7 ms·

How can we develop transformative tools for thought?

104 points by kristianc 7y ago

anotheryou 7y ago
Maybe UX complexity is a problem, too. Tools for thought have to become like prosthetics, an extension of your body. For this it has to have really low friction. Here we have two problems: 1. the ideal power user UX has some learning curve for newcomers which hinders adaption a lot; 2. I prefer typing, but it's hard to beat a piece of paper in ease of input and flexibility of arrangement, type of input (graphics, text, plots, mindmaps, tables all there without any toolbar to click). The 2nd point makes me think that a general interface is just really hard and specific ones can be done really well already. Examples of tools for thinking I like: - org mode: suffers a lot from (1.) and is not that generalized (more is possible, but little effortlessly). - Deepl Desktop App: real AI collaboration for translation. The UI could be more streamlined, but is fine for now: hit ctrl+c+c, check if the translation is ok, click "insert" and your sentence is translated within your document. It translates 90% as good as I do, I catch the other 10%. - Soulver/Opalcalc/NaSC: a bit like excel but all calculations always visible (trade offs: not working for tabular data, not as flexible as programing) - wikis: amazing for long term and collaborative storage, problems with scaling, relevance, discovery and organization - paper: great input flexibility, hard to refactor, no computer assist (calculations, data etc) - jupyter notebooks: finally mixing documentation and programming, sadly problems with structure and state - grafical programming languages (touch designer, pure data, bitwig grid etc.): predominant in the art-world, I think because you can tweak values and connections best.
- classified 7y ago
  > paper: ... no computer assist ... the greatest drawbacks being that you can‘t show relevant ads, personalize the UX, or gain deep insights into your users‘ needs.
- sansnomme 7y ago
  Org mode doesn't suffer from 1), Emacs does. Org-style editors like Workflowy, Notion, and now Roam are all high growth apps with a large user base.
  - anotheryou 7y ago
    But org-mode lives in emacs. But yes, part of it can be done in a "user friendly" way and other tools do so :). Most notably outliners and task to agenda aggregation.
  - tincholio 7y ago
    Org is a lot more than just outlining, etc. Do any of those tools have literate programming, tons of export backends, GTD, Agenda management, etc?
- anotheryou 7y ago
  I usually use paper when I do math or logic and can build a crutch from geometry. You just can't draw well with a mouse.
- infinite8s 7y ago
  Does it really need low friction? You know what's another tool for thought? Literacy. It's something humans aren't able to do easily without a lot of effort to scale the initial complexity, but once they so it opens up vast new vistas of thinking that were previously inaccessible.
  - infinite8s 7y ago
    I think the flaw in all these approaches to human computer interaction (in particular for cognitive augmentation) is the idea that it should be easy and intuitive.
- buzzkillington 7y ago
  >- jupyter notebooks: finally mixing documentation and programming, sadly problems with structure and state Solved amazingly well by orgmode babel/tangle.
  - anotheryou 7y ago
    Does it force the whole file to be executed? I mean jupyter notebooks for machine learning e.g. also profit from state, because to work fast you often really don't want to recalculate something that took half an hour. (If it takes hours I guess you should store it to a file anyways...)
    - buzzkillington 7y ago
      No, you can execute one cell at a time. In different languages.
robenkleene 7y ago
As a former UI designer, this is a gross misrepresentation of what happened with Sketch (and to a lesser extent, Figma): > Consider, for example, the way the program Sketch has eaten into Adobe’s market share, after duplicating many of the best features from several of Adobe’s products, perhaps most notably Illustrator. And consider the way Figma is now eating into both Sketch and Illustrator’s market share. Both Sketch and Figma have done this without needing to make an enormous investment in research. That’s a big advantage they have over Adobe. Leading up to the time that Sketch emerged, UI designers were complaining constantly to Adobe to provide features to support their work. All of us were using a tiny subset of Photoshop's features crammed into the corner (the vector editing features). Remember Adobe actively killed Fireworks[0] the exact product specialized for UI designers that should have prevented Sketch from emerging. I consider Adobe leaving an opening big enough for Sketch to emerge one of the biggest mistakes I've ever seen a