My takeaway here is that this just further proves the excruciating reality, that bringing new innovative technology to the market is inextricably incentivized to involve litigation, politics, infighting, backstabbing etc... in other words normal business practices in a highly competitive market. Competition for dollars just fundamentally makes this a reality.
I certainly wouldn't offer that there is any better system/process for this, but it remains disheartening for idealistic (some could say naiive) technologists that this is the case.
I think it's worth keeping that in mind when thinking through how to create the "next big thing", or even some next small things.
I wonder how much it of the problem was the Innovator's Dilemma and how much was the marketplace and technical infrastructure not being ready.
Dick Tracy had a smartwatch, but it took a while to actually bring it to market.
---
Edit: after reading to the end of the article, there were further difficulties: the units cost over $6,000 (adjusted to 2019 dollars), also there were supply chain problems from their competition (not sure how seriously to take this one).
Reading the article my impression was that it was a device well ahead of its time.
The CPU was some oddball thing much faster than its contemporaries that no doubt cost and arm and a leg. The battery technology wasn't ready so they had to compromise on life. The cellular modem technology was first gen: slow, expensive, unreliable, and without an Internet to connect to. The demo had them sending a fax!
Plus they had to build the OS from scratch with an entirely new paradigm (no keyboard!). And it had the unfortunate first gen pen-interface mistake of focusing heavily on handwriting recognition tech that never worked properly on that generation of hardware. Even today its pretty iffy, and virtual keyboards are the preferred solution. The only company that kind of got away with it was Palm, and only because they invented a new alphabet that would be easier for the computer to recognize.
It also goes to show just how much of design is not only coming up with the ideas, but realizing when the technology has advanced to the point where they can be practically implemented.
A software solution would be that graffiti Android keyboard, a hardware solution would be the FXTech pro 1, which is among the first landscape Android slider smartphones to come out in years.
That's not my takeaway - I think they just critically mistimed the idea and didn't have much business sense to realize it. Even if their experience was bs-free and they were extremely well capitalized, they wouldn't have succeeded because the tech wasn't anywhere near ready.
90s tech (batteries, cpus, screens, etc) could only reasonably produce single function hand-held devices like phones and CDs. Even late 90s tech like in Palm Pilots could only make pretty crummy experiences that had limited mass market appeal.
The Palm Pilot was pretty brilliant. It was so cheap! And as a result I imagine it got a lot of people used to the idea of being able to do "mobile computing", and what kind of activities would work well in the form factor.
I can still remember the sound of my m100s clicks.
I had a Handspring Visor. They were pretty trendy in 1999-2000. My friend had the Ricochet cellular modem and we were jealous he could get email on it. But there were apps similar to Yelp with databases of restaurant reviews. It would sync in the cradle, attached to your PC, to get the latest data over the Internet. Imagine an iPod touch with no radios.
Yeah these devices did check some boxes for various features and capabilities, and as a late 90s tech geek I wanted one badly because they were clearly 'the future', but the fit & finish in the UX just wasn't there for breakout success. That was largely a function of the state of hardware.
I also had a pseudo-iPhone experience with my Treo in 2005-07, with an add-on bluetooth GPS and various apps for MP3s etc. It did the job in a way, but it was clunky, and the web experience was subpar to say the least. But it was a sign of things to come.
Trendy is a good word for it. I won mine at some trade show raffle and I know lots of tech folks who all wanted one during a certain period. However I also remember that, among other things, the cradle sync requirement made them not really all that terribly useful. Mine became drawer-ware after a while and I don't think that was all that unusual.
I am persuaded that invention effectiveness in the market is almost entirely a function of business acumen, and almost completely divorced from the technical capability. Although a total lemon will fail.
I'd say you brought your takeaway along already, i.e. your conviction on how the world operates - and you "took it away" again since all that we can "discover" is usually a subset of what we believe already.
>In 1991 when "pen computing" was all the rage, a start-up company called "Momenta" came out with a "pentop" MS-DOS based pen computer with a pie-menu-like "command compass", whose user interface was done in Smalltalk.
>Remember that this was in 1991, the same year Go Corporation finally released PenPoint, and a year before the 1992 founding of Palm Computing (which took over the Pen market for many years) and the release of Microsoft Windows 3.1 in 1992 (the first version of Windows that wasn't intolerably irritating), and the release of Microsoft Windows for Pen Computing in 1992 (which was a big flop). So there was a huge amount of excitement around pen computing.
[...and more about how their founder confused Smalltalk with Lisp, plus links to photos, reference manual excerpts, and more info about the Momenta pen computer...]
I had (and used) an Amstrad PenPad for a number of years (until I replaced it with a Sharp Zaurus 5000D which ran Qt on Linux). The down-side to both was a lack of connectivity!
