Catching up is hard to do
"Down doobie doo down down..."
Remember that Neil Sedaka song? No? Maybe it's just me.
Anyway, I was aghast to realize that it's been nearly a month since my last blog entry. It's been a rather busy month, so let me catch you up...
Spanish Dev Lab
Last month's dev lab in Spain was very successful. By the end of the two days, Ayman and I were able to help all of the ISVs in attendance to upgrade their projects to VC++ 2005, with just a few dangling work items to complete the conversions. Everyone was happy with the new performance and security benefits and almost everyone was interested in using IJW interop to use managed code feature from their native C++ code bases. There was one fairly major hang-up early on, though, which was that somehow they attendees were not informed that the presentations would be in English. Ay de mi! However, we were able to work around this issue by talking slowly and explaining thoroughly. Luckily, the code itself crosses lingual boundaries, once you get used to variables being called TrabajadorDelThread and whatnot. :)
Visiting home
I visited family in the Bay Area a couple of times in the last month for the holidays. It's particularly difficult this year because it's likely to be my grandfather's (who is like a father to me) last, as his cancer is starting to get the better of him. However, we're happy for the time together while we have it.
Bay Area ISV Summit
Also in November we held a two day summit with large Bay Area ISVs that depends on VC++ for their business. The days were filled with lots of technical sessions on stuff like debugging, native/managed interop, x64 targeting, security, and the like. We also did a panel discussion where we talked about the future of Visual C++ from both the Microsoft and the ISV points of view. In addition to the presentations it was also an opportunity for us to talk to one of our most important market segments, large ISVs, about how they use VC++ today and how they'd like to use it in the future. The #1 request theme? Tools to manage and understand large C++ code bases.
Why people use C++
One of the other things to come out of my customer meetings over the last month is clarity around the reasons why people use C++. I boil it down to three reasons:
- A substantial existing investment in C++ means that continued use of C++ makes the most business sense. Note that this doesn't mean that the C++ code is somehow less desirable or legacy code. In most such cases, C++ code *is* the active, living code that embodies the organization's intellectual property and assets.
- A desire for a greater degree of control or flexibility than the CLR can offer today. While the CLR is a viable platform for an ever-broader range of application types, it's not yet the right choice for everyone. In particular, software systems requiring tight integration with the OS or hardware, a high degree of control over memory & performance characteristics, or minimal dependencies all tend to be native code scenarios today.
- A need for platform neutrality. We work hard for standards conformance so that customers can write standard C++ code to target Microsoft platforms as well as other platforms, such as Mac or Linux.
Visual Studio 2005 SP1
We finally nudged SP1 out the door, which we announced last week on the team blog. Several hundred fixes for VC++ went into the service pack, so I highly recommend it to anyone using VC++ 2005. If Vista is your development platform, we're working an update to SP1 that includes a number of additional fixes specific to running VS on Windows Vista.
Getting a new PUM
Within the next week, our new Product Unit Manager, Bill Stow, will be leading the VC++ team. Bill replaces Ronald Laeremans, who has been the acting PUM since David Burggraaf left in the spring. I'll try to get Bill to post to the VC team blog when I see him.