Client-side tasty soup

stewInteresting pieces brewing on client-side of web applications these days that I’m starting to investigate:

  • Javascript frameworks, such as ember and angular, providing an efficient object and data environment
  • Gesture libraries, such as hammer.js, for interactions such as pinch to zoom
  • Responsive design best practices to enable scaling UI adaptations from smartphones to desktops
  • Plus JQuery and JQuery UI for specific interactions

Looking forward to being able to formulate a client-side toolkit that can be matched to backends built on Django.




Narrowness in business

A friend at Intel used to say, “no matter how narrow you make something, once you get close to it, it looks pretty wide.”   My goal in 2012 is to get a sufficiently narrow web application business off the ground by myself.  No venture capital, no angels, no recruiting, no house of cards.  I’ve done that all before, but today it seems like a) it’s best to be able to move quickly and b) virtual private servers are very inexpensive.

Meanwhile, my narrowness guidelines for 2012:

  • Focus on providing a communication channel that is new, simple,  predominantly mobile, and “obvious in retrospect.”
  • “Faster, better, cheaper” rather than “brave new world.”
  • Browser-based (forget about web apps on Android and iOS).
  • Focus on feedback mechanisms and monitoring tools.
  • Avoid developing new domain knowledge.

Android un-Marketing vs. iPhone Marketing

The biggest difference so far in Google’s Android business development strategy vs. Apple’s iPhone business development strategy is that Google has un-marketed to experimenters, developers, and companies, while Apple has spent heavily on image-based advertising to techies and early adopters, and taken advantage of brand extension from other Apple product categories.  Android has a great spec sheet; but not much of an image.

The iPhone capitalized on the opportunity for iPod brand extension in Apple’s retail presentation and word-of-mouth. The iPhone is not a ‘better’ cell phone — the cellular service providers and handset vendors had commoditized the category, and smartphones were still struggling to be taken seriously; there is nothing that can be spectacular or distinguishing about a cell phone now, except maybe negatively — if too big and heavy, or its battery doesn’t last long enough. So the iPhone is instead a cool expansion on the iPod that includes cell phone capability.

One of Google’s biggest advantages with Android is its portfolio of partners that includes everyone from China Mobile and NTT DoCoMo to HTC and Motorola.  It’s also one of Google’s biggest challenges. Even if one of these partners were to introduce an Android phone that one-upped the iPhone with cooler hardware, it would not be a complete product (in the classic marketing definition of Bill Davidow) like Apple’s iPhone.

Maybe Google’s best bet is to re-define the competition by having not one complete product (and image-anchored) Android that is better, but thousands of tiny, splintered, un-marketed, open-source ones.

If that’s the strategy, Google is going to have to be a lot more aggressive in pushing an open source strategy.  Recent introductions like a limited Python, Lua, (soon Ruby) programming framework with Google’s Android Scripting Environment (see reality check blog post by Mike Riley at Motorola), or a Native Development Kit (some C/C++ programming and library access for focused performance re-coding of Java apps) on top of Google’s Android Java platform is effort in the right direction. 

Google will know it has momentum when there are open source developer forums and wikis that are driving Google’s Android work, rather than the other way around.

Ok, ok, so maybe un-marketed doesn’t really mean there is no marketing strategy and tactics, but, like un-conferences, it does mean that everything happens with far greater leverage and focus, and in a fraction of the cycle time.

See also “How do you sell an Android phone?” on

Initial Impressions: iPhone 3G

Rather than a general review, these are initial impressions on some aspects of the new iPhone of particular interest to me and also relative to the competitive market context.

U.S. TV ads during the Olympics have been touting it as twice as fast, which probably is a good, uncomplicated message for that audience, but maybe not quite what was hoped for (EDGE is around 160Kbits/sec and “3G” on UMTS phones you would hope to be running 1-2 Mbits/sec, so 6-12X would be a lot nicer).  Twice as fast basically means you’re still looking for wifi access (see my previous iPhone 3G posting for discussion of Apple’s coup with AT&T wifi at Starbucks), unless you’ve just got to have that email or web page, so most people I know are turning off “3G” (Settings -> General -> Network:  3G -> off) to save power.

The iPhone 3G system software is somewhat more sluggish and the Safari browser more prone to crashing than on the previous iPhone, but neither of these things is a major problem.

