One of the most viral activities on the Web is sharing links. It is fast and easy, and a good way to communicate ideas. What started out as something people did via e-mail and bookmark-sharing services like Delicious, is now moving to Facebook, Twitter, and other social broadcasting services. It is just so much more efficient to share a link once with all your friends and followers than to send it to each one individually.
Twitter is especially suited to sharing quick links, but its 140-character limit has perhaps done more than anything else to propel forward the use of URL shorteners. These take long URLs and turn them into shorter ones that usually redirect people back to the original. So for instance, http://www.techcrunch.com/2009/04/06/revolution-money-raises-another-42-million-for-alternative-payment-service-nobody-is-using/ becomes http://tinyurl.com/coflho or http://bit.ly/q3Sl9 or http://digg.com/u1LRR.
There are more than a dozen such services, including TinyURL, bit.ly, Snurl, tr.im, is.gd, and the new Diggbar. The better ones offer tracking stats. One of them, bit.ly, just raised $2 million. Nobody really likes them, but they are a necessary evil. How else are you going to share links on Twitter without having the URL take up half the message?
It may be more complicated than that, however. Joshua Schachter, the founder of Delicious, thinks they are downright evil. Schachter writes, “The worst problem is that shortening services add another layer of indirection to an already creaky system.” In other words, they slow down the Web. He gives several other reasons why they are bad as well. They add a whole new slew of middlemen to the equation, and these links become dependent on the continued existence of these startups or even the whimsical changes in their terms of service. URL shorteners make links opaque, which spammers love.
They also add an unnecessary extra step to what should be a fairly simple message. Some, like Digg’s new Diggbar, also steal link juice from the original destination by wrapping the Website in a frame rather than redirecting to it. That just messes with the whole link structure of the Web. If I am linking to your story using a shortened Digg URL, Digg gets the credit, not your Website. Most URL shorteners don’t do this, but If Digg is successful with its new feature, they may follow suit. If that becomes an accepted practice on the Web, it would create all sorts of complications for the search engines in terms of duplication and making sure the underlying link gets the proper ranking.
There is a simple solution to all of this. Services like Twitter could simply do a better job of incorporating links into their design by allowing users to hyperlink existing words in their messages, without wasting space by displaying the actual URL. This is how FriendFeed handles the issue. Or it could carve out a separate place for links outside the message (perhaps through a “link” button at the bottom of each Tweet). The only reason to keep the URL within the message itself is for SMS messages, and for those perhaps Twitter would be better off creating its own URL shortening service that can become the standard, or buy one of the existing ones. If it ever does go the acquisition route, bit.ly might be a leading candidate. It was created by Betaworks, the main investor behind the startup Twitter purchased last year (Summize) when it realized it needed its own real-time search engine.
So are URL shorteners necessary or just evil?
I think it depends on what purpose the service offers…if any. BurnURL is an offering that I’m involved with, and the layer we’re adding actually adds value for readers and for publishers. We’re taking a digg or a like to the next level with Mood Mining, and we’ve built our service based on feedback. We’re not hearing the “evil” side of things.
Firs things first, what is this – http://spy.tech...com/signin.php?
I’d tell you, but then we’d have to hunt you down and erase your memory. It is definitely not a URL shortener, though.
I had gone for StartupWeekend over the weekend and was impressed with one of the team’s that created a tool to share multiple links using one url. It kind of reminded me of the DiggBar
check it out guys, http://hubb.me …
They are bad only for twitter. Twitter should come up with [link] feature, which directly hyperlink to target.
Very interesting
At least you responded honestly without any PR bullshit.
+1 to schonfeld
The main issue for me is that they are very unsafe … it is just to easy to get redirected to a phishing or insecure site through a shortener … there must be a better way than having to choose between a long url and one that is completely opaque
Drew,
I’m okay with it as long as you aren’t framing another publisher’s content.
I only consider the ones that implement iframes to be extremely evil.
But here’s a better explanation along my line of thinking on this topic and why I believe publishers need to come together now to break such framing techniques before they infest the web again.
http://tomuse.c...oney-publisher/
I disagree that framing another web site is evil. Is letting the user choose which browser to use “evil”? What if it has other features and links beyond the underlying web site?
URL Shorteners that frame are no different – they are a form of web-based client that can add additional features to an underlying web page.
So what if they are advertising supported? It’s up to users to either choose to use them or not weighing whether they get value from them.
The only evil scenario (in my book), is claiming a link is to one site, when it actually goes to another.
