Saturday, May 23, 2009

Short links

Tip: for long links in comments (or emails) you can shorten them at:
http://snipurl.com
or even better:
http://tr.im
The latter has a very useful widget for it.

Update: I use them because long links sometimes don't work well in emails or in web pages with long columns, and it's time-consuming to make html links (I don't use html in email anyway).
But TTL points out some points against them, see comments.

Update: I've finally made myself that macro I've meant to make, to make hyperlinks. (I have to paste this as a picture, otherwise the code messes up.)


3 comments:

Timo Lehtinen said...

When Tim Berners-Lee introduced the World Wide Web in 1991, it had one feature that we hadn't had until then: network wide hyperlinks in free form text. Before WWW, we had been writing the addresses to Internet resources out as part of our prose, to be copy-pasted by the user. (You can still see these in, for example, many FAQs that haven't been updated since then.)

Now you are suggesting that we go back to the times before the World Wide Web. But not only that, you are suggesting that we make our references indirect, which is an even worse scheme than what we had in the 1980s. For, using a redirection service such as snipurl or tr.im is a bad idea for the following reasons:

1. It gives the user no visual cue as to where the link points to. For all the user knows it might lead to a malware site that hacks into their computer. In contrast, in a direct address, the user immediately sees if it's a YouTube link or a link to a news medium, etc, allowing them to make an informed decision on whether to follow the link or not. (For active hyperlinks, the browser shows the URL on the status bar.)

2. The reference credit goes to the redirection service, not the author/publisher. We all strive to create compelling content so that people would link to us, which then would cause the search engines to rank us higher, bringing in more visitors, which then would help us sell whatever it is we are trying to make a living of. When you use a redirection service, you are not giving proper credit to the author/publisher. It makes no difference if we are talking only about blog comments. A web page is a web page. A link is a link.

3. These redirection services are fly-by-night operations set up by some bored webmaster on a whim. There are tens of them, they come and go. As soon as the redirection service you have been using disappears from the net, all your references stop working, even if the original resource is still there.

There is one place where the use of a "short URL" redirector is justified: Twitter. But even there it would be better if the Twitter company were to provide that service. We shouldn't have to rely on a 3rd party provider.

The above points have been articulated by many people many times before. I am sure you know about them already. It's just that when you advertise those redirector services in your post, you make it sound like it would be some kind of a technological advancement, or in general a good idea, to use them. Which it is definitely not.

Timo Lehtinen said...

There is one more thing:

4. The major redirection services install long cookies in your web browser. For example, the cookie installed by the tr.im service expires in 2011. Because a redirector is a form of interceptor, what they can do is track your movements across those sites, compile a profile, and connect that profile with any site you visit.

Now, as I said, most of these redirectors are small and harmless hacks, but over the years a few have become big: for example tinyurl.com and, more recently, tr.im and bit.ly.

Ever wonder what the business model of these sites is? It takes considerable server resources and bandwidth to continue to provide such a service after it becomes successful. And they do it for free, out of good will?! They don't even have ads on their website when you go fetch a new URL, and as you say, they offer widgets where you don't even have to visit their site.

Perhaps the cookie lifetime offers a clue here.

Timo Lehtinen said...

Finally ...

5. It would be trivially easy for the operator of a redirection service to not in fact redirect you to that site at all, but instead pose as that site, intercepting your keystrokes when you enter your password etc. (many people use the same password across sites).

Now, there's no particular reason to assume those sites are run by bad guys, as there apparently are other ways to monetize the service, at least for a more successful redirector (see #4 above).

But what happens when a not so successful redirector site exchanges hands to an anonymous buyer from, say, Russia or China?

It would not be a stupid idea to make it a policy to never deference indirect URLs.