Read on Twitter
It’s @threadapalooza — time to spontaneously do a 100-tweet thread on anything! 
I’d hardly call myself an expert on this subject and by subject I mean prolific tweeting. So I’m just going to tweet about what interests me the most — tech, startups, building & scaling companies
I’d hardly call myself an expert on this subject and by subject I mean prolific tweeting. So I’m just going to tweet about what interests me the most — tech, startups, building & scaling companies
2) Not sure where this is going or if there’s going to be a clear narrative arc, so buckle up for the 
A note to preserve your timeline (h/t @anthilemoon here): if this doesn’t sound interesting you can mute this thread and no hard feelings.
A note to preserve your timeline (h/t @anthilemoon here): if this doesn’t sound interesting you can mute this thread and no hard feelings.
3) Technical problems are actually economics problems of not allocating enough resources to solve the problem for a meaningful amount of people from which you can extract value to justify the initial outlay
4) What better example of this than the COVID vaccine development this year? The timeline below just shows Pfizer and Moderna.
One small step for man (particularly the ones on the frontlines) with the vaccine, one giant leap for mankind to show us what is even possible.
One small step for man (particularly the ones on the frontlines) with the vaccine, one giant leap for mankind to show us what is even possible.
5) With a decade in tech behind me, it’s safe to say I have been “enchanted by what technology can do in partnership with users”
(a lovely quote by the inimitable @PlayNiceInst)
(a lovely quote by the inimitable @PlayNiceInst)
6) One of my standout conversations this year was an @interintellect_ salon hosted by @Thomas__Arnold about how cutting edge technology like AI and robotics melded with something as varied and personal as care work in intimate spaces.
7) Do machines give us a way to understand human activities like care and more deeply help understand what being human is?
8) With any type of automation, whether that’s a simple workflow or building flying robots, we ask:
- What do humans do best? What do machines do best?
- What should humans do? What should machines do?
- What do humans do best? What do machines do best?
- What should humans do? What should machines do?
9) Will a machine replace you? “Computers are tools, not rivals,” said Peter Thiel said and elaborated in a nifty 2x2 matrix (which should bring tears of joy to any management consultant’s heart I imagine).
10) Complementarity not substitution then. “Better technology in law, medicine and education won’t replace professionals; it will allow them to do even more.”
11) When I think about technology and the future, @doctorow's quote comes to mind: "You have to be both pessimistic enough to believe that things will get worse if left unchecked [and] optimistic enough to believe that if you take action, the worst can be prevented."
12) Talk about convergence of ideas! Peter Thiel in a similar vein as @doctorow when being optimistic and biased to action called it Definite Optimism. "To a definite optimist, the future will be better than the present if (s)he plans and works to make it better."
13) What humans can do best vs. should do best is an endlessly fascinating topic. We think of moral norms and competence and should tackle it from various angles.
14) From @Thomas__Arnold's freewheeling @interintellect_ discussion: care work has emotional dimensions to navigate, both for the patient & caregiver. On the other hand, a robot assistant has no expectations on its being and so you can tap into that lack of ability creatively.
15) With caregivers of dementia for instance, repetitive conversation with the patient might be mentally and emotionally taxing. An automated presence could provide an assist - no emotional reaction here helps the patient, so does consistency of responses and no dip in stamina.
16) Automation doesn't have to look like a sci-fi movie to be effective. Even starting with a whiteboard with answers to the common questions asked to help dementia patients is a low tech yet effective start. Minimum Usable Product FTW!
17) Fascinating to hear about @TapoBhat's research on assistive robot feeding systems targeting people with spinal cord injuries and an entire lab around caregiving robots (!).
18) Social robots, collaborative robots...mostly don't work. Some lessons:
- Long-term clear use case is key
- Calling artists to simplify and help with better design
- Robots as they are today, being in people's homes, already evoke emotional responses https://spectrum.ieee.org/automaton/robotics/home-robots/anki-jibo-and-kuri-what-we-can-learn-from-social-robotics-failures
- Long-term clear use case is key
- Calling artists to simplify and help with better design
- Robots as they are today, being in people's homes, already evoke emotional responses https://spectrum.ieee.org/automaton/robotics/home-robots/anki-jibo-and-kuri-what-we-can-learn-from-social-robotics-failures
19) When building a robot caregiver that feeds you ( @TapoBhat's work), constraining the problem is key as there are way too many factors. Every user is different so universal design is out for something as complex as caregiving, intuitive for humans but tough to fully recreate.
20) But with any endeavor at automation, fully duplicating the human experience wouldn't make sense. How do automated systems represent themselves responsibly and accurately?
21) More fascinatingly, what new experiences do robot caregivers and the like open up for us? How do we creatively tap into the gaps and make the lack of similarity between robots and humans work for us?
22) Could careful automation help get a more personal touch in an assisted scenario? Rote tasks done by robots and automation, caregiver has more time to connect with the patient.
23) With something like eating, such an intimate activity, the robot feeding system tool be an extension of the patient's arm. Could this reduce embarrassment for and empower the care recipient?
24) One of the things that subtly blew my mind from @Thomas__Arnold's mediated discussion on automation in care, was how many of the problems were not technical but rather business / economic fundamentals.
25) Purely economic-focused discussions we have when building companies are just as relevant when building new innovations that could change lives. Who knew?!
