SyncML Tests Remote Management for Wireless Devices
With Douglas Heintzman
Chairman of SyncML and
Manager, IBM's Pervasive Computing
Strategy and Standards Group
by Vance McCarthy
Later this month, the SyncML Initiative will begin the first testing of SyncML Device Management products for interoperability. These tests, if successful, would lead to the first wave of remotely upgradeable wireless devices (cellphones and PDAs) using a device-independent spec.
The testing, which will take place in Las Vegas, Nevada, could add an entirely new level to the term "smart devices," as updates, reconfigurations and even deployment of custom GUIs could be as simple as a mouse-click -- and done from thousands of miles away. Currently, reconfiguring or updating wireless features requires the changes to be made manually into each device.
On the eve of these important tests, Integration Developer News spoke with Douglas Heintzman, chairman of SyncML
Before we begin, a couple of quick notes:
[SyncML is the open standard that drives data mobility by establishing a common language for communications between devices, applications and networks. A multi-vendor wireless initiative formed by IBM, Lotus, Ericsson, Nokia, and Openwave among others to bring better wireless connectivity and manageability to cellphones, PDAs and mobile clients. SyncML recently elected to fold into the Open Mobile Alliance (OMA) where other notable enterprise vendors, including Microsoft and Sun, are members.]
[Also of note: Heintzman, in addition to his role as SyncML chairman, is also manager of IBM's strategy and standards group for pervasive computing.]
The SyncML Interview
IDN: SyncML has been working on Device Management for some time, haven't you? What is the significance of the mid-September engineering meetings?
Heintzman: Our work dates back about a year, when we stood up January 23 and showed a proof-of-concept demo of 5 clients, with 5 servers offering 5 functions like setting graphics, clock settings and language outputs. Then, we offered to let members of the audience just call out what server, client or setting they wanted changed, and we would do it from a central console in real time. It really was a good demo of the SyncDM spec, and now we're moving down the road so that this month we'll be setting interoperability testing and requirements to let vendors build and test their products to work with the spec.
IDN: And what would the end result of SyncDM compliance allow?
Heintzman: The goal will be to allow N[any]-to-end remote management and update of mobile clients from any point in the connection -- the server for multiple client updates or the individual client.
IDN: Will SyncML be able to help vendors deliver vision at the September event, or will there be another step?
Heintzman: At September's engineering event we will put a handful of companies that are readying commercial products into the engineering compliance testing phase and see if they can interoperate with the testâ€¦and with one another's devices and software. If all goes well, we'll do a formal certification process for Interoperability compliance with the DM spec and start to 'logo' the products that comply at a November event we have scheduled in France.
IDN: How big is the demand for these SyncDM features? Do you feel they will have a material affect on the industry?
Heintzman: These DM features are hotly anticipated by the carrier community. Bottom line, the carriers expect a real revenue impact. First, because it will save them large staffing and support costs that go with custom configuring of phones that has to take place today. But also, they expect in a short time these DM features will make it easier to sell users on adding/changing features to their phones. After all, if by pushing a simple "update function" button a user can subscribe to a new service, get his phone debugged or even change the language that their phone displays text in, people might be more likely to use those features. Right now, cellphones are difficult to change. The screens and dial pads are small, and the features need to be burned in [to the circuitry]. And so, 99% of the time, the only time a user changes his setting is when he gets a new phone.
IDN: Interesting. So, I would imagine there would sooner or later be an enterprise manager and/or developer demand for the SyncDM features? When do you think we could see that?
Heintzman: That's an interesting question. Certainly, we see the telecom carriers and wireless providers as being the primary customer for these services at the beginning. But there is no doubt the same DM technology could be used to update and reconfigure laptops, make pre-configuration for mobile users much easier and even let developers add new features on a local server first to test it, and then with a couple of mouse clicks deploy that new service to every mobile user that might want it. That's a big change from having to program each user's PDA, laptop or whatever, one at a time.
IDN: So, when and where do you expect the first commercial implementations or FCS of the SyncDM spec?
Heintzman: The first will no doubt be the horizontal, high volume carriers, like a Vodaphone who might want to let their millions of cellphone subscribers get a new screen or new features, like 4-way calling or something like that.
IDN: Aside from cellphone centric features, like conferencing and call-forwarding could SyncDM be used to also update PDAs and other digital devices with services that a remote enterprise user might need, like parts of a web services security feature for authentication or identity?
Heintzman: The idea behind the SyncDM spec is that the device management capabilities is feature agnostic, but the capability to add or modify device features over the air is the important element. So SyncDM is meant to be a garneted delivery system for coordinating servers and the remote client. It supports the most popular wireless protocols, including WPS, OBEX and HTTP. As far as new applications we could support, we are now looking at spec'ing out how to support digital rights management.
IDN: Who is interested in that?
Heintzman: Well, the music companies are interested in making sure that a person listening to music has the right key.
IDN: That doesn't sound so different, in concept at least, from making sure an enterprise user has the correct rights for downloading or reading a file based on an enterprise server back in the office?
Heintzman: Certainly you could view SyncDM as enabling those kind of functions, although we wouldn't get directly involved in writing those specs or applications.
IDN: There also seems to be growing interest from enterprise vendors in SyncML. For instance, what can we take away from the fact that both Oracle and Apple have recently joined your association?
Heintzman: With the addition of Oracle and Apple, we are certainly starting to see a broadening of support from the major cellphone companies to enterprise vendors.
IDN: Are there any specific enterprise-centric implications arising from these companies joining?
Heintzman: Well, having Oracle join, for instance, lets us dust off some work we had been doing to define an SQL definition for SyncML protocols.
IDN: Interesting, What would that do?
Heintzman: Well, as mobile devices become application platforms, enterprise managers will need to make sure that users are working on synchronized data -- that is that data going into the device as well as into the server is synchronized and updated. With Oracle as a member, that allows us to work through some of the nitty-gritty details of putting together such a spec for an SQL embedded schema for mobile devices that will let those devices map to enterprise-based SQL databases.
IDN: Well, that's truly a remote access/web services type application, isn't it? But doesn't SOAP provide the kind of core transport needed for that?
Heintzman: SOAP is OK, but there will be situations where we want to combine SyncML and SOAP technology. This would let us move SOAP structures back and forth between the wireless and the wired enterprise. It would also let developers leverage some of their core message-oriented code.
IDN: I know that IBM and Microsoft are also members of OMA. Once that SyncML group falls under the OMA umbrella, do you see any other areas where SyncML might get more active in researching support for various web services standards from OASIS or W3C?
Heintzman: We need to interact with web services standards to allow mobile payments, for example. And, we're already pulling together some simple technology and protocols from the IETF for that purpose. So yes, there are some really interesting connections between these wireless and enterprise groups over web services. And as time moves on I expect we'll see more.