The decentralized nature of the web, as it is, is under serious threat from initiatives like the Accelerated Mobile Pages (AMP) project spearheaded by Google. Search engines are becoming content repositories for large sections of the web. Many major web publishers have already ceded control over their own websites and handed over a treasure trove of data about their audiences to Bing and Google.
The Redirect AMP to HTML extension for Firefox, Microsoft Edge, and Google Chromes gives users a choice to opt-out of a centralized web in favor of the open and decentralized web.
Just about everything about AMP is controversial from how it’s trying to supplant the web and potential for censorship to technicalities like its URL structure and compliance issues with privacy and even copyright infringement issues.
How does anyone legally justify copying and redistributing so many webpages without a distribution license with any of the publishers? I’ve explored this topic extensively in a new article.
AMP Caches copies the entire AMP document that a publisher produces for their website and redistributes it directly from servers under the search engine’s control. They’re like central repositories of web content from participating websites.
This mostly cuts the publisher’s own infrastructure out of the content delivery process, and also hands over information of everything that happens on those pages to the company that operates the AMP Cache.
Regular websites can’t compete fairly with AMP directly on the platform’s main selling point: page loading speed. AMP has an unfair advantage as search engines will preload AMP pages before the user even clicks on them in search result pages; resulting in a great increase in perceived page loading performance over regular webpages.
As a web publisher, I feel the pressure to adopt AMP to please Google and their algorithms. Many web publishers are on board with AMP. This website already uses many of the techniques that AMP cache providers use.
However, I can’t compete with already-having-been-loaded-in-the-background or out-compete the likes of Microsoft and Google on infrastructure. (Although my CDN is trying their best.)
I believe that the web should be decentralized, that Google doesn’t need more information that it already has, and that the limitations that AMP imposes on web developers are outrageous.
In my opinion, the right direction for the web is refocusing on reducing bloat and more conservative web development. The incentive and monetization system needs to be addressed while we’re at it. Giving up and moving everything to Google’s servers isn’t the answer (except if you’re Google, of course.)
As no one else took the time to ask web users what they wanted, I’ve created an extension for Firefox that lets you opt-out of AMP. Just install it and it will point you to the regular web version of any AMP page your browser encounters (when available.)
Using the extension will help drive eyeballs (and ad revenue) from AMP and over to the publishers’ websites where it belong. It sends a signal to publishers that we want the web to remain open and decentralized rather than under the control of a few companies that offer “free” caching servers for “everything”.
The extension is especially useful if you install it on Firefox For Android as you’re most likely to encounter AMP pages while searching with Google on a mobile device. The extension is also useful on Firefox for desktop if you regularly share links from your mobile to your desktop device, or if you frequent social media websites (where many users unknowingly share AMP links) on your desktop computer.
Note that in some cases the AMP page will still be loaded before the extension has the opportunity to identify it as an AMP page and redirect you to the regular web variant of the page. The extension isn’t suited for people on an ultra-tight, slow, or expensive data quota. The current version of the extension doesn’t disable preloading of AMP result pages on Google Search even those pages won’t ever be displayed.
The Redirect AMP to HTML extension is an open-source project and you can report issues or get the source code on GitHub.