MSN Direct Watch – Part 2

Some more info on the SPOT watch: It is much better than I thought it would be. I am getting used to wearing it now, and it is not that much of a distraction.


The messenger thing is cool, although the response time is unpredictable. Friday I sent myself a message and it took about 4 minutes to arrive. Saturday I sent myself a message and it took 45 minutes. Although, the problem could be from where the watch was when I sent it. For some reason there is a complete dead spot in my house, if I set my cell phone down in that area the batteries will be drained within a few hours. I cannot get any radio reception in that area either, even with a FM desktop radio. Since the watch runs on the FM band, this makes sense and the watch was reporting that it had a 0% signal strength. I went to another part of the house and the reception went up to 85% and immediately got the IM.


I am getting more information that I thought I could. I am not sure if I will ever need to know barometric pressure, but it is there. I can get stock prices as they change throughout the day, and there is even a nice little graph of stocks over the past few days. And it has already paid off by reminding me that I had a meeting at 11AM, and when that meeting was going long it reminded me of a lunch meeting at 12 noon. Usually I need to go back to my office to look up the next meeting, but right there, on my watch, was the room I was meeting at.


Scott and Andy left behind a few links in the comments section of the last bit I wrote about the watch, one with a link to info about the MSIM protocol, the other a review (with more info than is here). I will need to look up how to programmatically send myself an IM. I have been keeping a daily record of the Amazon sales rank of my book since it was released in an Excel chart. Every day I go to Amazon, look up the info, then I record it in my spreadsheet. It would be nice if I could run some app that gets this info from the Amazon web service and sends it to my watch. I was thinking that IM would be the best way, but it would also be really cool if MSN had some web service that I could call where I supply the unique ID of the watch and a passport user/ID, then a sting to send. Maybe I will need to track down a contact in the MSN group and see if I can talk them into doing this (although, I suspect this is how they do it, but it is not documented).

Comments (14)
  1. andy says:

    That ‘area’ in your house is near the alien propulsion unit that brought you to this planet. I guess you forgot that you hid it in the wall, right?

    Either that or you have the ‘Skibo’s triangle’.

    My brother had something similar in his last house.


    Watch sounds awesome.

  2. Craig says:

    Hmmm, that would explain why I bleed green.

  3. Craig Eddy says:

    Chris Sells has some IM classes at They work like a champ.

  4. Apparently some of them are not obviously dorky. And seem to work for messaging, etc….

  5. James McDisi says:

    Apparently the guys at Ambient Devices agree with you about being able to add APIs. I’ve got an Ambient Orb (admittadely not a watch, but they run a narrowband network like Spot’s) and you can add whatever content you want to it. They were snapping back at Gates about releasing another closed system when he announce the sale of Spot watches (

  6. John Happy says:

    I’ve been wearing the Suunto n3 for a couple weeks. The news, weather, stock info and different watch faces are well-presented and interesting … the atomic-clock accurate timekeeping, calendar and instant messaging make the technology ‘worth it’ to me. The addition of more channels will make the watch an even greater value. I wouldn’t call the n3 perfect (the alarm volume could be a little louder) but I’m not at all disappointed about the money I spent.

  7. David says:

    Craig, any more info about IM API and method for sending custom data to the watch?

  8. Craig says:

    David: I have heard some rumors, but nothing definate.

  9. David says:

    Craig I’m using the imcli classes from and trying to ‘reverse engineer’ the specifics of the PGD command that MSN messenger uses. Fairly easy to get packet-level output that seems to match precisely what the IM client is doing…..but no success so far.

    Working upstream of the PGD command to determine if there’s anything unique in the authentication or locator phase of the communication.

    Any thoughts?

Comments are closed.

Skip to main content