One point he’s making, that many people miss, is that there was a time when people learned to memorize large bodies of information for ready recall. Note the oral traditions of indigenous peoples, the exam system in historical east Asia, or even the way western schools used rote learning until recently.
Socrates is partially lamenting the expected loss of this super ability. And he was right: it is rare to memorize things reliably now since we have so much reference material
I think he underestimated how many ever-changing programming apis we would need to “learn” today — otherwise he might have more sympathy for the use of reference material
> Socrates is partially lamenting the expected loss of this super ability.
This 'super' ability, while impressive, offers far too little benefit for the effort necessary to achieve the memorization. A scholar who has a bunch of notes and books at his disposal will almost always outdo a scholar who has memorized 1% of the material contained in written form. That's why we don't memorize things and rely on externalized storage of knowledge. It is far more effective.
Humans aren't smart enough to parse complex things without a ton of memorization. You can look up knowledge, but you can't look up how to look up a specific set of knowledge. Expanding that core is still necessary no matter what age you live in and requires you to remember things.
So the scholar who focuses on looking things up might be more effective today, but he wont get any deep knowledge in any domain, while the scholar who focuses on putting things in his mind will probably be less effective today but get much deeper understanding of domains. For deeper domains like math or physics the only way to get anywhere is to embed a ton of knowledge in your brain, there is just no way to get anywhere with shallow look ups.
The lookups aren't necessarily shallow. They are as deep as is necessary for the scholar at his current task.
The undeniable truth is that if memorizing everything Homer-style was effective in sciences, people would be doing it, achieving great results. But it doesn't happen, and it is easy to understand why: It is not effective.
In our field, we don't need to remember every single parameter of an API we use, we just need to know where to look when the necessity arises, while we focus on more important things.
In addition, a scholar who has a bunch of notes and books at his disposal will actually end up remembering useful parts of it all too. However, the scholar will wasted significantly less time rediscovering what he vaguely remembers and bothering colleagues whether they do happen to remember.
Yes, because memorization will happen naturally, of the things most important for such scientist. But if such scientist would exert himself to remember everything verbatim, he'd end up remembering only a tiny bit of information and will likely end up accomplishing nothing else.
Socrates is partially lamenting the expected loss of this super ability. And he was right: it is rare to memorize things reliably now since we have so much reference material
I think he underestimated how many ever-changing programming apis we would need to “learn” today — otherwise he might have more sympathy for the use of reference material