[fumigator]

Your cloud service does everything except for file backups, load balancing and can't handle huge file copies. And for reliability and responsiveness, requires a hybrid private-public cloud. Which begs the question as to what's the point of moving to the cloud in the first place.

[Rjevski]

Is this a troll or you are really blaming Amazon for a fault in a shitty router?

[joshribakoff]

I had the same reaction. If I need to support users with shitty routers should I stay away from amazon? My users will just blame me not their router, or switch to my competition who is not using Amazon since it works. My users aren't going to read a long explanation of TCP/IP. Some users have no choice like a school or work firewall/router. If my website doesn't work and the competitions does what do you think happens next if I don't stop using s3? The real question is how many users have a "shitty router".

[Rjevski]

Well S3 is popular enough (you don't realise it but a lot of sites use it under the hood for static assets, file uploads, etc) and this is the first time I've heard of this issue, so the number of routers so shitty that they break S3 (and only S3) would be pretty insignificant. There will be more routers that break the general web (no matter which site) than those only breaking S3.

[CupOfJava]

I don't think he read the article, just the title.