Loading ...
Sorry, an error occurred while loading the content.

Re: [XP] Embedded systems ARE bridges

Expand Messages
  • Bill Kelly
    From: Joshua Macy ... [...] Thanks for your considered reply. You ve got me thinking about this topic a little more broadly than
    Message 1 of 34 , Nov 7, 2001
    • 0 Attachment
      From: "Joshua Macy" <j_amused@...>
      >
      > I actually think that most of it (all except the regression test
      > part) is straight-forward; certainly if a customer presented any piece
      > as a story I wouldn't blink (although on Windows I might curse about
      > the back-out function, if I had to code it myself; on the other hand
      > most modern installers seem to create a remove script).
      > The regression test stuff is more blue-sky, but I would settle for
      > hooks for running an xUnit suite.
      [...]

      Thanks for your considered reply. You've got me thinking about this
      topic a little more broadly than before. I guess it's interesting,
      we are a CD-ROM encyclopedia product, probably at the opposite end
      of the spectrum from something like a compiler update. We have an
      update mechanism that's more centered on data updates than program
      updates - though we can of course handle program updates too via
      auto-updates.

      We don't currently allow an "undo" of an update. I'm not sure under
      what circumstances a user would want to undo an update in our case,
      unless something went whacky with a program update and things stopped
      functioning properly. We are basically a fancy search & browse tool
      when you get down to it, operating on a static database. In the
      event something somehow did get messed up as the result of an
      update, tech support would recommend a re-install (I'm pretty sure.)
      . . . Well, I'm not sure where this is going other than to say it
      would be interesting to have some statistics on how many tech
      support calls we'll receive specifically because of some snafu with
      the updates. My guess is, though, that it should be few, as again,
      most of what we'll be updating will be the data...

      Anyway, thanks, I'm still thinking about it. :-)



      Regards,

      Bill
    • Bill Kelly
      From: Joshua Macy ... [...] Thanks for your considered reply. You ve got me thinking about this topic a little more broadly than
      Message 34 of 34 , Nov 7, 2001
      • 0 Attachment
        From: "Joshua Macy" <j_amused@...>
        >
        > I actually think that most of it (all except the regression test
        > part) is straight-forward; certainly if a customer presented any piece
        > as a story I wouldn't blink (although on Windows I might curse about
        > the back-out function, if I had to code it myself; on the other hand
        > most modern installers seem to create a remove script).
        > The regression test stuff is more blue-sky, but I would settle for
        > hooks for running an xUnit suite.
        [...]

        Thanks for your considered reply. You've got me thinking about this
        topic a little more broadly than before. I guess it's interesting,
        we are a CD-ROM encyclopedia product, probably at the opposite end
        of the spectrum from something like a compiler update. We have an
        update mechanism that's more centered on data updates than program
        updates - though we can of course handle program updates too via
        auto-updates.

        We don't currently allow an "undo" of an update. I'm not sure under
        what circumstances a user would want to undo an update in our case,
        unless something went whacky with a program update and things stopped
        functioning properly. We are basically a fancy search & browse tool
        when you get down to it, operating on a static database. In the
        event something somehow did get messed up as the result of an
        update, tech support would recommend a re-install (I'm pretty sure.)
        . . . Well, I'm not sure where this is going other than to say it
        would be interesting to have some statistics on how many tech
        support calls we'll receive specifically because of some snafu with
        the updates. My guess is, though, that it should be few, as again,
        most of what we'll be updating will be the data...

        Anyway, thanks, I'm still thinking about it. :-)



        Regards,

        Bill
      Your message has been successfully submitted and would be delivered to recipients shortly.