4 ms·

> edge thinkers understand the canons of their trades and practices deeply Yes but not from others and not from books. If you look at correspondence between so

by comboy 4y ago

> edge thinkers understand the canons of their trades and practices deeply Yes but not from others and not from books. If you look at correspondence between some brilliant math and physics minds they are often discovering that they arrived at the same things independently. I'm not advocating reinventing the wheel, but life is too short to learn what all other smart people had to say on given topic. You end up being expert on what was written by whom instead of learning the thing. I'm sure you've met those people. Subtleties of new discoveries are ignored because "well obviously, John wrote about that in 1928". Plus you get stuck in local minima along with the rest, because for even smal improvement it may be necessary to use completely different framework, a way of looking at the thing. And just knowing vocabulary that others have used you are already stuck within their framework. We think in patterns, we name them. If you were thought a name for some range of colors between red and orange, you would think of it as something clearly distinct than both red and orange, just as you are unlikely to accept that brown is just dark orange.

andsoitis 4y ago
> but life is too short to learn what all other smart people had to say on given topic. You could just reach out to those people who are on the cutting edge and discuss your ideas… they’ll accelerate your work without you having to read / research the entire corpus yourself.
- comboy 4y ago
  It doesn't scale, only a few chosen can do that and I think that's what Wolfram is trying to do here. But when you are using a different framework, learning it and creating a mapping is usually just too expensive for other experts. So you get ignored until you can no longer be ignored.
  - andsoitis 4y ago
    Wolfram could easily reach out to the top 10 experts in the field to bounce ideas off of each other. The cumulative time would be nothing in comparison to going down dead-end paths he could've avoided. I'll add something more: having dialog with others help you iterate and explore new surface area much more rapidly than sitting at your desk or walking in your garden alone.
jacobolus 4y ago
> If you were thought a name for some range of colors between red and orange, you would think of it as something clearly distinct than both red and orange, just as you are unlikely to accept that brown is just dark orange. What does this mean? People have come up with all sorts of wacky names for “hues between red and orange”, but since they are not in common currency they are generally ambiguous or incomprehensible. The clearest names for these hues are reddish orange, orange–red, orangish red. Brown is “just” dark orange (the range of hues considered “brown” for dark colors doesn’t precisely match with the range of hues called “orange” for lighter colors, but close enough). If you want a reasonably unambiguous English-speaker-comprehensible system of color names, you could try the ISCC–NBS system, https://en.wikipedia.org/wiki/ISCC–NBS_system https://en.wikipedia.org/wiki/ISCC–NBS_system
- comboy 4y ago
  It means that if you have these names internalized (redorange for example, I'm sorry for not using the proper names), then if you see berries in that color then you recognize them easily - those are the only redorange berries I know. Otherwise these are just another red berries to me, I have to notice differences in shape for example. How you divide reality into patterns influences a lot what abstractions you can come up with on top of that. E.g. take evaporation. It's much easier to discover it if you divided things into solids, liquids and gases than if you divided things into fire, earth, water and air. It can get much more subtle like with intro/extroverts or asperger/autism in psychology. It's really hard to compare things and find new patterns between them once you put them into some buckets. You can much more easily find patterns within a bucket. But among different buckets, it's hardly possible, and it seems extremely hard to ignore some pattern once you've learned it.

by comboy 4y ago

andsoitis 4y ago

> but life is too short to learn what all other smart people had to say on given topic. You could just reach out to those people who are on the cutting edge and discuss your ideas… they’ll accelerate your work without you having to read / research the entire corpus yourself.

comboy 4y ago
It doesn't scale, only a few chosen can do that and I think that's what Wolfram is trying to do here. But when you are using a different framework, learning it and creating a mapping is usually just too expensive for other experts. So you get ignored until you can no longer be ignored.
- andsoitis 4y ago
  Wolfram could easily reach out to the top 10 experts in the field to bounce ideas off of each other. The cumulative time would be nothing in comparison to going down dead-end paths he could've avoided. I'll add something more: having dialog with others help you iterate and explore new surface area much more rapidly than sitting at your desk or walking in your garden alone.

jacobolus 4y ago

> If you were thought a name for some range of colors between red and orange, you would think of it as something clearly distinct than both red and orange, just as you are unlikely to accept that brown is just dark orange. What does this mean? People have come up with all sorts of wacky names for “hues between red and orange”, but since they are not in common currency they are generally ambiguous or incomprehensible. The clearest names for these hues are reddish orange, orange–red, orangish red. Brown is “just” dark orange (the range of hues considered “brown” for dark colors doesn’t precisely match with the range of hues called “orange” for lighter colors, but close enough). If you want a reasonably unambiguous English-speaker-comprehensible system of color names, you could try the ISCC–NBS system, https://en.wikipedia.org/wiki/ISCC–NBS_system https://en.wikipedia.org/wiki/ISCC–NBS_system

comboy 4y ago
It means that if you have these names internalized (redorange for example, I'm sorry for not using the proper names), then if you see berries in that color then you recognize them easily - those are the only redorange berries I know. Otherwise these are just another red berries to me, I have to notice differences in shape for example. How you divide reality into patterns influences a lot what abstractions you can come up with on top of that. E.g. take evaporation. It's much easier to discover it if you divided things into solids, liquids and gases than if you divided things into fire, earth, water and air. It can get much more subtle like with intro/extroverts or asperger/autism in psychology. It's really hard to compare things and find new patterns between them once you put them into some buckets. You can much more easily find patterns within a bucket. But among different buckets, it's hardly possible, and it seems extremely hard to ignore some pattern once you've learned it.