Technology autopsies

Jean-Louis Gassée has been at the centre of the technology industry for at least the past 30 years. He worked at Apple though the early Macintosh years, founded Be Inc. – a now forgotten OS and workstation company that focused on multi-media prowess and was chairman at PalmSource.

He recently published a meditation on why Palm, BlackBerry, Nokia and Microsoft failed in the smartphone sector – it makes a really good read, I have linked to it under more information. But there a few details missing, I suspect for the ease of storytelling. I’ve added them below as additional accompanying notes for his essay.


Reading David Wood’s ‘biography’ of Symbian makes you realise how from the early years the OS was kludged together into something fit for purpose.  Moving Symbian on was a major issue, one that Nokia knew they faced. It was perplexing why Nokia couldn’t get Maemo right. I had used a developer model Nokia N950 and it was an impressive piece of kit – a symbol of what could have been.

A second part of Nokia’s problems were hardware related. Nokia Networks and phones had thrown their lot in with Intel on WiMax for 4G, rather than LTE championed by Siemens, Ericsson and NTT Docomo.

That put them in the wrong camp to do business with Qualcomm and its SnapDragon processors for modern smartphones. Nokia’s engineering brain trust had been completely wrong-footed. It also explains why valuable time was lost merging Nokia’s next generation mobile device OS with Intel’s similar project. Ironically, this operating system now powers Samsung smartwatches – which is a testament to its ability to squeeze real-world performance out of extremely low powered devices.

Texas Instruments a long-time Nokia supplier, pulled out of the mobile embedded processor market in 2012, which would have had implications for Nokia’s much vaunted supply chain, in particular chip pick-and-place machines.  One could see how these operational problems would have rippled through the engineering organisation.

Nokia actually had a prototype iPhone-esque device running by mid-2004, but were afraid to make a leap of faith

“It was very early days, and no one really knew anything about the touch screen’s potential,” Mr. Hakkarainen explained. “And it was an expensive device to produce, so there was more risk involved for Nokia. So management did the usual. They killed it.”

I had used touch screen devices since 1999, but it is hard to explain how transformative a responsive capacitative touchscreen interface was in comparison to everything that had gone previously.


By 2002 Palm had acquired Be Inc. presumably because they realised that mobile computing needed to have a modern OS for its underpinnings. Palm had previously looked at moving its OS over on Symbian as by 2000 the PalmOS was creaky.  PalmOS at that time ran on a low power version of the Motorola 68000 series processor that powered the first Macs in the mid-1980s to mid-1990s. The OS was migrated to an ARM processor for use on mobile devices.

Its PalmSource subsidiary was spun out of the business to better build an eco-system of licensees. The work of Be Inc. made it into a modern version of Palm OS called Cobalt in 2004, but this was not used by Palm or anyone else. Cobalt covered multi-tasking, better security and better multimedia.

PalmSource acquired a Chinese mobile Linux company in early 2005. PalmSource was sold to ACCESS of Japan.

ACCESS Linux offered the Palm interface running on top of a Linux micro-kernel and functionality for mobile networks etc.  ACCESS Linux was ready to go in 2006 prior to the launch of the iPhone.  While there was collaboration with NEC, Panasonic  and NTT Docomo there hasn’t been an ACCESS Linux powered device launched.

Instead Palm launched its WebOS in 2009. WebOS was slow and sluggish to use. Part of this was because the device was under-powered compared to competitor products.  So despite having an interface which had many of the pieces in place Palm had at least three gos at the software and still failed badly in terms of execution.


Gassée rightly points out that Google giving away its OS left Microsoft’s business model for Windows Mobile disrupted.

However, truth be told Google did a poor job of signing all the disparate Chinese manufacturers onboard and fully legit on Google. Many Chinese handsets had not gone through official channels for compatibility testing (CTS) and do not have a Google Mobile Services (GMS) license.  Google historically hadn’t bothered to scale to address the international aspirations of Chinese tier two and tier three handset makers.

Building a partner eco-system in the west would have been challenging. Microsoft had too many skeletons in their closet and their partners didn’t do too well.

  • Nortel was a historic Microsoft partner in wireline telecoms prior to going bankrupt in 2009. Where companies have PC / phone integration  it is built on the knowhow Microsoft gained with Nortel on VoIP PBXs
  • Motorola had a Microsoft Windows Mobile-powered smartphone the Motorola Q. That worked out sufficiently well that Motorola abandoned it and focused on Android devices
  • Sony when it was co-branded Sony-Ericsson used Windows Mobile for its Xperia phones for two years as it recognised that Symbian had reached the end of the line. Eventually Sony-Ericsson moved over to Android in March 2010, the company has struggled to remain relevant in the mobile market
  • Sendo was a start-up founded in 1999, they signed an agreement with Microsoft to be the company’s go-to-market partner for their Smartphone 2002 mobile operating system. The deal gave Microsoft a royalty-free license to Sendo’s designs if the company went insolvent. There was a legal dispute when Microsoft used Sendo’s designs to create the first of the Orange SPV phones made by HTC
  • After making Windows phones from 2009 to 2013, LG said that there was no demand for Windows Phone devices and moved its portfolio exclusively over to Android where it competes with a respectable performance against Samsung

Microsoft’s name in the telecoms industry is mud. To add insult to injury its Skype VoIP application is a direct competitor to carrier voice minute businesses on both wireless and wired connections.

More information
Blackberry: Meditation At The Grave | Jean-Louis Gassée
Nokia’s New Chief Faces Culture of Complacency | New York Times (paywall)