The Xanadu Dream | Hacker News Reader

The Xanadu Dream

codinghorror.com

42 pointsbdfh4215 comments

He is the archetypal borderline autistic, non-conformist, free-thinking technologist.

This is incorrect.

Like most programmers, Ted's reach often exceeded his grasp.

He can program, but he's not a programmer.

The above text is excerpted from the definitive 1995 Wired article on Project Xanadu, which is still as electrifying to read today as it was then. The hubris and sheer scale of the Xanadu dream are at turns both inspiring and desperately, hopelessly out of touch.

Desperately and hopelessly out of touch perhaps for you. The reason it's inspiring is the same reason that working at a great company or in a great computer lab or on great software is inspiring. You're working on something where your philosophy is embodied by the software.

The hubris and the sheer scale are called planning ahead. Instead of solving problem X, Nelson wanted to solve problems X, Y and Z all at once. Some problems need to be tackled all at once and Nelson was right about this. How many implementations of payments and micropayments have we seen that have failed on the Web? They fail because they aren't integrated with the Web.

But even that alone was enough to build a functional and useful internet for the world

The Internet was already useful without the Web.

Consider the reality of what's actually possible, what people can understand, and what us all too human programmers can practically implement. It might not be the Xanadu you dreamed of -- heck, it might even suck -- but it'll at least have a fighting chance of existing in reality rather than fantasy.

If all we do is build good-enough software, we'll end up just applying patch after hacky patch to it to make it do the 100% thing. Again, sometimes you have to design with the whole of the problem in mind. Turn the corner cases into normal cases, in other words.

Off-topic... I'm sick and tired of having to defend the history of computing. I had to defend Smalltalk yesterday, and I've had to defend Lisp and Scheme against people as well. I had to defend friggin' Java against C...I shouldn't have to do that.

I'm sick and tired of having to defend the history of computing.

Um, it's codinghorror. The author has no reputation for knowledge, insight or understanding. He's just a guy who talks about stuff. The programmer on the street, if you will.

"Every Xanadu service provider"

I would much rather live with broken links than rely on the centralized infrastructure that would be required to implement anything like Xanadu (which would ultimately end up being a government, university, or business run system anyways). Sure, Facebook Connect and Microsoft Passport are commercial "poison" in terms of centralized identification and authentication, but at least we have the ability to choose or implement our own systems.

> than rely on the centralized infrastructure

I wonder why you think such infrastructure must be central. It's certainly not part of the spec.

Bingo,

Nelson anticipating many of the features of the web was great. The extra features he "anticipated" were exactly what made Xanadu impossible to implement as well undesirable to implement (see RMS' The Right To Read).

The extra features he "anticipated" were exactly what made Xanadu impossible to implement

And yet everyone wants to implement all of those features eventually...funny how that works.

Isn't that how it always happens? Someone will come up with a grand vision that is impossible to implement given the current technology or environment. Someone else will strip it down to the essential bits that are achievable. The next person will add another feature or layer that adds value to the network as a whole (and which would be valueless if it were not for that first, dumbed-down version), etc.

aka "Worse is Better"

While I always been a great fan of Ted Nelson, his vision does not include distributed cryptographic filestores, where documents do not reside on individual servers, and where individual servers are in fact completely invisible and redundant.

>his vision does not include distributed cryptographic filestores, where documents do not reside on individual servers, and where individual servers are in fact completely invisible and redundant.

The vision of the Web doesn't include those either.

IIRC one of the Xanadu hackers invented or reinvented the idea of the cryptohash of a document as a self-verifying identifier for the kind of lookup you mention. Are you sure that was not in the vision, even if Nelson himself was not so technical? (How I remember this: it came up on a mailing list (cap-talk?) when people were looking for prior art to challenge a patent. I may have it wrong.)

Yes, it was one of the Xanadu hackers who came up with that one (but it was never published and stayed internal to that group.) Post-Xanadu the idea migrated to Joule, Inc. then to Electric Communities, then MojoNation, and after that it just became the way things were done...

"shipping is a feature"

Yes indeed, it's the most powerful feature your software have.

Hmm... what about working?

Shipping comes first. You won't find out if it really works without shipping it.

Reply on news.ycombinator.com