One of the first harsh lessons I learned right away is that this type of remote is not forgiving to a novice.  If you don’t know what you’re doing, be prepared to have a lot of patience.  Despite the wizard’s attempt to outline the steps needed to get my equipment working and trying the remote out with a simple “All Off” activity, the wizard tool (Pronto Configurator) kept giving me a “There is an error with the configuration data. No valid configuration can be generated.”  Suffice to say, I wish that it would have specified what exactly the error was, and a great deal of restraint went into:

• reinstalling the software,
• installing ProntoEdit Professional to take a look at the code database (but not necessarily knowing what to look for),
• performing a lot of code “learning” for the code database, and
• being bewildered as at some point, the “My Database” set of codes wouldn’t populate with entries and I could not open alternate databases such as the One1R database.

After spending nearly 12 hours (spanning three days) tweaking and testing and revising, I was finally able to call my little project done!  Based on that experience, here are some lessons learned:

• Where possible, test out and stick with the Factory Database codes.  Only teach the codes that really don’t exist in the database and always duplicate the component before you start adding or changing things around.
• Expect a lot of trial-and-error when looking for a code set from the Factory Database.  Sometimes the codeset ID will map to a device’s model number, and other times, not at all.  I found that relying on the Component Type helped somewhat, but it would be useful for Philips to include information such as “Codeset xyz works with ACME G-500 and likely similar devices.”  That way, a user can at least test out codesets for devices that you might recognize as part of a product family.
• Think long and hard about your “Activities” (which are in essence your use cases).  I ended up making four Activities that involved multiple devices and mapped buttons that coordinated all of the devices in concert, but I also added four Activities to correspond to each piece of equipment.  Those activities’ sole purpose was to mimic all of the commands possible with just the device remote.  Where a hard button would not cover a command in an obvious way, I duplicated it on the list of additional “Screen Functions.”  This goes a long way in keeping me from reaching for the original remote.
• Make sure you download this little piece of software (here) provided by Philips.  It should save your sanity if you get the “There is an error with the configuration…” message.

Two hardware improvements that should be made:

1. The silver scroll wheel is a great idea, and its implementation is most of the way to where I think Philips should be.  Instead, I would suggest that the wheel either take the form of a thumbwheel next to the screen (scroll and click, like a Blackberry), or else to simply lower the profile and introduce raised bumps for the cursor arrows.  I found that pressing the cursor arrows usually shifted the scroll wheel to another Activity that should not be active.
2. Along the same lines for the scroll wheel, introducing a discrete instead of a continuous wheel (with slightly more resistance/friction) to give more feedback on the menu item selected and also let the user know of accidental movement of the scroll wheel.

Recommendation: Change the scroll wheel to be more flush with the remote’s surface and introduce a bit more tactile feedback.

It’s clear that a lot of thought and effort was put into aligning this product with the marketplace offerings and distinguishing it from the rest of the Pronto lineup of products to offer consumers a great choice.  With some touchups on the minor hardware and software issues, this would be a no-brainer choice for a universal remote without a touchscreen.

Scaling the controls to the content is extraordinarily important and the problem here is that for 10 functions, one product uses 5 buttons and the other uses 3. For example, if the door knob of a door is too small, many people can’t grasp it and apply the right amount of torque to open the door. In this case, Apple might have found the ideal physical control system while minimizing physical controls with the 2G. However, they might have mistaken necessity for an opportunity to be too clever since having fewer controls seems to have made some functions more difficult to access. In other words, the density of functionality with one control method should be kept to a minimum.

Recommendation: Merge the audio feedback system in the 3G with the physical controls of the 2G. It keeps the control systems consistent and simple to understand.