Tablets, Android, fragmentation and marketing versus product development

Posted on January 14, 2011


The Android market is “fragmented”. Also, it is “not as good an experience as Apple”. Criticasters on the platform blame the OS itself. I think that is not where the pain lies.

Bottom line is, that it is really frustrating to have relatively low cost hardware with a lot of potential hacking possibilities and not being able to utilize them.

The real pain with Android

I think the real pain with Android is this:

  1. Too many different chipsets – Making devices “unique” and as a consequence making the device-specific Android version locked in. As a result, many cheap – below 200 USD – Android phones that will enter the market will run Android 1.6 as that version probably supports the chosen – and likely outdated – chipsets the best
  2. Not enough power yet – The slumbering giant that is ARM is only now waking up to unleash processors running multi-core (2 or more) and on clock speeds approaching the 2GHz
  3. Not enough RAM – Until 2010 the Android devices – including tablets – on average delivered 128 or 256 MB of RAM.
  4. A lack of hardware drivers / No support for USB devices – What is an operating system that does not allow you to print or attach a scanner, harddrive, web cam or printer to it?
  5. It is not finished yet – See item #4

Linux and hardware drivers – a similar story

When trying Ubuntu on my Fit-PC2 I felt the pain of having to jump through hoops to find hardware support for the GPU that runs inside that machine. It is a low-power 3rd party piece of hardware licensed by Intel. No hardware drivers means: no support for your hardware.

You can not just install Linux on any type of hardware. Even if it is a Intel based machine.

The lack of drivers and the hardware-lock-in is one of the reasons why tablets like the Archos 7 will only run a very obscure Linux version called Amstrong and ARM RISC versions of Ubuntu will not – for a while.

Apple and hardware drivers

Apple made their business simple. Since they build the hardware AND deliver the OS they can work around the issue of: “is your hardware supported by the drivers in this OS?” This one thing gave them a big advantage in a long-term support for all – or at least – most of their devices. Only having to choose out of 4 instead of 40 chip sets reduces the  work and the overhead.

Drivers and USB support

To make an USB device work, you need hardware drivers as a “man in the middle” to translate the data that that hardware expects and delivers to your system and the software you run.

As most hardware is based on a limited set of chip sets, and / or run standard drivers or standard protocols as a fall back scenario, it is possible to create “generic” drivers that do 90% of any basic job.

Still: someone has to write and produce these drivers for that specific platform. Also: Android will have to support access to the hardware layers that make USB, Bluetooth, GPU rendering and whatever you offer. Then: you need to have some internal or external drive, stimulation or motivation to invest in the creation of these drivers. Where it be hacker / nerd idealism, a practical or a commercial one.

For now (Januari 2011), Googling Android USB support / Android USB drivers mainly results in post asking: “when will they be there?” or “where to download USB drivers to connect your device to Windows”

USB support on some devicess like tablets like the Archos line currently includes: keyboard, mouse, USB storage devices like memory sticks and Harddrives (with FAT32 as NTFS is not yet supported).

Non-standard hardware and the disposable quality of phones and tablets

The ARM chip clock only recently started to speed up. My 2009/2010 Archos 7 Home Edition sports a 600 MHz single core RISC processor. Produced about 12 months later, my 2010 Archos 101 sports a 1 MHz single core processor. Where were these chips three years ago?

About six months after the first 1GHz tablets, the 2011 series of higher end tablets will sport a Dual Core Cortex A9 with clock speeds up to 2 GHz.

Due to the hardware-lock in of the Archos 7 Home Tablet, Android got stuck at version 1.6. As Archos already released the next generation of tablets in the last quarter of 2010 the question is if the Archos 7 will ever get past that version.

In other words: in less than a year, and with a very little amount of RAM, my 2010 Android tablet is already outdated and ready for the museum of stuff that has been. Would this have been a Intel Atom tablet, or a tablet supporting more generic and more widely supported chipsets for the GPU and USB my options would have been bigger, installing some version of Ubuntu or whatever Linux to compensate the lack.