You can see this page framed in my URL shortener which adds a real-time chat feature.
http://go2.me/375
If that’s your intention to invite your friends to a real-time chat about a web site, why shouldn’t you be able to use a shortener like Go2.me?
So Mike how would you feel if I framed your frame? Do you see the problem now? That pushes us back to the late 1990’s when every website was just a damn frame within a frame.
My memory is quite different from yours about the web of the 90’s. Many web sites were built using Framesets for their own navigation and internal structure. The main problem with these was that frames broke the model of one URL to one resource. This is quite different from having one web site frame another. If a site would do this with the intention of fooling a user that they are actually looking at the original site, then I’d agree that this is dishonest and undesirable.
You’re quite welcome to “frame my frame”. Of course, in the extreme, the frames dominate the screen and leave nothing for the content.
But I believe in a rich Internet ecosystem, where we have room for mash-ups and many kinds of “multi-source” solutions. Framing one web site with other is just one form of this kind of mash-up; and I don’t see anything “evil” about it.
Some content producers may want to control every aspect of their readers’ experience. But lets not forget about the rights of readers to choose how THEY want to consume content, including sharing it with their community.
I think successful content producers will embrace all the ways in which users want to use their content including social bookmarking, full RSS feeds, 3rd-party API’s and mash-ups, as well as allowing sites to frame their content…
Mike:
You’re missing the point: no one will come to your site if it is nothing but a frame within a frame within a frame, etc.
“You’re quite welcome to “frame my frame”. Of course, in the extreme, the frames dominate the screen and leave nothing for the content.”
My point exactly, thank you.
Im first! (joke)
I guess its very good idea to hyperlink words. Than you get more space for tweets and you can actually hover over the link to see where it goes.
Shouldnt be that difficult to implement in existing system.
I don’t think anyone will be able to monitize URL shortening.
I think Delicious and Digg may think it’s evil because Bit.ly is taking some market share.
I guess I’m only second
just to ad more, it seems there is a problem on Twitter when you share the story and somebody just repack it in new short link with his own username. The original user is simple wiped off.
Maybe it would be possible that this featuer where you hyperlink a word could include the username of the person who originally started the process which can be seen when you hover over the link.
For example:
By @ArringtonCrunch: http://www.breakingnews.com
So original contributor would get more credits (its still possible to repack it but I guess it takes more effort)
You raise good questions. However, I don’t necessarily agree with the statement “nobody really likes them.”
Before Twitter came around, I used TinyURL all the time in emails where I was forwarding a link, and found it to be a great convenience.
Overly long URLs tend to get chopped in half by line breaks in email, which can be problematic–especially for less savvy Web users who might not get that they have to re-construct the URL. In these cases, I never worry much about the permanency of the solution, because the shortened URL is just a means to an end–getting someone to the right page.
Thanks for the article.
I have a little PHP & MySQL expertise. So I registered my own short URL ($12) and found a free PHP script (Short URL). My own short URL service. Link to this post: http://skre.us/z.php?29
In more than one way, creating your own service is a genius idea. Imagine the link rot around the web once these services get bored, run out of money or shut down for one reason or another!
Frighteningly a lot of people don’t realise that using shortened URLs in blog posts is a bad idea.
Short term communication like Twitter: Fine, use a shortened URL
Long term web material like blog posts: Use the long URL, don’t risk it
I assumed that they were intentionally leaving out hyperlinking (and related features) so they could charge for these upgrades in the future. On the other hand, Twitter seems to crash quite a bit…perhaps these upgrades would be too much for the system to handle.
If you’re looking for something a little more silly, you can have the world’s shortest URLs by using Unicode:
http://➔.ws/诮
via http://tinyarro.ws
You score kudos points for not using the phrase “link juice”. Unfortunately, this post doesn’t score well
Other than that; I agree. URL shorteners seem to come out neutral in the positive/negative ballance.
Another big issue with shorteners is the masking of “bad” web sites. You never really know where a shortened URL is going to direct you. I’m fairly picky with what sites I’ll visit for a number of reasons: network security, employer limitations (ie. keeping my job by not going to porn sites), and the general dislike for certain sites that I’d rather not the Google Analytics get the wrong impression by me going there.
Perhaps all these URL shortening sites need to work with browser developers so that a mouseover on them shows the real URL it will be directed to.
You may find this useful:
http://lifehack...d-shrunken-urls
The blame, as far as having to shorten links for Twitter, goes to Twitter themselves. They could simplify the entire process by adding “add link” feature of some kind and letting users insert a link as large as necessary into their “tweets” without having to eat up precious 140 characters.
That would cut out the middle-ware guys and provide “credit” to where credit is due, on the originating content provider.
