I understand that other people have made up this idea that "hotlinking" is bad, but it never made any sense to me and I never got on board that train. People "hotlink" stuff off my blog all the time and I think that's great. I put the photos up there to be looked at; if people want to share them, I read that as a sign that I took a good photo. If I wanted to stop people from looking at my photos in certain ways - when the referer comes from a site other than my own, for example - then I would configure my web server not to serve out the bits under those conditions.
If the Onion doesn't want to serve out their CSS file under certain conditions, they should configure their server not to serve the bits under those conditions. If the Onion chooses instead to configure their web server to serve the bits to anyone who asks for them, people should feel free to ask for them.
But it's not "people" asking for them, it's one specific site causing browsers to ask for them unwittingly. Embedding images is less egregious if attributed correctly, but still discourteous and possibly a violation of copyright unless it is encouraged by the site.
However, hotlinking CSS is essentially copyright violation (for lifting the design wholesale, essentially creating a derivative work). There is simply no excuse for not writing and hosting your own CSS.
As the others have said, when hotlinking/embedding, you are profiting from another system's bandwidth and storage resources. Why don't I just hook up your water tap to my house? You should have put a lock on it if you didn't want it to be used for that purpose.
I must point out that hijacking resources from another server is in strong contrast to scraping for personal use, whitehat research, etc. It's one thing to send arbitrary HTTP requests to a server that will answer them. It's another entirely to cause others' browsers to do so without permission, especially in a way that is not explicitly announced and attributed, or accrues to your benefit at the other server's expense.
It doesn't make any sense to me to get all worked up about "rude behavior". The internet is full of people who want to break into your box and take over your system. If you have a server facing the internet you have to expect that people are going to be ruder than you ever dreamed possible, and they are going to do this all the time. Complaining about it won't change anything; just fix your server so it doesn't do things you don't want it to do. I understand that there's this whole big social thing about "hotlinking" which has grown up now, and I've heard all the reasons people use for it, I just look around at the state of the internet and think it's a trivial thing to worry about. Just fix your config and get on with life. Browbeating someone else for using resources which are advertised and provided to the public for free seems like an amazing waste of time.
Just don't freaking do it. It pisses people off to no productive end (as opposed to personal scraping, whitehat access, etc., which also piss people off, but serve a useful purpose).
Also, copyright provides more rights to creators than just restricting copies and distribution. There are also public performance rights, which I would argue are violated by embedding another site's resources into your own site without permission.
> If The Onion has a problem with the amount of data they are serving, perhaps they should configure their server to reject those requests; putting the onus on the end user not to issue HTTP GET unless a certain non-technical context exists seems.... odd.
No clue what you're talking about. For one, we're talking about website creators here rather than end users, and it's these website creators' failure to issue an HTTP GET that's considered bad manners.