Evolution of an OS – on steroids

What is important to remember is that Android is still evolving. From being an obscure limited OS for phones, bought by Google in 2005, in only two years time – with 7 versions, from 1.0 in September 2008 to version 2.3 in December 2010 – it is evolving on speed and steroids to become a full blown OS for tablets and netbooks in the years to come.

So yes: consequently it is a very unstable market with a lot of next versions being released, a lot of devices left behind in the wake as “outdated” while version “X minus one” of that OS is not even available on phones or tablets in the market.

However: what we do not see with Android – as mentioned – is the support for hardware drivers. Something that Microsoft got right from the beginning. The lack of drivers will be – until solved – the biggest slow-down for Android in the next years. It is the constant and main reason for me to keep my eyes open for the first release of a proper Linux version that WILL run on the devices I own.

Immaturity and the hope for stabilization

With ARM waking up and delivering (the designs and licenses for) high speed multi-core processors – but also focusing on a more complete solution, we might see some stabilization coming. This will work when:

  1. Updates on the OS are backwards compatible – also capable to run on my phone / tablet / netbook of 2 years ago. Linux has a solution for drivers by using scripts (run using “sudo”) to download and install the missing parts so why not use that for Android too?
  2. Updates are aimed to be even more optimized – lean and mean instead of moving towards the bloatware that – for instance – Windows has become in 20 years time. An OS does not really change that much under the hood. So instead of making it more demanding, work on taking away bottlenecks that bog down the OS and increase the speed of things.

Marketing and the priority on things

As a geek, your main goal is to make new things things better, faster and more productive. It is like a sportsman with only one main purpose in life: to break your own personal record and – if possible – all records made by others in the past as a consequence. Higher benchmarks, better compilers, better approaches to the same problem, moving the solar system to another Galaxy, opening gateways to new universa and so on.

Marketing looks at the image you produce. As a marketed product you do not have to do anything. You do not have to have any talent in any direction, as long as you look cool and create this kind of promise that one day you might…whatever. Marketing is about selling as many boxes from A to B as possible in the shortest amount of time. And it does not really matter if these boxes contain oranges or phones. It does not matter if these phones are a complete crap or the most amazing thing you have ever held.

In most cases, crap is better than quality as crap with crappy parts is more cheap to produce and can deliver a higher margin on sales per unit.

Who cares that you can not upgrade, or Flash crashes or does not run on it? It was never promised in the first place.

In most cases the things that drive Marketing are the same things that bog down the evolution geeks love to see hapen with their work. For Marketing the main question is: why sell a completely new thing “that nobody understands” if you can milk something that is there for 10 years and can deliver revenue for another 5? As long as your competitor is not doing it, why would you?

Hence: when the competitor DOES come out with something you have not, suddenly everybody wants to jump the same bandwagon and making blind copies are more rewarded than pulling that thing to a next level and leave the competitors in your dust-trail.

The reason why most innovation is slowed down and even stopped or reversed in Marketing Driven companies is due to two things:

  1. “It is not proven to deliver any money” – so we will not take the risk
  2. “The competitor is not doing it, so why would we?” – being the first is not always resulting in being the victor

Google with Android – at this point – is not Marketing Department Driven. Otherwise the fast release of Android versions would not be there. From a Makering Department point of view, three successively better releases of a newer version of the same product in the same year does not make any sense. Google with Android are a bunch of geeks with enough money and visibility to get away with this. And I salute them.

I also hope that Android goes IPO itself and becomes an independent sister – for reasons of survival. When Google goes down, or becomes Marketing Department Driven instead of geek: “lets see what we can make up, build and deliver this week” driven, Android can still continue to evolve and explode in new directions.

It is good to have something new and very vital that is not Windows, OSX or Linux and still can go in any direction imaginable.

