daverea.com

I’ve been blessed with what parents might call an “active imagination” – and when you cram this imagination into my head alongside an engineering degree and a penchant for Libertarianism, a lot of interesting ideas come out. Many of these ideas are for businesses, products or activities – and these are almost invariably (except maybe when they happen at 3AM) scribbled into my Palm Pilot for future reference.

But sometimes, my inner geek produces some fun ideas that might even be worth sharing. While working with the aforementioned Palm Pilot (which is actually a Sony Clie) today, such an idea materialized somewhere between my ears, and I figured this is as good a place as any to jot it down…

As I tapped and scribbled on my PDA, it occurred to me that I’ve had this particular PDA for quite a while – in fact, I’m pretty sure I bought it during my third year of engineering school, which would make it somewhere in the neighborhood of six years old. Honestly, I’m amazed it survived that long! (of course, it did go back to Sony for a broken screen.) (twice.) (shut up.)

I decided to take a quick cruise by Palm.com and see what they’ve been up to lately. As it turns out, there are plenty of slick, sexy palm-sized toys to be had from everyone’s favorite PDA manufacturer, but something wasn’t sitting right with me. I got the same feeling I get when I think about people downloading music from iTunes or buying proprietary software. What feeling you ask? Why the sinking, sorrowful feeling that accompanies impending vendor lock-in.

Vendor lock-in is the common name for the Pottersville Pattern, which is what happens when you buy proprietary software or hardware or what-have-you, and then down the road you decide you want to switch systems and you realize you’re stuck with the original vendor. A great example of this comes from my dear-ol-Dad, via the year 1995. As a dedicated Mac user for years, he had used a HyperCard stack (geez, remember those?!) to manage his hundreds and hundreds and hundreds of contacts. I remember spending hours manually entering business cards into this early PIM for him. But when he decided he wanted to switch to another piece of software, vendor lock-in reared its ugly head. Unfortunately, there was no way to export the raw data from the HyperCard stack into a common, open format. Nor was there any way to import the HyperCard data into his new program. Eventually, we found a way to make the transition by purchasing one or two other pieces of software and doing multiple imports and exports, but it was a pain.

But some vendor lock-in is more than a pain – it can be extremely costly, or even prohibitive. The solution to vendor lock-in? Common standards. There are lots of good open, common standards out there for vendors to use for storing data, exchanging files, and the like. While it’s not exactly “open” (in the “free as in speech” sense of the open-source idealists) vCal and vCard files are good examples of this. Just about every PIM worth its salt can export and import vCard and vCal files nowadays, so it’s a lot easier to move your massive collection of digital business cards from one to another as your preferences change, your software gets stale or your provider goes out of business.

So how does this relate to my PDA? While I was a big fan of Palm five years ago, when I bought my Clie, I’m not quite as big a fan these days. The main reason for that is that there’s only one OS that will work on their PDAs: PalmOS. I can’t go off on the big, magical Internet and download an alternative to PalmOS that will happily run on Palm’s hardware, the way I can for a HP iPAQ or Dell Axim. While these devices were designed to run a certain buggy, complicated, cumbersome Microsoft OS, they can easily run an open Linux-based alternative like OPIE or its cadre of brethren.

But creating those alternative OSs was no picnic – it took a lot of people a lot of effort to reverse-engineer the interfaces of the iPAQ, and then figure out how to burn new firmware in, and then write firmware that the highly-customized little computers could actually run in some sort of stable, usable fashion. All those people worked without pay, and with few if any thanks for their efforts…because they believed you and I should have an alternative to running closed, proprietary code on our PDAs.

How much easier would it be to create free-software-based handheld OSs if the hardware were just as free and open as the software? How much faster could these very smart, but very resource-constrained people work if they didn’t have to reverse-engineer every new piece of hardware that hits the market? How much better would a device work that’s developed and integrated as a complete system, rather than the efforts of a software team riding atop the completely isolated efforts of a hardware team?

Thus, in the course of just a few minutes spent pondering during my lunch break, the idea was hatched. It’s not a new idea – people have certainly tried “open source hardware” projects before. And there’s certainly plenty of folks out there trying to find less-proprietary ways of achieving palmtop computing. Hell, there’ve probably even been a few folks who’ve thought of (and published) this exact idea. And this is what makes the open-source community so great: there are at least a few folks out there who’ll readily get their brain around a given idea, and maybe some of them can even lend some talent toward making that idea a reality.

The kernel of the idea (geeky pun intended) is this: build a PDA-like device from scratch, and collaborate with open PDA OS coders to make it as readily-integratable as possible. Keep the design of the hardware – right down to the schematics and the simulation results – as open as possible, and use as many open-source or open-source-compatible tools to create that hardware as possible. Some folks will buy and use the thing as a PDA, whether it’s in kit form or fully-assembled. Others could download the design, change it to suit their needs, and manufacture their own version. Still others might opt to build or buy some incarnation of the hardware for some purpose completely removed from handheld computing – maybe they’d use it for industrial automation, maybe they’d use it for robotics, or maybe as the heart of a car PC. But the point is, it’s meant to be extensible, as open as possible, as low-cost as possible and as free (as in speech) as possible.

In reality, that’s no small task – creating what essentially amounts to a single-board computer isn’t easy, and packaging it with a battery, power management, interfaces, a display and a touch screen is just as challenging. Oh yeah – did I mention it has to be at least in the same solar system as Palm and iPAQ when it comes to price? But since the BOM (that’s Bill of Materials – a parts list – for those of you in Rio Linda) would also be open, even the components used could be optimized and altered by The Community.

So let’s see. We’ll start with an Intel PXA271 processor, since it has just about everything we’ll need on one chip – a fast ARM core, some StrataFlash, some SRAM and DRAM, an LCD controller, and all the peripheral interfaces a geek could want. Add in a power supply circuit (maybe Intersil, Maxim, TI or any of a dozen other manufacturers would have something for us here) and a nice LiIon battery and you’re ready to nail down your form factor. Can’t forget the display – a TFT from Optrex would probably do the trick. Next, we’ll hit up the case – how about a nice little hand-held case from OKW? And so it goes… add in the USB-OTG, the Ethernet, the Wi-Fi and the Bluetooth… hook in a CF interface, and maybe SD/MMC too… then hold onto your butts and let the software team go at it.

There’s no shortage of material to work from out there – many SBC manufacturers post their schematics on their web sites, and there’re plenty of reference designs to use for ideas, too. Manufacturers’ appnotes and datasheets provide a wealth of information, and where these don’t say what to do, collaboration between smart hardware engineers and smart software engineers will. I could go on…

The part of me that wants to build a car PC would like to go buy a copy of Eagle and start laying down tracks right now. The part of me that knows that’s a crazy idea wants me to go fix up an adult beverage and spend my evening relaxing like “normal people” do. And the part of me that just wants a PDA that isn’t quite so proprietary is just sitting back watching, saying “gee, I hope something comes of this someday!”