Live data from Hacker News

The Problem with AMP

80x24.net

271–280 of 353 posts

Re: The Problem with AMP

#271
post #89
post #70

Earlier quoted context omitted.

I've seen ordinary users get frustrated with it, but they don't understand why. They don't know what AMP is, and they don't realize they are on an AMP page. For example, the URL bar says they are on Google but they are reading a Slate article. The web is already confusing enough, and AMP makes it worse. Many ordinary users think they are reading content from Google, like that famous CNN interview where a lady said, "…

> Which leads to the question, why is google doing this? They, you, could easily promote AMP pages while not masking the real URL! Perhaps to allow the content to be served from a CDN (over HTTPS), without requiring the site to CNAME over their domain to Google. If webmasters are willing to CNAME over their domain to a caching proxy, then a less intrusive design is possible[0], such as the one recently announced by C…

The AMP CDN doesn't even really help with caching - when I implemented AMP at launch for some large sites, it was strongly implied that we couldn't expect any reduction in calls to our servers, and we didn't notice any reduction when AMP went live. From my perspective it appeared to be a walled garden counter-measure to FB Instant Articles and Apple News.

Re: The Problem with AMP

#272
post #70

Earlier quoted context omitted.

I've seen ordinary users get frustrated with it, but they don't understand why. They don't know what AMP is, and they don't realize they are on an AMP page. For example, the URL bar says they are on Google but they are reading a Slate article. The web is already confusing enough, and AMP makes it worse. Many ordinary users think they are reading content from Google, like that famous CNN interview where a lady said, "…

i am curious about the numbers, though i do not know internally or externally if there are numbers showing this. As a power android user, i kind of like the rss-reader-esque look and feel of the amp links. It also opens blazing fast. With these in mind, minor inconveniences (like the address bar) are probably something i wouldn't mind.

If only Google had a nice RSS Reader view which was blazing fast, and allowed you to view an uncluttered, non JS version of an article, while still correctly attributing it to the original provider, while making it easy to link to the original source.

Such really simple syndication technology could even have a catch TLA. I can't really think of one, but I'm sure there is something.

Re: The Problem with AMP

#273

I am unfamiliar with AMP, so I apologize if this question is based on a misunderstanding. The article says that all links in AMP pages begin with " https://www.google.com/amp/" right? Isn't this an invitation to bad actors of every flavor to produce AMP pages which will, to the users who know to look at their address bar and note the source domain, appear to be legitimate?

100% and it's already being exploited

Re: The Problem with AMP

#274
post #87

Earlier quoted context omitted.

It's not an AMP thing per se, it's how Google handles AMP in search results.

No, its explicitly an AMP thing. Google serves the user a cached version of your page directly from its servers, hence the google.com/amp/ url

True, but you can serve the AMP page directly from the host.

Re: The Problem with AMP

#276
post #258

Earlier quoted context omitted.

The crazy thing is that AMP is open source and this is still happening. I've been tracking their GitHub and what is happening is that while yes, the source is open, they won't let you add anything they don't like. Oh, and if you try to fork AMP you fail the official validators and suddenly your site isn't AMP anymore. What is happening in this case is that open source is doing nothing to stop AMP from being proprieta…

> I've been tracking their GitHub and what is happening is that while yes, the source is open, they won't let you add anything they don't like. Well, thats basically how Android works. And with the upcoming policy changes (in CDD and GCM) things will get even worse.

Could you tell me about those changes?

Re: The Problem with AMP

#277
post #70

Earlier quoted context omitted.

I've seen ordinary users get frustrated with it, but they don't understand why. They don't know what AMP is, and they don't realize they are on an AMP page. For example, the URL bar says they are on Google but they are reading a Slate article. The web is already confusing enough, and AMP makes it worse. Many ordinary users think they are reading content from Google, like that famous CNN interview where a lady said, "…

i am curious about the numbers, though i do not know internally or externally if there are numbers showing this. As a power android user, i kind of like the rss-reader-esque look and feel of the amp links. It also opens blazing fast. With these in mind, minor inconveniences (like the address bar) are probably something i wouldn't mind.

Your opinion is here is valid, but how would you address the criticism that this isn't solving the real issue, and is obviously a power-grab?

Re: The Problem with AMP

#278

Earlier quoted context omitted.

it does, because we care about a number of things, including: 1. Technical implementation (and alternative ways of implementation). 2. Freedom (keeping things within google's walled-garden or not) Your average user (who doesn't necessarily know anything about software) doesn't really care about 1, and probably doesn't care about 2. They care about speed of the page, data consumption of the page, and maybe the look-an…

99% of internet users are average users. All your concerns are fine but at the end of the day when you want to read the news or whatever, speed matters.

Then why doesn't Google rank speedy pages?

Re: The Problem with AMP

#279
post #221

Earlier quoted context omitted.

> They are a business, doing exactly what they are supposed to be doing. Violating anti-trust laws?

Anti trust prosecutions work only if harm to consumer is proven - harm to other companies doesn't really work for successfully anti trust prosecutions. There's a reason no US attorneys bring anti trust lawsuits against Google and that's because Google would very likely be able to show their choices benefitted consumers (eg. say saved 10 mins and 50MB data a day across 200 million consumers without charge)

I'm an individual, not a company. I make web pages. Lots of other individuals also make web pages. It's not like only companies make websites and consumers only use websites.

Re: The Problem with AMP

#280
post #209

Earlier quoted context omitted.

Because they don't have to move faster.

Yep - this has popped up a few times on HN. Google has zero incentive to improve certain aspects of their business or products (customer service, in particular!) because of their dominant position. For things to improve/change at all is a bit of a stretch, but for this to happen at pace is almost fantasy

It's funny to see Google as the "slow, bloated, evil corporation that doesn't care" while MS is actively trying to shrug off that stereotype.

15-years-ago-me is confused.

Post reply on HN