Hacker News new | past | comments | ask | show | jobs | submit login
Catching crumbs from the table (2000) (nature.com)
46 points by nether on Dec 26, 2016 | hide | past | favorite | 5 comments



One thing I think is particularly interesting about this idea is the ways in which it already applies today.

Working with overseas-produced chips with no or poor English documentation is a kind of hermeneutics. The ESP8266 is a popular chip but mostly because of the (now) enormous body of third-party English-language instructions. The RTL2832 is perhaps an even better example - nobody knew it could even do SDR for a long time. In the west we rely a lot on third parties to bridge the gap with the real hardware hubs in Asia.

Popular science is another example. We can see lots of videos of people trotting around the LHC handwaving about how it works. Is that really useful if there's basically zero chance of the audience getting involved in particle physics? And yet widespread opposition to evolution and climate science shows that teaching science to non-scientists can be a wise investment.

And, of course, the infamous "programming by poking" metaphor from Sussman: "You grab this piece of library and you poke at it. You write programs that poke it and see what it does. And you say, 'Can I tweak it to do the thing I want?'" In other words, programming by hermeneutics. And yet, this approach solves certain problems far more quickly than building from the ground up.

I see these as an essential artefact of specialisation; we're not smart enough to be particle physicists, and computer scientists, and chip engineers. We can't understand everything, and as these fields deepen, they only get more incomprehensible. I'm sure most observations by one serious physicist to another would go totally over my head, despite my relatively decent amateur physics knowledge. It's maybe not a stretch to say that at some point to push the frontiers of physics you will need to be a physicist from birth.

So it seems we don't necessarily need a jump from human to metahuman; hermeneutics is important already, and going to become more so as our specialists become more specialised. Greg Egan once wrote about passing important information through a kind of enormous transhuman game of telephone, where each kind of intelligence translated the concepts into a form its nearest intellectual neighbours could understand, and so on until it spread to everyone. Perhaps that's where all of this ends.


Well put.

Except:

> And, of course, the infamous "programming by poking" metaphor from Sussman: "You grab this piece of library and you poke at it. You write programs that poke it and see what it does. And you say, 'Can I tweak it to do the thing I want?'" In other words, programming by hermeneutics.

I'm afraid that this is sometimes the excuse for not documenting software -- sort of a, "Hey, here it is, just poke at it, you'll figure it out!" kind of attitude. Eat your veggies, rotate your tires, document your code.


In my land we call this system : Stack Overflow.


Thank you for your thoughts. I thought they were very well written and totally becoming of what makes hacker news great.


I don't know how many people realise this, but Nature puts out great little SF short stories. If the interaction between humans and “metahumans” tickle your fancy, I can recommend Peter Watts' Blindsight/Echopraxia as well.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: