[HN Gopher] Algorithms are changing what we read online
___________________________________________________________________
Algorithms are changing what we read online
Author : pseudolus
Score : 39 points
Date : 2020-09-14 15:47 UTC (7 hours ago)
(HTM) web link (thewalrus.ca)
(TXT) w3m dump (thewalrus.ca)
| carabiner wrote:
| The algorithms are motivated by ad impressions. Solution: seek
| out newspapers that are funded by subscriptions, not ads, the
| ones with stiffest paywalls. These would be Financial Times and
| WSJ. Chomsky called the FT "the only newspaper that tells the
| truth" because the readership and authors are aligned in their
| goal to make money from the news.
| zepearl wrote:
| I personally never read articles of FT nor WSJ - I don't even
| try to "click" on those links/references as I know that I won't
| be able to read them => I'll never have any chance to
| "understand" if I would like to subscribe to those newspapers
| :(
| rtx wrote:
| As a consumer, I enjoy and pay for human curated content. Maybe
| publishers just need to ride out the analytic craze.
| octodog wrote:
| I quite enjoyed this article and found the author's perspective
| on the inherent internationalism of art resonate with me.
|
| However, I'm not sure that the issue he is discussing is mostly
| caused by algorithms. It seems to me that the
| internationalisation of newspapers and media organisations is the
| main culprit. As the author says, if you can read art reviews in
| the NYT, why bother with 'small time' critics? Personally, I
| share the author's view that having a local perspective is still
| valuable but it seems clear that a lot of people don't.
| jseliger wrote:
| I saw this essay in Arts and Letters Daily
| (https://www.aldaily.com/) and added it to my blog's next links
| post, with a variation of this comment: I've never heard of this
| guy and yet his work sounds like just the sort of thing I'd like
| to read: I'm not interested in most of the standard political and
| pop culture stuff that's endlessly written and re-written.
| Unfortunately, he doesn't seem to have a link list of his recent
| works anywhere, at least that I can find. His website
| (https://russellsmith.ca/) appears to be pretty generic, and its
| RSS feed seems to have last been updated in 2015. How are we
| supposed to find his work and follow him, if we are interested in
| his work (and I am, after reading this)?
| motohagiography wrote:
| Can recommend highly, particularly his books, "Confidence" and
| "How Insensitive." I read his column in the Globe for years.
|
| When you compare what he writes to the content that gets pushed
| to us like in that Social Dilemma documentary in another thread
| (https://news.ycombinator.com/item?id=24468533), smart people
| are writing, they're just getting drowned out. Content
| algorithms are like rating food on how efficiently it can
| deliver sugar and caffeine with no other criteria for "good."
| mellosouls wrote:
| This article rings true to me.
|
| In particular I've noticed a steady stream of woke articles being
| pushed my way despite me having little interest in that sort of
| thing.
|
| These things are supposed to be fed by our preferences, but while
| I'm happy to read across the board brow-wise, most of my reading
| is of the deeper articles and yet genuinely intellectual essays
| are rarely offered up.
|
| It's frustrating as a reader, and interesting to hear about it
| from the view of the writer.
| AlexandrB wrote:
| I know what you mean. After all the tracking data collected and
| ML models trained the primary metric for suggesting content
| still seems to be "stuff other people are
| reading/watching/playing/listening"
| jacobr1 wrote:
| As someone that used to work on recommendations systems ...
| the problem is that the signal is just so robust. Stuff other
| people like really is one of the stronger ways to predict
| what you will like. Most additional data sources that allow
| us to customize further and identify what sub-groups you
| might belong to and thus prioritize highlighting things of
| relevance to your subgroups tend to empirically show lower
| performance for the overall population. The only thing I've
| seen work is focus on the "most popular" things for the most
| important demographics (such as those with money to buy
| things) which might overfit for that demo, and underperform
| overall, but still performs the best on some higher metric
| like revenue.
|
| I commented a while back on someone asking "Why recommend me
| a ladder if I just bought a ladder, couldn't the system
| understand that people usually only buy one ladder?" But the
| fact is, the commenter was wrong about his priors. Actually
| someone who buys a ladder is indeed more likely to buy
| another one ... even if most people that buy ladders only buy
| a single ladder. The fact that you bought a ladder, puts you
| in the cohort of people that buys ladders, which means you
| are more likely to buy another one than someone who isn't on
| the record buying a ladder.
___________________________________________________________________
(page generated 2020-09-14 23:00 UTC)