What I learned from a 1993 issue of WIRED I found in a book box on the sidewalk:

[mattbell]

I didn't know it was from 1993 at first, but I was startled to see a picture of Sonic the Hedgehog on the cover, so I opened it to a random page and saw an ad for an Apple tablet -- no, not the iPad, the Newton MessagePad. The ad copy was very wordy compared to modern Apple, but the feeling was more or less the same. "We have a shiny and brilliant product that will become your indispensable new best friend"

Here are some things I learned by reading it.  I knew a lot of these already, but it's good to have them validated:

- Lots of technologies and products have been around a *long* time and are riding a long, slow curve of decreasing cost and increasing ease of use and features. No one knows when they will pop over a threshold and begin to see market traction... it could be within a year or it could take another 20.  Some things I was surprised to see for sale:
-- Massively multiplayer online gaming
-- Serious games for adult education
-- 3D mice
-- Automatic business card scanners with OCR to automatically read the text on them
-- A Google Maps equivalent (with info and reviews of businesses, directions, etc) that ran off the desktop.  Get one city on a floppy, or all of them on a CD-ROM!
-- TVs with face recognition that would tell who was watching.  (For developing a more accurate TV viewership rating system)

- Software is unique in that just about any software tool (eg a spell checker) goes from expensive to completely free and integrated into a larger package within a few years.   This happens to a lesser extent with electronics and much less with anything that has a significant materials cost. 

- Offering a free, unlimited plan and then switching to an a-la-carte plan will piss off your customers. (Prodigy went from unlimited email for $15/mo to "50 free emails + 25c per additional email per month" in 1990. Users revolted.)

- WIRED didn't see the Web coming at all. It was all about closed, walled-garden online experiences.

- On Steve Jobs:  "NeXT is like [a] model of Steve's head: brilliant, charming, but with no input jacks"

- It's really freaking weird to read a hype-y tech magazine full of ads for cool, futuristic devices and then seeing a phone number to call to learn more about them. Maybe 25% of the ads had an email address you could use to contact them.

Comments

[madbard.livejournal.com]

Wow. I'd forgotten how old Wired is!

[madbard.livejournal.com]

(This is Desiree's friend Mike from L.A. I followed the link on your FB page. It didn't even occur to me until after I'd responded that I'd transitioned to Livejournal.)

[nasu-dengaku.livejournal.com]

Hey, good to hear from you. I just added you as a friend.

[merovingian.livejournal.com]

RAD.

[sandrak.livejournal.com]

Yeah, they had OCR back then-did you have the misfortune to try to use it? I did! In fact, I was so frustrated by it that I have no idea of the current state of that technology-I was traumatized! As far as tablet pads-they were all the talk of Comdex 1991-but they were all heavy, monochrome with no backlight. That was the same Comdex where I asked a music software booth about storing songs digitally so you could just play them off the computer; I was told that it would require about ten mb per minute. Given that my desktop had 175mb hard drive at the time (which was pretty normal), I knew it would be awhile... Must have been pre-MP3

[nasu-dengaku.livejournal.com]

No... I didn't make use of OCR until freshman year of college (1998). I had no idea there was a big tablet boom around 1990.

I'm guessing that in 1991 computers weren't fast enough to decode an MP3 in real time (even if the MP3 format had existed back then).

[easwaran.livejournal.com]

OCR, directions, and face recognition are all software problems that have taken a long time to reach the level of succeeding at their goals. I suppose your adoption curve from "expensive to completely free and integrated" probably starts only once the software actually succeeds at its goals. Thus, OCR and directions didn't catch on for the general public for a while - I assume that in 1993 OCR was only good for recognizing ZIP codes, and directions would have a lot of inefficiencies (and they still do, but they're good enough now to actually use most of the time). As I understand it, face recognition is still basically unusable except possibly in cases where there's a fixed and relatively small pool of possible faces to choose among.

[nasu-dengaku.livejournal.com]

Yeah... all the machine learning domains have problems that get a lot easier or harder depending on the number of items to be identified and how much different examples of each item tend to vary. A 1993-era OCR could probably recognize letters and numbers in a narrow range of typical fonts. The Nielsen face-recognition box only had to distinguish between a few different family members and could spend minutes analyzing any given image.