alanlamm

Criticism and downvotes u have received aside, I like the idea! consolidating the recommendations adds value. This is smthg I would use.

Agree - in Brazil I also saw a few non-compliant paid political ads that should have been taken down, and there has been significant controversy re. professionally generated fake news on Whatsapp & fb having tilted results

Agree. You see this clearly with videos on there that are hard to close - bad UI, but surely bumps up their total hours of video served stat.

But I I get the odd feeling using Facebook that the content is not even so data-optimized for engagement (because surely if that were the case I’d find my feed more interesting than it is) but rather that it is optimized for revenue in ways that are actually negative in terms of real engagement. Example - if I see an interesting article and stop 5min to read it, that’s real engagement for a significant amount of time, but I might see only a couple of ads in that time. Whereas If the content is really dull and I’m just scrolling mindlessly, I’m seeing a lot more ads, and the gap between how interesting the ads are and how interesting the non-ad content is is much smaller.

Does anyone else get this feeling? Ie that facebook is almost intentionally boring?

Also important to have the date when the video/screenshot was taken and be able to filter by it.

As a small indie app dev - Great resource! Will use it. My thoughts -

1. Agree that use of space on screen could be improved. 2. Need to be able to rewind/replay videos to see interactions well 3. One way I think about patterns is the following 4 buckets

- consensus among whatever major apps like fb, google are doing whether u like it or not, because that defines the ‘standard’ people are used to, and there is a good chance it’s backed by a ton of UX research $$.

- novelties / cute details / visuals/ conceptually elegant design that add differentiation & perceived value, but on the other hand often suck up dev time and end up not being the ideal in terms of functionality.

- design that adds usability, even if it is considered not elegant or redundant - E.g captioning icons

- design that nudges towards intended behavior.

Of course great design does all 4 (HQ trivia is a good example), but more often there is a real trade off and a lot of the design pattern sites I see focus too much on #2 and a bit of #1 - at least I think it would be a really useful segmentation to have e.g “we’re showing you this pattern because it looks cool & different” vs “this has become standard” vs “this is simple & works really well”

4. a more comparative view of the apps (A does this this way, B does it this other way, etc) would be more useful than the individual ‘A does this, isn’t it nice’. I realise some comparison is possible if you select a tag (eg settings) then scroll through the apps, but not ideal yet

1yo Brazilian full feature length film that had a cinema release. 52k views - example of a professional but small creator being pirated. https://youtu.be/ZvQljiVg-xU

Another Brazilian feature film about 6yo - 3.4M views on a profile whose name translates to “watch films online”, with a FOX intro at the beginning. https://youtu.be/r8KGMyfXXHo

1h old Showtime comedy special taken from TV, 1M views - https://youtu.be/Wo27OxaEB3w

If this is above board, what am I missing?

No, I don’t have evidence. But I find it unlikely that HBO wants grainy versions of full TV episodes (eg filmed from a VHS recording playing on a TV) on the page of some random YT uploader (not an HBO page or influencer or etc) even if somehow in exchange for tiny CPM royalties. That is only one of several examples I have seen. Perhaps in the case of HBO they could be doing a better job given their size & resources, but for smaller creators it would be a challenge and why should the burden be on them anyway?

I regularly see a number of videos of clearly copyrighted content that have been up for years, with several million views, for fairly mainstream search terms and that would be relatively easy to filter algorythmically (real example - via the characteristic sound & image of the HBO intro in a pirated tv episode). Hard to look at that and believe that Google is doing everything it can to bring them down. That they reject copyright violation reports that come from anyone other than the actual rightsholders (sure, they’d have to make sure someone is not reporting by mistake - it is not a perfect, but still a valuable signal, and if they really wanted to they could encourage it) is further evidence. Further, where these specific videos are not monetized via ads, they bring traffic to YT which is then monetized in other videos, therefore this pirated content is being used commercially, even if indirectly. Like many here I’m no fan of copyright law, but that is no excuse for a multi-billion listed company to make a fortune deliberately flaunting it and then play coy.

Try doing that as a non-multi-billion dollar company and your app gets rejected. The guidelines specify that in-app-content MUST use the Apple in-app-purchase system

Surely Google gets more data from Android. But it does get much more data on iOS than Apple would have you believe. Case in point - Apple ostensibly prevents apps from uniquely identifying devices other than through ‘safe’ ad and vendor-specific ids. But just implement Google Firebase Anonymous Authentication in an app and try to get it to forget your id. you can delete the app and all other apps from same vendor, reset your ad id, set ‘limit tracking’ delete all google apps, clear google logins & passwords on your device, update the iOS, and when you reinstall it still knows who you are. Short of wiping the phone completely (which I haven’t tried bc of the hassle) Google appears to not only uniquely identify your device in a pretty sticky way, but to sell & give away this capability to anybody who wants it.

One issue Im surprised never comes up in discussion of buybacks vs. dividends is, apart from the tax issue, whether buybacks drive stock prices up because of how large orders to buy stock at market price interact with the outstanding order book - ie that in a buyback you buy specifically from those of the current shareholders who attribute the lowest valuation to the stock at any given time. To simplify lets say 2 mutual funds each hold 50% of the stock of a corp. Mutual Fund A values the stock (ie is willing to sell at) $100, and mutual fund B is willing to sell at $120. Other potential buyers value the stock at $99 or less. The company decides to buyback 50% of its stock. It buys all of A’s stock, and thereby drives the new market price to $120. Meaning it drives up the market price even before you consider the effect of reducing the # of outstanding shares (fewer shares > higher EPS > higher stock price at a given P/E multiple). Of course, in the real world stockholding is less concentrated, the gap between valuations in the order book wouldn’t be so wide, shortsellers would also play a role and over time one would argue that efficient-markets-hypothesis factors in. But still... any thoughts?