tech company make. It was plainly obvious that there was product category there, so many people were talking about it that to me as a designer it was deafening. Not to mention, if you were paying the slightest bit of attention to graphical apps for artists, you would have noticed Acorn[1], a Photoshop competitor released in 2007, three years before Sketch, maintained by one person, based on the support Apple was aggressively adding to AppKit for this type of application. If you were an expert in this market and you didn't realize that someone was going to use that same toolkit to make a UI tool, which so many people were asking for, I don't know what to tell you. And of course that's exactly what happened with Sketch. This isn't a story about Sketch copying Adobe products, it's a story about Adobe's product leadership making a series of mistakes that left a gap in the market. [0]: https://en.wikipedia.org/wiki/Adobe_Fireworks https://en.wikipedia.org/wiki/Adobe_Fireworks [1]: https://flyingmeat.com/acorn/ https://flyingmeat.com/acorn/
- anewguy9000 7y ago
  ive never heard of acorn before, does it have smart fill or tool for adjusting raw images? for me, those features are what separates photoshop from everything else
- andymatuschak 7y ago
  Hi there. I'm one of the article co-authors, and I was definitely one of the Photoshop-anchored designers agitating for Adobe to add UI-centric design features. I think you're right that it was enormous mistake to leave this market open. Unpacking how this comment relates to the claim in the article: That section's claim is that these mediums are often common goods, so it's hard to build moats around one's investments. Extending your comment to more explicitly attack the central claim, I think you're suggesting that if Adobe _had_ paid attention to this market, they'd be able to defend their position, and Sketch/Figma never would have been relevant. Maybe they would have been able to more quickly create a "spinoff" app because they could adapt existing code, or maybe they could take advantage of their existing network effects (e.g. through educational channels). And, more broadly, that the general claim is weaker than it appears: media inventors can often find ways to defend their turf through mechanisms like this. That might be true! It's hard to adjudicate counterfactuals like this. I'd be curious if you have any lenses that might help us think about that. Given that we're on Hacker News, I'll offer: one reason why startups often outcompete incumbents is that line extensions like this are counterintuitively often _harder_ for the well-resourced incumbents because of inertia, culture, etc.
walterbell 7y ago
The article does a good job of connecting business incentives to interface design or lack thereof. Arguably, some of the most influential "tools for thought" today are videos, thanks to mobile devices/data & Youtube. They offer: - distribution: global at low cost - accessibility: low barrier to create/view - funding: ads, patreon, product placement - creation: video editor + smartphone - composition: screenshot/record anything - annotation: free-form text/audio/overlay - UX: any structure the creator can imagine The diversity of youtube "thought UX" examples shows what's possible when creators are unleashed from restrictive tool pre-conceptions. Twitter is a social network where an early minimalist UX lead to user-driven innovations, with later attempted formalization by the platform vendor. Google has not much ventured into content creation tools for its video platform. Apple's iPad provides tools for freeform markup of screenshots -> PDF, with the LumaFusion video editor rivaling PC apps for one-tenth of the cost. Perhaps universities have studied online videos for UX patterns employed by creators to communicate complex topics? Which ones deserve formalization in tool workflows, while leaving room for ongoing experimentation?
CKN23-ARIN 7y ago
Tools that empower individuals on a mass scale are threatening to existing power structures.
- anotheryou 7y ago
  Any evidence for active discouragement of tool building? I actually don't see that for general purpose tools, only where censorship or free speech come in to play.
  - robenkleene 7y ago
    Point blank I think Mac App Store restrictions, iOS restrictions, and the current decrepit state of AppKit are all a systemic approach by Apple to prevent new powerful tools from emerging that threaten their current market dominance. Ben Thompson wrote the canonical piece about this situation[0], but I think this paragraph from a different article is the best summary[1]: > The reality for Jobs and Apple was that the company’s users needed Office (along with Adobe’s products) more than they needed a Mac. I’ve long argued that being in this position is a big reason why Apple hasn’t enabled sustainable apps: never again would a software developer hold Apple hostage. The irony, though, is that when it came time to launch the iPad Pro, Apple had no one else to turn to. [0]: https://stratechery.com/2013/why-doesnt-apple-enable-sustainable-businesses-on-the-app-store/ https://stratechery.com/2013/why-doesnt-apple-enable-sustain... [1]: https://stratechery.com/2015/from-products-to-platforms/ https://stratechery.com/2015/from-products-to-platforms/
