3 ms·
Reading "always from authors of proven worth" raises the question of how new authors might come to develop proven worth. Most, obviously, wont. Some, however,
by dredmorbius 5y ago
Reading "always from authors of proven worth" raises the question of how new authors might come to develop proven worth.
Most, obviously, wont. Some, however, will. Seneca himself was one of them.
But this remains an interesting problem and question:
- How do we assess quality?
- How do we cope with a superfluity of new information?
- What are the costs of failed assessments and promotion (and there will be failures).
- How many 2nd chances should be afforded to those who've betrayed trust, and worse, attention?
- How do we cultivate new and good interpreters and fabricators of the universe we find ourselves in?
I often call this the vetting problem. There's an essay by Arthur C. Clarke in which he discussed "the servant problem, Sri Lankan style", of how to offer (and interpret) letters of recommendation for household help. (The post-script to the essay is both amusing and insightful.) This appeared in one of his essay collections, though I don't recall which one.
It's also a problem in any activity in which agents and agency exist: the principle-agent problem, regulatory capture, corruption, infidelity, betrayal, and the like. Initial assessments are useful but not definitive; ground truths may change and mistaken assessments may prove durable and even potentially valid. Some truths are self-generating ("if you want to have trust, give it", and rather similarly for distrust), some are self-defeating.
- kashyapc 5y agoThank you for the essay reference. You raise many excellent points. I don't have answers to all the nuanced questions you ask, but on evaluating quality: my approach is to research how involved and "invested" the author is on the topic they're talking about. Also, I'll simply read some articles and chunks of paragraphs from their book to get a sense of tone and competency. I realize these are all value judgments of some kind; but it is inescapable. A journalist writing about "the benefits of neuroscience"? I'll skip it. I'd rather put the likes of Antonio Damasio, Robert Sapolsky, or Michael Gazzaniga—all of whom have spent decades studying cognitive science—under the rubric of "proven worth". Also, their writing is accessible for the non-specialist and specialist reader alike. Note though, this doesn't mean that the journalist's writing doesn't have any value. It simply doesn't meet my bar of quality (again, value judgement). Another rough rule of thumb is, usually, books from academic presses have better value-to-noise ratio than popular presses. Even this is slowly degrading, with universities having heavy-duty marketing departments to drum up their books.
- dredmorbius 5y agoGood points as well. I'm somewhat hesitant to invest too deeply in a now-stale HN discussion (a site dynamic I ... dislike), but another element I use, and one that addresses TFA's focus, is what I've been calling "BOTI": Best of the interval. That borrows from a 43 Folders / Ticker File model (or more generally, a round-robin database). The idea is that after a given interval (hour, day, week, month, year, decade ...), you review a list of possible candidates and select which you think are best. Ideally, there's some opportunity to revisit earlier potential candidates, and there should be an option for hidden gems to enter the pile as well --- items originally passed over but on consideration found deserving attention. That's the basic idea. In practice, review itself turns out to be difficult, though there are shortcuts. On HN (using Algolia), you can enter a blank time-limited search to return the top submissions of the previous day, week, month, or year. Or specific earlier periods can be constructed (I've posted a shell script that will generate the URLs for this using Algolia's seconds-since-the-epoch timestamps). HN's choices aren't always optimal, but they're OK, and are cheaply accessible. I'd like to have a number of filters and systems I could tap similarly. The other element I'd like to include is reputation-grading sources. That is, which consistently turn out high-quality content, and which either don't, are variable, or straight up contribute noise?[1] So sources (authors, editors, translators, other contributors, publishers) accrue a reputation over time, and can see that admusted. What happens with repeated selection, filtering, and assessment tends to be quite high quality. That said, there's still far more published in a year (or posted online in a day) than could be read in a lifetime. Roughly 1 million books/year, and 5 billion Facebook items/day.[2] A well-read person likely reads on the order of 10 books per year over the course of their life (some read more, many less).[3] An interesting question in an age of abundant ... culture / entertainment / literature / drama / documentary ... becomes what constitutes a common cultural heritage or touchstone? Is it ... the Marvel Cinematic Universe? Star Wars? Shakespeare? The Bible, Koran, Torah, Tao Te Ching, or Upanishads? And how criticial is it that there is, or is not one? (I suspect that this was the bigger question that TFA was leaning at but never quite got 'round to.) My own awareness of my limited capacity to assimilate culture occurred as I walked into my univerity library the first time. Its local phyisical collection then numbered about 5 million volumes (with 15--20 million or so across multiple affiliated campuses), and ... I was never going to access more than a small set of those. What I realised and discovered was that a mix of intentional seach and serendipitous sampling could prove quite rewarding. Some sections of that collection I came to know fairly well, most I utterly ignored. ________________________________ Notes: 1. Some noise is tolerable, and none probably means a source isn't trying. But an indifference or insensitivity to error also indicates a source is highly inconsistent, if not outright deceptive. I'm inclined to the view that intentional noise and deception are quite probably worse than no information (especially if the source is aware of how to mask that deception). Dante put liars in the penultimate circle of Hell for a reason. 2. Just for grins: if a typical FB post contains 120 characters of text (I'd found similar results on Google+ a couple of years ago, and Twitter-length posts seem ... frequent), and we have 6 characters per word, then 5 billion items -> 100 billion words. At 500 words/page, that's 200 million pages, and at 250 pages/book, about 800,000 books' worth of content posted to Facebook daily. (How much of that is original or nonduplicate content is another matter, and much of it, likely about 1/3, is predominantly visual w/ images or video). By comparison, "traditional" publishers in the English language were publishing about 300,000 books/year from the 1950s through the 2010s (Bowker, assignee of ISBNs, keeps statistics). 3. At a book a day, from age 10 to 85, or a reading career of 75 years, an ambitious reader might complete just under 28,000 books. Or less than one tenth of one year's traditional publishing output.