My favorite two innovations are downloadable apps and the multilingual operating environment.  The first thing I downloaded was a WordPress client (no, this post is not being written on the iPhone.  Desktop typing is faster than for an 親指俗人 oyayubi zokujin (“thumb tribesman”) using the phone.  But that brings me to the language part, which actually means the system language, the keyboard, and the region.  For example, switching to Chinese (mainland characters or traditional) puts most everything in Chinese and then you can set the region as China, Singapore, Taiwan, etc.

The keyboard, in the case of Chinese, can be “pinyin” (typing in Roman alphabet, which is contextually converted to Chinese characters) or “handwriting.”  These techniques have been around for a while but work great implemented on the iPhone 3G.

I have to admit I naively thought there was no way that handwriting was going to work with a finger tip dragging across a capacitive-sensitive screen rather with a stylus on a touch sensitive screen (like my previous favorite cell phone, the Chinese Motorola A780 — see way below).  I’m not sure why I thought that, because a common way to clarify to someone which character you mean (in a Chinese language context) is to “write” the character you mean with your finger on the palm of your hand so they can “see” it written. Well, contrary to what I thought, it works great!

Twitter posts in many languages, yes!  Once I started trying some Twitter posts in Japanese and Chinese from the phone, I moved on to German and French, which are almost as cool because predictive interpretation and correction are used there, too.  You type “sein konnen” and it’s changed, like “fail whale, oder warum Twitter Posts trotzdem toll sein können.” Same thing with French diacritics and accents.

In the competitive context (writing as I watch a Verizon ad with an LG iPhone-alike being rolled out and pasted on the side of a building out my office window), adding downloadable webbish apps like WordPress almost, almost, quasi open sources the iPhone, which I think is a key strategic front on which a competitor “could” make their offering(s) bigger and broader (maybe Android phones or Ubuntu on Intel MIDs as they evolve into cell phones will have an opportunity here), and adding lots of languages that work extremely well (esp. Chinese with 600 million cell phone users) both are going to put the iPhone brand just way out there in a way that took the iPod much longer to accomplish.

iPhone 3G

The new iPhone becoming available July 11 will have 3G for faster web and email connections and longer battery operating time. Meanwhile, where’s the best place for 4 million iPhone users (not counting the 2 million that are in China or elsewhere) to demo to potential buyers? Starbucks. Apple has pulled off a great move in the U.S. with free AT&T wifi through Starbucks, significantly reducing the barrier of locked or for-pay wifi access points. This also will do more for Starbucks’ stock price than Starbucks could manage by changing their menu. There is, er, the small matter of a contract with T-Mobile that needs to be rationalized. Apple, AT&T, and Starbucks could perhaps share equally the cost of splitting the difference with T-Mobile.

Since everyone I know who has an iPhone is planning to switch to the new one, and everyone I know who doesn’t have an iPhone is planning to get the new one, it appears that Apple can do no wrong.

When the iPhone launched at the end of June last year, I made a bet with a friend about which would happen first (and by the end of 2007): iPhone clones from Chinese cell phone makers, or iPhone liberation by open source software.

Well, there have been clone sightings (“Top 10 iPhone Clones“), and a lot of hacking (“iPhone Hacks” – it was a matter of days before samba and ssh were functional on it), but I don’t think you could argue that either of these outcomes has occurred.

Now Sprint is going to take a $100 million run at AT&T and Apple with the Samsung “Instinct” iPhone clone. Sound familiar? I think it’s going to be “Indistinct.” In 2005, the year that iPod sales really took off, Creative Technology’s CEO said he was going to spend $100 million to compete with the iPod. A year later, Creative had decreased market share. By that time, “podcasting” was hot and “iPod” was the category. I’d say it’s too late for Sprint and Samsung to compete on feature-vs.-feature comparisons and ad spending because the iPhone has already redefined the smartphone category. It will take another redefinition.

An “open phone” could be that redefinition, but the hardware will have to be cool-better (OpenMoko has not been able to execute on either aspect; and the Google Android cell phone platform is Java, not sufficiently open source, and may have bitten off too big a piece of the stack; maybe evolution of the Intel MID project with the Atom processor on Linux will be the source of a broadly usable open phone platform).

