PHOTOOG Photography writings by Olivier Giroux


My Strategy For API Wars : Let the Dust Settle

This is in weak response to this piece, itself derived from this other classic piece.

For my part, long lists of new and shiny APIs fill me with skepticism and are ever my invitation to side with inertia. That, and marketing smoke and mirror games drive me away from the new and shiny.

Seriously guys, touting "the creation of stronger customer connections" as a Vista feature is like claiming people's souls will open to my own insecurities about self-fulfilment if only I made an animated toolbar widget. Really? We'll connect on a deep level like that?

In my short experience with this industry I've found that anything new that has any technical merit always meets with three things from the community:

  1. Complete Apathy. It's irrelevant, just another kind of round wheel.

  2. Massive Enthusiasm. It's better than sliced bread, and peanut butter.

  3. Crippling Fear. It'll doom us all, black helicopters cometh.

Equally brilliant people will find themselves arguing any of these points at the same time. I'm no exception, I do this too. When there's equal noise on all sides, though, logic favors the moderates.

So is Microsoft going down in flames with all these new APIs? Will it reinvigorate the giant?

Maybe either, but today I choose optimistic apathy: I think the overall effect will be hard to perceive but some interesting stuff might emerge. Some of the new APIs will rock, some will be rocks. Those developers who write software for Windows only are likely to pick up the better ones, while those who write cross-platform code (not least of which are game developers) will ignore all of them until the adaptation libraries start to leverage them.

This is going to be my fate as well. Against my personal preference I have to write cross-platform software by day, so the argument is a little moot for me. I won't use any of it for a long time, and it'll have to be pretty darn compelling for me to write a seperate code path for Vista (I am planning on it for one software however).

A note on the adaptation layers: I'm really looking forward to the enhanced MFC, and opportunities for VC++ libraries to leverage new Vista kernel features. Backwards compatibility is the way to my developer heart - likewise for many/most of my fellow Nvidians.

Filed under: Uncategorized Leave a comment
Comments (4) Trackbacks (0)
  1. Olivier,
    Thanks for the link. I do agree that with time, these issues will likely work themselves out. The hard part is trying to identify things that could be useful in the near term that won’t simply disappear over the long term.

  2. Btw, here is a follow-up post from the Visual C++ PM.

  3. Sorry, I missed these comments. Yes, I check out Teixeira’s blog on a regular basis. I enjoyed that exchange quite a bit.

  4. Plenty of people moved on from ‘waiting for dust to settle’ to ‘not waiting for Windows to display it’.

    My 2 story-telling cents,

Leave a comment

No trackbacks yet.