[linuxandrew]

Google/Blink consistently pushes for bad things like EME, FLoC and Topics API.

They have a very different vision of the web where users live in a corporate playground and complex browser engines which only a few large corps can manage.

At this point what is making the web a better platform? The web has feature overload, even with features like Server Push which are seldom used. FWIW I think there are some exciting possibilities and new features for the web, especially around the P2P space, but I don't think it's in Google's interests to push for that at all.

[rektide]

It's a huge company of hundreds of competing ideas.

I think most no one has any respect or appreciation for the circumstance flocs & topic Api was raised in. The dogfucker skanks at Internet Advertising Bureau were actively pushing government regulators to replace cookies with some gobshit anti user trash, far worse. I genuinely feel for Google. No one sees or knows any of the other context going down at the time, but all eyes are on the team of like 40 trying to find some way to preserve some privacy, in an org & task that is the hugest fucking lightning rod for attention & negativity, being the most visible & one of the most hated companies on the planet.

I agree that Google seems to have kind of lost the will to fight for a lot of good shit. There's still tons of great Google initiatives, but if someone can't tick it off their OKR within 8 months & call it a raring success, the effort & the team has seemingly no backing, no one with real principle intelligence or spine to keep the really really smart good shit going. That's just not a reasonable time frame for adoption. The web's early adopters take 3 years, minimum, for most interesting capabilities, and there's seemingly no one anywhere with that kind of patience for rolling out. Fuck this industry. This is why we can't have nice things.

[kelnos]

Google is a huge company, yes, but the number of people who drive Blink development is relatively small (of course it is; if most of Google was focused on or even had significant input into Blink, they'd both fail).

I think you have it backwards. It's not that Google has lost the will to fight for good, it's just that Google is a corporation, and their "don't be evil" mantra was only a thing when they were small and it was convenient and profitable for them.

Google has not given a damn about doing "good shit" or about privacy for well over a decade now. Their entire business model is predicated on surveillance capitalism. If they ever truly were the "good guys", they have not been that for most of the evolution of the modern web.

Google is not our friend. They suck, for entirely predictable capitalistic reasons, and their stranglehold overv web standards needs to stop.

[quantumwoke]

While I think your viewpoint is interesting, the language prevented me from appreciating it fully. Would you mind expanding on your opinion on privacy and Mozilla/Apple’s role in the standards process a little more?

[rektide]

It feels shallow to me to let oneself be rebuffed by such petty small barbs. I try to follow the "Highest Form of Disagreement," to find the best interpretations of things first. Rather than let small fry shit cause roadblocks in my mind.

Your ask about Moz/Safari I've said quite a few words on elsewhere in this post & on others. I think the far more interesting topic is what a bunch of jackal villains the IAB is. I cannot stress enough how hard a time Google has had trying to preserve any privacy on the net when there is a huge lobbyist group close to regulators pushing so hard to end user privacy. These people have the worst most anti-user outlook imaginable, are up to absolutely no good. My strong language is a just a start on describing how awful the IAB is & what sinister monsters Google has to go to the mat & wrestle to try to preserve user privacy in a post 3rd party cookie world.

Personally I don't think it's right that I get flagged for my previous reply, but I'm glad to have made a better go at my reply this time. https://news.ycombinator.com/item?id=35565707

[charcircuit]

>pushes for bad things like EME, FLoC and Topics API

EME is better than browsers having to implement their own proprietary APIs for DRM. If EME didn't exist DRM would still be used by sites like Netflix.

>FLoC and Topics API

These are better for privacy than learning interests by tracking via third party cookies. These are moves to retain the positive uses of the web while increasing people's privacy.

>At this point what is making the web a better platform?

WebGPU released recently and provided big speedups to GPU intensive use cases.

>but I don't think it's in Google's interests to push for that at all.

What's the benefits to users or server hosters? Will it improve the user experience? Reduce latency? Save costs? If peer to peer features provide value I don't see why they wouldn't be interested. Peer to peel has its own set of drawbacks so there are many uses where it isn't a good option.

[sanitycheck]

> If EME didn't exist DRM would still be used by sites like Netflix.

(Playing devil's advocate just a little here...)

...and maybe it'd be cumbersome/difficult/annoying for users, opening the door for big changes in the landscape.

Spotify-like streaming services for music (basically the same stuff everywhere, just choose where you get it from) only exist because they had to compete with the ease of getting DRM-free music free from P2P services.

The acceptance of easy standardised DRM for video has led to movie streaming services being the modern equivalent of the old cable networks. You want to watch X? You must subscribe to Netflix. But Y? Y is only available on Disney. Z? Amazon.

Personally I don't care. I've implemented EME and proprietary DRM playback numerous times, and I don't subscribe to any streaming services because I find 99% of TV/movies to be not worth my time. For people who do care though, EME is probably net negative.

[pjmlp]

> WebGPU released recently and provided big speedups to GPU intensive use cases.

On the contrary.

WebGL was going to get OpenGL ES compute shaders, contributed by Intel three years ago. Google blocked the effort with the reasoning WebGPU compute was around the corner. Again three years ago!

Due to politics between browser vendors, we have now yet another shading language to learn, and because Rust is fashionable, naturally it moves away from classical shading languages into a more Rust like syntax.

It arrived six years too late, so it represents Vulkan, Metal, DirectX when they were released in their version 1.0.

Still it doesn't fix the issue that after 10 years of WebGL, there is no reasonable debugging story and no Web games that can match PlayStation 3 graphics.

[rektide]

What an echo chamber. People are so desperate to hate, want to hate. Any outside view is just blasted away on this complex situations.

I do personally think p2p is absolutely key to unlocking a future where users are not so beholden. It creates a much more connected web, versus the ultra-federalized model. Yes there are drawbacks but simply turning our backs on connecting people, deciding the web is just going to stay hosted forever and ever, is a huge denial of galaxies of potential. We won't know how far we can go until we try, until we begin.