Apple iPhone vs the FIC Neo1973 OpenMoko Linux Smartphone
August 23rd, 2007
Daniel Eran Dilger
Frequently compared to Apple’s new smartphone, the OpenMoko FIC Neo1973 is described as the free and open software community’s alternative to Apple’s officially closed iPhone platform. Here’s a look at what it really is and how it compares to the iPhone.
OpenMoko Linux Software.
Proponents of open source software describe the Neo1973 OpenMoko as the “Anti-iPhone,” because unlike the iPhone, it isn’t tied to AT&T for service, nor is it sold as a closed platform reliant upon Apple for software updates.
OpenMoko software is a Linux distribution paired with additional phone-related software all offered under the GPL. It promises an open platform that anyone can use to write their own phone apps and hack upon as desired. If you find a missing feature or a bug, you have the potential to solve the problem yourself.
The project shares some similarities with Nokia’s Maemo project, set up to deliver a Linux-based line of Internet Tablet products, although Nokia’s product only works as a mobile web browser tablet, not as a phone.
OpenMoko was started last year inside of FIC, a Windows PC maker located in Taiwan. Linux proponents convinced FIC that selling a phone as an open developer kit could result in broader sales of hardware for its…
Named after the year the first mobile phone call was placed, FIC’s Neo1973 phone, designed to run OpenMoko, was announced just prior to the iPhone in November 2006. It was originally slated for delivery in the first quarter of 2007 for just $350. How could FIC hope to deliver it so quickly and so cheap?
The phone wasn’t custom designed for Linux. It is a Windows Mobile unit codeveloped by the Chinese government as a mass produced people’s phone. It uses a GSM/GPRS module running the proprietary Nucleus OS. In the OpenMoko version, Linux runs the device’s phone-related operations like a small handheld computer connected to a GSM phone interface via a serial port.
The “free and open Anti-iPhone” rhetoric surrounding the FIC phone is therefore marketing drivel to sell a Chinese Windows Mobile device as a hobbyist kit for phone hackers. Why use a version of an FIC Windows Mobile phone rather than any other phone on the market as a target for the new Linux distro? Really, there’s no reason apart from the fact that mobile software requires tight integration with the hardware it runs on.
The existing complication of running Linux on any random PC hardware would be dwarfed by attempts to run a Linux phone software distro across a range of different mobile phones.
While PC hardware is largely a commodity and all works in a relatively similar way, the world’s various smartphones are all built upon custom hardware designs that vary considerably. Even among phones developed for Microsoft’s Windows Mobile reference platform, there are plenty of gotchas involved with different vendors’ implementations.
OpenMoko therefore isn’t a new “open phone,” it’s merely a version of Linux designed to run on a specific vendor’s proprietary implementation of Windows Mobile. Buying an FIC phone to run OpenMoko is like buying a Dell Windows PC to run Linux. You’re not changing the world, you’re merely funding development of Microsoft’s platform while giving yourself the opportunity to work with community software.
Where’s the Anti-iPhone Advantage?
When you buy an iPhone, you get $500 worth of hardware and make a commitment to use AT&T. You also get a commitment from Apple to deliver you regular security updates and new applications for your hardware.
When you buy a Neo1973, you pay $450 for hardware from FIC, and have no commitment to anyone but the GSM provider “of your choice.” In the US, that means AT&T. The only other US GSM provider is T-Mobile, which doesn’t provide GSM coverage on standard frequencies. T-Mobile may possibly be able to resell AT&T GSM service to you on slightly different terms, but that’s the extent of your real “freedom” in terms of service.
So much for the glorious freedom of choosing your own provider. Any real choice is prevented by the proprietary differences between GSM mobile providers like AT&T and the CDMA2000 providers like Sprint and Verizon Wireless who don’t offer GSM service at all.
The other “freedom from commitment” you get with the Neo1973 is the freedom to write your own software. In fact, you have no choice, as FIC has no particular expertise in writing software. You are entirely dependent upon the OpenMoko community to deliver, maintain, and secure your phone software.
