Calling All OS Designers: Redesign "Save As..."

It's time to redux "Save As..." Keyword tagging is everywhere and there is a resurgence in metadata collection, now that benefits are being realized. Indeed applications have included metadata forms for generations but no one ever uses them because they have always been hidden from our awareness and critical task paths. In other words, you have to go out of your way to collect metadata. But this doesn't need to be the case. We learned while working a very complex information organization problem for the FCS Lead Systems Integrator that it is entirely practical to insert minimal steps into the task of bookmarking in order to create a much richer and more easily shared collection. It wasn't a matter of automated versus manual tagging since senior subject matter experts were already using browser-based, hierarchical bookmarking techniques to keep track what they were finding while conducting their research online. (More on that later.) Instead, it was a matter of suggesting that, "Hey, since you are already actively browsing and bookmarking, would you mind if we pop up a window and collect a few key words?" The cost was the number of keystrokes and/or clicks to enter key words plus one to submit a pre-populated tag form.

And so this is what I am suggesting: We should add this kind of metadata collection to the "Save As..." task. Of course it can be opted out, but Save As... is the semantic equivalent of browser-based, hierarchical bookmarking. I won't argue the pros and cons of keyword tagging versus hierarchical cataloging. I know from experience that both have value and that one is much more versatile than the other. I would just ask, how many times do you find yourself saving files with names like:

"Arxiv_electron_force_depth3_foregraph_select_highly-referenced.bmp"

Maybe it's just me and since I can really only speak for myself, I will: I damn sure would enter a few key words during the Save As... process if only I had the means. It needs to be universal, too, or as close as we can get given the hodge-podge of technology on our machines.

I Can't Believe I'm Thinking This...

I just stumbled across a post from Bruce Sterling on Wired's "Threat Level" blog: "Threat Level is fascinated by the National Cyber Range." I found it because I happened to have read the BAA last week and I was now doing a little homework for a proposal. Let me first record that I applaud Mr. Sterling's calling attention and his quippish commentary. I have to admit that I've had many of the same thoughts over several years of working in this business and even today as I consider implications of potentially working this concept. But given rationalization, a most useful tool and frequent resolver of guilt, I am compelled to counter Mr. Sterling's remarks.

I'll start with the obvious: DARPA invented the Internet; the thing Mr. Sterling is using to exercise his freedom of speech (in the wide open, by the way - something that will get you jailed or killed in other places in the world). Brilliant and now legendary inventors like Vinton Cerf worked for DARPA. Two prominent California universities participated, too. Even then the Internet started with a trepidatious title: "Galactic Network," but here we are.

But enough of the obvious. It is easy to pique our fears by tapping into what is surely a bottomless well of sarcasm and criticism, particularly when one is not bothering to give treatment to any other relevant possibilities. At very least, an entirely contrary position owes some consideration of likelihood. Generally from there we do a good job of protecting ourselves from those downsides, particularly, believe it or not, where our privacy and rights are concerned. A more appropriate criticism would be one that counters the supposed benefits an endeavor such as this program may suggest to offer. Since none are suggested by Mr. Sterling I will have to come up with my own.

I have to start by asking myself, "What existing problems can this solve?" Or, "What questions can this answer?" Or better yet, just plain, "Why?" I know I have to consider these questions in "DARPA-hard" terms. It seems naive to reduce this to a matter of...what? DoD's attempt to CONTROL EVERYONE AND RULE THE WORLD!? ...muhahahaha! (I get sarcasm.) I am absolutely certain that the kind of folks who would work on a program like this are smart enough to know that the concept of people exchanging information in the form of data packets over wavelengths has no constraints on infrastructure. There is always peer-to-peer.

I do suppose it is worth making explicit, by law, the fundamental right to operate your own network on your own infrastructure...

If there was ever an obvious example of a lab experiment having massively unforseen consequences beyond the lab it is the Internet. Ironically, the very spread of the Internet itself must have all of the same characteristics of the spread of an Internet virus or else Internet viruses would lack the very means attributing their basic behavior. Spread of cell phone technology is even more pervasive. And it is all fantastic. But there is some really bad, bad shit that happens on OUR Internet - the one conceived of, invented, and owned by civil people - and it has nothing to do with privacy.

Spam is beyond out of control. It may be intractable. DARPA does intractable. Who would tolerate hundreds or thousands of phone calls daily? Why is Internet spam so disproportionately leveraged as an advertising medium compared to junk mail? Is it because spammers think it's better marketing? Or could it be because criminals are exploiting our Internet for personal gain?

Why would DARPA want the "ability to rapidly generate and integrate replications of new machines?" Could it have anything to do with the existence of multimillion node botnets wreaking havoc?

Then there's this CNN report of researchers demonstrating the feasibility of hacking and destroying critical infrastructure.

So, I just don't know. What exactly is wrong with trying to fix that stuff? Why can't that be DARPA's reason?

How about this one: "The ability to replicate large-scale military and government network enclaves." De facto, the military and government have every right to pursue that ability for themselves if they think they need it. Er, ourselves...if we think we need it. Whether we like it or not the military and the government is us and us and us. But, hey, wouldn't it be great if DARPA invented a revolutionary new way to replicate [FOO] large-scale network? Like the ones on which our banks run.

And this one: "The ability to connect to distributed, custom facilities and/or capabilities as necessary to incorporate specialized capabilities, effects, or infrastructures." Hello, system integrators. DARPA should call Apple for this one.

And: "Realistically replicate human behavior and frailties." Well, that's just DARPA being DARPA-weird.