Smartphones -> Web Phones

It’s been obvious that the iPhone sets the new gold standard for smartphones, and simplifies the definition of smartphone as web phone. But other cell phone competitors so far seem content (Michael Malone’s analysis is that they’re unable to respond to a risk-oriented product strategy) to ignore the competitive threat.

A January 14, 2008 article in The New York Times by Miguel Helft (“Google Sees Surge in iPhone Traffic” [at Christmas]) notes that the iPhone is the first cell phone on which web browsing is useful enough to generate significant traffic on Google. Even though Symbian and Windows-based phones are 63% and 11% of the worldwide smartphone market vs. Apple’s 2%, traffic from iPhones surpassed traffic from them on Google over the holidays. Wow!

I’d add to that the notions that “it’s the browser, stupid,” not built-in applications, that create the iPhone’s potential for value; and that it’s wifi that enables you to actually realize that potential value.

Looking at my own usage, I get little use out of the vast majority of the iPhone’s built-in applications. I’m using the browser as my platform (ok, except for camera, alarm clock and, absent VoIP telephony, the “phone.”) With the 1.1.3 version of the iPhone software, to which I upgraded yesterday, I’m even more aware of the browser focus, because now I could move unused applications off the main screen and out of the way.

In addition to not using built-in applications like the iPhone’s “Calendar,” wherever I have a choice between “mobile” and “desktop” versions of browser access to an information service, as is the case with Google Calendar, I choose desktop. The iPhone hits the sweet spot in terms of display size. Together with being able to stretch or squeeze the display to zoom or unzoom, what’s great about the display size is that you can have the convenience of fitting it in your pocket without giving up the convenience of being able to read web pages.

So how long will it be before other cell phone competitors make wifi standard and go to an iPhone-style display?

Simplicity as a Market Strategy

WordPress, Google Calendar, and Apple’s iPhone have a disarming simple-mindedness that is one of the keys to their market leadership.

In certain respects they’re so simple you have to wonder why a lot of people wouldn’t just be infuriated and give up on them, and yet their simplicity is cute in a way that encourages customers to advance their cause way beyond what is normal.

In the case of WordPress, an open-source blogging tool which you use either by installing it on your own server or by creating an account on a WordPress server, one element of its simplicity is that every new blog entry you write is posted by default with the posting date/time to the top of a stream of entries going down (or back) from the present. Further, many WordPress templates provide a calendar widget on which entry dates automatically are highlighted with links to these entries. That’s basically it; there isn’t much more to the usage model, although there are a lot of tweaks and management you can carry out fairly intuitively by poking around in the WordPress dashboard page.

With WordPress, there’s a lot more you COULD do, but it isn’t necessary. So I and millions of other users recommend it to everyone contemplating starting a blog because it’s so simple that there’s very little downside in doing so.

Google Calendar’s simplicity is that there’s very little to using it and you can access it (log into your version of it) from a browser anywhere — home, office, laptop, cell phone (you do have a cell phone with a web browser, don’t you?) — so it extricates you from the 90’s problem of synchronizing your calendars, some of which would have been Outlook. Like WordPress, the basic model in Google Calendar is ridiculously simple. Just click on the calendar to pop open an entry line and type something that seems like it could be interpreted as an entry, say “11/2 12pm R. Cheung – Sungari 1st and Yamhill” and it will show up in the calendar at that date/time from any browser.

Google has not needed to do anything to market Google Calendar (which has been “BETA” for a long time), because, again, I and millions of people recommend it to everyone we know as a calendar solution, because there is very little downside and every chance that people will stop needing support for Outlook. In a lot of cases, people may still be required to run Outlook in their offices, but Google Calendar can blow right past all of that by word of mouth.

In the case of Apple’s iPhone, most of the functions are laughably simple. For example, the “camera” tool has no zoom or any other controls, and the “iPod” tool has no back or forward control, so all you can do is start/stop play, etc. But this also makes it impossible for me and (now) over a million other people not to demo it to everyone who asks, because it is so simple to demo.

Some cell phone makers seem to have expected they could ignore Apple’s market entry, because there really are no new functions on Apple’s phone and the thinking was that what you can do on a cell phone is pretty limited, anyway. But by making the Apple cell phone basically ALL DISPLAY, and a dramatically larger display, and finger operated, which is incontrovertibly cute, it would be demoed like crazy compared to other cell phones.

