Slashdot Mirror
Don't Write Them Off: A Palm Retrospective
An anonymous reader writes "OSNews' managing editor Thom Holwerda has posted a lavish five-part retrospective on Palm, covering its history, user interface, internal technology, and competition. Holwerda first pays tribute to the pioneers of automatic handwriting recognition, including two remarkable stylus tablets (connected to mainframe computers) produced by RAND Corporation during the 1960s. The action picks up a couple decades later as Jeff Hawkins implements a handwriting recognition engine for his employer, the makers of the high end GRiD compass (MS-DOS) laptop. Hawkins dreamed of developing handwriting recognition for a device small enough to be carried around in one's pocket and cheap enough to be sold to a mass market. Along the way he had an epiphany: instead of trying to recognize the user's natural handwriting, why not create a simple alphabet that could be recognized reliably by the software? When Bill Gates entered the game, Hawkins had another big idea: why not compete against the Microsofts of the world by having fewer features, instead of more?"
The handwriting recognition part is chock full of screenshots and video demos of early recognition systems, too.
I still keep my appointments on a Palm Zire. The simplicity is hard to beat, and I mastered the stylus alphabet many years ago. It's a little thick in my pocket, and many people have asked me about my special cellphone.
I loved my Palmpilots - I still own two even now, but their real problem from a programmer's perspective was the overly restrictive 64KB model and having to use a database for all storage. That meant everything had to be especially written for the palmpilot and it was difficult to include support in a cross-platform programs. Not impossible, but it felt like it was Palm or everything else at times. As the devices got better and better, the tiny memory restrictions just got more ridiculous.
Also, there was some developer support issues - for instance, I remember at the time when people wanted to create pdb files offline and the official response was "we don't know what the pdb format is because the MS runtime handles that" was just ridiculous. Obviously, it was possible to reverse engineer the format, but a company not having documentation on its one and only file format isn't great.
That said, the API was well thought out and nice to use. Just different to everything else.
I wrote a couple of programs using PocketStudio (Pascal based RAD tool) for a Tungsten C. One of them used the IR interface to show status and do configuration on a PIC-based solar power system. The other used WiFi to talk to a model railway signal and turnout control computer.
The old mono palm-pilots were fantastic and I still haven't found an app that can hold a candle to the Palm suite. The battery life was phenomenal and it's only recently I've had a backlit e-reader again.
Vale.
POKE 36879,8
When I had... Pilot 5000, I remember finding on a palm forum some fantastic renders of 'what if' type devices. Someone had taken a palm pilot and said 'ok, 10 years, what are we likely to get'. Full colour, glass screen, thin, internet access. I so wish I'd saved them somewhere I could find again. It was some impressive designs for the future at the time, but of course in retrospect. I think they hit the nail on the head though, if only they were still about, perhaps some of the 'look and feel' lawsuits going on might have prior art renders.
Waiting for an amusing sig.
Palm's handwriting innovation wasn't technological. It was psychological. They managed to convince consumers that it was cool and hip to learn to write in a way that the device could interpret. That made the technical aspect much more manageable.
That's one of the reasons the Newton bombed. Apple tried to build a system that could interpret natural writing but that's an incredibly difficult thing when writing styles are as unique as fingerprints. I didn't even bother waiting to get my hands on a demo unit at the launch because people were walking away complaining that it was impossible to get the Newton to accurately recognize anything written on it. That had been the Really Big Thing Killer Feature and it was underwhelming. So I'm supposed to walk this thing through learning my writing? Ain't nobody got time for that! Somehow, Palm convinced people to learn how to write all over again. I tip my hat to their memory.
I'm no Apple fanboi, but the Newton's (old jokes aside and it was tightened up immediately and in ensuing OS updates) implementation was a miracle to behold back then in '97 or so. Palm had that mess "graffiti" which I always resented because it made me learn to write all over again. I liked a machine that comformed to me, not the other way around.
www.chihuahuarescue.com- Help to end dog abuse, abandonment and cruelty
Decent article, although the conclusion (that Palm should have built a new device based on Cobalt or any of mess coming out of Access) is misguided.
The story of Palm is not too far flung from the story of Apple in the 90's. Architecturally, Palm OS was similar to the classic Mac OS: handle-based memory management, limited multitasking (using event loop tricks), and no memory protection. Both products were simple, elegant, and popular in their heyday.
But eventually, the designs that worked well for old 68000 processors with miniscule amounts of RAM and processing powers started to get creaky. Both Apple and Palm switched chips (PowerPC and ARM, respectively) with emulation layers, which extended the lifespan of the aging OS even longer. Both experimented with licensing the OS to other manufacturers, with poor results for the bottom line.
Apple started writing a "next-gen" version of Mac OS, called Copland. Palm/PalmSource started writing Cobalt. Both attempts were ultimately too complex and mismanaged to actually ship, despite all their "promise". Apple tried to buy BeOS, but Be wanted $200 million. Palm bought Be in its decline for $11 million and change.
Apple finally succeeded in putting classic Mac OS to rest by switching to a modern UNIX-based OS. The lesson is that writing your own OS from scratch is freakin' hard. Palm started fresh with webOS, built on a Linux core. Ultimately, thanks to years of mismanagement and stagnation during the times when the company was making loads of money in the 90's, it was too late for Palm. And then they had to compete with the original Apple: Apple.
On top of that, a number of ex-PalmSource employees had long ago ditched Access and went on to work on what would become Android ... which included using the Binder IPC technology originally developed at Be.
Palm had the first integrated smartphones, the Treo series. Camera, PDA, net connectivity, music and media all in one.
They were far ahead of the first iPhone in terms of features. Stereo bluetooth and copy paste were there way before iPhone implemented them. Palm had stuff iPhone never will have, like hardware keyboard, SD card slot, user-replaceable battery, install any program, multi-day battery, and actual physical buttons.