Seems obvious, however I suspect part of the problem is Twitter’s architecture. I am sure their devs are not excited about wanting to find out if their servers could handle the extra bit of information added to each message.
Content providers should offer their own url shortening services.
For example, this TechCrunch post http://www.tech...l-or-just-evil/ could put a “shortened link” next to its ShareThis button. The url could be TechCrunch.com/cX4y.
If I can rely on TechCrunch maintaining this URI: http://www.tech...l-or-just-evil/, why can’t I depend on them redirecting to it TechCrunch.com/cX4y?
Those hosted URL shorteners are coming, which is bad because then you end up with a million of them, and you won’t know which ones to trust.
He’s got a point.
And I think he didn’t meant to create millions of hosted services,
he spoke about that each site will provide its own short reference to every page\article (not as a service to create short links of arbitrary pages)
Add such feature to WordPress and you already solved 50% of the problem.
correction: TechCrunch.com/cX4y should redirect to http://www.tech...l-or-just-evil/. my bad.
Dont think new outright URL shorteners will be able to grab big pieces of marketshare from tinyurl unless they add new simple and value adding features. Bit.ly realized this. Newcomer Pagetweet.com takes the biggest steps into a new direction, by offering the ability to leave an individual comment – rather than sending a blank link. http://p8g.tw/?ovu . This could be the new direction
Whatever about the rights and wrongs of short URLs I just love this guy who got annoyed when a “friend on IRC commented on my use of long URLs, and told me to use TinyURL instead”
His response is genius – http://freakinghugeurl.com/
evil evil! Solution is just around the corner: http://revcanon...al.appspot.com/
“Or it could carve out a separate place for links outside the message (perhaps through a “link” button at the bottom of each Tweet).”
Yes.
I know that some adops departments have used URL shorteners to fit into Ad Serving platforms with URL limitations.
Also, I wonder what would happen if TinyURL or the like just went down. Millions of dead links just floating around.
The DIGG version is brilliant from a selfish perspective- think Dr. Evil- “how can we increase the amount of traffic to our site….” but potentially horrible from a community standpoint- wait, I am getting no link credit? WTF digg. However, it will most likely go ignored and everyone continues on their merry way.
URL shortners are big problem, spammers use them all the time.
yeah, e-mail is a big problem too, spammers use it all the time.
won’t somebody think of the children?
The problem is that people implicitly trust bit.ly, TinyURL and the like. I know better than to click on the “Vaigaagra for @2″ email…
Fascinating.
There is a lot of hype around this issue the past 10 days or so, and some of it seems a tad hysterical to use here at tr.im. But not too much. To be concerned is reasonable.
Most links, by far, get the most of their activity within the first 24-48 hours of their lifecycle and then drop off after that. This concern of a potentially broken web seems overblown.
I would suggest that some services out there are just irritated that we are compiling similar data to theirs for a very very small fraction of the cost they have incurred. And, after you hit a certain number of users, statistically, the data is very similar to the larger more established operations. This must be frustrating to some of the entrenched larger players.
As for Twitter buying bit.ly or whomever, this seems unnecessary on its face. What would be better would to be for Twitter to define a URL shortening API fiat “standard”, and then force any URL shortener to adopt it for inclusion on Twitter’s website and API, and then let the users decide. This way Twitter could support almost all shorteners with very little work on their part.
–ejw
Eric Woodward
Email: ejw@tr.im
URL shorteners are evil and have no point… if you want a short url, then use short url… It is useful on places like IRC or instant messengers where HTML is disabled. However it has no reason of being on web apps such as twitter.
Perhaps web browser vendors could agree on a free, open standard algorithm for shortening URLs. Then the address bar in all said browsers would just transparently handle a shortened URL encoding. Something like:
shortened link = http://MD5-sumURL-link, current-date]
.. or something similar. Since the algorithm would be open, there wouldn’t be any dead-startup issue, as the redirection wouldn’t be done through any company.
I’m not sure how you’d do this without some centralized database. If all you’re doing is hashing URLs, the obvious problem is that hashes aren’t short!
The advantage of a centralized DB is that you can ensure no collisions, but still maintain relatively small URLs.
Facebook is also starting to do this. But on another note, the Digg toolbar absolutely ruins the browsing experience on my Blackberry. I’ve already stopped visiting any links that use it.
The digg toolbar allows you to disable it on all links using a cookie. Useful if you have cookies enabled.
Click the little arrow next to the “X” button that hides it.
Wanted to contribute my thoughts just a bit.
