COM isn’t (Don’s) love anymore …

Don Box says good-bye to (D)COM here and explains why Indigo is the future.

Comments (6)
  1. R says:

    quote: "Microsoft software architect Don Box said the company will not invest much more in Component Object Model (COM) and Distributed Compound Object Model (DCOM)"

    Well I hope somebody invests something into it between now and 2006/7 🙂 Some of us are still using (or at least integrating) COM.

  2. FrankPr says:

    Rest assured that COM will be supported for a looong time to come (as will be Win32 etc., even after Longhorn has arrived). But the money goes into support and bug fixing, not into developing it further. The strategic priorities have changed.

  3. From the article:

    "Box stressed that COM and DCOM are not dead."

    To say that COM is not Don’s love anymore, who’s to know? But the thing that Don said way back when is that "COM is love", not "I love COM".

    The idea behind "COM is love" was that COM was a great way to interop between all different sorts of code. For example, Developmentor classes at the time would show you how to write a Java class, and connect it up to COM. We know more now than we did then, and we’re moving forward with interop into areas that weren’t really well defined a few years ago.

    So, is COM dead? Well, I guess it depends on how you define dead. Can you still write COM code, interop with it, expect it to work as designed? Sure – not dead. Should I take a brand new dev project and base it entirely on COM? Probably not.

  4. Sami says:

    I think there’s even better explanation for the phrase now, in the light of service orientation. The saying "COM is love" actually summarises pretty well the rise and fall of COM. COM components are so much in love with each other that they know too much of each other’s intimate details, and it is exactly this deadly embrace that has given birth to service orientation. I wonder if Don Box appreciates the irony?

    If COM was love, then what is SOA? Friendship?

Comments are closed.

Skip to main content