    - anotheryou 7y ago
      fair point, thanks. Not my universe I guess...
    - zepto 7y ago
      This seems like a bizarre point of view - AppKit may not be getting as much investment as some other elements, but the overall platform has received massive investments. What would Apple be doing differently if they weren’t hypothetically trying to hold back tools for thought?
carapace 7y ago
If I had more time I would write a better comment, apologies. - - - - I doubt it requires big research to make good tools. Inventing "zero" wasn't done by committee. In any event, the research has been done. It is steadfastly ignored. E.g. something as simple and straightforward as Jef Raskin's "Humane Interface" applied uncreatively would measurably improve the majority of user interfaces out there. There's a near total disconnect between good UI design and what gets inflicted on IRL users. Evidence: every change ever made to every app or web app that made the users scream. Heck, "Google backtracks on search results design (techcrunch.com)" is on HN front page as I type this. At best people are fucking UI/UX to make a buck, but most of them do it because they suck at their jobs. This isn't limited to UI design. E.g. MLs have had type algebra for decades and yet it's only just barely gaining traction among "professional" programmers today in A.D. 2020. Instead of flying cars we have Rube Goldberg machines made out of other Rube Goldberg machines. That's why "When small groups of motivated people [try] they make rapid progress." Because what we're using now is refried merde. It doesn't scale either because people are stubborn about giving up crap[1] -or- it just hasn't been let. Canon killed the Cat. https://en.wikipedia.org/wiki/Canon_Cat https://en.wikipedia.org/wiki/Canon_Cat Dynamicland may be fantastic but you can't buy it for money. ([1] This might be the real problem: People just don't give a fuck. Thinking takes energy- calories -and so, as an evolved animal, if you can get by w/o it, you will. "The Internet is for porn.") I've known what I wanted to make ever since reading Ted Nelson's "Dream Machines" and that was twenty-odd years ago (it was published way earlier, that's just when I got ahold of it.) Engelbart's "Mother of All Demos" was even earlier. (I've been playing with a very simple UI demo that incorporates elements from Raskin and from Wirth's Oberon OS. Its backing store is a git repo and all changes are autosaved. It's dreamy. Raskin warns that excellent UIs are addictive in the sense that once you get used to them it's like withdrawal if you have to go back, and he's right. Mac, Windows, X et. al., it's all so painful and clunky now.) So yeah, with a little work, and standing on the shoulders of giants and not on their toes, you can make something fantastically better than current tools/OSs. It's not hard. It might even make money in the marketplace, eh? - - - - > What's needed is the development of a powerful praxis, a set of core ideas which are explicit and powerful enough that new people can rapidly assimilate them, and begin to develop their own practice. We're not yet at that stage with tools for thought. Speak for yourselves, eh? Here's where I think they fall into the same trap as so many people do (philosophers, psychologists, designers-of-tools-for-thought, etc.) when talking about thought: They do not define the term. What is thought? (As an aside the only place I've seen that can begin to claim to have a concrete mathematical theory of intelligence amplification is "Introduction to Cybernetics" by Ashby.) Anyhow, the folks who have the goods on the structure of subjective experience work under the rubric of "Neurolinguistic Programming" (the other NLP). Unfortunately the school of thought is still seen as pseudoscience! Oh well, what a world... Anyway, the "machine code" of thought is called "submodalities" in NLP jargon, an (outmoded) model of "microcode" is called "strategies" but that's deprecated (I don't know how to succinctly describe the replacement models, and many of them don't have names), etc. Using the patterns and algorithms discovered or reified by the NLP folks you can rapidly and easily reprogram your mind. [2] At that point you don't really need external "tools of thought" but if that's your thing you're in a much better place