I saw a presentation about Go and PenPoint in the early 90s, probably at a Usenix conference, and I was very impressed. I was sorry it never went anywhere.
My other PenPoint story comes from a business trip I took to Japan in 1994. One weekend, I went to a gay bar in Tokyo (GB, in Shinjuku) and met an American who was working for Go and was in Japan to work on Asian language input (which was an interesting problem). I mentioned that I had a friend who was working on Hobbit software and it turned out my friend and this guy had exchanged email the week before! Two American gay nerds meet in a bar in Tokyo and find they know another gay nerd in common. Crazy.
OMG! Do you know the dude in the Terrapin Tacos interview who said he is there five or six times a day, and sits up all night thinking about how good the food is? They pick the rocks out of your beans.
I'm surprised that this article doesn't mention General Magic for context. It's really crazy how much of this was invented and reinvented but the infrastructure and performance just wasn't there during the 90's in a way to support mass adoption. It was close enough to build something, but all of those somethings failed until WiFi, cell technology and Moore's law supported small enough gadgets. If you haven't watched it yet, I highly recommend the documentary on General Magic. https://www.generalmagicthemovie.com/
My roommate and I went to that COMDEX in 1992 specifically to see this device. When we got to the booth they wouldn’t let us in, so my roommate, thinking much more quickly than me, told the dude at the booth entrance that we were developers (in fact we were juniors in the BYU CS department). Our status immediately switched to “VIP” and we were escorted inside, handed a device, and left on our own to play with it for half an hour.
I thought I was seeing the future of computing.
Unfortunately, Microsoft was also at COMDEX, in force, with its “Pen Windows” product. Which was clunky and stupid and very clearly only meant to crush GO. And it was clear even then to me as a 20 year old that that strategy was going to work.
I was working for a chip startup in the early 90s - we moved into office space in Sunnyvale that had recently been vacated by the GO corp (I think) .... above my desk were dozens of pencils embedded in the ceiling tiles ... some poor soul sitting there as the company company was dying has sat there throwing them up until they stuck ... hopefully he or she retired with both eyes .... I removed them before they came down on their own
>> “They started with the iPhone, and went to the iPad,” Kaplan says, referring to Apple. “But we were starting with the iPad and moving to the iPhone.”
That’s not true, from a development side.
Jobs stated that the development of the iPad started before the iPhone.
I worked there until very close to the end, and the collapse of GO/Eo provided me with a priceless instant network of friends at dozens of companies around the SF Bay Area.
I met my spouse and some of my best friends through GO/Eo, and to this day still greatly value the connections made there for my personal and professional networks. Silver linings.
> AT&T eventually took control of EO, which it bought in mid-1993, and GO, which it purchased outright in early 1994. These acquisitions created a confusing corporate structure.
Yeah, AT&T was on a buying binge at that time. I got hired there in late '93 and started in '94. In addition to GO, they also bought NCR and a few other things, I think. All of it fell apart within about 2-4 years when they spun off Lucent and NCR into separate companies. It was pretty much a mess!
And the iPad of the 80's would be the Apple //c as featured in the movie 2010. It was seen being used on a beach, complete with LCD screen, and this was 1984.
There's a couple of issues though, the //c did not run off battery power, so you'd need a long cord, and the LCD was terrible, I don't think you'd be able to read it in direct sunlight. But if you have a long phone cable to go with the power cord, maybe you could have sent a fax? Or at least a telex.
You had a few early laptops like the DG/One [1] that could run off of battery by 1984. But there was really no way to communicate without plugging into a phone line. (You could maybe have used a radio modem in principle but you mostly had to connect to landline phone service to communicate.)
Way before my time, but I got to play around with one in 2000 and really liked the form factor.
There were also glorified calculators like the 1980 TRS-80 PC-1 (https://www.old-computers.com/museum/computer.asp?c=937&st=1). Never used that particular model, but that style of keyboard/display in a palmtop computer is not very ergonomic IMO.
That list may look long, but it is probably far from complete. For example, at the 2017 VCF PNW I got to play with a 1980s "glorified calculator" Lisp machine handheld. Can't remember the manufacturer, but it is definitely not on that list, and the only reference I was able to find with a quick web search was this: https://en.wikipedia.org/wiki/Talk:Lisp_machine#Sony_and_Lis...
So there were a lot more portable computers around in the 1980s than you suspect.
In the early 90's, I remember seeing a dude typing away furiously on an Apple //c at a coffeeshop on Haight Street in San Francisco. I did a double take when I realized he didn't have a screen or a plug. It turns out he was a stenographer, just practicing his typing.
