While Gabe did a great job covering the news about Citrix Synergy Barcelona this year , I wanted to chime in on some other thoughts after attending the conference.
Last year at Citrix Synergy in Barcelona, the head of the Citrix “Office of the CTO” Martin Duursma ran a second day keynote along with the other Citrix CTOs. It was an odd enough session that I wrote about the session twice, immediately after the event and a few days later after I thought about it more, on this site. This year they CTOs were at it again in Barcelona. I am very glad they came back. Last year, we arrived expecting a typical Keynote, but got something very different. It was a bit awkward and not being a typical keynote we weren’t sure just what it was. This year, Martin made sure to make the session a little bit more of a keynote with an underlying theme to follow.
In the multi-hour session I was taken by how Citrix’s thinking has evolved. If you look back 5+ years, the Citrix mantra was one of “Access”, meaning remote access to your (shared) desktop. Access from anywhere, and Citrix does that pretty well. Lately, it has been access from any device, and about a higher quality end user experience. And then we got this whole VDI, cloud, BYOD, and whole lot of other interesting, yet not quite solid, things thrown at us. For quite some time I have been saying “it’s all about the data”, and I think that Citrix is getting closer to that understanding.
To start with, previously Citrix talked about “any-any-any” (meaning any user, any device, from anywhere). Last spring at Synergy San Francisco I lashed out about that at one of the Geek Speak sessions. Users today are more connected, with more devices, and they share much more information with other people. So when Martin put the words any-any-any on the screen I cringed and tweeted “no, it’s many-many-many”. What I didn’t know was that a few minutes later Martin would say the same thing after walking the audience there with him.
Users access many devices simultaneously. I don’t know who first came up with the term for geekheads of “STD”, short for “Schlepping Three Devices” (I heard it first from Labana), but there are a lot of people at a conference like this with a Notebook, iPad, and Smart Phone with them. And the phone stays on all of the time. So these users are accessing their stuff simultaneously from multiple devices. This has a tremendous impact on how we treat data. While not spot on, Citrix is moving in the right direction. But what direction might that be? Here are some of my thoughts…
When you only have to worry about serialized access to data, we can build systems that work on a “check in, check out” synchronization model. As soon as we have parallelized access, the synchronization model needs to be much smarter. This is where we are heading, and the systems have a lot of work to do get this right. I think Citrix is getting this, and are fitting ShareFile in as an enterprise solution. I use many cloud storage solutions and they each handle the basics, but we have a long way to go.
To complicate things a bit more, it isn’t just about synching all of my data. There are differences we need to take into account.
· I am someone that lives on a Notebook. I want all of my data there, whether online or offline.
· I also use a lot of other OS’s (mostly VMs) when I work. And I want access to the data there also, Most of the major cloud file storage platforms can handle this well for me, synching the data to those devices I connect in from without me thinking about it too much. But so many of those devices I rarely access and don’t really want all of that data automatically synching.
· But I certainly don’t want all of that data synching with my smart phone. Some of it, I need when offline, such as contact lists. But storage and bandwidth to sync is limited on a phone so most of it I like having access to on a demand basis. This is where real-time data synching – streaming – comes into play. In this model, the main synchronization tells the remote device about the data. Think of this as like having the file directory without the actual files present. I’m sure someone would use the term virtual data for this (but not me). When I ask for a particular file, that data syncs down. This is kind of what App-V 5 is doing with cache-less client streaming (although a bit more extreme as the streaming is sub file-level).
· I want to share data subsets with others also. The folder-level sharing techniques used by some of these services works OK for now. But the notion of identity is going to be a problem eventually. What do I mean? Each of us uses multiple identities on the internet. And it isn’t just a work identity and a personal one. Identity based access models break down when this happens.
· Just how would I separate out everything I do as “work” or “personal”, anyway? I have many “work colleagues” that are also “personal friends”, and they interact with me similarly.
And what about the apps? From my laptop, I am going to consume a document using Microsoft Word, whether I am reading or editing that data file. But not on my phone. What I would love is for the technology to recognize that I don’t edit a word document there and don’t have the office app. When I want to access that file, something should PDF it for me and send that down for viewing. Maybe I can call this “right-right-right” (The right data, at the right time, in the right format).
Ah, but how to orchestrate all this? Do we use a central intelligence model that understands my usage by watching how I access this data, or (as a rip-off of “user installed apps”) is it a “user directed data” approach? That’s for people that get paid to think about these things to figure out. Both are viable approaches, but it takes a thought process that starts with the data, moves out to the apps, and finally the user (not the other way around). As I’ve been saying for a while now, it’s all about the data!
Tim Mangan is an independent consultant at TMurgent Technologies, LLP.