to make them. (I.e. if you want to make mind-machine interfaces and such like, or design audio/visual/kinesthetic UIs and UX flows.) Here you're going to want to pick up a copy of Scott McCloud's "Understanding Comics: The Invisible Art" to get a handle on the finer points of cognition and storytelling. See also Brenda Laurel's "Computers as Theatre". [2] For example there's a spelling "strategy" that involve using visual memory to recall the word (as a picture) and a kinethestic check for correctness. In other words, remember the word, feel that it's right, then read it off from the picture in your mind. Most good spellers use that strategy, most bad spellers do something else. You can take someone who is bad at spelling, teach them to use the proper "strategy", and suddenly they are good spellers. It's analogous to replacing a buggy spellcheck subroutine with a correct one. Spelling isn't hard, we have just been teaching badly. We are in a transition time in psychology similar to when alchemy became chemistry.
avivo 7y ago
This is incredibly important — but I actually the focus of the piece may be too narrow. They concentrate on tools for learning quickly and effectively as an individuals. Tools for thought are especially crucial for enabling groups to grok and tackle complex problems. Imagine if every discussion of a complex societal problem was situated within both a: A) Problems map: spelling out concrete sub-issues B) Systems diagram: illustrating how parts of the system interact (useful for understanding levers, forces, feedback, etc.) It seems non-ideal to try to "solve" intermeshed problems if we don't even try understand both of the above. It's like driving while blind. Perhaps this isn't "tools for thought" but "tools for seeing"? Is anyone working on this or doing it well? *Somewhat excerpted from https://twitter.com/metaviv/status/1219732275174133760 https://twitter.com/metaviv/status/1219732275174133760
wrnr 7y ago
I am going through the quantum computer exercise, but one point of improvement is that it's awfully wordy. A lot of time is spend on saying they are saying how they are saying what they are saying. Better to stick to their own advice of writing concise and to the point.
melvinroest 7y ago
This is why I created doodledocs.com It is a humble prototype [1] for: what if the web was simply made of paper and you have a stylus? Collaborative drawing Drawing over websites Drawing over images (and making them transparent) Those were important features I came up with. Currently I’ve noticed the following by using it that it helped me as follows: - it is super easy to talk about websites because you can draw over them and circle things and so on. - By line drawing images and playing with the transparency of my pencil, my drawing skills improved - It has been the quickest way to communicate about math if you also use voip with another program [1] While I find it transformative, it is that just a little bit. Also, it works best in Chrome and since my implementation of webrtc is totally free (no TURN server), it doesn’t work with corporate wifi. Also while collaborative drawing should work quite well collaborative web annotation is really buggy (you need the same viewport).
pieterk 7y ago
Lots of thoughts already on this page, hard to focus a discussion but I'll try to get to the essence. The mnemonic medium tool is fantastic, it forces our brain to keep an idea current; we tend to naturally forget stuff if it isn't necessary to our survival. That the software "knows" when you're about to forget something seems to be what's novel here. The tool is always paired to an individual though, so I bet this will have to be personalized down the line. What do I need to be reminded of at time X? As for the What... I feel like I can only learn one "neural diff" at a time. Which one totally depends on what I already know: some times I learn nothing, other times there's too much new information unrelated to anything I already know. Ideally the tool would know what I know, compress a whole essay, and "program" me with a story composed of something like the following operations: a) add a new connection between concepts you previously thought unrelated b) change a connection between two concepts, they are not related how you thought c) insert a new concept Of course the hard part is knowing what I already know, but I could tell you. PS. This whole ordeal seems to be related to what good teachers and advertisers do: first get to know the audience, then tell a story with good rhythm and spacing. PPS. The key to augmenting our thoughts with technology is trust. Our nature requires our brain to keep us alive. We can learn to externalize thought processes, yet if that externality can't be trusted, our instinct will be to unlearn our connection to it.

