Live data from Hacker News

The web sucks if you have a slow connection

danluu.com

521–530 of 622 posts

Re: The web sucks if you have a slow connection

#521
post #10

Joey Hess (joeyh) has been writing about this for a long time (because he uses dial-up at his home). Here is a recent thread about a 2016 blog post on this: https://news.ycombinator.com/item?id=13397282

> Please, please, if your site requires AJAX to work at all, then retry failed AJAX queries. Anyone here have information on the most reliable heuristics to do retries? Or information on the implementations used by say Gmail or Facebook?

I'd say that blanket advice to retry is not terrible but not great either. First ask, is the ajax call important enough to warrent a retry at all, if it is, did the response give any useful information about why there was a failure? If yes, did it tell you that something was wrong with the request? Then don't retry as it's not going to work the second time, or third time either. Did the request timeout? Again, think about whether the request is important enough to hit a potentially already overloaded server.

Also, as a side note, any page that becomes unusable because an ajax request failed to return has some really broken design. Ajax retries are not a solution for that, go fix the design instead.

Re: The web sucks if you have a slow connection

#522
post #515

Earlier quoted context omitted.

It's really not about a privileged youngster, it's about business priorities. Most of the time the cost benefit ratio doesn't justify the effort to optimise for high latency high packet-loss connections. Let's say 1% of your potential users use such connections. It only makes sense to support them if your total userbase is a large enough number. For Google, it's a no-brainer. For other sites, it's something to consid…

What is an ageism? People are always inventing new ways to get offended...

I've heard about ageism (like racism, but for people of different ages) since around 2000-2001. Really around the time baby boomers started getting close to retirement age and some companies decided it was a better deal to fire them or lay them off than pay the pensions they had earned, and also with the DotCom boom where startups would only hire 20-somethings. It's not really a new term.

Re: The web sucks if you have a slow connection

#523
post #499
post #485

Earlier quoted context omitted.

How are you distinguishing between Giffen and Veblen goods? If you define Veblen in such a way that all Giffen goods are Veblen and then say that disqualifies them then of course you'll find there are no Giffen goods.

According to the article: > To be a true Giffen good, the good's price must be the only thing that changes to produce a change in quantity demanded. A Giffen good should not be confused with products bought as status symbols or for conspicuous consumption (Veblen goods) Veblen goods = Goods for which demand rises with price because they are status symbols Giffen = Goods for which demand rises with price - Veblen Good…

Isn't a Giffen good then just something where one infers quality from price? I've seen that happen many times with my own eyes, so hard to believe they've never been identified. It's possible to price something so cheap, people assume there's a catch.

Re: The web sucks if you have a slow connection

#524

Earlier quoted context omitted.

I love that they are keen on holding compatibility with archaic SMS but not so much with slower connections.

1. SMS isn't archaic. 2. 140-char limit isn't about SMS.

1. 1992 in archaic. 2. Officially it was, not it's just stupid and unneccessary.

Re: The web sucks if you have a slow connection

#525
post #499

Earlier quoted context omitted.

According to the article: > To be a true Giffen good, the good's price must be the only thing that changes to produce a change in quantity demanded. A Giffen good should not be confused with products bought as status symbols or for conspicuous consumption (Veblen goods) Veblen goods = Goods for which demand rises with price because they are status symbols Giffen = Goods for which demand rises with price - Veblen Good…

Isn't a Giffen good then just something where one infers quality from price? I've seen that happen many times with my own eyes, so hard to believe they've never been identified. It's possible to price something so cheap, people assume there's a catch.

> Isn't a Giffen good then just something where one infers quality from price?

No, it's an inferior good (in the economic sense) in which the (negative) income effect of a price increase outweighs the success substitution effect.

What you are describing is a good that has a positive elasticity of demand with respect to income (or, technically, two different goods, because the higher price represents a different good altogether - one with a higher status symbol).

Re: The web sucks if you have a slow connection

#526
post #292

Earlier quoted context omitted.

I live in a major city, have an iPhone 6S with good LTE coverage according to benchmarks, etc. and still routinely have AMP take 15+ seconds to render after the HTML has been received. I don't know if that's Mobile Safari applying strict cache limits or an issue in Google's side but the sales pitch isn't delivering.