So far, that community hasn’t delivered a functional phone, let alone competition to the iPhone. However, you wouldn’t know that from the posturing made by Linux users who commonly describe OpenMoko as if it is a functional product that rivals the iPhone.
The Anti-iPhone Disadvantages.
The flipside of the freedom to write your own software is the obligation to maintain your own phone software. The overall complexity of administering security policy and navigating through the uncharted territories of its experimental far reaches have resulted in Linux making little headway on the PC desktop.
Why would a mobile platform be any different? For starters, OpenMoko users are likely to use a standard build of the software, limiting the variety and scope of problems a Linux PC user might have when running one of the scores of different Linux distros aimed at the desktop.
However, OpenMoko is only one of several attempts to use Linux on mobiles. Motorola has its own Linux-based platform that is closed to third party development outside of its J2ME environment. TrollTech started its own open mobile platform that shares a lot in common with OpenMoko apart from having a hardware backer. Access markets a Linux mobile platform that once attempted to replace the underpinnings of the Palm OS.
These projects–among others–mean OpenMoko isn’t the future of Linux in the mobile space, buy only another shot at delivering a workable Linux phone implementation.
That sounds a lot like Caldera’s OpenLinux desktop platform, which looked very promising ten years ago but went nowhere. Various other Linux platforms targeting the PC desktop have similarly struggled to find a market over the last decade. Despite plenty of advancement, Linux on PCs has run into the same barriers that hindered OS/2, NeXTSTEP, BeOS, and Apple’s Mac OS.
Erecting a Platform.
Simply finding an OEM that will sell Linux is tricky enough. Buying a PC–or a Mac–and installing Linux on it is certainly a way to run Linux, but represents no sustainable market for Linux as a platform. That’s because the volume market required to fuel a sustainable platform demands a critical user base running the same software, as well as software able to run on that platform.
What is described collectively as “Linux” is really a lose confederation of rival platform components that all run on top of the Linux kernel. “Linux users” all use slightly different platform variations, whether based upon the K Desktop Environment built upon TrollTech’s Qt development framework, or GNOME, a rival effort started in 1997 by GNU to make sure the graphical environment for Linux was all done under the GPL. GNOME is built upon the GTK+ development framework.
In addition to these two main branches of Linux’ graphical application development, there are also variants such as ROX and Xfce, bare bones versions of the X Window System such as Blackbox, and much fancier versions of desktop environments such as Enlightenment, which adds its own extensive support for customizable themes, advanced graphical effects, and user interface ideas.
The problem with all these choices is that they shatter the already limited Linux user base into smaller rival camps. Since Linux users see this factionalism as a feature, there have been few efforts to standardize on any single reference platform. The result is that different application developers pick the environments they want to target, forcing users to either run multiple window managers, or choose between different groups of Linux apps.
GPL vs the GPL.
Even worse, the exceptional efforts expended by individual groups of Linux developers frequently go unused and end up wasted because of the fractured user base of Linux users. That’s one of the very problems in software that open development under the GPL and LGPL was intended to solve.
In theory, if GPL software is abandoned, anyone can take the code and reuse it. However, in many cases, significant efforts are expended, then left unused by both users or other developers, simply because the underlying value of the code isn’t recognized or because it can’t be reused for practical reasons.
In 2000, the KDE development team released KHTML as a web rendering project. Had Apple not known of that code base, KDE’s efforts would only ever impact the group of users choosing the KDE desktop environment. Apple made improvements to the code, developed an independent glue layer that released it from dependance upon the Qt frameworks, and released its changes back under the LGPL.
The result was that KHTML is now available to a much wider group of users. Nokia used Apple’s KHTML-derived Web Kit to build the mobile browser in the N95 and its other advanced smartphones. Mac developers make extensive use of the Web Kit to render all sorts of content in their apps; it even powers Mac OS X’s Widgets.