First of all, the claim that it makes spamming or linkbaiting easy may be true but most users are only clicking links like this on Twitter where they follow someone and trust the link is something relevant to the tweet it succeeds.
Additionally, this reminds me of the early days of Micro-blogging. blogs like TechCrunch would put these services head to head saying it’s too much and none of them solve x & y. My gut feeling is that there will be one victor in this shortening race and it’s not going to be the site with less characters but the site with better analytics and tracking.
Finally, I preface all of my links. Like this one. http://twitter....atus/1465461332
I mention what i’m linking to, bracket it with FLICKR so the reader knows the destination and then a link. I like bitly analytics the most.
Fast forward a year from now: there will be one choice for URL shortening that everyone users, analytics will be better, trust will be higher and tools like PowerTwitter for Firefox will actually expand those links in your Twitter stream (a tool I highly recommend).
you tell me, since you used one in your twitter post about this article: http://twitter....atus/1465295318
URL shorteners are pointless, especially on web apps where you could simply do shorturl. It might be useful in places like IRC or MSN messenger where HTML is disabled, but it is totally useless on web apps (like Twitter).
I wouldn’t go so far as to call a shorten url services filthy but they’re definitely dirty.
Without Twitter, there is no necessity for such services.
And when these kind of services will start to monetize all the traffic going through them,
it will become realy ugly.
Twitter will do us a great favor if they will adopt one of your suggestions above.
What will be next? Maybe we’ll start to zip archive our entire text, so we’ll be able to post within 140 characters an entire blog post.
(Hey.. wait a minute….. maybe it’s not such a bad idea…)
LOL, how the hell would any of these things make money?
I’ve used URL-shorteners for some time in emails. They are almost necessary there.
They are a nasty part of Twitter, and I agree with some the comments here.
We all want incoming links to our sites, and when someone wants to twitter about something wonderful at my site or blog, and they use a shortened URL, I don’t receive the benefit of a better Pagerank from Google.
If a user of Twitter is tweeting using Tweetree.com, that site will resolve the shortened URL and reveal the ultimate destination of the link. I like that.
It would be better if Twitter made no discount for URLs or @thenameofatwitterer. I’m @RobertB on Twitter
I second this motion. That makes a lot of the points against URL shortening moot – you can turn to a 3rd party when a site or blog service doesn’t have a shortening option, and then they can deal with the loss of hits and search engine ranking.
…that would be Alan’s suggestion that URL shortening come from the content provider and bear the site’s domain name.
The link rot argument is bogus. URL shortener links have 99% of their traffic within the first 24hrs of a URL being posted. It’s all transient data, this is the real time web…
They’ve been useful for embedding urls in email/IM/README/twitter etc. It’s the only way to provide tracking analytics in non HTML media.
It’s pretty easy for crawler/url/bookmark aggregators to resolve these links, especially if the url shortning service provides efficient API for that (batch resolve.)
The Digg bar can be easily circumvented by destination sites using the usual javascript tricks.
The sites (Twitter) or apps (Destroy Twitter / TweetDeck)) on which these shortened URLs get posted should arguably expand them for the end user. I use a Greasemonkey script to do this for me:
http://userscri...ipts/show/43826
http://userscri...ipts/show/32115
When it comes to Twitter, I hate not being able to post a link w/o using a URL-shortener! 140 characters are not enough for this, and think they should really add a new feature, like Pownce had, where you can add a link, separatly from the 140 chars. Same goes for IMG’s and so on.
URL shorteners are bad. Why do we even have to shorten them in the first place? Restrisctions = BS
I’m not a big fan of the DiggBar, but I do understand what they’re after by implementing it.
Again, Twitter should add a AddLink section of some sort, with a ‘Visit Link’ link or an icon..
Sharing links in twitter is by far the best I’ve ever seen. Companies like http://www.boilingpage.com take advantage of these links that are shared in twitter and bring the hottest pages on the web based on how many people share the same links in Twitter – which makes complete sense. In fact the top result in Boilingpage’s home page is about “how useful the URL shorteners are” – here’s the link: http://www.boil...s&n=1059994
Even if none of these URL shorteners exist in the future, you can find the hot links that were once shared in twitter in http://www.boilingpage.com
So funny ! This post is exactly the thought I had this morning ! It was kinda my “thought of the day” and here’s an article on Techcrunch… Smart minds meet
I agree, sites like twitter could do a better job with URL and not force people to use such services as URLshorteners.
What a waste it would be if any URL shortener site would go down, thus unlinking countless links from their destination. Do you think in such an event that the service would provide a kind of backup file to the public before it went down ? There, all the links and destination would be specified in order to “save what could be saved” and avoid the lost forever.