104 points by kristianc 7y ago

Maybe UX complexity is a problem, too. Tools for thought have to become like prosthetics, an extension of your body. For this it has to have really low friction. Here we have two problems: 1. the ideal power user UX has some learning curve for newcomers which hinders adaption a lot; 2. I prefer typing, but it's hard to beat a piece of paper in ease of input and flexibility of arrangement, type of input (graphics, text, plots, mindmaps, tables all there without any toolbar to click). The 2nd point makes me think that a general interface is just really hard and specific ones can be done really well already. Examples of tools for thinking I like: - org mode: suffers a lot from (1.) and is not that generalized (more is possible, but little effortlessly). - Deepl Desktop App: real AI collaboration for translation. The UI could be more streamlined, but is fine for now: hit ctrl+c+c, check if the translation is ok, click "insert" and your sentence is translated within your document. It translates 90% as good as I do, I catch the other 10%. - Soulver/Opalcalc/NaSC: a bit like excel but all calculations always visible (trade offs: not working for tabular data, not as flexible as programing) - wikis: amazing for long term and collaborative storage, problems with scaling, relevance, discovery and organization - paper: great input flexibility, hard to refactor, no computer assist (calculations, data etc) - jupyter notebooks: finally mixing documentation and programming, sadly problems with structure and state - grafical programming languages (touch designer, pure data, bitwig grid etc.): predominant in the art-world, I think because you can tweak values and connections best.

classified 7y ago
> paper: ... no computer assist ... the greatest drawbacks being that you can‘t show relevant ads, personalize the UX, or gain deep insights into your users‘ needs.
sansnomme 7y ago
Org mode doesn't suffer from 1), Emacs does. Org-style editors like Workflowy, Notion, and now Roam are all high growth apps with a large user base.
- anotheryou 7y ago
  But org-mode lives in emacs. But yes, part of it can be done in a "user friendly" way and other tools do so :). Most notably outliners and task to agenda aggregation.
- tincholio 7y ago
  Org is a lot more than just outlining, etc. Do any of those tools have literate programming, tons of export backends, GTD, Agenda management, etc?
anotheryou 7y ago
I usually use paper when I do math or logic and can build a crutch from geometry. You just can't draw well with a mouse.
infinite8s 7y ago
Does it really need low friction? You know what's another tool for thought? Literacy. It's something humans aren't able to do easily without a lot of effort to scale the initial complexity, but once they so it opens up vast new vistas of thinking that were previously inaccessible.
- infinite8s 7y ago
  I think the flaw in all these approaches to human computer interaction (in particular for cognitive augmentation) is the idea that it should be easy and intuitive.
buzzkillington 7y ago
>- jupyter notebooks: finally mixing documentation and programming, sadly problems with structure and state Solved amazingly well by orgmode babel/tangle.
- anotheryou 7y ago
  Does it force the whole file to be executed? I mean jupyter notebooks for machine learning e.g. also profit from state, because to work fast you often really don't want to recalculate something that took half an hour. (If it takes hours I guess you should store it to a file anyways...)
  - buzzkillington 7y ago
    No, you can execute one cell at a time. In different languages.