In response to the LGPL working as intended to salvage and adapt software for broader use, the extreme fringe of GPL adherents complained that Apple hadn’t gone far enough beyond its legal obligations and badmouthed the company for participating in their GPL version of open source.
It should come as no surprise why many commercial developers avoid any use of GPL and GNU afflicted software projects. Linux represents a confederation of many individual groups working at cross purposes, like a convention of volunteer militia minutemen with thousands of guns all pointed at each other’s feet.
Getting Things Done with A Unified Platform.
One of the first things Steve Jobs did upon returning to Apple was to scuttle the idea of custom themes and move Mac users to a single integrated platform with a standard look and feel. While Macs can run both classic Mac software and new software built upon native Cocoa and Carbon frameworks, users always live in the same consistent environment.
Macs can run X11 apps and Windows apps in special environments, but those are bolted on as special use cases to support the few users with those needs. That has resulted in the Mac platform growing into a strong, unified user base that has created a demand for applications that target its unified selection of native frameworks.
That also helps to explain how Apple can garner the attention of major developers while only having a 3% share of the entire world’s PC sales. It also explains why Apple’s platform is growing faster than Linux and PC growth in general. Apple is using the same strategy to deliver its iPhone.
Scattered When Struck.
While interest in Linux among PC makers will likely advance as vendors grow increasingly wary of paying for Windows licenses–particularly on low end PCs–this will have little overall impact because Linux continues to be fractionalized among groups that can’t agree on the direction they want for the loosely affiliated platforms they each represent under the broad definition of Linux.
Linux’ PC platform problems are even more problematic when applied to consumer electronics. While Linux is ideal in server and embedded applications where platform interoperability issues are different, a mobile phone platform involves issues similar to a desktop PC; smartphone users have increasingly sophisticated demands for mobile applications.
Applications for OpenMoko phones–the FIC Neo1973 is supposed to be the first of several phones supporting the distro–won’t work on other Linux phones, and vice versa. As with Linux on the desktop PC, that means interest in Linux on mobiles will again be shattered among various incompatible variants, just as Unix was fractionalized into various camps during the 90s, paving the way for Microsoft’s unchallenged advancement.
FIC’s Missing Links.
OpenMoko isn’t just rivaling other Linux smartphone variants; it’s also racing against the clock and competing against existing platforms including Symbian and to a lessor extent, Microsoft’s Windows Mobile. Ironically, buying an FIC phone to run OpenMoko supports Windows Mobile as much as it does Linux.
In addition, while FIC announced its hardware and software project months in advance of the iPhone and promised to ship a quarter earlier for $250 cheaper, it hasn’t been able to hit its intended targets. The phone still isn’t ready yet, and doesn’t plan to be functional until at least October. The price has gone up to $450 for the beta release hardware, and is missing a lot of features expected in comparable phones.
In contrast, Apple delivered the iPhone at the promised date, price, and feature set it advertised in January. The Neo1973 is 160% of the thickness of the iPhone and 136% of its weight. The iPhone has twice the system RAM, but more importantly has 4 or 8 GB or Flash, compared to the quarter GB of Flash in FIC’s phone. Since the Neo1973 uses all its system RAM to run Linux, it has to run additional apps from separate Flash cards.
More Missing Bits.
The FIC phone has no support for EDGE, so users will only have the much slower GPRS for data. All of the crying over the iPhone’s lack of support for 3G data service will look like a party compared to the reaction of users to the glacial GPRS data service on the Neo1973. The existing model also lacks WiFi support, although it is promised to appear in a later hardware revision. The FIC phone also lacks a camera of any kind.
Like the iPhone and the LG Prada, the Neo1973 uses a fullscreen touchscreen display in place of physical buttons. It has an enormously high resolution 640 x 480 VGA display packed into a 2.8“ screen, compared to the iPhone’s 3.5” display running at 320 x 480. Despite having a higher resolution display closer to a small mobile laptop than a smartphone, its screen isn’t multitouch. It is a conventional Palm Pilot-style pressure-sensitive screen intended to be primarily used with a stylus.
