Hacker News Reader

You misunderstood him. His point was this - say in 1900 they made 100 kettles, only 10 of which were well made. Those 10 were passed down as heirlooms while the other 90 were consigned to the scrap heap. Now 117 years later people go looking for old kettles, look at those 10 and say "wow, they don't make 'em like they used to".

In reality we're making them possibly better than we were earlier.

Some percentage of kettles were not well made and did not last. All the kettles that did last were well made. It's wrong to look at the population of kettles that lasted, observe that they are all well made, and conclude that all kettles made long ago were therefore well made.

i think you are missing their point, which is talking about selection bias