robenkleene 7y ago

As a former UI designer, this is a gross misrepresentation of what happened with Sketch (and to a lesser extent, Figma): > Consider, for example, the way the program Sketch has eaten into Adobe’s market share, after duplicating many of the best features from several of Adobe’s products, perhaps most notably Illustrator. And consider the way Figma is now eating into both Sketch and Illustrator’s market share. Both Sketch and Figma have done this without needing to make an enormous investment in research. That’s a big advantage they have over Adobe. Leading up to the time that Sketch emerged, UI designers were complaining constantly to Adobe to provide features to support their work. All of us were using a tiny subset of Photoshop's features crammed into the corner (the vector editing features). Remember Adobe actively killed Fireworks[0] the exact product specialized for UI designers that should have prevented Sketch from emerging. I consider Adobe leaving an opening big enough for Sketch to emerge one of the biggest mistakes I've ever seen a tech company make. It was plainly obvious that there was product category there, so many people were talking about it that to me as a designer it was deafening. Not to mention, if you were paying the slightest bit of attention to graphical apps for artists, you would have noticed Acorn[1], a Photoshop competitor released in 2007, three years before Sketch, maintained by one person, based on the support Apple was aggressively adding to AppKit for this type of application. If you were an expert in this market and you didn't realize that someone was going to use that same toolkit to make a UI tool, which so many people were asking for, I don't know what to tell you. And of course that's exactly what happened with Sketch. This isn't a story about Sketch copying Adobe products, it's a story about Adobe's product leadership making a series of mistakes that left a gap in the market. [0]: https://en.wikipedia.org/wiki/Adobe_Fireworks https://en.wikipedia.org/wiki/Adobe_Fireworks [1]: https://flyingmeat.com/acorn/ https://flyingmeat.com/acorn/

anewguy9000 7y ago
ive never heard of acorn before, does it have smart fill or tool for adjusting raw images? for me, those features are what separates photoshop from everything else
andymatuschak 7y ago
Hi there. I'm one of the article co-authors, and I was definitely one of the Photoshop-anchored designers agitating for Adobe to add UI-centric design features. I think you're right that it was enormous mistake to leave this market open. Unpacking how this comment relates to the claim in the article: That section's claim is that these mediums are often common goods, so it's hard to build moats around one's investments. Extending your comment to more explicitly attack the central claim, I think you're suggesting that if Adobe _had_ paid attention to this market, they'd be able to defend their position, and Sketch/Figma never would have been relevant. Maybe they would have been able to more quickly create a "spinoff" app because they could adapt existing code, or maybe they could take advantage of their existing network effects (e.g. through educational channels). And, more broadly, that the general claim is weaker than it appears: media inventors can often find ways to defend their turf through mechanisms like this. That might be true! It's hard to adjudicate counterfactuals like this. I'd be curious if you have any lenses that might help us think about that. Given that we're on Hacker News, I'll offer: one reason why startups often outcompete incumbents is that line extensions like this are counterintuitively often _harder_ for the well-resourced incumbents because of inertia, culture, etc.

walterbell 7y ago

The article does a good job of connecting business incentives to interface design or lack thereof. Arguably, some of the most influential "tools for thought" today are videos, thanks to mobile devices/data & Youtube. They offer: - distribution: global at low cost - accessibility: low barrier to create/view - funding: ads, patreon, product placement - creation: video editor + smartphone - composition: screenshot/record anything - annotation: free-form text/audio/overlay - UX: any structure the creator can imagine The diversity of youtube "thought UX" examples shows what's possible when creators are unleashed from restrictive tool pre-conceptions. Twitter is a social network where an early minimalist UX lead to user-driven innovations, with later attempted formalization by the platform vendor. Google has not much ventured into content creation tools for its video platform. Apple's iPad provides tools for freeform markup of screenshots -> PDF, with the LumaFusion video editor rivaling PC apps for one-tenth of the cost. Perhaps universities have studied online videos for UX patterns employed by creators to communicate complex topics? Which ones deserve formalization in tool workflows, while leaving room for ongoing experimentation?

CKN23-ARIN 7y ago

Tools that empower individuals on a mass scale are threatening to existing power structures.