The FIC phone also uses a mobile mini-port audio jack rather than a standard audio headphone port, and uses USB 1.1 for syncing, ensuring that data sync and file transfers are as painfully slow as the 2001 Nomad audio players that were blown away by the speed of Apple’s first generation of Firewire iPods. The iPhone uses USB 2.0, which is forty times faster than USB 1.1.
Despite all these problems, the FIC phone is now only $50 cheaper than the iPhone. Add one 2 GB SD RAM card, and the FIC Neo1973 costs as much as a 4 GB iPhone to obtain, despite still having half the Flash RAM and lacking all sorts of iPhone features, including such things as its ambient light sensor and the proximity sensor that turns the screen off as you hold it to your face.
OpenMoko’s Anti-iPhone Complaints.
Despite all of those missing features, the OpenMoko group says the FIC Neo1973 is “fairly similar” to Apple’s iPhone. The really significant difference is that its OpenMoko software simply doesn’t work yet, despite being six months overdue. It can’t reliably browse the web, function as a video iPod, or even place calls, according to OpenMoko’s own progress reports.
The group complains that the iPhone uses a “proprietary” dock connector rather than using one of the mini-USB connectors (there are at least two standard mini-USB connectors to choose from). That conveniently ignores the fact that there is a huge market for iPhone and iPod compatible devices; the only restriction posed by the openly documented iPod Dock connector is that companies wanting to build commercial implementations are expected to pay Apple a royalty fee to sell them as “iPhone compatible.”
OpenMoko also touts the Neo1973 as having an edge over the iPhone in that users can open the FIC case with a torx screwdriver, while cracking open the iPhone “voids warranty.” The group should really point out that it only risks voiding the users’ Apple warranty, and that FIC obviously doesn’t warrant that users hacking on its phone will get a replacement if they fry or damage it. If one wants to hack on their phone, the Neo1973 doesn’t really offer anything that the iPhone doesn’t, apart from support for Microsoft’s Windows Mobile.
The Open iPhone Community.
In fact, there is already a much larger community of iPhone users than OpenMoko/Neo1973 users. There is also more effort behind building software hacking tools for the iPhone, and a complete development toolchain is available from those interested in building upon the iPhone’s OS X software.
Anyone who wants to install Linux on the iPhone could apparently do so, just as hackers did for the iPod. Apart from political or philosophical reasons, there is little reason to actually do so, as the BSD-based OS X running on the iPhone already works, whereas OpenMoko doesn’t.
OS X delivers a set of advanced development frameworks very similar to those in the desktop version of Mac OS X. Developing OpenMoko applications is a lot like writing apps for BeOS; one can, but nobody will use them, and no market will ever develop around them.
Sun has already attempted to hijack OpenMoko and convert it into its own version of “open,” which happens to be Java-based. While it only involved press releases and a Photoshopped graphic of the Neo1973, Sun clearly views OpenMoko as a beast it can push into service delivering Java in competition against the iPhone.
Development is easy when you simply expect the community to do it for you.
Deliver First, Brag Later.
OpenMoko should really tone down its anti-iPhone rhetoric. Its claims about the iPhone include too much outright misinformation, and its support from FIC is really nothing to brag about. The company is a knockoff hardware cloner infatuated with Microsoft.
The group can only hope FIC will respect the GPL more than it does other intellectual property. FIC currently sells a feature phone listed with “i-Pod function.” How much effort will Linux proponents invest in FIC’s business model, which really just subsidize the development and sales of Windows Mobile phones?
OpenMoko contributors might ask themselves what they’re really contributing toward: an open future for mobile phones, or yet another splintered, delusional faction pretending that Linux is a product, and not a project.
Like reading RoughlyDrafted? Share articles with your friends, link from your blog, and subscribe to my podcast! Submit to Reddit or Slashdot, or consider making a small donation supporting this site. Thanks!