Hypertext creator says structure of World Wide Web 'completely wrong'

What?

Classic revolutionary speak ("throw off the shackles of the old") but without offering a revolutionary solution. The solution we have may not be the best possible (and how can we know whether it is?), but it works damn well. A working solution will always beat out a purely theoretical one, and we've seen this time and time again in web technologies.
We don't need rhetoric, if Mr. Nelson would like to offer up a working solution I'm sure we'll all be here to try it out and adopt it. Until then, the commentary is pointless.

What's nice about file structures in general is that they are hierarchical. (Files and folders are also a pretty descriptive analogy.)

Sure, you can introduce new concepts to the system (e.g. using high-level object graphs instead of files) or replace them with something simpler (e.g. a flat, file tagging system like in a CMS). That's fine for a single, independent system. But when you want to *combine* environments from multiple vendors, it's hard to scale.

For example, 3rd party components start stepping on other components in a flat system. And they can easily abuse the power of a more complex system. The file system is a great compromise, since components can retain their own model of how files should be laid out (in their own directory) without restricting (or being restricted by) the models of others.

It's still around because it works.

He can bash it all he wants. It's easy to attack. But it's extremely difficult to produce something better. What's worse is that if two new systems emerge, and they happen to be incompatible with each other (but compatible with current file systems), then we have a severe divergence. It doesn't matter if they're both far superior, since adoption of both new systems would make it difficult for people using them to work together. The common ground is now gone.

I understand the core issue to be - the software giants are not thinking enough to change the old traditional way of using information into something revolutionary rather painting new colors with each version. This is a step forward, yes, the system still exists coz it works but how do we even know there is something better without trying. I believe the question is what is the effort / contribution by these giants to innovate the computing apart from providing a new attire, and everytime new enhancements are done to their software you need better hardware ultimately borne by the consumers.

As a builder and implementer of low level systems which includes operation systems, robotic controllers, and automation system controllers I have seen my fair share of truly bizarre systems that are outside of the normal computing environment (Desktop, Web, Phones) at we use every day. These company’s and people have found ways of doing things that are completely different. Be it the networking of devices (DeviceNet for example), or control systems that never ever crash (you have a system crash on a production line and it could be the end of someone’s life), or ways of visualizing networks, data, and process (use of flow graphs to show process or networked devices and their running states), or ways of interacting with devices (I installed my first touch monitor for an assembly line years before I ever saw one as a consumer product).

Mr. Nelson’s words of “thin veneer makes people think that they are in control of the device” are so true. The current trends are all moving away from controlling physical devices, networks, and information to controlling virtual elements that may or may not have physical representations. As well there are many more traditions left behind from the 40s and 50s then just files and folders. Most of these traditions would be better off left in past. They were developed for different hardware, different form factors, and different social norms then we have today, let alone tomorrow.

I completely agree that next big thing (whatever that may be) will more and likely come from an advanced user that is fed up with “user interfaces [that are] so simple that in an emergency, a beginner is able to understand it within ten seconds.” It maybe time indeed that someone did something radical and new again that will change the computing environment for ever.

Ted Nelson has attention deficit disorder. His hypertext concept is a perfect example of ADD thinking; links in a document to easily distract the user away from the document before you finish reading it.

It is my belief that the wide spread use of his concept in Hypertext Markup Language (html) has made ADD a learned behavior for an entire generation that might not have ADD had the world wide web not become popular.

Do we really want someone with ADD to redesign how operating systems and the web work now so it more closely resembles how he thinks?

I agree with some of Nelson's viewpoints, vague though they may be, but I also agree with this: "without offering a revolutionary solution". He apparently fails to even elaborate on how these things that are "wrong" should be.

There is a reason some of these "archaic" systems persist. They are simple, robust and maintainable.

Is he trolling?

"Another tradition is that one file equals one document."

Or maybe we need to look back to 1945, when Vannevar Bush published the article "As We May Think" in the Atlantic Monthly (there are links to the article - look it up - I didn't want to distract anyone from what I say by providing a link ;-)

He talked about the construction of "named trails" that provided a guided path through the body of the literature, that included the ability to link to portions of another document, or transclude the appropriate fragment in place in your trail. I called these things VT's (Vannevar Trail) - pronounced veetee.

I don't think we have this functionality yet. Remember this article was 1945 - and the "technology" discussed at the time was desks, microfilm, and some means to provide electronic links between documents (not file folders and files).

I think Ted wrote some papers about how even Bush had this wrong - but It has been a while since I read. Interesting to look up Ted Nelson's papers to see what he was attempting to do - far nore than could fit in this small article.

I think I wrote an article about linking for ComputerWorld. Look in http://cyberspace-industries-2000.com/Publications.html if interested.

Increadingly, today, documents are collected together into "Folders". A folder may be implemented by a directory, but it need not be. And many users have forgotten that a folder is actually inside some other folder somewhere. They tend now to think of "My spreadsheet folder" floating all by itself on the hard disk.

The reason that one file = one document is that life is a mess otherwise. ZIP files let you put more than one document in a file, but then it's harder to get at the document you want. External links allow you to merge more than one file into a document, but when you move it, back it up, or send it to a friend, you tend to lose pieces of it. That is why, on the lowest level, a single file works as the most reliable way to store a single document.

Whether that "file" is stored on a hard disk, or in the cloud, or in a database, does not matter. But it should still be an unbreakable unit.

Ted Nelson says that Microsoft, Apple, and Linux all work "exactly the same way". Well, I'd like to hear what works differently. And, more severely, Ted's excuse why his alternative isn't popular. Traditional ways are used because THEY WORK.

There has been lots of innovation in the computer field. And most innovations fail for valid reasons. We had touch screens in the 80's; they failed because after an hour your arm falls off.

“user interfaces so simple that in an emergency, a beginner is able to understand it within ten seconds.” Really? We've had automobiles for a hundred years, and drivers who have only ten seconds of experience kill people! The complexity isn't in the user interface, it is in the process being controlled. It takes a kid a long time to learn to walk, to run, to ride a bicycle; why should anyone expect that kid to learn to operate a computer in ten seconds?

Never heard of Ted Nelson before? What about Vannevar Bush? Check out this video on Mastering Information Through the Ages: http://f4a.tv/ik0YU6

Frankly, I think this is bad reporting. Stringing a bunch of quotes together is not good journalism. How about some context or depth?

Is it just me or did anyone think this was Henry Starling from Star Trek Voyager "Futures End"?

See the side by side here:
http://chimac.net/2011/04/17/twins-at-birth/

HUH? WTF does this mean? Does he want to see ribbons and beams connecting documents or does he mean that today there are ribbons and beams? And how would one refer to another bit of info in an email or other document? Just strike it out and make the connection between sentences? whaaaaa ?
I'm sure there's more to the meaning. - Too bad this is Crap reporting -

"“[My approach] would be entirely different from today's documents where you look at one page at a time and you can see a ribbon or beam connecting documents together,” he said. “Having to refer to a paragraph and a sentence in an e-mail is just so barbaric when you could just strike it out and make the connection between sentences.”

Is it just me or do most old school computer revolutionaries still hold these idealistic ideas of the web as some instant accsess information tool free of any problems? There will arlways be bugs and holes and things people would like to be able to change and the easy solution would just to have a mac or windows system thats open source and moddable possibly with a more user friendly interface in regards to the mods. I feel like what he says is certainly true in some regards i just wish the reporter would have been less spacey writing this.

digged google 30 mins, finally i get what i need.Thanks!