anotheryou 7y ago
Any evidence for active discouragement of tool building? I actually don't see that for general purpose tools, only where censorship or free speech come in to play.
- robenkleene 7y ago
  Point blank I think Mac App Store restrictions, iOS restrictions, and the current decrepit state of AppKit are all a systemic approach by Apple to prevent new powerful tools from emerging that threaten their current market dominance. Ben Thompson wrote the canonical piece about this situation[0], but I think this paragraph from a different article is the best summary[1]: > The reality for Jobs and Apple was that the company’s users needed Office (along with Adobe’s products) more than they needed a Mac. I’ve long argued that being in this position is a big reason why Apple hasn’t enabled sustainable apps: never again would a software developer hold Apple hostage. The irony, though, is that when it came time to launch the iPad Pro, Apple had no one else to turn to. [0]: https://stratechery.com/2013/why-doesnt-apple-enable-sustainable-businesses-on-the-app-store/ https://stratechery.com/2013/why-doesnt-apple-enable-sustain... [1]: https://stratechery.com/2015/from-products-to-platforms/ https://stratechery.com/2015/from-products-to-platforms/
  - anotheryou 7y ago
    fair point, thanks. Not my universe I guess...
  - zepto 7y ago
    This seems like a bizarre point of view - AppKit may not be getting as much investment as some other elements, but the overall platform has received massive investments. What would Apple be doing differently if they weren’t hypothetically trying to hold back tools for thought?

carapace 7y ago

If I had more time I would write a better comment, apologies. - - - - I doubt it requires big research to make good tools. Inventing "zero" wasn't done by committee. In any event, the research has been done. It is steadfastly ignored. E.g. something as simple and straightforward as Jef Raskin's "Humane Interface" applied uncreatively would measurably improve the majority of user interfaces out there. There's a near total disconnect between good UI design and what gets inflicted on IRL users. Evidence: every change ever made to every app or web app that made the users scream. Heck, "Google backtracks on search results design (techcrunch.com)" is on HN front page as I type this. At best people are fucking UI/UX to make a buck, but most of them do it because they suck at their jobs. This isn't limited to UI design. E.g. MLs have had type algebra for decades and yet it's only just barely gaining traction among "professional" programmers today in A.D. 2020. Instead of flying cars we have Rube Goldberg machines made out of other Rube Goldberg machines. That's why "When small groups of motivated people [try] they make rapid progress." Because what we're using now is refried merde. It doesn't scale either because people are stubborn about giving up crap[1] -or- it just hasn't been let. Canon killed the Cat. https://en.wikipedia.org/wiki/Canon_Cat https://en.wikipedia.org/wiki/Canon_Cat Dynamicland may be fantastic but you can't buy it for money. ([1] This might be the real problem: People just don't give a fuck. Thinking takes energy- calories -and so, as an evolved animal, if you can get by w/o it, you will. "The Internet is for porn.") I've known what I wanted to make ever since reading Ted Nelson's "Dream Machines" and that was twenty-odd years ago (it was published way earlier, that's just when I got ahold of it.) Engelbart's "Mother of All Demos" was even earlier. (I've been playing with a very simple UI demo that incorporates elements from Raskin and from Wirth's Oberon OS. Its backing store is a git repo and all changes are autosaved. It's dreamy. Raskin warns that excellent UIs are addictive in the sense that once you get used to them it's like withdrawal if you have to go back, and he's right. Mac, Windows, X et. al., it's all so painful and clunky now.) So yeah, with a little work, and standing on the shoulders of giants and not on their toes, you can make something fantastically better than current tools/OSs. It's not hard. It might even make money in the marketplace, eh? - - - - > What's needed is the development of a powerful praxis, a set of core ideas which are explicit and powerful enough that new people can rapidly assimilate them, and begin to develop their own practice. We're not yet at that stage with tools for thought. Speak for yourselves, eh? Here's where I think they fall into the same trap as so many people do (philosophers, psychologists, designers-of-tools-for-thought, etc.) when talking about thought: They do not define the term. What is thought? (As an aside the only place I've seen that can begin to claim to have a concrete mathematical theory of intelligence amplification is "Introduction to Cybernetics" by Ashby.) Anyhow, the folks who have the goods on the structure of subjective experience work under the rubric of "Neurolinguistic Programming" (the other NLP). Unfortunately the school of thought is still seen as pseudoscience! Oh well, what a world... Anyway, the "machine code" of thought is called "submodalities" in NLP jargon, an (outmoded) model of "microcode" is called "strategies" but that's deprecated (I don't know how to succinctly describe the replacement models, and many of them don't have names), etc. Using the patterns and algorithms discovered or reified by the NLP folks you can rapidly and easily reprogram your mind. [2] At that point you don't really need external "tools of thought" but if that's your thing you're in a much better place to make them. (I.e. if you want to make mind-machine interfaces and such like, or design audio/visual/kinesthetic UIs and UX flows.) Here you're going to want to pick up a copy of Scott McCloud's "Understanding Comics: The Invisible Art" to get a handle on the finer points of cognition and storytelling. See also Brenda Laurel's "Computers as Theatre". [2] For example there's a spelling "strategy" that involve using visual memory to recall the word (as a picture) and a kinethestic check for correctness. In other words, remember the word, feel that it's right, then read it off from the picture in your mind. Most good spellers use that strategy, most bad spellers do something else. You can take someone who is bad at spelling, teach them to use the proper "strategy", and suddenly they are good spellers. It's analogous to replacing a buggy spellcheck subroutine with a correct one. Spelling isn't hard, we have just been teaching badly. We are in a transition time in psychology similar to when alchemy became chemistry.

