No. 155: 👔 73% of U.S. Workers Plan to Freelance This Year - Fiverr
+ 🎧😬 Podcasts aren't so smart
Hiya,
Happy Sunday.
It’s a sunny, frosty day here in London.
Perfect weather for revisiting yesterday’s hi, tech. special about Microsoft Azure and OpenAI:
As we will see below, ChatGPT has a lot more human involvement than you would probably think.
Elsewhere, Fiverr has conducted a nationwide survey of 2,000 U.S. workers in partnership with Censuswide.
66% said they’re already on, or plan to join, a freelancing platform in 2023.
That’s a sign of a tricky economy, for sure (53% feel less secure about their income vs. 2022) - and also an acceleration of an ongoing trend. People want to have more control over their labour - and the proceeds from that labour - to soften the blow of potential layoffs.
Just as significantly, they want the flexibility to deliver their work remotely.
This time in hi, tech.:
Amazon closes its Charity “Smile” program - but why did it even start it?
Reddit launches Community Points
The human cost of ChatGPT
Why podcasts aren’t so smart
Amazon Closes its Charity “Smile” Program
Amazon says “the program has not grown to create the impact that we had originally hoped” and, well, the only way to understand our world is to stick on your cynical hat.
I saw this explanation the other day and it really is worth reading:
In case you didn’t read it: the Smile program was a way of forcing users to go direct to Amazon.com, cutting Google Ads out of the loop. Amazon paid a token amount to the charity; much less than they would have paid to Google for the same traffic.
For what it’s worth, I have a grudging admiration for this. It is a smart tactic and if the end result is that charities get something and Google makes less, so be it.
Reddit launches Community Points
Each sub-reddit can create its own community token and reward creators for their contributions.
It’s an intriguing development, because Reddit is building from a position of strength here. It already hosts vibrant communities and these rewards could elevate individual sub-reddits to a higher status. Plenty of other companies have put the cart before the horse by introducing meaningless “tokens” on top of limited substance.
Some of Reddit’s ideas are gimmicks (buy coins and gifs), but the communal voting and rewards for creating improvements to the group could incentivise innovation.
ChatGPT’s success comes at the expense of poorly paid Kenyans
If a technology looks like magic, you can bet some poor people’s labour has been exploited to conjure the trick.
A horrible company called Sama, which has been used by Meta to exploit workers in Nairobi, contracted Kenyans to remove toxic content from ChatGPT’s training content. Needless to say, they were not paid or trewared well, and:
After the project of removing toxic content on ChatGPT was completed, Sama closed its Nairobi office, rendering all 200 content moderators jobless
📚 hi, tech. is also reading
Podcasts aren’t as smart as you think - FT
This is one of my favourite articles this month, because it says what I think. Look, I might as well be honest about why I’m sharing it here.
If you’ve read hi, tech. before, you’ll know that I don’t believe learning should be easy. I don’t even think reading a newsletter should be easy. If it’s too easy, you see, the knowledge won’t stick. My dense prose is for your own good.
The FT article reasons that we do all want to learn and improve.
We don’t want to put in any effort, however.
This demand for quick-win knowledge is met by the flim-flam machine of superficial supply.
That's how we end up with snappy content that promises to do the hard work for us. A quick shot of rejuvenating espresso, not a flask of viscous Bovril.
And, of course, we end up with podcasts.
As the author notes:
Naturally, there are multiple ways to learn and if they work, they are valid.
Some of us will retain more information from an impactful documentary series than we would from reading a dry book, for example. There are excellent podcasts that use a discussion format to probe at ideas in a way that original texts rarely do.
If there is a hierarchy, it should be built on the strengths of each medium to improve the recipient’s mental acuity, not on its perceived “class status”. There is room for them all, but to replace books with podcasts is to replace a Rembrandt with an NFT of a Rembrandt.
That’s why I would ask:
What are the lessons that have stuck with you for years? How do you learn best?
I’m sure I am not alone in saying the following:
My mind’s landscape was carved at university, where I had to read and read and read, then write, then defend what I wrote, then listen to other perspectives and discuss. And then, do it all again the next week.
I remember those books remarkably well.
I don’t quote Enlightenment figures in hi, tech. just because it’s cool to do so. It’s also because they’re always chattering away in my head.
I can recall fun facts from the history podcasts I listen to. The rest skims across the surface of my consciousness, “Or like the snow falls in the river, A moment white--then melts for ever” to quote Robert Burns in a timely fashion.
Dare I say it, I think I even listen to podcasts for company at times. In the same way that people have the TV on constantly, even when not actively watching it, I will have a podcast on that I’m only partially tuned into.
Contrastingly, I am reading what I am sure is one of the best books ever written right now: An Immense World, by Ed Yong. I read it slowly, often going back to re-read passages, and I could tell you pretty much everything that happens in the book. If you’ve been near me recently, you’ll know this to be true.
In fact, my sister called round to see us at our new flat last week. I was standing in the bedroom and looked out the window to see she was on the street, looking back up. As we went for a walk, she asked why I was gesticulating so wildly to my partner when she saw us through the window.
“I was telling her about my new animal book”, I replied sheepishly.
It’s that good. It’s about how our senses create our world, or our Umwelt, as the eggheads call it. We are extraordinarily good at completing tasks - when we are focused.
The same applies across the animal kingdom. Bats can use sonar to navigate the most intricate obstacle courses at high speed. They also fly straight into cave walls when they get lazy.
The main point here is that knowledge requires focus and it requires focus because it is worthwhile.
The medium matters.
Books (and newsletters) demand focus in the way a podcast does not.
Light entertainment rarely delivers a weighty punch.
I am intrigued by new ideas on this front, nonetheless. This week, Prezi announced some new interactive graphics that tutors will be able to add to lessons.
Sure, books are great. They demand that we engage; they offer “guided creation” as Sartre put it (yes, I studied him at uni), but we must join the dots for ourselves.
That doesn’t mean our only recourse should be to return to the library, wonderful though that would be. Instead, we can interact with content using new technologies. The key is to focus on the substance first, then use the style as a necessary means of delivering the former.
Many podcasts are a placebo, but audio will play a key role in future pedagogy.
As long as we are ready to focus.
I'm a little baffled at this idea of podcasts as a substitute for reading. It's a different medium. You might as well talk about the TV as a substitute for reading - no doubt people have, but to me that completely misses the point. It's a different method of information delivery with different goals and appeals.
I'm a huge book lover, but not everyone is. If books turn them off, why can't they listen to podcasts? Or, shock horror, do both xD I find it hard to make the time for podcasts because I'm rarely in a position where I'm doing something that's rote or mindless enough that I can concentrate on what I'm listening. I prefer reading because I can control the speed of it, but I have huge affection for a couple of podcasts I listen to, including 'We're Not So Different' (a mediaeval history podcast) and 'Lingthusiasm' (a linguistics podcast). And I have learned a ton from both, especially WNSD. Maybe that FT writer is just listening to bad podcasts?