So in all three of these cases, simplicity is part of a strategy of redefining the market’s expectation of how to carry out a task in a way that the standing competitor(s) can’t begin to match for marketing-less viral promotion.

iPhone Commentary

Competitive Positioning:

“The coolest upscale smartphone you can get; applies minimalist design with a vengeance.”

As it happened, when I purchased my iPhone the first week they were on the market back in June, at the same time I bought and then dropped two other cell phones with very different positioning:

– AT&T (HTC) 8525 phone running Windows and supporting UMTS high-speed data connections on the cellular network, as well as wifi. Curiously, Windows was a constant “in your face” nightmare on this phone. In contrast, I set my wife up on HTC’s Windows-based 3125 (on Cingular -> now AT&T) last year and, after an hour or so of configuration to get rid of things she was not going to use, it has been great. I really wanted to like HTC’s phone because of the speed and because HTC makes a wide-array of very slick cell phones, but they’re all sold with Windows.

– Motorola A1200 “Ming,” which is the third generation of Chinese-designed Motorola cell phones running Montavista Linux. Curiously, there is only a slower GPRS data connection on the A1200. Its predecessor, the A780, my main cell phone for the past year, had the faster EDGE data connection. I really wanted to like the A1200 and find it useful, but Motorola’s failure to open up its open-source based platform and dropping of an intermediate speed for an even slower speed data connection just killed the deal.

– iPhone — well, it’s the one of these three phones that I decided to stick with, but if it weren’t for wifi support, I would have taken it back, too, in favor of sticking with my Linux Motorola A780 (bought it on eBay; came out in China in 2004), on which reading web pages over AT&T EDGE data connections is no better and no worse. But the network connection for voice is better on the iPhone and the A780 was becoming unreliable, shutting itself off or freezing for no reason. Side note: Apple should have a marketing group whose only charter is to open up wifi access in public places, especially the coffee chains and airports, without sign up, registration, monthly fees, or whatever. Just show them the NY Times story from 9/18/2007 about The Times giving up on charging to read the paper’s columnists online and instead going to an advertising-supported (aka Google) model. The 4-5 top reviewers that Apple lined up for initial product reviews basically all said “nice phone; AT&T EDGE is terrible.”

Evaluation of features (aka “What do you demo?”):

Panorama from Portland's West Hills - iPhone photos turned into panorama on Linux using hugin + autopano-sift + gimp+ Camera (see panorama from a series of photos above), photo browsing, finger-based operation, wifi web browsing and email, great display, generally keeps going two days on battery charge with moderate use; one day with high use.
– Need to type? Forget it, you are going to read not write email on this. Youtube? Not really, the iPhone has a very limited subset on old technology. Music? I had classical CDs and Chinese and Japanese language recordings on my A780, but haven’t tried getting that on the iPhone, yet. The biggest minus is the AT&T EDGE data network: was barely adequate 5 years ago; now it’s a national embarrassment; we’re about 20th internationally on cellular data speeds, as well as digital cable bandwith, but that’s all another whole topic. Furthermore, back on the Apple-controlled part of what doesn’t demo so well, the information model is old-school in the sense that you’re looking at information that you’re going to try to sync between applications on the phone and applications on the desktop. “Web cloud” apps, like gmail, Google Calendar, Flickr, Google Maps, Google Docs make more sense, but that’s not where this phone concept is.

Also, unfortunately, only a subset of AJAX (“Web 2.0”) capabilities are supported in the iPhone’s web browser. So Google Docs don’t work. Google Maps don’t work. Gmail sort of works.

Channels and Strategy:

Finally, even though the phone’s software is based on 50+ open-source projects, the phone is not set up to facilitate open-source software, but instead to lock up the software environment.

That is missing a big opportunity, I think. Meanwhile, the open source community is busy working to “liberate” the iPhone on the one hand, and Chinese / Taiwanese electronics manufacturers (the iPhone is manufactured in Shenzhen) are bringing out clones.

The New York Times just published a thought-provoking commentary by Randall Stross (9/16/2007, “A Window of Opportunity for Macs, Soon to Close”) asking why Apple is thought of as doing well when they only have 3 percent of the computer market now vs. 14% more than 20 years ago. In comparison, Apple’s iPod product line has been much more dominant. What will be the case with the iPhone? In each of these three areas, Apple basically has had a strategy of trying to redefine the category up a notch (DOS PC -> graphic Mac; generic MP3 player -> “iPod” as the category; smartphones -> “the” iPhone) and to control very tightly the evolution of marketing messages in (or by also controlling, like 185 Apple stores in the U.S. vs. HP computers for sale in 23,000 U.S. retail locations) the sales channel.

