3 ms·

I really wanted AMP to be to HTML/CSS what asm.js was to Javascript, a subset with better performance. What other alternatives are there? The Gemini project se

I really wanted AMP to be to HTML/CSS what asm.js was to Javascript, a subset with better performance. What other alternatives are there? The Gemini project seems too barebones and requires its own browser.

dheera 4y ago
I actually kind of liked AMP because it regularly circumvented paywalls and those fucking annoying newsletter popups.
- waynesonfire 4y ago
  Makes sense. That must have been the "loss of control over user experience" that publishers complained about.
- arcticbull 4y ago
  Unfortunately it hijacked scrolling, navigation and search which more than made up for the lack of annoying newsletter popups. I personally use the Amplosion plug-in for Safari on iOS which nukes AMP pages from orbit transparently, substituting their standard variant. [1] [1] https://amplosion.app https://amplosion.app
- busymom0 4y ago
  I use archive dot org and archive dot is for bypassing paywall. The dot org one works best usually in sites with JavaScript.
- AlexCoventry 4y ago
  I liked it because it dangled the possibility of author signatures on content, via RealURL.
tomcam 4y ago
> What other alternatives are there? 1. Fantastic content. If you have that pretty much nothing else matters. 2. Hand-tuned CSS and HTML
- hyperhopper 4y ago
  Also, a lot of AMP isn't even bad. You can self-host the js for it, follow the spec and restrictions, etc, and still get all the speed upsides except for the js being cached from Google. I've done this on many personal websites and it is a easy brainless way to get a fast simple site.
shakow 4y ago
> What other alternatives are there? Correctly design your website. Website obesity is a human problem, not a technical one. If asm.js solved a technical problem, AMP is basically just a very strict and unavoidable linter.
MaxBarraclough 4y ago
The kind of people who publish on Gemini are the kind who already made their website fast. The way I see it, Google just need to penalise bloated websites far more harshly. There's no real technical challenge here - we already have fast browsers. The challenge is one of incentives.

dheera 4y ago

I actually kind of liked AMP because it regularly circumvented paywalls and those fucking annoying newsletter popups.

waynesonfire 4y ago
Makes sense. That must have been the "loss of control over user experience" that publishers complained about.
arcticbull 4y ago
Unfortunately it hijacked scrolling, navigation and search which more than made up for the lack of annoying newsletter popups. I personally use the Amplosion plug-in for Safari on iOS which nukes AMP pages from orbit transparently, substituting their standard variant. [1] [1] https://amplosion.app https://amplosion.app
busymom0 4y ago
I use archive dot org and archive dot is for bypassing paywall. The dot org one works best usually in sites with JavaScript.
AlexCoventry 4y ago
I liked it because it dangled the possibility of author signatures on content, via RealURL.

tomcam 4y ago

> What other alternatives are there? 1. Fantastic content. If you have that pretty much nothing else matters. 2. Hand-tuned CSS and HTML

hyperhopper 4y ago
Also, a lot of AMP isn't even bad. You can self-host the js for it, follow the spec and restrictions, etc, and still get all the speed upsides except for the js being cached from Google. I've done this on many personal websites and it is a easy brainless way to get a fast simple site.

shakow 4y ago

> What other alternatives are there? Correctly design your website. Website obesity is a human problem, not a technical one. If asm.js solved a technical problem, AMP is basically just a very strict and unavoidable linter.

MaxBarraclough 4y ago

The kind of people who publish on Gemini are the kind who already made their website fast. The way I see it, Google just need to penalise bloated websites far more harshly. There's no real technical challenge here - we already have fast browsers. The challenge is one of incentives.