Two years on: revisiting a conversation with Exchange development chief Perry Clarke

Two years ago, I spoke with Perry Clarke, the development chief for Exchange (both on-premises and cloud) to discuss the current state of Exchange and how its technology was likely to evolve. We know how the influence of the cloud has affected the way that product development occurs and how quickly things now happen. I thought it would be interesting to look back on how Exchange has evolved in the period.

The first topic we discussed was the reputation for poor quality that Exchange 2013 had accrued at the time. It’s true that Exchange 2013 had a rocky start and that the first few cumulative updates contributed their own problems to the mix. The problems have largely disappeared as Microsoft mastered the cumulative update process to a point where recent updates have been “boring” because they are high quality. I hear fewer complaints from customers about the quarterly cadence too and it seems like most have become accustomed to keeping Exchange updated.

Quality has carried through into Exchange 2016. It is the first release in years that is a solid deployment candidate from the initial release. I think much progress has been made here and customers can look forward to deploying Exchange 2016 without having to wait for the traditional service pack to appear. Just think of Exchange 2016 as Exchange 2013 SP2 and all your worries about deploying RTM software will disappear.

At the time, Perry stressed the fact that on-premises customers gain heavily from the work done to refine the operations of Exchange Online inside Office 365. Microsoft obviously does the work for their own benefit but it’s remarkable how much technology has been transferred from Exchange Online to Exchange 2016, especially in the high availability space. More will come in future Exchange 2016 updates, notably the ever-expanding archive mailbox.

Moving to the cloud was a big part of our conversation and we debated how much of the Exchange installed base would end up using Office 365. Perry’s position was that it didn’t make business sense for the entire base to move to the cloud and that those who believed that this was the case were “nuts”.

Perry and I reached consensus that about 40 percent of the total Exchange installed base would remain using on-premises software for “at least the next several years”. This feeling remains true but there’s no doubt that the adoption of cloud email is accelerating. At Ignite 2015, a Microsoft speaker claimed that 35 percent had already made the transition and the statement that accompanied the launch of the Microsoft Graph in mid-November said that 60 million people are active users of Office 365. Taking this into account, Exchange Online is now likely to host a couple of hundred mailboxes (of all types - user, shared, inactive, site, group, and resource) with some 60+ million users active on a monthly basis.

Perhaps the 25 percent to bring us up to the figure that we thought would remain on-premises will move in the next two years. After that, who knows. In any case, the point Perry made was that Microsoft has made a huge investment to support both cloud and on-premises platforms and to ensure great interoperability between the two. The launch of recent Office 365 datacenters in the U.K. and Germany is an example of the continuing investment, with the creation of Deutsche Telecom as the trustee for data in Germany being of particular interest in terms of resolving issues around data privacy in that country. It's also worth noting that Microsoft is leveraging the investment in Office 365 datacenters to create a common fabric to host both the Office 365 and Outlook.com workloads.

We also spoke about the notion that email was dying. Today, even though some believe that tools like Slack will take people away from email, it’s pretty clear that email remains the single most important tool for business communication and that Microsoft continues with a high “level of investment that customers are making to do email every fraction of every second of the day”. In fact, if anything, the increased focus on delivering great email clients on mobile devices has increased that investment as seen in the Acompli acquisition in November 2014 and the progress in the Outlook apps since.

Speaking of mobile devices, we discussed Exchange ActiveSync (EAS) and its influence on the market. Perry had a strong opinion that Apple would not have been as successful with the iPhone in the enterprise if they hadn’t supported EAS. Since we spoke, we’ve seen Microsoft do much more to support iOS clients for enterprise customers, leading to the recent “iPhone Pro” name used by Satya Nadella at the Dreamforce conference to describe an iPhone running a range of Microsoft apps.

EAS remains the de facto standard for mobile device connectivity to Exchange and not much has been done to enhance the protocol (version 16 is now available). But work behind the scenes has gone on to bulletproof the server against the actions of rogue (malfunctioning) clients.

Interestingly, Perry said that the “arrival of much more capable mobile devices has made it harder for platforms that use a "sync-based approach" (i.e., ActiveSync) to keep up to date with the evolution of functionality on the server”. The approach at the time was to use the OWA for Devices clients to gain greater exposure of server-based functionality by running the Outlook Web App in a browser container on the client. That approach failed and has now been junked.

Perry’s view that Microsoft would take more control over the mobile experience and not rely so much on the clients delivered by EAS licensees such as Apple by “providing its own mobile clients will be "an increasing part of the [mobile] story” remains a big part of Microsoft’s strategy. The build-it-ourselves approach was replaced by the Acompli and Sunrise acquisitions to deliver the best email clients available on iOS and Android.

Sadly, less progress has been made on Windows Phone and the Outlook Mobile client available for 8.1 and the new universal app available on Windows Mobile 10 are not as modern looking as the other apps. Because EAS remains limited in terms of the server functionality exposed by the protocol, since 2013 Microsoft has created a suite of individual apps to deliver mobile access to functionality such as Delve and Office 365 Groups. These apps use the Office 365 REST-based APIs instead of EAS. Future development is likely to exploit these APIs because of the investment that Microsoft has made in the Graph and the Unified API for Office 365, but quite how that impacts on-premises software remains to be seen.

Even so, there’s no evidence that EAS will disappear any time soon. I hold to my interpretation that EAS will continue as the lowest common denominator for mobile device connectivity to Exchange. The replacement of the browser-based OWA for Devices strategy made sense because it has delivered some great clients, but that doesn’t take away from the position of EAS.

In summary, over the last two years we have seen:

Making any prediction about how technology will evolve in the future can be challenging. It seems to me that Perry’s views have largely been fulfilled since 2013. But then again, if you’re the guy driving Exchange forward, you’d expect that to be the case.

Follow Tony @12Knocksinna