Dan Fabulich

The word "cool" has been cool since the 1920s, and it's still cool today. "Cool" has outlived "hip," "happening," "groovy," "fresh," "dope," "swell," "funky," "bad," "clutch," "epic," "fat," "primo," "radical," "bodacious," "wicked," "ace," "bitchin'," "smooth," and "fly."

Google Play’s new “discount offers” will charge higher prices in older app versions danfabulich.medium.com/google-plays-n… Users who don’t upgrade your app will pay the wrong price for in-app purchases @AndroidDev @GooglePlayBiz

@getSendy I'm a Sendy customer. I've been emailing you at hello@sendy.co all week trying to notify you about a critical security issue. Are you getting my emails?

@googledevs There are only two workshops on the schedule, both of them for AI. Are there no workshops at all for Android, Web, or Cloud?

Big updates are coming at #GoogleIO, May 20–21. Get a first look at AI advancements, Android tools, new ways to build across web and cloud, and much more. Explore the program ✨ → goo.gle/3EHjCWY

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat Why manage adapters in N frameworks, which all the frameworks are forced to be on the hook to support consuming WCs? It’s a prisoner’s dilemma in which legacy companies are defecting, making life worse for FWs and hurting their user’s performance.

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat As for missing the point, consider this: I think you might be right that some legacy teams would pick WCs for lead components in a design system because they reach the best cost/benefit ratio, but only because they assume that Ryan, Rich, Evan, and Meta will pay their taxes

FWs play an IMPORTANT role. Custom elements don't solve the same issues. We still need FWs. I think FW authors see WCs as a tax cause they wanna support compiling to CEs... but I don't think WC fans care about that.

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat WCs are the worst thing that can possibly work. They work as poorly as zebra striping frameworks, even when they're "working." Even in legacy land, there are superior approaches. (6/6)

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat What should those design systems with leaf nodes do? Design components in a lightweight framework, and maintain wrappers in the N frameworks they need to support. Consider rewriting the component N times for N frameworks, and/or transpiling. (5/?)

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat But if you're just composing multiple components from multiple architectures alongside each other, where each one controls its own top-level sibling div, every framework can handle that trivially. WCs again bring nothing to the table.

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat There's no utility for that use case at all. Everyone wants to own their entire subtree, and no one more so than Shadow DOM. If you want to zebra-stripe legacy X with legacy Y, with legacy Z all you can do is write adapters, and they all suck, WCs perhaps more than most.

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat Framework-to-framework adapters aren’t pretty, and don’t support fine grained reactivity, but neither do WCs! And WCs have that attributes/properties problem, making it pretty gross to pass complex data structures from one WC to another.

@WickyNilliams @Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat I think even there, the advantages are overblown. WCs do make it possible to mix frameworks, but they don’t make it performant. (You’ll have duplicative frameworks!) And the frameworks themselves can all embed other frameworks with adapters.

@Rich_Harris @code_e_pendant @LeaVerou @claviska @youyuxi @RyanCarniato @justinfagnani @zachleat I think the mistake is the opposite direction: people assuming that WCs must be valuable because they’re standardized. @LeaVerou has called for us not to throw out the baby with the bathwater, but there is no baby. It’s nothing but taxes, just so we can be “standard.”

I mean sunk cost fallacy is a thing, right? We can't tear stuff out because the web is the web, but we don't have to keep doubling down on something with such deeply embedded flaws. What I'd most like to see is a bit more humility from the people pushing this stuff, and a recognition that something being a 'standard' isn't in and of itself a reason to embrace it.

@patio11 @jasoncrawford IIRC, your AdWords spend was break even before you began running A/B experiments. The experiments turned rapidly into four figure and then five figure cash flow. This is not how it usually works. Why did that happen to you and nobody else I’ve heard of?

@patio11 @jasoncrawford My point is: 2009 Bingo Card Creator was a vaguely similar firm. (Bingo is a game, too.) And yet somehow BCC made money on Google AdWords, enough to feed your family, unlike any indie game dev I’ve ever heard of. Why?

Targeted advertising is good actually

@patio11 @jasoncrawford For example, I have hundreds of games for sale on the Google Play Store. You'd think I could run a Google ad targeting ppl who paid money in one of my other games, inviting them to buy my new game. But I'm always outbid by mobile casinos, who apparently don't care about targeting

@patio11 @jasoncrawford It still baffles me that this worked for you. I've tried for years, but I've never run an ad on Google or Meta that even came close to break even. It's widely understood in the indie game industry that there's no way to make money buying ads, and that everyone who tries it fails.