Google vs. iPhone vs. Asia vs. U.S. cellular providers

Kevin Delaney’s Wall Street Journal Online blog entry from May 31, 2007, on “The iPhone Needn’t Fear Google, Yet” points out that Google’s cell phone strategy is not to have a phone product per se, like the iPhone, but rather to evolve a services platform.

With the impending iPhone launch, anyone who’s been using cell phones in Asia the past several years has to wonder, as the Japanese technology business magazine ASCII did in February, what the big deal is. Browser phones without physical keyboards? That’s already been mainstream there for some time. Sliding screen content with your fingernail? Same deal. 2.5G connection speed using EDGE for wireless data? Are we missing something? 3G has been operational for some time in Japan. Why would you want Apple’s phone at three times the price of 2004 model web phones on eBay? Well, because it will have an apple logo on it. Still, probably at the top of the desired improvement list is 3G, according to Ben Charny’s “Apple Changes the iPhone, But Critics Want More Still,” June 18, 2007 WSJ Online.

On the other hand, Apple’s entry is great because it will help loosen the silo death grip of most senior management in U.S. cellular service providers. They know their business model will have to change, but nobody wants to blink first. Managers who aren’t getting paid to put the current business model at risk are actually letting themselves be quoted, according to the Wall Street Journal’s lead story on June 14, 2007, to the effect that they don’t want to blow owning the silo this time like they did with the Internet. Hello? Some large companies can survive forever without realizing what business they’re in. Verizon thinks they’re going to make money as content providers instead of as service providers?

Meanwhile, the Chinese and Indian cellular markets are rapidly becoming 5 times the size of the U.S. market. I kind of think that the market-driven model over there, where the handset, the service provider, and the services platforms act, sell, and interact with the customer quasi-independently and quasi-cooperatively, is what will eventually take hold in the U.S. as well. So in the Asian context, Apple’s handset is nothing new. But in the U.S. market, it’s the break in the dike.

ZTE to start on 3G Pilot in China

It’s been reported in Light Reading that ZTE (中兴通讯) and a consortium of other companies are in the final stages of concluding implementation contracts with China Mobile, the largest cellular provider in China, for year-long pilots of a Chinese 3G cell phone system based on TD-SCDMA. It appears to still be a goal that a TD-SCDMA system will be running in Beijing and a few major cities in China in time for the 2008 Olympics in Beijing. Perhaps both TD-SCDMA and HSDPA will be deployed in China on a scale sufficient to move equipment and handset makers well down the experience curve and gain flexibility in their strategic positioning in the global cellular market. Now that NTT DoCoMo’s 3G in Japan has over 30 million users and 100+ U.S. metro areas have had Cingular’s 3G service available in the U.S. since last fall, it will be interesting to see if the Chinese market finds any new approaches, especially open-source based and driven by user-content, to accelerating development and use of 3G applications.

Restructuring the Global Wireless Information Market

Probably the four biggest potential changes that I think will have an impact on the global wireless information market over the next few years are 1) wifi and other short-range wireless, 2) moving cell phone-based information applications onto open source, 3) the size of the Chinese and Indian markets, and 4) 3G cellular data.

Five years ago Japan and Scandinavia were the “happening” cell phone markets. While most people in U.S. couldn’t conceive of typing emails on cell phones with their thumbs, there were already 30 million people in Japan using cell phones for messaging and web-based services. It was just more convenient to have Internet access on your phone. And in the Scandinavian countries everyone was doing SMS short messaging. Cell phone markets worldwide were quite different from each other, primarily due to one combination or another of culture, policy, and the state of prior communications systems.

Today, we’re a long way from homogenization, but potential changes over the next few years go across local markets and will have a huge effect.

Wifi in cell phones may just mean lower and lower average cost for more and more value on cellular service. As users get more and more for less and less, market growth may be fastest for the cellular providers that are quickest to incorporate wifi. It’s a chance for new leaders. Current leaders like Verizon in the U.S. and China Mobile in China may resist breaking down their “silo” business model where they want a percentage of any monetary transaction that happens over their service, like using a cell phone to buy using PayPal.

