Linket and the Web Killer
In the early 1990s when the Web was in its infancy I was scanning the bulletin boards. These predated the Web and were mostly accessed by modem. Back then the (World Wide) Web was a small arid place mostly inhabited by unwashed geeks and your truly. The websites were typically at universities (ucla.edu, mit.edu) or tech firms involved with research (hp.com, ibm.com, cisco.com). Few today know that domains are older than the Web. 1989 was when the Web was invented as a way to easily access information at different computers across the globe. The initial reaction to Berners-Lee’s proposal was positive but muted. Few envisioned what it could be today.
If you grew up with the Web (you are under 30), understand that it was a different mindset in earlier years. Today the largest US firm is Amazon, worth $2 trillion. Behind it are lesser(!) behemoths like Google (~$1 trillion) and Facebook (~$0.8 trillion). These are creatures of the Web and the Internet. Of course the Web is essential now.
But in 1991 there was no way to safely do ecommerce. You could not buy anything on the Web by merely typing your credit card number into a webpage. https was still years in the future. Ecommerce was so little thought of that the concept of 1 click buying was unknown. Something to keep in mind when you hear protests against Bezos and his 1 click patent.
The Web of the early 90s was mostly read only. Brochureware in the put down parlance of some detractors. The firm behind the website could not sell anything directly. But they could have a catalog of its products. Other websites let you write comments, aka blog. Though Jerry Pournelle, who claimed plausibly to be the first (prominent) blogger, always detested blog; he thought it was ugly.
Other detractions were: The audience is too small. (True.) Loading webpages is too slow. (True. 56kb modem.) You had to write webpages by hand in text editors. (True. No Wordpress. No Wix.)
But 1 person, whose name I have long forgotten, came up with a killer objection. It was elegant and seemed unanswerable. I can demonstrate it on this webpage. He said I have a webpage with this link hp.com. You click it, thinking you are going to Hewlett Packard’s site. No! You end up at IBM’s site. This is a fundamental property of the Web and HTML. The author of a webpage has absolute control over what she writes as the visible part of a link and of the URL in the link. (He used other choices for the 2 domains.) An unscrupulous salesperson for IBM could do this to mislead readers and take them to IBM.
So, he said, if we do this WWW thing, we can make webpages riddled with lies. This cannot be stopped. How can you believe what you read on it? Surely his objection would kill the Web.
Looking back from 2020 and after enduring 4 years (no more!) of Trump and his Fake News we might have some sympathy for this 1991 view. Was it just ahead of its time?
There were echoes of the objection in what might now be seen as a more innocent time; the spam wars of 2000–10. Spammers discovered that sender addresses could be modded because mailer programs did not check this. A spammer sending from mailer5.com might say he was joey@wsj.com. Today mailers have tightened down their emissions so this rarely works. But for several years it ran rampant.
Reading the 1991 replies to the objection, no one had anything effective. No technical refutation. But the years went on. People ignored the objection, if they ever knew it, and slowly we got ebay.com. hotmail.com, etc. Today the Web is hugely built out.
The deeper point is that this is an example of the cliche about perfection being the enemy of the practical. Most webpages are essentially correct. Most authors are not trying to actively mislead readers. And the rise and value of Google is in its ranking of websites. If a website is truly useful, other sites will link to it, so its Page Rank rises. This makes a positive feedback loop that rewards the virtuous. Hopefully. Sites that mislead will not. Plus there are now review sites that let readers comment on and rank other sites. Another type of feedback loop.
Of the 2 replies, only the second was anticipated in contemporaneous responses to the 91 posting. The first reply is of course the 1 sentence elevator pitch for Google. No one saw this in 91.
How does Linket enter into this? One objection to linkets was this. Here is a linket as a clickable link in a webpage [Flowers]. That is the visible part. The invisible part is something like http://linket.info/ask/…=Flowers. At linket.info is our Registry database. It replies with the raw Internet address of a website that does not use a domain but a linket as a guide to let users find the site. What if the visible [Flowers] link goes not to the site of the owner of the [Flowers] linket but to another linket, owned by a competitor.
The objection is unwittingly the objection from 30 years ago to the Web itself. Just as the Web grew in spite of this, so too will linkets.
Dr Wes Boudville (wesboudville@gmail.com)