A hillbilly never-used-a-computer-a-day-in-his-life relative of mine has one. He bought it new, but never even turned it on.
The last time I saw it was about five years ago, and it was in absolutely mint condition, even though he keeps it under a tarp with a bunch of junk in a barn.
For decades he kept promising to send it to me since he has no use for it. Recently that changed to "I'm gonna use it some day" so I no longer have any hope of getting my hands on it.
My MessagePad 2100 could do AppleTalk over it's serial port and wired Ethernet with a card. There was also a pre-wifi DigitalOcean device[0] that could do wireless AppleTalk.
There was IrDA (infrared based communication) that could be used for serial connections to other Newtons, computers printers and cell phones. It was slow though, so the better step up was an actual serial connection to either a modem or a cell phone (acting as a modem).
Later versions of the OS (2.1 for sure, and I think 2.0 could run a few specific Ethernet cards) had Ethernet (10Base-T, and technically 10Base2 I think) and Wifi (802.11b only) drivers written for them.
Not out of the box and probably not for the first couple years at least and I don't think wireless (i.e. cellular... there was no wifi back then) ever. There was a serial modem dongle but that still required a land line.
The closest thing we had to wireless networking was infrared serial to either other devices or a computer. It generally didn't work well.
The Newton could connect to some online services via wired modem. It could also do e-mail and with some additional software PPP and TCP/IP. Apple sold a wired serial modem for the Newton and several PCMCIA models were also available. It was possible to use a cell phone to dial out, treating it like a serial analog modem. I don't know anyone that did that but it was technically possible.
There were a few good network apps for IRC, newsgroups, and even a few limited web browsers.
With Newton OS 2.x people got WaveLAN (pre-standard) and later a few 802.11b PCMCIA cards working. It was all a complete pain in the ass and only the most dedicated Newton fans even bothered.
Sounds like you stuck with it longer than I did. I can only speak to the original MessagePad 100 since that was the only model I bought. I still have the Apple modem dongle lying around here somewhere which is what I was referring to. While it worked, I never found it a terribly practical peripheral: slow even for its time and it devoured batteries.
The Apple modem had its own AA battery but it did devour them. I had one but only used it a handful of times and each time was a hassle and ultimately disappointing.
Other comments have mentioned some options that the released Newton had, but the Newton actually had network support during development; we just couldn't work out how to fit it in the package and make it affordable.
When I first joined the Newton team the hardware guys were testing a spread-spectrum network that supported roaming devices with automatic discovery and connection. At that point all the working Newton hardware was big clunky naked circuit boards, so they tested them by ribbon-cabling them to Macintosh SEs that they wheeled around on carts to go from one network to another.
There were quite a few devices in the same category. None of them really succeeded but I think a lot was learned from them which then could be used from the 2000s in when things started to take off.
Cellular data--or at least widespread WiFi--was pretty much a prerequisite for this class of device. That's one reason that the Palm Pilot was always something of a niche device. The need to constantly sync it made it a lot less useful than it would otherwise have been.
(I'd argue that background syncing over WiFi or cellular networks is also a reason why Podcasting 2.0 is a lot bigger than Podcasting 1.0 was.)
Similar to that AT&T campaign, I was struck by the prescience of Qwest's "Ride the Light" commercials: "All rooms have every movie ever made, in any language, anytime, day or night." This was in 1999, when Netflix had barely started mailing DVDs.
> "We could have had the iPhone 10 years earlier, at least."
The fate of prototype techs is almost always to be remembered with these kind of wistful statements about how they could have been [market leader] except for [reasons].
These statements betray an ignorance about just how category-defining products come into being. Making even a moderately successful, let alone category-defining product is super hard and it's nowhere near enough to just have an idea, whether represented on a napkin, or in a demo video, or even in a tangible prototype. Ideas are cheap, even ideas that are rather prescient.
Making successful products also requires TIMING, LEVERAGE, and EXECUTION.
TIMING - build something too early and you waste capital fighting uphill against a market not ready for the idea. Product-market fit is not always possible for any given idea at the current moment in time. For tech, things like battery density, speed, screen tech, internet bandwidth, cost etc are critical to time correctly.
Example: Youtube and Netflix appropriately timed the cost of storage and internet bandwidth so they were ready to go when streaming was just becoming cost-effective, versus earlier efforts like Broadcast.com. It wasn't a leap to conclude video streaming was a good idea even in the 80s or 90s, but it needed to be timed.
Additionally, timing things culturally matters a ton. People need to be ready for an idea, to incorporate a new product into their life. The window of what's cool or acceptable is a moving target, what's uncool now may be cool in 5 years. It's often other interim products that shift this window.