As for open source, I think it will rapidly become the way cell-phone based, enterprise-style information flow is done — things like field service dispatch, on-site sales quotes for customers, basic CRM applications — will go to open source because the quality and turn around time for innovation will be better than for close-source products. If improvements can be made overnight, and there are millions of users and thousands of developers networked on projects, the application space can grow pretty quickly. That’s great for the cellular providers because right now their enterprise application business is weak

The bottom line on China and India is that whatever happens there is going to be influential everywhere else. With 500 million and 150 million cell phone users as of the end of 2006, China and India are not exactly out of room to grow, either. So what are some of the things that are different there? Motorola sold 1 million Linux-based cell phones in China in Q2 2006, a $50 phone with low-power electrophoretic display by Motorola started shipping in India at the end of 2006, and there are plenty of good wireless engineering design shops in China and Taiwan experimenting with different feature sets and capabilities on shorter and shorter development cycles.

Finally, 3G and China in a certain sense are one and the same topic. Because China has been stalling on 3G service to give time to develop a perhaps licensing-independent 3G technology – TD-SCDMA – Chinese business will be in a position to control the market. The size of the market means everyone else will have to be a player on TD-SCDMA.

Linux Projects Should Totally Concentrate on 3G Cell Phones

With the introduction over the past month of the Cingular 8525, the Cingular Samsung Blackjack (both running Windows Mobile 5.0, incidentally), plus several LG and Samsung flip phones sold by Cingular, such as the SGH-ZX20, as the first 3G (UMTS -> HSDPA) cell phones in the U.S., it seems obvious that Linux projects should totally focus on 3G, because that’s where open source will bring the greatest value proposition for consumers, enterprise application buyers, cell phone makers, and wireless service providers.

The Cingular 8525 is made by HTC in Taiwan (宏達國際電子股份有限公司 -> 宏達電, the 8525 is the HTC P3600 phone), known by the XDA-developers project as the “Hermes” — the Hermes Linux port wiki on is the focus for Linux effort on that phone.

As far as I know, a Blackjack Linux project hasn’t been started yet.
An article from a year ago on by Luke Kenneth Casson Leighton, a prime mover of the projects, on “Hacking HTC’s Windows CE phones with Linux – a progress report,” is still a good summary of the issues and motivations for hacking Linux on cell phones. It just all adds up to creating more value faster at lower cost.

Just as it’s obvious that open source should concentrate on 3G cell phones, it’s also becoming obvious that open source can help open the flood gates by swarming to help one wireless service provider in each major national market gain competitive advantage through open source, just to get the system out of balance. Looking at this thing in terms of competitive moves, that’s something open source can do but the competition can’t, so that would be good. I’d suggest Cingular in the U.S., NTT DoCoMo in Japan, China Mobile in China, and Bharti Airtel in India.

Maybe a wiki to facilitate 3G focus and carrier concentration would be helpful.

A Business Professional Goes Shopping for Cell Phones

Objective: Light, slim, long battery operation, sync with contacts and calendar on the desktop.

Considered: Nokia E62, Blackberry 7130c, Cingular 3125, Samsung Blackjack — all at Cingular.

I did not include the Cingular 8525 because a) she won’t care about fast Internet connections (802.11g and 3G network) until she’s used 2.5G (GPRS/EDGE) for a while, b) it’s too big, and c) it’s too expensive.

I did not include Verizon because even though the cellular service on Verizon seemed good the past two years and Verizon was top rated for service when I switched to them from AT&T (gnashing of teeth) the instant number portability happened, Verizon said they were getting off BREW and opening up their phones “real soon” two years ago, but didn’t, and I was never able to find a second retail sales person who knew anything after the first one quit.

I did not include Nextel, since acquired by Sprint, because my wife’s organization was not happy with them for their failure to provide any initiatives to help them improve their value proposition.

The Blackjack was cool but too short on battery operation, per forum comments, plus she doesn’t think she needs email and web on a cell phone, yet, plus the massive Cingular and Samsung marketing campaign for the holiday shopping season had just started, but hadn’t been noticed by her. The E62 and the Blackberry, like the Blackjack, sort of fit in the “Q envy” category but after putting hands on the phones it was clear that the Cingular 3125 — the lightest, thinnest, longest battery operation flip phone that could handle syncing — was the best choice for her needs.

