The Disposable Laptop Project – reducing the cost for the AppSense user base

One of the things that I have been driving here at AppSense for some time now is something that I call the “Disposable Laptop project”. 

Let me set a scene;

At AppSense we have made use of the Remote Desktop Services and XenApp technologies since inception of the business back in 1999 (OK, so the technologies had different names back then, but in essence delivered a desktop and application set to the user population – much like they continue to do for us today :-) ).  One of the significant factors for our using this technology was that we would reap the rewards of low cost computing with regards to the end point devices that we used.  Back in ’99, we had a dozen or so thin client devices in the business (probably cost $500 each back then) that didn’t ever require any IT support time really at all.  IF they broke down, we replaced with another device with around 10 minutes downtime for the user, and we were off again.  I say IF, because we still have most of those devices today, and if it weren’t for their inability to deal with scrolling etc in the likes of Microsoft Excel, they would still all be in use.  What a fantastic return on investment we have had from those devices. We all know this story, since we most likely all have made use of the same technology and reaped those same rewards over the years.

As time has moved on however, our challenge has morphed, because our sales and marketing folks now have differing needs from the IT systems that we provide to them, and the key change is that of mobility.  So as a result, where in 1999, 75% of our employees used thin client devices accessing a Server Based Computing solution (the remaining 25% being developers with powerful laptops or workstations), we find that TODAY, only 20% of our workforce operate solely from a thin client terminal, with laptops now being almost the defacto for new starters in the business.  So over the years the cost to support and maintain our workforce has gradually been creeping up for all the usual reasons (even though the acquisition cost of the mobile device is slowly dropping);

• Administrative requirement to enble the user to add printer drivers etc
• User installed applications causing stability issues on the laptops
• User data and security – how to manage our data and back it up suitably
• Laptop theft / damage causing downtime for the user while they await a new device – only to find their data was not backed up suitably and that they would have to create the document(s) from scratch / memory

I don’t think these challenges are any different from those that other organisations also deal with on a daily basis, but that does not detract from the pain caused for the end user at the time, and the cost that it is implying onto the IT group, and hence the AppSense business.

The reason that we were so able to reap significant rewards was simply that the SBC solution allowed us to deliver a standardized desktop to the user population that enabled them to do their job.  The fact it was standardized, meant we were far more effective with our use of IT resources. 

Back in 1999, nobody spoke of standardization and my analogy of Lego (see recent posting – “Standardization / Lego blocks – crucial for managing desktop infrastructure?” – https://appsense.wordpress.com/2009/07/22/standardization-lego-blocks-crucial-for-managing-desktop-infrastructure/) just plain would not have worked out so well, since the closest anybody came to standardization was low level imaging of servers and applying a new server name and the all important SID to each new “imaged machine”.

So, now the scene it set, let me get back to the point ! :-)

As we step forwards, we are now seeing technology arise that allows the Lego analogy to function and will allow me to drive my new project within the company that I know will have significant value to most other enterprises – the project is simply this – “The Disposable Laptop Project“. 

I want to continue to be able to provide IT in the form of a mobile device (as appropriate of course)  to our user population that enables them to deliver upon their job function.  However, I want to do it in a way that is (from the IT Group’s perspective) as controlled as the deliverable back in ’99, but that offers the freedom of today, where the user “feels ownership” over their corporate laptop, having the freedom to personalize it as much as they do their home PC.

By using the Lego approach, we can ensure that we have;

• Single Standard Operating System
• Core application set layered on top of the above
• User installed applications managed as a separate layer on top of the above
• User personalization to ensure look and feel of all the above remains personal to the user
• User data management – managing the data that all of the above are actually (and ultimately) there to serve

As a note I accept that it most likely will not be possible to only have one shared OS image, but the point remains – we have (for example) just 4 images to manage rather than 1,000’s.

By enabling the management of each of these layers individually, we are able to quickly construct the mobile device operating environment on behalf of the user seamlessly.  This makes the mobile device no different from a hosted virtual desktop incidentally – I could be talking about either environment or BOTH of course.

So, in the situation where user Oliver has just had his laptop stolen from the trunk of his car, the IT group can allocate Oliver a new device, provision an OS and the core apps to the device.  As soon as Oliver logs on to the device for the first time, his own user installed applications will be applied, as will be his personality information as well as the data that is ultimately used by all his applications.  The point is that within 30 minutes, Oliver will be set to work without needing to worry about any of the things that enable him to work in the first place.  This is a huge saving for Oliver as he does not now need to go off and find the tools / utilities that he requires, download all the data (worse still, re-write the data), and then spend days making the whole environment look just like it did yesterday.

You get the picture – so how is the project getting on in AppSense?

 Well, my own laptop TODAY has a base OS delivered to it, the applications are actually locally installed (soon to be delivered to the device via an application virtualization technology) I have my personalization dealt with by one of our own products (Environment Manager) and I curently employ standard Microsoft Offline Files within Vista to deal with my user data.  I am also making use of a very early user installed applications product that AppSense are working on (demoed recently at the BriForum event in Chicago) to manage my own user installed aplications.

Thus far, I find that the offline files are OK, but I am unable to store Outlook PST files on them since they will not synchronize as I have them open most of the time.  This for me is a bit of a showstopper as the files that are possibly the most important to me will not ever be synchronized and hence in laptop failure / theft, I would still be left high and dry :-(

My personalization follows me beautifully between the laptop (MS Vista), the XenApp solution (Windows 2003 Server) that AppSense have at my disposal and the hosted virtual desktop solution (MS Windows XP) that I also make use of occasionally.  If we look at that again quickly, we may just notice the subtle point here – I am able to use both V1 and V2 of Microsoft profiles here and share my settings across the two without actually worrying (or indeed caring) that there are two versions in use.   This is actually a very key point for the Environment Manager solution since our user population are not ALL Windows XP OR Windows Vista, hence we need to cater for both.

The final piece for me is the user applications, and the new product I mentioned is managing this well enough for me presently to accept that I now have the very first “Disposable Laptop”.

The places that we need to work on now are the user installed applications and better management of the actual user data itself.  Once these two key components are under the belt, the cost for mobile computing within AppSense will have dropped significantly and the end users will be receiving a far superior service from the IT group. 

Watch this space!