Live data from Hacker News

Discuss HN: Amazon is going to win heavily aren't they?

news.ycombinator.com

11–19 of 19 posts

Re: Discuss HN: Amazon is going to win heavily aren't they?

#11
post #2

Walmart is doing a terrible job. Progressive degradation has largely been forgotten because just about everyone is using a full-featured browser. Most web developers stopped caring and I'm one of them. Chrome won the browser wars. If you don't want to use it, enjoy your bad experience.

Foolishness. Firefox is a better experience than Chrome.

Re: Discuss HN: Amazon is going to win heavily aren't they?

#12
They pretty much have already, haven't they? Everything they do is to minimize the friction in the typical online ordering process: simple website with decent search, 'free'/fast shipping, relatively easy returns/support, decent/numerous reviews, low if not lowest prices, and a huge inventory. I can put an asterisks next to each of those items where it isn't completely true, and I can easily find competitors that do some of those better, but nobody does all of them better.

Amazon gets a lot of hate, mostly deserved, but if you told me 20 years ago that I would be able to sit at a computer, quickly search for and find a huge number of products, find a price that often beats regular stores, and have the item show up same day or next day, I would've never believed it. It seems like every move they make is insane at first, but it turns out to be a brilliant method of capturing more of the market. Paying a 'membership fee'? Insane. Running your own fleet of delivery vehicles? Insane. Yet, here we are. I see more Amazon vans in suburbia than Fedex and UPS combined.

Re: Discuss HN: Amazon is going to win heavily aren't they?

#13
I recently looked up how far back the wayback machine captured my personal website. I was pleased to find the top ten lists I made as an 11 year old [0] in the year 2002, but two things really surprised me:

- I could list 10 games I liked for each of Gameboy, N64, Playstation, Xbox, and PS2 (how spoiled was I?)

- every one of the amazon links from 19 years ago still resolve to the product in question, so I can still buy a copy of Ready 2 Rumble Boxing for PS2 if I so wish [1]

I was a little dissapointed in my younger self tho to find they were not affiliate links.

On the subject of compatibility tho, the fact that shopping and checking out on Amazon.com works without javascript inspired me to learn how websites work the “old fashioned” way

[0] https://web.archive.org/web/20020609084356/http://www.colten...

[1] https://www.amazon.com/exec/obidos/ASIN/B00004U5VJ/qid=10127...

Re: Discuss HN: Amazon is going to win heavily aren't they?

#15
The best thing about Amazon is that it is still an old school website... like none of SPA crap that everyone (including places like Reddit that have no business being a SPA) keeps shoving down our throats. Amazon is fast, responsive and does page reloads when you click around. I can't imagine in 2021 that is seen a big UX win, shows you how far the UX on the web has fallen.

Re: Discuss HN: Amazon is going to win heavily aren't they?

#16
For this very reason I am trying to buy old mobile phone so that I could test my web app on it. At present I am looking for either older versions of Android phones(very difficult to find) or use phones with Android GO or KaiOS. But both Android GO and KaiOS have "modern web browser" so that is the dilemma I am having while choosing a mobile phone.

For the laptop/desktop, I try to stick to older time-tested front-end libraries such as jQuery and Bootstrap.

I read an article recently where a developer mentioned that he was very easily able to "port" his desktop web app to low-cost small form-factor mobile devices because he is using Svelte. I have not used it, but if developers want to target low-cost devices(though with modern browsers) then they could try Svelte. But to support old browsers/devices I think sticking to simple old time-tested libraries helps.

Re: Discuss HN: Amazon is going to win heavily aren't they?

#17
post #8
post #7

Amazon.com seems to still supports TLS 1.0 and 1.1, older less secure protocols. Newest operating system no longer enable TLS 1.0 or 1.1. Without three hacks an Ubuntu 20 won't allow any connections with those ciphers. While on the other hand older operating systems like Windows 7 don't understand TLS 1.2 and can't be upgraded. So apart from HTML when it comes to secure HTTPS connection all security advice is to no l…

Following security advice blindly is never a great idea. I don't think modern browsers will fall back below TLS 1.2 anymore (or at least not automatically), so offering support for TLS 1.0 doesn't impact them. Then the question is what do you want to do with older browsers? Do you want to give them a browser error that users probably can't understand or do you want to let them in and shop?

Definitely let them shop. The per-account security issues created by old ciphers will be a rounding error in Amazon's risk management budget.

Re: Discuss HN: Amazon is going to win heavily aren't they?

#18

Software made by developers who don't care about the product. Willing to sacrifice thousands of people to eliminate a pain point from their dev workflow. That and resume driven development. It's not about the product, it's about your resume.

I won't blame the developers. They'll do what the leadership asks them to do.

I do often see promotion-driven development, which too is an artifact of the challenges from leadership ultimately.

Re: Discuss HN: Amazon is going to win heavily aren't they?

#19
The "long tail" can be quite expensive to support and maintain. For a small shop, testing on all those old devices is probably not feasible without significantly affecting their overall velocity. For a large company like Amazon, hiring a few people to handle just that, full time, is a no-brainer. For Walmart and Apple, they have other business models to support their existence, so they do not see the point.
Post reply on HN