We need self-verifying hyperlinks.
—
Today links provide, the worlds most used mode of discovery and retrieval. “click here” to subscribe, visit us “here” etc,…
But what exactly is a “link”?
A sampling of what I’m sure is an acceptable definition by most, gives us:
noun
- a link from a file or document to another document or file, typically activated by clicking on a highlighted word or image on the screen.
Initially this seems to be a perfectly fine definition. Definitely it conforms to what most conceive a hyperlink to be.
—
For those interested in a more formalized descriptions, feel free to check out the specifications, such as here, written in 1995 and here, written in 1998. — for everyone else, know that those documents go into great detail on how links should be “expressed” or “written” as well as many “types” of links etc.
—
To complete fleshing out links, we should address one last concept, “Addressability”. This is essentially the idea that in order for something to be “linked” to it must be locatable. This can mean being “locatable” through one or more addressing schemes — very much like your house is addressable by it’s “postal address”, as well as, amongst close friends and family, by come over to “my house”.
Your house can be found and “targeted” as a destination, and thus can be linked to, or referenced.
—
So now, perhaps for the time being, we can agree that a hyperlink is:
noun:
- A mechanism by which a link can be described between two end points, be they documents, sites or any other addressable thing.
- In HTML and other enabled environments hyperlinks, are generally endowed with “click” or “activation” behaviors and subtle styling to alert a user of its presence.
Right?
Close enough?
Are we tracking?
Lies! All of it!
Forgive me, I could not help being a bit hyperbolic… In all seriousness, there is one thing missing, one aspect completely overlooked in all the specs and all the implementations.
Let me explain, let us look at the hyperlink, <href=”http://www.google.com/answer-to-life.html”>.
Let’s forget about “activation” behaviors, head and tail semantics… Let us simply consider to what that hyperlink points to.
Well we can infer, it’s an HTML document.
We can infer, by the content in its url, that is holds the secret of life, just there on the other side of the link.
Well my friend, no, no it does not.
Today’s hyperlinks, are nothing more than stated paths to some place somewhere… maybe.
They are no more links than the hairs on a cats face brushing up against one’s leg.
They are meant to be, all these things… but they are but whiskers which may have at one time or another brushed up against the leg of a document at some time, or maybe not, perhaps they have always just been whiskers pretending to connect with a document.
You see hyperlinks are broken, they are at the whim of the systems on top of which you happen to be surfing on. They are pretenses, statements of fancy and have functioned as they have due to convention and common sense.
—
Using the house analogy above: imagine that your house could be relocated. Imagine you moved it across the street, or down the block. Your old street address will no longer “point” to your house. The address may still be a valid address, only now it points to an empty lot.
That is why todays hyperlinks are broken. They are thin, frail and impossible to verify.
—
And so they have been party to massive scale bait and switching. You yourself have probably witnessed this in your own travels, the infamous 404 or “not found” message. It just so happens we call these broken links, when the page is no longer there…
But what happens when the end point works, the address works, only the document has been switched on us — well it’s business as usual, but the user is totally unaware incapable of detecting and responding in any meaningful way.
This happens all the time, its how spammers hijack a users browsing experience; There you are researching how to make mom’s favorite cake for her 50th birthday — BAM!!!! YOU ARE INFECTED, DOWNLOAD OUR…
Nana clicks on a link to her bank, from an email while at a coffee shop, logs on and does some online banking. By the time she gets home, her life savings have been drained from her accounts. All because hyperlinks showed her the way.
—
The good news. At least now we know ( us two ).
The great news, it can be fixed, by adding to the spec and the implementation mechanisms for hyperlinks to become verifiable.
For until this happens, HTTPS, DNS and all the security in the world… can’t make a hyperlink incapable of lying.
—
I can think of a few mechanisms… I think a self-verifiable mechanism would be best. The method should reside in the link itself… or perhaps in the head of the link… however it should definitely, not exist at the tail, the target document, as that would be a disaster.
Anyways those are my thoughts on the matter, I look forward to hearing yours.
