Hacker News Reader

Comment by Trufa | Hacker News Reader

Completely tangential, I never understand why people use URI instead of URL. He obviously means URL and goes an abstraction lower. You could also say without the corresponding internet uniquely scheme or whatever.

Sorry for the rant. But I am missing something?

View on HN

I'm the author of the blog post. It's a good question but, yes, you are missing something. Mostly me being an idiot.

https://shkspr.mobi/blog/2020/03/it-is-spelled-url/

I usually spell it upper-u upper-r lower-L specifically to annoy both sides of the argument.

(I'm not sure why the submitter substituted the l for I.)

> (I'm not sure why the submitter substituted the l for I.)

Sounds like they fell victim to your deliberate trolling. Is that not an expected outcome?

I, for one, am amused at the implication that we should trust an inaccessible file:// link...

LOL! That's hilarious, love the URl, you could take it a step further with UR1.

Anyway, digged your article, hope I didn't come off as pedantic, thanks for taking it with humor :)

Probably because the blog url does as well: https://shkspr.mobi/blog/2020/07/any-claim-without-a-uri-sho...

The difference between URI and URL is practically meaningless except to a very small circle of specialists. Noone uses the name "URI", even when they are talking about one. Except if they explicitly want to signal their membership in that circle of experts.

Am I wrong to suppose that L is locator and I is indicator, thus I being a more broadly term and a somewhat nerdy term for Source?

L is locator, I is identifier. URLs are a subset of URIs. Both start with a schema (something:...), but URIs lack the information about the location and way to get at a resource. The canonical example for identifier-but-not-location is an ISBN or DOI. That interpretation however requires a very peculiar interpretation of "location", because imho "where can I read about that" can precisely be answered by "ISBN ...". The usual copout "but you don't know where the next library will be" also makes all the traditional URLs non-locations, because you then also don't always know where the next browser is or how to operate it.

The rest of the responsible RFC is also not very useful: https://danielmiessler.com/study/difference-between-uri-url/

Being too lazy to go ready RFCs on a Monday morning, is what a valid schema consists of spelled out? Is `isbn://0441569595` a valid URI? If so, is there a standardized list of schemas?

Read the article, someone already read the RFC for you. To your questions: Yes. Almost, urn:isbn:123-4567-89 (URNs are something I neglected to mention to keep things simple). Yes, the IANA maintains a registry.

URIs can lack the information. Not necessarily.

I is Identifier.

It's broader in the sense that every URL is a URI, but the converse is not true.

URI somewhat nerdier because it's lesser known.

uri encompasses/superset of both url and urn (there was also urc that went no where). urls have a "scheme" (http/ftp/gopher/etc) to access the resource, while urn is a persistent non-location based type of name identifier e.g. urn:ietf:rfc:2648

All URLs are URIs (as other people have noted), but the distinction is less important as time goes on. I think W3C's "URIs, URLs, and URNs: Clarifications and Recommendations 1.0"[0] does a good job explaining the reasons why. Additionally, it mentions the proposed, but never accepted, Universal Resource Citation (URC). Sounds like just what social media needs these days!

[0] https://www.w3.org/TR/uri-clarification/#classical

In the general case of an old newspaper which may not be scanned and online, a URI (i.e., an indentifier) is called for; in the case of an old newspaper which has been scanned, a URL is called for.

The distinction is that an identifier identifies a resource; a locator states where it is located.

'My grand-father's diary, 4 July 1930' is a resource identifier (but not a URI); urn:isbn:0-486-27557-4 is a URI, but not a URL; https://news.ycombinator.com/ is a URL.

Saying URI instead of URL may be technically correct sometimes but it does seem needlessly confusing.

Almost like it's done intentionally to get smug satisfaction when being asked "what's a URI?" I guess that can be a bad faith interpretation but come on who doesn't know plenty of people like that in real life.

Reply on news.ycombinator.com