3 ms·

If people raise objections, that suggests that they are having issues with understanding. I know I've been confused by it before, and I usually feel compelled t

If people raise objections, that suggests that they are having issues with understanding. I know I've been confused by it before, and I usually feel compelled to use modifiers to distinguish between 'literally millions' and 'actually, no-exaggeration, literally millions' when I want to be understood. That doesn't necessarily mean we're moving in the wrong direction. Language evolves to satisfy speakers' criteria, so it will gradually improve median utility. But any change is going to have cost, and confusion could be the price we pay for progress. Alternately, you can imagine a situation where 1% of the population derives a huge amount of utility from a word, while 99% only get a small benefit from another definition. Improvements to the median in that scenario can have negative consequences in aggregate. Or you can see that this particular usage has been confusing people for a hundred years, and the cost of change may outweigh any potential benefit. I think the position that natural language will just take care of itself is just a touch hand-wavey. Human interests and linguistic-evolutionary movement are often coaligned, but sometimes diverge- the same as with human interests and economic growth.

selestify 5y ago
> I think the position that natural language will just take care of itself is just a touch hand-wavey. Why not? It's done just that for countless millenia. Cavemen without Ph.D.'s in linguistics managed to develop languages for communication just fine. Now, they could be further optimized, sure. But the default of taking no conscious action at all has historically worked out just fine for linguistic evolution.
- Stupulous 5y ago
  In the last half of the last millenium, we achieved widespread literacy, invented the printing press, made publishing available to the public, and invented the internet. I don't think we can assume the process that worked before this tech will definitely work now.
  - Stupulous 5y ago
    On reflection, I'd like to retract this. I was reading quickly and didn't understand what I was replying to. It is not my belief that, left to its own devices, natural language will become unusable. It is probably true that we can get away with letting nature run its course. I do object to using this as an argument for avoiding optimization, but I don't think I'm in disagreement with the parent there. Apologies for my error.

selestify 5y ago

> I think the position that natural language will just take care of itself is just a touch hand-wavey. Why not? It's done just that for countless millenia. Cavemen without Ph.D.'s in linguistics managed to develop languages for communication just fine. Now, they could be further optimized, sure. But the default of taking no conscious action at all has historically worked out just fine for linguistic evolution.

Stupulous 5y ago
In the last half of the last millenium, we achieved widespread literacy, invented the printing press, made publishing available to the public, and invented the internet. I don't think we can assume the process that worked before this tech will definitely work now.
- Stupulous 5y ago
  On reflection, I'd like to retract this. I was reading quickly and didn't understand what I was replying to. It is not my belief that, left to its own devices, natural language will become unusable. It is probably true that we can get away with letting nature run its course. I do object to using this as an argument for avoiding optimization, but I don't think I'm in disagreement with the parent there. Apologies for my error.