I’m talking about the software side of things. Generative “AI” seems to be a “brute force” approach to artificial intelligence - just throwing hardware at the problem instead of finding a better approach. Given the limitations of GenAI, it just feels crazy to keep going this way. Like a sunk-cost fallacy. These are just my thoughts though, not a real scientific analysis.
I have to doubt the cost efficiency too.
It’s strange that the concept of efficiency seems to have been abandoned. Is consumption of vast computing resources no longer seen as indication of a design flaw?
Upvotes/downvotes are unfortunately a fundamentally flawed concept. They originally served as an superior alternative to forums’ previous sorting method of most-recently commented, but they are far from flawless themselves.
My ideal alternative would be some kind of customisable sort order chosen by the user that uses some kind of sentiment analysis of the text to find the kind of posts the user is interested in. For example, you could sort by whether post look serious or joking, how long they are, ratio of words to hyperlinks, etc. Could also filter out ragebait and similar rubbish.
Of course I can see downsides - performance considerations, and it would only work for text posts and comments, but it’s just an idea off the top of my head.
If you want to message them about it, now is the time most likely to work.
Quite the opposite in fact. Microtransactions offer the promise of fun, but never deliver, because in order to incentivise users to purchase them, the player must feel like the game is 90% of the way to being fun and that tiny additional purchase will get it there.
It’s like the cartoon image of the donkey rider holding a carrot on the end of a rod. The donkey keeps moving to try to get the carrot, but never quite reaches it.
Looks like Skylight is app-only, no website.