Example: Google's launch of consumer-focused Google Glass and the immediate blowback that resulted.
Finally, pioneers get arrows in their backs. You reveal your grand ideas too early and you've just paid for a ton of concept development R&D that others can just pick off for free.
LEVERAGE - Ideas require business leverage to succeed in market. Things like brand leverage where you're already trusted in the market and have a history of delivering quality, the ability to scale manufacturing to meet demand, business relationships and scale to get access to special components or lower component costs, a well-oiled organization with deep talent bench, etc.
Example: The iPhone would not have been what it is without the iPod. Apple leveraged its talent, its history of manufacturing iPods, its suppliers, and its huge consumer appeal. It also used that to negotiate with AT&T to have complete control of the UX and to get unlimited data plans.
EXECUTION - Lots of high-concept ideas are lousy at the exact details and execution of the idea. Cool, you can send a fax from the beach, but under what conditions? Is fax even the right modality? Is the UX cumbersome, or require a laborious setup process? Its often the little things that separate market defining products from abject failures, even when they check the same boxes.
Example: Galaxy Fold is great high concept product that doesn't execute on its promise because its too fragile and creates a screen crease.
It's also possible to do a great job on details, but within the wrong constraints. You can be efficient (doing things right) but still not be effective (doing the right things). Specifically in the case of this 'iPad of the 90s', they didn't have the enabling technologies to make a great experience - things like fast refresh rates, multi-touch, beautiful screen, etc. They probably did a fantastic job with the pen, but the pen was the wrong constraint.
Basically the creators are severely downplaying the timing, leverage, and execution required to make products successful, and seem to think this cool but barely-launched concept could have been the most successful product on earth (iPhone). I doubt the team would have been able to produce anything more than a PalmPilot or Treo, both of which were OK products for their time, but ultimately are just mostly forgotten preludes to the product that DID define the category.
other prior but non mainstream commercial products are demonstrated in many Alan Kay talks (sketchpad and the other diagram/pcb cad tool I forgot the name of)
Yeah, no frequency capping. I got all Apple Card ads, which tarnishes Apple Card brand value. This feels like a dollar store theoutline.com. Brutalism doesn't work for long-form content, imho.
Not sure but there have been lots of idiotic use cases in tech demos of recent years. Consider the demos of the impossibly stupid Microsoft Surface Table, which you must use to order cocktails at a bar, or browse photographs in a really inconvenient way.
That was 2008. More recently, a rational and critical look at 99.9% of proposed use cases for blockchain are things that nobody wants.
Another thing that literally nobody wants but people keep trying to market: folding displays.
Actually there’s whole business models that are based on nonsense. Like the idea that people will pay someone to drive over to Subway and bring back a terrible $3 sandwich, and that this scheme will somehow be systemically profitable.
Maybe not a $3 Subway sandwich but the delivery of pizza in particular was pretty widespread before the current VC cash bonfires so I'm not sure it has to be "nonsense."
I expect a lot of people 20 years or so ago would think that the idea you could get many, many types of things delivered to your door in a day or two for a reasonable price pretty out there.
The delivery aspect of the pizza business does not need to be profitable as a stand-alone endeavor, whereas the delivery part of Uber Eats does. And it's the driving part that doesn't make any sense. Obviously people have been delivering food in dense and compact cities like New York for decades.
- Call local pizza joint, tell them to make a pizza
- Pizza joint cooks pizza
- Pizza joint tells young kid/local burnout to bring me pizza
- Hand young kid/local burnout $$ for pizza, $$ for tip, have pizza.
Now:
- Tell a publicly traded company to to tell your local pizza joint to make a pizza
- Pizza joint cooks pizza
- Multinational logistics monopolist-wannabe tells young kid/local burnout to bring me pizza
- Take pizza from young kid/local burnout.
You may notice a couple insertions that don't seem to make a whole lot of sense, but nevertheless are taking a significant percentage of the, well, not the pie in this case. The pie isn't getting here faster nor does it taste better. Neither the pizza joint nor the young kid/local burnout are making more money. It isn't more convenient.
Taking cash out of the transaction legitimately does reduce certain risks. That is an advantage to those who aren't reliant on cash, but note that neither Uber nor Grubhub are required to do that.
Remote surgery. A doctor is vacationing in The Bahamas and gets an emergency message that a patient needs to go under surgery or will die. The doctor pulls out a briefcase and opens it up, where there is a screen and interactive controls where he can remotely perform a life saving surgery
I certainly wouldn't offer that there is any better system/process for this, but it remains disheartening for idealistic (some could say naiive) technologists that this is the case.
I think it's worth keeping that in mind when thinking through how to create the "next big thing", or even some next small things.