Email complexity and next generation middleware
This morning I have been watching the battery life of my Nokia N95 drain rapidly away. It is 2pm and I have less than 40% remaining. When I unplugged the device at 7am, it was fully charged.
It is not the handful of calls I’ve made or my few minutes of web browsing which have caused this, but my push email client, which is working away in the background updating my mailbox throughout the day.
The effect on battery life is extraordinary and it has provided a stark reminder of just how difficult it is to deliver a great mobile email experience.
Over the last 10 years I have tried numerous solutions: from the early SMS-to-email gateways (anyone remember this on the Nokia 2110?) to the mobile Gmail application. I must have used 30 different products in this time and only 2 have managed to deliver what I’d term a ‘good experience’: the BlackBerry and the Danger-powered T-Mobile Sidekick.
The product currently killing my Nokia battery is Mobile Mail, a white-labelled solution developed by Seven and offered by network operator Three UK as part of its X-Series package.
Battery life aside, there is a lot to like about this system. The setup process was very simple – I just entered my email address and password and it figured out all the settings without any intervention from me. It is included free within my X-Series tariff and therefore offers tremendous value compared to the cost of a subscription product like the Blackberry. Mobile Mail also works with remarkable speed, with emails sometimes appearing on my mobile before they arrive in my desktop inbox.
However, all these fine characteristics make little difference to my long-term chances of using the product. The simple interface, the efficiency of the client/server processing – even offering it to me free – mean nothing while the battery life issue continues to deliver a broken experience.
It is now 3:08pm and I have had to plug my N95 in to recharge after multiple ‘Battery low’ warnings. Today I am in my office, with access to a charger. Tomorrow I may be on the road and anything which kills my mobile halfway through the working day is simply unacceptable.
An otherwise good mobile user experience, fatally compromised by inefficient use of power.
I’ve always been tempted to think that most mobile email architectures are basically the same, but having spent some time searching for the best experience, it has become clear there are huge variations in efficiency.
For instance, the few hours usage provided by my N95, Three and Seven combination contrasts with the 3 days I get from my T-Mobile Sidekick.
Battery life is frequently cited by Blackberry users as one of their key reasons for their continuing loyalty to the brand.
It is a great example of the difference between a technology-driven approach and a user-driven approach. Neither the Sidekick nor the Blackberry use the latest open operating systems and they favour 2.5G GPRS radios over the latest 3.5G HSDPA connections. If one were to glance at their specifications, the N95 would out perform the Sidekick and the Blackberry in almost every respect. However, it delivers far-and-away the worst mobile email experience.
The Sidekick and the Blackberry focus on the aspects which are most important to the user: great battery life, reliability and ease of use. Their architecture is built from the ground-up for a usage scenario in which a device needs to constantly confer with the server to see if there is email waiting and, if so, download it in the background. To do this in a power efficient manner is actually a considerable engineering challenge and one that few companies have managed to crack.
It’s no wonder mobile email is such a litigious area. Research In Motion, manufacturers of the Blackberry, famously settled a suit with patent-holder NTP for USD 650m last year. Visto has just reached an agreement with Microsoft after it alleged the software giant infringed some of its patents. Clearly those who think they’ve cracked the email experience want to protect their technology.
I think this is going to become a very interesting area. The enterprise usage scenarios are obvious, but there is also growing demand from consumers for mobile email access (witness RIM’s recent deal with Vodafone to bring Blackberry to the mass market and Visto’s consumer deal with T-Mobile).
However, it is worth noting that this is not limited to email. The same kind of asynchronous data transfer which makes for a great mobile email experience is also a crucial component of many new mobile data applications: caching RSS feeds, photo sharing, downloading music and video content. All of these services benefit from having an intermediary layer able to sort out problems like dropped connections, working in off-line mode and the battery unexpectedly running out of power.
The middleware expertise gained by mobile email providers is going to be enormously valuable to the industry when developing the intelligent layer between the handset and the world of web applications.
It’s no wonder Microsoft is rumoured to have paid USD 500m for Danger, the company behind the T-Mobile Sidekick. I think time will prove this to be a very smart acquisition.
Power requirements and the inherent unreliability of wireless network connections will mean that many core mobile applications will need an intelligent middleware layer for some years to come to ensure consistency of experience. Ownership of this layer is likely to become an important strategic battleground. It is not difficult to see Nokia’s Ovi taking on this role or the likes of Google and Yahoo evolving their mobile platforms to offer this kind of functionality. Microsoft is now well placed after its Danger acquisition and Research In Motion have some very valuable assets in their Blackberry server platform.
Network operators should take note: those that are thinking intelligently about their long-term future in the mobile value chain would do well to look closely at how they can implement a generic middleware layer within their network and open it out to new third party applications.
What email solution do you use on your mobile device? Do you have two devices: one specifically for email and the other for web browsing? What other mobile applications do you wish worked as reliably as your Blackberry email? Post your comments below…
Nokia proposes a super battery with which you can reload 3 phone batteries when you are off mains outlet Nokia Power Pack DC-1 and a smaller Nokia Extra Power DC-8.
Personnally i use Gmail because the prez on a small screen is very good and the spam is fully filtered out.
Yes I have to connect, but it is quick and discreet enough and i don’t like to be “disturbed” for “small” info.
Just as the previous reader, I have chosen Gmail. All my other mailboxes are forwarded there and I’m using the Gmail native client (Java app). Google is filtering all spam quite well. The features I miss in the phone are group message management and file attachments.
Doesnt the VZW BlackBerry support 3G services and stil provide excellent battery management?
Re: the comment on the Verizon Wireless Blackberry, it is certainly possible to provide good battery performance and push email over a 3G network (note that Verizon’s network is CDMA-based rather than the UMTS used by my N95). I haven’t used the Verizon version myself but I wouldn’t be surprised to see that it offers good power management – the design of the the Blackberry architecture is inherently power efficient.
It is an interesting article. I just start using a device with Seven’s email client. Haven’t paid much attention on the battery life yet. I’d like to study more on this battery life and mobile application topic. Where can I find more info on some common practice in designing a software with reduced power consumption?
The gap in thinking of putting an email or other app on a phone and draining the battery is like having a wifi hotspot without electric plugs. Just not thought through from the user perspective.
I use Momail (a Swedish company that’s recently opened up in the UK and a few other European markets) as it’s free, their data optimisation is excellent and they use the phone’s native client. Presumably your problems are not so much with email being a power drain (as IMAP-IDLE properly implemented is very data efficient) but with the power cost of keeping the wireless connection open? Whilst I hope for some fantastic breakthrough in battery technology, I suspect the only near-term solution is to wean ourselves from the belief that we have to know the moment a new message arrives, rather than checking every few minutes. After all isn’t that kind of ‘instant’ notification and response what IM is for rather than email?
Regarding mobile e-mail experience, I use my iPhone’s native mail client on several IMAP mailboxes and find the overall user experience convenient. Like Larry, I still prefer to “pull” e-mail when I have availability rather than be pushed. Plus the synchronization that IMAP provides is very convenient, as well as the possibility to move messages into folder, with the only caveat that if you end up with too many folders (which is the case of one of my mailboxes, that I have been managing and segmenting over the past 12 years), it becomes very tedious to “thumb scroll” and locate the folder you want to file your message into.
Now regarding the patent war around mobile e-mail, it may be worth looking into what SetNet is doing in that area…
[…] very nature of wireless technology mean this middleware layer will only grow in importance. I have written about this previously and stand by my assertion that companies which succeed in developing an effective architecture to […]