LoadWB's Journal: Where are we heading with computers?

Every so often I come across someone talking about the next big thing in computers. From what I remember, we went from a single personal computer, to multiple personal computers, to desktop convergence (entertainment PC,) to hand-held (ala PDA,) to hand-held convergence, so on and so forth.

From what I understand now, we need:

A single convergence computer which will either
  a) act as an entertainment device playing media from another machine (LAN or WAN server)
  b) act as a server from which other convergence PCs (or simple PCs) may play media

A hand-held multi-purpose convergence machine which acts as a personal organizer, cell phone, digital camera, audio recording device, multimedia playback device, and also be able to interact with desktop convergence PCs/servers via Bluetooth, IrDA, WiFi, or other wire/less networking interface.

Wow. Can I get mine in a watch, with fries, please? (Funny enough, I have a Casio watch which takes color pictures at QVGA... alas the model I have only displays grayscale at the face and has been obsoleted by a model with a color display!)

(Oh, and let's not forget the DRM... that's the large shake.)

ANYWAY... the battle now is who is going to take us there, how, and when.

As a system admin by trade, programmer and hardware hack by hobby, I see a bigger problem. How are we going to achieve the infrastructure?

Turning our PCs into dumb terminals for displaying content seems to be the big thing. But that doesn't remain inexpensive, and it also requires (with the advent of massive networking operating systems like Windows, Linux, MacOS X, et al.) that the user become an administrator. Wow, again. My grandmother just wants a computer to send email, browse QVC, and get pictures of the family. She doesn't need to be concerned about service packs, critical updates, dependencies, and of course malware.

So that leads to the next logical step, making the system administration-less as far as the user is concerned. This happens by moving the OS, at least the bulk of the OS, to the content provider. Ala WebTV or other (TV) set-top boxes.

Discussion of the security implications of the client OS and network transmissions is beyond the scope of this entry, and an exercise for the reader.

To continue... That's nice and all, but the WebTV-alike system is also limited, and inherently untrustworthy.

The biggest concern out there is a person's data, information, and identity. So, here we have moved from the "fat" client (ala AOL on a personal computer,) to a "thin" client (ala WebTV-alike set-top unit.)

Let's press on. In a previous post I made I discussed the potential for a "chunky" client. It's not fat, it's not thin, it's kind-of husky... CHUNKY. What's this all about?

Well, let's take the good parts of everything so far:

The centrally administrated system offered by the thin client. The ability to keep one's information local and private (admittedly, mostly, though this can be changed through implementation, a piece-of-mind thing.) The ability to perform "fun", useful, or productive tasks.

errr Sort of a minimal boot environment with terminal services and a back-mapped drive.

CHUNKY CLIENT.

The idea is the the "server" operating system is maintained by a provider. The user can choose which information is stored on the "server," and which information is maintained locally. This gives piece of mind that sensitive information is not kept in an active location. False sense of security, it might be, dependent upon total system implementation.

The client will have a hard drive, a basic operating system, and some some kind of media drives (remember, we want convergence, too) for multimedia playback and recording. Should the hard drive fail, the system will boot a stripped-down OS from the server. PXE-ish.

Now, what about the fun, useful, and productive aspect? Well, we all like software. We all like games. And hell, we all LOVE to work. At the server applications are available for people to use... word processors, graphics editors, web browsers, games, etc. A virtual catalog should be available which allows users to select software to "install" to their profile. Such apps may already be installed by the "server" admins, but not available to a user profile until purchasing, licensing, whatever.

I like the idea of distributed computing. Something like this system could be run on my Amiga 1200 50MHz 68060... but where are we going to find such, uh, robust hardware these days? The average client hardware will more than likely be pretty beefy in the CPU arena, and probably have a GPU as well. Some applications may need to have parts run at the local client level (consider games with heavy CPU and graphics requirements.) Parts which would not only tax resources at the far end, but also the bandwidth in-between.

And so on.

Hey, this is only a journal, not a thesis, so I'm dodging some bullets here.

So, in the end we have the Chunky Client System:

Centrally operated and maintained
Local storage capabilities
Local peripherals (printers, storage devices, etc.)
Far-end application execution
Distributed processing, including local execution
Virtual catalog of applications available to user

I don't know of any kind of system like this which exists. I'd like to think that I'm one of the first to articulate it. I'd like to think that one day this kind of thing will come about.

Hey, can we program on a system like that? Give us BASIC, dammit :) Oh, distributed LOGO. Okay, that's enough now. eerrrr one last hit... Visual RPG-II.

Okay, I'm done.

To reach this kind of communistic technology nirvana our current thinking of technology needs a rethinking. We cannot stick to our current pragma of operating systems, mice, and men.

Oh, and marketing for this system. I think that such a system could run hand-in-hand with the traditional PC for many years. Many of us still want the freedom of modifying, tinkering, and doing whatever the hell we want with our hardware. Such freedom simply cannot be afforded with the Chunky System, at least not easily and without risk. But there is a large part of the populace which doesn't care, doesn't know, doesn't care to know, doesn't have the time to know, or some combination thereof.

I'm really done now.