oh, that's neat" dynamic.
This sounds fantastic and is something I've broadly thought about for a little while. A little while back someone else linked a concept of what they had done with the idea -- initially a paragraph was visible, and some of the words had a coloured box around them which could be clicked to expand on that term, all the while maintaining a flowing prose. So, I think a slightly different goal to what you describe (it was technical writing I think?), but similar enough. Unfortunately I didn't save it and have previously (frustratingly) spent at least an hour looking for it to no avail, so I'm curious to see what you come up with!
If you ever find it again, send it my way please! Indeed it would be very useful for complex concepts explanations. Definitely not do tabloids or anything like that — though it might... help?
It's essentially the reverse way of what writers do when they write the "tweet version" of their essay — plus all the steps in-between.
@ your last sentence: Check out the InVision link I've added to the OP!
Reminds of https://getcoleman.com/ and this other portfolio site where each word could be expanded until it became a large essay, can’t remember its name.
Actually, I think news articles don't have clear-cut layers, but actually gradually dive deeper.
The original motivation were layout considerations for printed newspapers with limited space. With all articles pre-written like that, an editor could arrange them on the final layout and cut off at the end at discretion while keeping the most relevant information given the remaining space.
That being said, I don't think this is too relevant for your idea, just an interesting anecdote.
I didn't but I just checked it out now. Interesting I'd say. They have the short-version and the max-version. The tool I'd like to make would allow you to do all the in-between versions
I like this idea! I'm building something similar at www.sivv.io. This is a community for knowledge sharing where users share / discover summaries of useful knowledge, ideas or advice that are structured into sections (e.g. background, key point, examples) that can be hidden or revealed depending on the preferences of the reader. The idea is that this forces authors to remove any 'padding' and allows readers to consume the key points as quickly as possible, in theory learning more while reading less. We are currently focusing on the topics of business, behavioural science, personal development, professional development, science & technology and wellbeing. You can sign-up to the beta version at www.sivv.io - any feedback would be much appreciated!
I actually prototyped something similar a long time ago. It allowed for authoring sections that could be composed into a number of views. The only difference is that I wanted to capture the different "facets" of an article as well - so you could have a tab that was just the code, or statistics, etc.
Example: a short, by words, explanation for a mathematical problem. Then, the possibility to expand to a more rigorous proof and so on. This would enforce a top/down type of writing that could benefits audience and creators. I believe it would easely apply to the great part of knowledge sharing.
This would be really good for scientific articles. They are usually filled with (important but noisy) details that make skimming them for the important ideas difficult.
This is how pretty much all writing online should work IMO, even if the 'deeper' version is just a citation or link to respected expert work. Some combo of this tool for writers and community-annotations a la Genius.
I've never heard of it but I looked it up for the last minutes and indeed, Layered Ink would be one of the ways to respect the pyramid principle! Thanks for pointing it out. Where have you heard about it?
Sounds really cool and reminds me of Jason Fried's writing class idea[1]. I think many people still view Twitter's brevity as stifling nuance instead of forcing users to really refine their ideas.
Wow. I love JF and, obviously, this idea as it's pretty much on the same basis. I'd love to send it to him as soon as it's done — if anyone can help me with this, I'd be more than grateful
Indeed, there are problems with Twitter, people take stuff out of context and like to be polarized. But just as you say, the first time I interacted with Twitter I loved the fact that I had to "distill" my idea.
Kinda like Vodka — you keep distilling alcohol until you purify it (or close to it). What you're saying about "refining the idea" made me think about this.
The problem? Tweets are easier to read than long-form essays, as they require less time commitment. If the content is not good on a long-form article, you'll find out way too late. With this tool I'm developing:
Layer 1 is the shortest version of your essay, the 1 min read — like a tweet. The idea boiled down to the shortest version
Layer 2 is the same text from layer 1, but with extras added here and there. What's already read by you is in black ink. What's new is in blue ink. This is the 2 min read version
Layer 3 shows everything from Layer 1 and 2 in black ink, but what's new is now in blue ink. and you keep doing that until you get to the full version.
I can post some screenshots here of my mockups, as I'm a designer. PM me if you find this intriguing!
——
Edit: Since people are showing interest, here's how I see it happening — https://invis.io/GQWINO2YKU2#/410298082_1_Min_Verison
The first thing that you see is the first layer (1 min version). Go right for 3 and 5 min version!
——
Edit 2: since I'm seeing the upvotes and the emails, I quickly made this sign-up form for the people who want to be updated when the product is done: https://layered-ink.webflow.io/
I would put up the https://layered.ink link but the domain hasn't been propagated yet.
@Admins — please do let me know if this is not permitted so I can take it down. Apologies if so.