26) Some basic questions on deciding what fancy technical thing to automate / build:
- Who are your customers?
- What key insights do you have from talking to them?
- How do you segment the customer base to ensure the product is useful?
- Which segment do you focus on and why?
- Who are your customers?
- What key insights do you have from talking to them?
- How do you segment the customer base to ensure the product is useful?
- Which segment do you focus on and why?
27) Another cool tech angle is @PlayNiceInst's research and work at http://theplayniceinstitute.com on merging tech and mental health of children and young adults through video games. More agency here for kids than prevention or therapeutic approaches.
28) AI bots can be helpful and playful simulation partners. They can be sparring partners for people, help iterate on their self image and identity over time and help grow with them. AI can show and unearth parts you haven't seen, minimize personal blind spots.
29) Heard about @SkipRizzoVR's work with using VR therapy for PTSD and trauma in military contexts. Taking vets through reenactment in VR with his #Bravemind project -- exposure therapy and therapy bots.
30) Therapy bots in particular are interesting. Shame drives a lot of mental health problems. The bot isn't projecting judgement and so any barriers to effective therapy are removed here. Yet another interesting place where an automated/bot approach has unique advantages.
31) Eliza by @zachtronics, a visual novel about an AI counseling service, is an interesting peek into what evolved therapy bots might look like in the future.
http://www.zachtronics.com/eliza/
http://www.zachtronics.com/eliza/
32) Therapy bots and visual novels aside, people self-soothe on the Internet, communities as quasi-group therapy. With the rise of new social and audio apps, worth understanding: how do these online groups work and what functions do they have for real socialization?
33) A fantastic question I've been ruminating on: How do we educate ourselves about what to ask of the technology we're building?
34) #RequestForStartup: A therapy bot / Alexa for you and your conversations, tracking what you're saying (slightly creepy yes), reviewing and surfacing insights about you.
35) In another fun @interintellect_ salon this year by @atroyn and @nickarner (on reimagining the computer as a machine for enhancing and scaling human intellect and creativity), they poignantly asked: When was the last time you had genuine joy at using a piece of software?
36) Games are a good entry point into changing how people think about computing.
37) Moravec's paradox --> logic & reasoning requires very little computation but simple, sensorimotor skills require enormous computational resources.
"Easy to make computers exhibit adult level performance...and difficult or impossible to give them the skills of a 1-year-old"
"Easy to make computers exhibit adult level performance...and difficult or impossible to give them the skills of a 1-year-old"
38) In computing, the complex is easy and the easy is complex. How should this shape our answers to what computers can do best and should do?
39) When automating, worth asking: what if taking tedium away while also reducing agency with the goal of making it easier for customers actually limits options?
40) When thinking of delightful computing experiences, helpful to look for where weird stuff is getting made. Where's the new frontier? @tiktok_us and @instagram are promising at the moment! Culture is being created there using creative tools, the best software allows for that.
41) Change & innovation point to the value of raising people's ambitions from this evergreen @tylercowen article: "It costs you relatively little to [raise aspirations of other people] but the benefit to them & to the broader world may be enormous." https://marginalrevolution.com/marginalrevolution/2018/10/high-return-activity-raising-others-aspirations.html
42) Helping create innovation = raise others' aspirations (à la @tylercowen) + inspire a healthy dose of "but look at these clowns around us" (thank you @AriannaSimpson for a new rallying cry)
43) What the most ambitious people choose to do with their lives has a profound impact on society, the economy and culture. We're constrained by ability but ambition sets the ceiling / outer limit both personally and for innovation at a macro level.
44) Speaking of ambition, I've quoted this @cdixon article aplenty! The pull of the next upward step on the current hill is so tempting. We overvalue near term rewards and this gets us stuck at local optimums, AKA not the highest hill in our terrain. https://cdixon.org/2009/09/19/climbing-the-wrong-hill
45) If you're talking about innovation and high ambition, startups are definitely at that intersection. Startups = Growth.
46) I reference this great @paulg article constantly: "Not every newly founded company is a startup. For a company to grow really big, it must (a) make something lots of people want, and (b) reach and serve all those people."
http://www.paulgraham.com/growth.html
http://www.paulgraham.com/growth.html
47) What do startups & technology have in common?
- "Rapid change in one area uncovers big, soluble problems in other areas"
- "Startups create new ways of doing things and new ways of doing things are, in the broader sense of the word, new technology"
http://www.paulgraham.com/growth.html
- "Rapid change in one area uncovers big, soluble problems in other areas"
- "Startups create new ways of doing things and new ways of doing things are, in the broader sense of the word, new technology"
http://www.paulgraham.com/growth.html
48) If startups = growth, then a target growth rate acts as a lodestar. Having spent this year heads down building something at @JoinAtomic and living this, I can attest to what an effective forcing function growth really is.
49) Startups die of surfeit not starvation. Less (things to focus on) is definitely better than more. And having growth as the one thing to focus on above all really helps.
50) Startups really only have speed and optimism on their side. So just do the damn thing. "Having to hit a growth number every week forces founders to act, and acting vs. not acting is the high bit of succeeding."
http://www.paulgraham.com/growth.html
http://www.paulgraham.com/growth.html
Because I can’t seem to tweet thread properly, the next half of this 100-tweet @threadapalooza journey is messily linked here: https://twitter.com/swathyprithivi/status/1339143558469578755