Cingular 3125 Customization Tips

Cingular 3125

My wife got a Cingular 3125 ( review) Friday after Thanksgiving. It’s a cool phone and very practical for people who “just want a phone” with long battery operation that can sync with their desktop Outlook contacts and calendar, especially if they don’t have (or have gotten over) Blackberry-envy.The 3125 is the first flip phone running Windows Mobile 5.0. It’s made by HTC in Taiwan, which has a number of phones with Linux projects on, but not yet this one, known as the “StarTrek”.

The following are customization tips. A lot of it is general Windows Mobile 5 topics rather than 3125-specific. None of it is a big deal. It’s just a new customer sitting down with the phone and saying “Could you find a way to…,” again and again at each turn for an hour, to someone who says “Maybe, let’s see…,” until a dog has been turned into a kitten.

Two useful forums for howto information are:



  • How to use Activesync to put files onto the phone or to manipulate files on the phone:
    Start Activesync (comes on a CD with the phone; installs to a Windows PC) by connecting the phone, wait until it finishes synchronizing, click on “Explore Device” -> navigate around in the Windows Explorer that opens up and drag from the PC desktop to the phone directory as needed.
  • Clutter reduction: Using Activesync, create an “other” folder in \device\Windows\Start Up and then drag into it all the other programs besides Calendar, Contacts, Settings, Win Media Player, Communications, a few others, reducing 3+ screens of icons to 1 screen.
  • How to get rid of the promotion of Xpress Mail on the home screen: Settings -> home screen background: pick “Windows Default” or “Windows Simple.” All the screens that start with “Cingular…” have the promotion.
  • How to customize the home screen wallpaper: In GIMP on Linux or Photoshop on Windows, make a 320×240 pixel JPG file, use Activesync to put it onto the root of the file system, then in Settings -> home screen background, select the file name.
  • “What happened? I can’t make a call/receive a call”: If the home screen says “Phone is off,” go to Communications and click on the phone icon, which will have an “x” on it, to remove the “x”
  • “Registry editor” needed for some configuration hacks below: PHM Registry Editor: Google it, download it (as a .CAB file, I think), use Activesync to put it on the phone, use the phone “file manager” to select and run the .CAB to install it.
  • How to turn off start up and shutdown sounds: use PHM Registry Editor: HKLM -> software -> HTC -> startup -> WAV –> set to none; and HKLM -> software -> HTC -> shutdown -> WAV –> set to none
  • How to change the name of the phone that shows up in Activesync: use PHM Registry Editor on HKLM -> Ident -> “Values” -> “Name” -> change
  • How to get sounds: for custom rings, etc. .MP3 files that are placed into the “sounds” folder are supposed to be available for ringtones, but I found that “ringtones” in Settings also has access to sound files in the root directory. I also found there was no need to convert to .flac or .mp3 if you’re in a hurry and don’t care about the space savings; a .wav music file worked just great.
  • Browser short cuts: I put all the predefined links she wasn’t figuring on using into an “Other” folder and added instead:
    • // cell phone-oriented weather
    • google
    • PDX arrivals // Portland airport arrival schedule
  • How to delete one large set of contacts and put on another:
    • On the PC, copy Outlook.pst (the Outlook data file) to a save backup location in case of error. Outlook.pst is in Documents and Settings\\Local Settings\Application Data\Microsoft\Outlook
    • Then, in Outlook, delete all contacts.
    • Now sync with the 3125, thereby deleting everything on the 3125.
    • Now move the new Outlook.pst file to the …\Microsoft\Outlook directory, overwriting the old file.
    • Now sync with the 3125, which will put all the new Outlook contacts onto the blank 3125.

    The same approach can be used with Schedule and Tasks. Seems like it should be possible to select all and delete in Contacts, but online searches just turned up people looking for ways to do it. Seems like it would be possible in Contacts -> menu -> 6 SIM manager, which allows you to select all and to delete, but all I could get it to do was to scroll rapidly through the entire contact list.

  • Quickest way to do sound file conversion: on Linux, use flac -s $in -o $out // $in and $out would be, for example, the wav to flac filenames
  • How to clear storage including contacts, schedule, email, music, photos, settings [think about it for a while before doing it, because that’s pretty much everything]: Start -> Accessories -> clear
  • How to do a soft reset: Hold down both soft keys + power
  • What are alternative browsers: Minimo (phone version of Firefox) might work. Opera Mini is great on the Motorola A780. A forum note suggested installing the version for the LG 225. [On 2006-11-28 Opera introduced Opera Mini 3.0; need to check it out and see if it’ll run on the 3125 WM5 environment.]

