English | 中文 | Русский | Français
2,856 Posts served
8,606 Conversations started
In the digest that sends me links to interesting online tidbits, the description of a recent ZDNet blog by Mary Jo Foley, provocatively titled What will next-generation multicore apps look like?, made me think it was just going to be another expert wingeing on about the same old things wrong with the current methods of software design that aren't going to work and how we need something new and better. You know the kind I'm talking about: all bluster and no real solutions. I was pleasantly surprised that Ms. Foley's recap of some the comments from Craig Mundie at the Microsoft FAM made it sound as if the discourse was well above the typical pundit gnashing his teeth on the sad state affairs within concurrent programming. Alas, skimming through the transcript of the Mundie talk, there were no real concrete solutions and plenty visionary artifical intelligence speculation.
However, while the "next-generation multicore apps" sound as if they will be something from an SF novel, it was abundantly clear that Microsoft has a good handle on why multi-core processors are needed and the pain that is going to be faced by the current and next generation of software engineers. These will be growing pains all around the industry, but we should be able fill that growth potential with something new and bigger and better than anything we've seen or thought up before this.
Intel has been crowing about this for a couple of years now. It's nice to see that someone as big as Microsoft "gets" it, too. I look forward to the new ideas and tools from the folks in Redmond.