The issue with hardware- and software manufacturers

It is a rare occasion where UX experts, a hardware- and software / OS manufacturer will sit together and say:

  1. How can we match? – the hardware with the OS and the drivers so that we can deliver the ideal package?
  2. How can we optimize? – both so that this new version will kick complete ass in the sense of:
    1. Usability / User Experience / productivity
    2. Speed / responsiveness
    3. Form factor / size / weight
    4. Hardware support / extendibility

How other people think about the market

In the past days I read two articles I will quote heavily in this post, as they state things more clear than I can.

From an article by Tim Gideon on PCmag on Android tablets and his experiences at CES I quote here:

Every step I took on the congested CES 2011 show floor last week, I had to be careful: I didn’t want to step on a tablet. They were everywhere. Remember MP3 players—specifically, remember how many they used to be? Within the past two years, every company not named Apple has basically stopped MP3 player production.

[…] was running Android 2.2—an OS that was not designed for tablets. Tapping the screen, using apps—it all feels like a lesser experience when compared to using an iPad.


As I walked around CES, the cautionary refrain of the giant from Twin Peaks played on an endless loop in my head: “It is happening…again.” Just like the MP3 player market at the middle of last decade, every company wants a piece of the pie. Just like that market, the field, over the course of a few years, will dwindle to a small group consisting of survivors, the clear winner, and some budget options that probably should just be avoided. So far, the clear winner is the iPad.


What Apple’s competitors should be paying for, behind the scenes, is a team of brilliant young product developers who can beat Apple to the Next Big Thing.

Looking at the approach that Apple seems to take on hardware AND software develop as a holistic process with full control over the hardware used, Marco Arment gave his opinion on the tablet market and how Apple is able to be on the better side of things on December 31. I take some quotes from that here, as it reflects my own ideas much better than I can write at this point:


Apple could solve the software problem exceptionally well, putting them in a strong position to release the first mass-market tablet that had a chance of being successful. So they did. And they ignored the other tablet problems, for the most part: they did what they could to minimize them where possible, but they’re all still very good reasons for many people not to buy iPads. Fortunately, there are millions of other people swayed by the appeal of this low-needs, tactile, and downright fun computer because of its incredibly good software.

But no other device maker today can solve the software problem as well as Apple did.

Most of them are hardware companies relying on someone else — Google or Microsoft — for the majority (or the entirety) of the software.

Neither Google nor Microsoft will ever be able to tailor their software to other manufacturers’ specific (and varied) hardware devices as well as Apple can (and does) with theirs.


These manufacturers aren’t software companies: they’re hardware companies that write software out of necessity. Apple is a software company that makes hardware out of necessity. The software side of a modern computing platform is far more difficult and expensive to create and maintain than the hardware. Anyone can cobble together the same processors, DRAM, flash, and radios as Apple, put them into a plastic case, and run a commodity OS on them with slight front-end customizations. But not everyone can create an entire software platform.

It’s not just a matter of interface design. Apple has built an entire ecosystem to support and enrich the iPad for both customers and developers. To be competitive, a newcomer to the tablet software market needs to replicate or sidestep the need for nearly all of Apple’s major efforts

From a more recent post, “Too much hardware choice,” Marco states his opinion more:


The manufacturers and carriers have very little incentive to maintain the software on devices that are still relatively new and under contract, because they want everyone buying the newest ones instead. We’re already seeing carriers and some manufacturers refusing to release new Android versions to handsets that were launched as recently as 6 months ago, even though most users bought them with 2-year contracts.

On hardware and vendors of Operating systems


Instead, we were greeted with a bevy of hardware choices that we neither asked for nor wanted. Rather than committing to one device — taking a stand and saying, “This is what we think is best, and we’ll support it for the next two years” — Microsoft punted, leaving the manufacturers to give us too many fragmented choices that will likely face many of the same issues we see in the Android market.