More Linux cell phones

FIC (First International Computer 大众电脑, Taipei) and Trolltech (Oslo) have announced cell phones that bring open source Linux closer to reality on cellular.

The FIC Neo1973 uses Linux 2.6.17, a Samsung ARM9 processor, a simple case design with a 640×480 2.8 inch touchscreen, and a UI based on a widget kit FIC’s own “OpenMoko” group wrote using GTK2+. There are open APIs to connect to the proprietary GPS and radio parts of the phone. Like the Motorola Linux phones, it patches quad-band GSM/GPRS multiplexing into Linux, but does so using open standards.

The Trolltech Greenphone is more of a development phone for Trolltech’s “Qtopia Phone Edition” UI platform. Using Linux kernel 2.4.19 and based on a Marvell (ex-Intel) XScale PXA270 (ARM) processor, it has a 320×240 touchscreen display. The price is $695 and you have to get it as part of an SDK.

The Greenphone and the Neo really have different purposes, because the two companies’ situations are very different. Trolltech is a software component supplier to many cell phone producers and apparently would like the Greenphone to shorten their customers’ “time to money” for each new phone, which would strengthen Trolltech’s position as a “platform” (QPE is also on Windows phones). FIC, on the other hand, is a computer device maker, as well as contract manufacturer and designer, with nothing to lose in cell phones. So their strategy seems to be to do, as fully and as quickly as possible, whatever is necessary to get legions of open source developers cranking on their hardware. Worst case, they fail with the end-user product but build a strong basis for supplying bigger producers who have better customer access. I guess you could say the Greenphone hands out matches on the street but the Neo lights a fire in a closet.

By linuxdevices count, there are now over 50 (mostly cell phones) Linux devices. Of these, I’ve used the NEC-Panasonic phones as rentals in Japan and they’re great. I also like my [not really] open source Chinese Moto A780. But if you were doing an enterprise application in the U.S. today, it would be pretty hard not to check out the Neo.

$350 for the Neo and a SIM card on a GSM cellular service like Cingular in the U.S. and you’re in business doing sudo apt-get for everything that runs on hundreds of millions of Linux systems.

A780 GNU/Linux cell phone

Motorola A780

Motorola has the first more generally successful GNU/Linux-based cell phone in China and Europe, and reportedly sold 2 million Linux-based phones, including the A780 and the newer A1200, in China in Q3 2006. Using MontaVista embedded Linux 2.4 and Trolltech’s Qtopia, the quad-band GSM A780 started shipping in August 2005. It’s reviewed on and

I recently picked one up second-hand on eBay from Hong Kong by way of LA. Once you download the Opera Mini browser to improve web performance, the 150 kbps EDGE (rarely falling back to 50 kbps GPRS) Cingular connection for data, such as email and web browsing, is nothing to write home about, but certainly ok for limited use, like looking up entries on Wikipedia while waiting for lunch in a restaurant. Fortunately, gigabit web phones are coming. Meanwhile, people are poking at the GNU/Linux side of the A780.

Using information assembled by Dino Kern at, I set up a telnet connection into the phone that works over Bluetooth or USB. It’s pretty cool to make the connection and issue dmesg and ls -latr to see what’s up on the phone, but it’s not really a true open source environment, yet.

Motorola probably could profit hugely from market leadership with open source phones (Japanese companies NEC, Panasonic, and NTT Docomo are also active in this space, as are Chinese companies Zhongxing tongxun [ZTE] and Datang) but I don’t think you can say that Motorola’s strategy is “open source” today. They’ve been building needed technology and talking around the periphery of open source for several years now, without making the management decision to “go native,” as noted in June by Eugenia Loli-Queru, editor of The official Motorola strategy appears to be defensive — the party line is that application developers should write Java applications on Motorola’s J2ME platform. Kinda booring!

Meanwhile, Harald Welte and others are working to create a flashable, open-source Linux 2.6 environment for the A780 and its successors through the OpenEZX project. The, Opie, and OpenEmbedded sites also have related information. It will be interesting to see what happens with Linux on (and around) cell phones.