Links are part of the internet history !
I like the metrics built into some URL shorteners. They can prove useful and to some extent are a necessary evil when trying to compress a message and URL into 140 character for Twitter. In a perfect world, I wouldn’t shorten my URLs, but until then, I’ll use ‘em when I feel the need.
I still only use email to share links with friends and family. I never had the need of a twitter or a facebook to fill that purpose. I don’t see why would other people do.
tiny urls have carved an interesting niche for themselves, and what is possibly an unwarranted trust level – it seems ppl are quite happy to click on them regardless. i remember certain slightly devious techniques that sites used to hide the destination urls showing in the status bar using javascript etc .. now they dont need to – tiny urls do it for them!
btw loved the description of the interweb as a creaky old system – ive always thought the web will one day implode, crushed by its own weight, a victim of its own success – and bulletin boards will rise from the ashes haha
URL shorteners increase the url service’s pagerank while diminishing the rank and presence of the site being referenced from the shortened URL. Using it to give a friend an easy link in email or IM or certain limiting situations like that is fine, but using them in place of a real URL on a webpage is not.
Security and phishing, having to do a HEAD to expand them sucks too, FF handles them well, twitter handles them horribly. Also what if they close up shop, what happens to those links?
I hate it when magazines print Short URLS rather then printing the full real url. I mean what if they get the address wrong or miss spell a address. Then there is the reader error. Is that a I or a 1, I can’t then type that url into google and find out what it was suppose to be. so I am screwed.
I find it absurd that people think there’s a business model for url shoteners. Twitter could easily put every url shortener out of business by adding a link button as aforementioned by other commenter’s. Better yet, they could allow you to write a message, insert your hyperlink, and reformat it as (link) + snapshot. URL shortners = No big deal. Next topic please, Tech Crunch.
I totally agree.
And of course URL short services have only ever existing since twitter came out…
The key words are “out of business.” These companies are obviously trying to build a business model around a simple shortening app that can be reproduced in a day’s worth of code. Whether this business model is to create a real time digg clone of the hottest links on the web, or to resell analytical data based on clicked links, it’s not worth a 2 mil investment that twitter can easily out due with something as simple as news.twitter.com. Maybe I’m missing something, because it’s just not making sense. I understand people used url shorteners before twitter, but that’s not my point. People are seriously investing in these things to generate a profit. That’s absurd….where is the long term value that makes these “companies” something to invest in?
Isn’t there another way? Previous comments from Russ Magee and David O may have alluded to something like this, but not specified it.
Rather than using long URLs which are shortened by these services, each destination site could use much shorter URLs; the domain name could be followed by just one short path element. For example: http://abcxyz.com/at543r.
Depending on the length of the domain name, the total length might not be as short as those from tinyURL etc.. But would be much shorter than many URLs currently used especially, it seems, by blog sites. Also, in some mitigation, because each site would have a much smaller namespace than a shortening service, the shortened element can be shorter when specific to a site.
Erik, I have said this before in the comments on TC somewhere related to this, but here it is again:
You said: “The only reason to keep the URL within the message itself is for SMS messages”
You really see yourself texting your buddy “go look at http://bit.ly/aSz2s” ?? you’d need to remember the short code for the URL, which, given that it’s a random set of characters, could prove harder than remembering the actual long URL that has a readable link.
URLs in sms text messages are at the least unlikely if you ask me.
Great article, glad to see TC advocate to keep our intarweb clean and fresh!
Why have them at all? Twitter could detect when a link is in the message, and make the whole tweet clickable. Save some precious characters.
Because a single tweet could have multiple links, and many certainly do have multiple hash tags that function as navigation/filtering links, so a single all-text link fails. So it would seem.
“They add a whole new slew of middlemen to the equation, and these links become dependent on the continued existence of these startups or even the whimsical changes in their terms of service. URL shorteners make links opaque, which spammers love. ”
He’s 100 percent correct and its why I never click on tinyurls. The people who don’t care must not understand much about vulnerabilities that can be exposed by hackers and spammers who may attempt to use these shortened urls.
Anyone sit with me in a class where the teacher or professor spent 3 minutes giving out a URL (aich tee tee pee, colon, forward slash, forward slash, double-u double-u double-u, dot…)?
URL shorteners have their place. If you’re giving out a 50 character URL verbally or in written form, you’d save a lot of headaches by taking a few seconds to shorten it.
The main problem is they are getting used more and more by default when they are not really needed. People are shortening urls from sites like youtube which are already pretty short and even with commentary are well below the 140 character limit for twitter.