avivo 7y ago

This is incredibly important — but I actually the focus of the piece may be too narrow. They concentrate on tools for learning quickly and effectively as an individuals. Tools for thought are especially crucial for enabling groups to grok and tackle complex problems. Imagine if every discussion of a complex societal problem was situated within both a: A) Problems map: spelling out concrete sub-issues B) Systems diagram: illustrating how parts of the system interact (useful for understanding levers, forces, feedback, etc.) It seems non-ideal to try to "solve" intermeshed problems if we don't even try understand both of the above. It's like driving while blind. Perhaps this isn't "tools for thought" but "tools for seeing"? Is anyone working on this or doing it well? *Somewhat excerpted from https://twitter.com/metaviv/status/1219732275174133760 https://twitter.com/metaviv/status/1219732275174133760

wrnr 7y ago

I am going through the quantum computer exercise, but one point of improvement is that it's awfully wordy. A lot of time is spend on saying they are saying how they are saying what they are saying. Better to stick to their own advice of writing concise and to the point.

melvinroest 7y ago

This is why I created doodledocs.com It is a humble prototype [1] for: what if the web was simply made of paper and you have a stylus? Collaborative drawing Drawing over websites Drawing over images (and making them transparent) Those were important features I came up with. Currently I’ve noticed the following by using it that it helped me as follows: - it is super easy to talk about websites because you can draw over them and circle things and so on. - By line drawing images and playing with the transparency of my pencil, my drawing skills improved - It has been the quickest way to communicate about math if you also use voip with another program [1] While I find it transformative, it is that just a little bit. Also, it works best in Chrome and since my implementation of webrtc is totally free (no TURN server), it doesn’t work with corporate wifi. Also while collaborative drawing should work quite well collaborative web annotation is really buggy (you need the same viewport).

pieterk 7y ago

Lots of thoughts already on this page, hard to focus a discussion but I'll try to get to the essence. The mnemonic medium tool is fantastic, it forces our brain to keep an idea current; we tend to naturally forget stuff if it isn't necessary to our survival. That the software "knows" when you're about to forget something seems to be what's novel here. The tool is always paired to an individual though, so I bet this will have to be personalized down the line. What do I need to be reminded of at time X? As for the What... I feel like I can only learn one "neural diff" at a time. Which one totally depends on what I already know: some times I learn nothing, other times there's too much new information unrelated to anything I already know. Ideally the tool would know what I know, compress a whole essay, and "program" me with a story composed of something like the following operations: a) add a new connection between concepts you previously thought unrelated b) change a connection between two concepts, they are not related how you thought c) insert a new concept Of course the hard part is knowing what I already know, but I could tell you. PS. This whole ordeal seems to be related to what good teachers and advertisers do: first get to know the audience, then tell a story with good rhythm and spacing. PPS. The key to augmenting our thoughts with technology is trust. Our nature requires our brain to keep us alive. We can learn to externalize thought processes, yet if that externality can't be trusted, our instinct will be to unlearn our connection to it.