Too bad the OS was dated...
I passed my whole college time with a PalmPilot and a Palm III and it was great to develop stuff on it and to carry around. I even remember connecting to the Net with the modem dongle which connected to the bottom port to check emails and browse online.
The worst thing about them was that they totally messed up my hand writing.
I owned two Palm devices. In their heyday these were great devices. In their heyday I would not have wanted any other type of PDA. But I feel no nostalgia whatsoever today for those devices. My old Palm devices don't hold a candle to my Android devices. There is nothing, absolutely not-a-thing, that my Palm devices did that my Android devices do not do better. Handwriting recognition? How about entering note using real-time *voice* recongition.
Palm's technology was incredible & ahead of it's time until it wasn't. The CEO and other C-team needs to keep pulse on the market and focus their organization and how they need to change to be relevant on the future.
For this they were out of tune completely. CEO fail. They were not able to execute on even simple improvements until way too late.
Not that it is uncommon, tons of examples of this exist today. Blackberry anyone? Maybe a Nokia device?
I used dozens of Palm OS devices for close to a decade, and I can say the article is well researched but misses the mark on several very important reasons why the platform went downhill. In short, except for the very first generations of Palm OS devices, the hardware and software was never in sync, either one or the other was lacking. At the very end, both were tired and had no place in the market.
The first Palm Pilots were ground breaking devices when they came out. The premise was backing up your data to the PC and having a disposable device to access it in the field. I sat on my Palm Pilot 1000 and cracked its screen. Later that day I picked up a used Palm Pilot 5000, synced it, and was back up and running as if nothing ever happened.
That attitude started to changed when the Palm V came out. It was an iconic design, with high quality materials, extremely thin and beautiful, but it also retailed for around $700 in Canada. And the hardware had issues, like the Up key being pressed by the cover and failing over time, the glued case that made replacing the battery very difficult and expensive. Today Apple fanbois do not seem to mind, but back then this was a big deal.
The Palm V was also the last bit of hardware where Palm was ahead of the game. Every single generation after the Palm V was far behind other offerings on the market, especially the Sony Clie. I had the SJ30, NR70, NR70v, NX70, NX90 and several TH55. Absolutely beautiful devices, with high resolution screens, very long battery lives, but like the Palm V, they were EXPENSIVE. The high end models were retailing in the $700-$1000 range.
And at the end the Palm OS was really showing its age. Connecting to wifi was was slightly less painful than pulling teeth, but it did not matter because displaying a simple website took the better part of a day. Also dealing with strange file formats became tiring. Besides the organizer functions, my main use for all those devices was taking notes and reading books. There was no way to upload a text file to a Palm OS device and display it as an ebook without first converting it to PDB. In 10 years I found exactly ONE utility that could perform that function, and I still have it somewhere on my PC.
Like the author I had the chance to play with a Palm OS device I forgot I had, the Sony Clie TH55. The OS is fast, the applications load almost instantly, and it has a certain beauty in its simplicity. But then you realize it cannot do any of the things we take for granted today, and all you're left with is nostalgia.
http://www.osnews.com/print/26838/Palm_I_m_ready_to_wallow_now ... instead of page by page view.
I was always really into the idea of portable "palmtop" computing. Back in the 1980's, I coveted the Radio Shack pocket computers. The thought of being able to carry around a device in my pocket that I could program on the fly was thrilling to me. In the 90's, HP came out with the HP 200LX which gave you a full MS-DOS computer in your pocket. Wow! Of course, this was the age of Windows, so if you wanted a GUI, HP had the Omnigo which was my personal favorite (it ran Geoworks GEOS on it). But, none of these really caught on with the general public. The HP200LX did have a strong cult following, but it's high price precluded wider adoption. A used one still costs over $250 on eBay, not much less than its original retail price. One thing was sure, though. Palmtops were the wave of the future, and Palm jumped in at just the right time. Their units were exceedingly popular, and I desperately wanted one, but I couldn't justify the cost for me.
Then, one day, an unexpected package arrived in my office. The unabomber had not been caught, yet, so I was a little suspicious, so I opened it. Inside was a brand new Palm Pilot Pro! A few months earlier, I had put card into a drawing for one of these at a conference, and I promptly forgot about it. After all, no one wins those contests, right? apparently, I defied history and won the contest. I immediately got the Palm III upgrade card (with an IR beam so strong, you could use it as a universal remote), and fashioned a screen protector out of an old transparency projector sheet I had lying around. I used that thing until it was worn thin. The development kit was rather sparse, but it got better, and there were other tools that became popular, like Pocket C. It's biggest limitation was the measly amount of RAM--only 2 MB. The biggest complaint I had about the unit was the battery--not the battery life, which would last weeks, but the whole power "system". It didn't have a backup battery when changing the alkaline triple A batteries. It merely had a capacitor that held the power for about a minute while changing them. Well, that capacitor went bad quickly, and I always had to resync after changing the batteries. Eventually, I soldered in a new one. The sync cradle made even less sense. Ideally, you'd have the Palm sitting next your desk as an extra calendar "window". But, you couldn't do that with the old Palms. Not only would the sync cable not power the palm in the cradle, it actually DRAINED the battery if you left it in there for any length of time! Nuts!
Still, I miss the simplicity of that little palmtop. It worked well and was quite reliable. I eventually traded it in to get $50 off a color model, which I still have, but it's not the same. It's sad how Palm just kind of disappeared. There's tons of software still floating around somewhere that is unusable. There's such little interest in the platform, that no one has even bothered to develop an emulator for Android or iPhone, which surprises me. It's almost as ig the palmtop revolution of the 90's never actually happened at all. It's certainly been mostly forgotten, even though many benefited from the technology.