Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

9m.no – Short URLs for the Unicode Age

9m.no

Post reply on HN
Print view View original

21–30 of 75 posts

Re: 9m.no – Short URLs for the Unicode Age

#21

Post by buttscicles »

mp4box wrote: post #7

Probably should block the same domain from shortening.else you'll get something like http://9m.no/婀蒾 (multiple redirects)

I created a simple url shortener[0] a while back and though I thought of this I hadn't considered people could add a '.' to the domain to get around that (so http://yl.io/1 becomes http://yl.io./1), which somebody was nice enough to figure out for me ;)

[0] http://yl.io

Re: 9m.no – Short URLs for the Unicode Age

#22

Post by wcchandler »

jebus989 wrote: post #19

All the usual HN nitpicking and obscure edge case discussion should probably be done in the light of ehamberg's own comment: "This is a horrible idea in many ways and was a quick hack for fun." [0] [0] https://github.com/ehamberg/9m

All these people had to do was click "about" at the bottom and find this nugget from the author:

"Made slightly – but not completely – tounge-in-cheek by @ehamberg."

Re: 9m.no – Short URLs for the Unicode Age

#23

Post by martin-adams »

jrajav wrote: post #11
martin-adams wrote: post #3

I'll be screwed if I ever have to retype the short URL. So no good for printing on packaging.

Why would you rely on any free third-party shortener for print to begin with?

Sounds like both you and I agree that it's no good for printing on packaging. I'm unsure why I'm suddenly defending its use.

Re: 9m.no – Short URLs for the Unicode Age

#25

Post by Pxtl »

SanderMak wrote: post #15
smoyer wrote: post #10

I got http://9m.no/𑂜൭ for my daughter's website, but I'm wondering whether this helps shrink character count when used for services like twitter. It's not at all readable, but since it's two (I think) 4-byte unicode characters, will that URL be counted as 13 characters or 21 characters?

Twitter shortens every URL regardless and reserves 23 chars (I believe) for each link in a tweet.

Obviously the correct answer is for Twitter to incorporate this approach.

Re: 9m.no – Short URLs for the Unicode Age

#26

Post by billpg »

berdario wrote: post #9
DoubleMalt wrote: post #2

I don't think risking url rendered as unicode boxes justifies the reduction of the url by 2 or 3 characters. Nice proof of concept though.

Not only that, but since it will be encoded in utf-8 and then percent-encoded the number of bytes transmitted will actually increase: something as simple as 邗諾 becomes %E9%82%97%E8%AB%BE , 18 characters! Still, as other mentioned it's a neat idea that if widespread would make all kinds of encoding mistakes pop up :)

But twitter will only count it as two characters. (I guess.)

Re: 9m.no – Short URLs for the Unicode Age

#28

Post by pornel »

billpg wrote: post #26
berdario wrote: post #9

Earlier quoted context omitted.

Not only that, but since it will be encoded in utf-8 and then percent-encoded the number of bytes transmitted will actually increase: something as simple as 邗諾 becomes %E9%82%97%E8%AB%BE , 18 characters! Still, as other mentioned it's a neat idea that if widespread would make all kinds of encoding mistakes pop up :)

But twitter will only count it as two characters. (I guess.)

Twitter doesn't pay attention to URL length any more. They will force all URLs through t.co even if that makes them longer.

Re: 9m.no – Short URLs for the Unicode Age

#29

Post by MereInterest »

martin-adams wrote: post #23
jrajav wrote: post #11

Earlier quoted context omitted.

Why would you rely on any free third-party shortener for print to begin with?

Sounds like both you and I agree that it's no good for printing on packaging. I'm unsure why I'm suddenly defending its use.

As I read it, you both agree that it is no good for printing on packaging, for two different reasons. Your original statement, however, gave the unspoken implication that other URL shorteners were good for printing on packaging. It was that unspoken implication that he was talking about, rather than the directly spoken statement.
Post reply on HN