I have all of those as well, and AMP takes Sounds like either a configuration issue on your end or maybe your wireless carrier.

Note that I did not say it always happens — when everything is cached, it performs as well as any other mobile-optimized site — or that it's specific to my device/carrier – it also happens on WiFi, Android, etc.

The problem is simply a brittle design which depends on a ton of render-blocking resources. The assumption is that those will be cached but my experience is simply that fairly regularly I'll click on a link, see the page title load (indicating the HTML response has started), and then have to wait a long time for the content to display. Many news sites also load a ton of stuff but since fewer of them block content display waiting for JavaScript, the experience under real-world wireless conditions is better in the worst case and no worse in average conditions.

Re: The web sucks if you have a slow connection

#527
post #153

Earlier quoted context omitted.

I'm guessing you've never downloaded porn from usenet over a 2400 baud modem.

I have not. That was a little bit before my time. The first modem I've had the luxury of using was a 33.6K.

Perhaps you shouldn't be asserting yourself as an authority on what did or did not happen if you are too young to have experienced it in the first place.

I'm very young, only 27, and know that I missed a full decade of early internet culture and can't speak to it. Even given that, I had a 14.4 and fondly remember downloading a music video for hours. (No porn on the modems personally, I think I was barely adolescent when we upgraded to cable)

Re: The web sucks if you have a slow connection

#528
post #428

Earlier quoted context omitted.

In Ethiopia's case, it's not so much the connection speed in Addis. There's a great deal of interference from the national Deep Packet Inspection filters that leads to timed-out requests, reset TCP connections, etc. JS-heavy apps make a lot of requests to background servers and should one of those requests fail, apps will hang. It's quite frustrating and I would often load pages with the console open to see which req…

I may get flamed for pointing this out either by people who are offended by the viewpoint, or by those who find it so bleeding obvious as to not be worth stating, but those page hangs (and I know exactly what you mean) are really down to poorly architected and implemented front-ends rather than an inherent flaw with JavaScript-heavy apps and pages. Any time you do an XHR you can supply both a success and a failure ca…

Nothing controversial here, it's common sense. Most web stuff is built by total amateurs figuring things out as they go.

Re: The web sucks if you have a slow connection

#529
post #447

Earlier quoted context omitted.

Both describe unexpected effects when trying to extrapolate from price and quantity involved in individual use cases. Jevons: Quantity required per use goes down, so you might expect total demand to decrease. Instead total consumption goes up. Giffen: Price per use goes up, so you might expect total demand to decrease. Instead, total consumption goes up. Either could increase consumption by displacing available subst…

And now with both of those things Baader-Meinhof is going to be triggering every hour for the next month at least

> And now with both of those things Baader-Meinhof is going to be triggering

You know that Andreas Baader and Ulrike Meinhof were the main founders of the terrorist organization RAF (Rote Armee Fraktion (Red Army Fraction)) in Germany. RAF was also the reason that grid investigation was used in the 70th (where lots of innocent people were accused wrongfully) after a RAF terror series, after which some constitutional principles were quashed. The German word for this is "Deutscher Herbst" (German Autumn; https://en.wikipedia.org/wiki/German_Autumn).

These experiences lead (indirectly) to the rise of a completely new party (Die Grünen; The Green Party) and are (besides the experiences with the two dictatorial regimes on German ground in 20th century) one of the reason why data privacy is taken very seriously in Germany.

Thus mentioning RAF, Andreas Baader or Ulrike Meinhof to (in particular older) Germans is perhaps like mentioning Al-Kaida, 9/11, Mohammed Atta etc. to US citizens.

Re: The web sucks if you have a slow connection

#530
"Google’s AMP currently has > 100kB of blocking JavaScript that has to load before the page loads"

Wasn't it that Google was claiming that by using AMP, you can actually make web pages load faster as it is a stripped-down form of HTML[1].

From what I am hearing from the author (Dan), bare html with minimal JS and CSS should (in theory/reality?) load pages faster.

https://moz.com/blog/accelerated-mobile-pages-whiteboard-fri...

Post reply on HN