December 12th, 2008, 12:15 AM
The future of the "Secure SCL"
I won't go into all the reasons why I think the current Secure SCL option does more harm than good, at least as long as it is enabled by default, but I'm curious about what's going to happen with this "feature".
I've read several blog posts by VC++ team members defending this, but on the forum for the recent VC++ CTP release, it suddenly seemed like it might get disabled by default in the future.
Are there any news on the current status there? Are there going to be changes, either in how and when it is enabled, or in how it is implemented?
December 12th, 2008, 12:54 PM
Re: The future of the "Secure SCL"
I assume you are talking about Stephan’s comments here: http://social.msdn.microsoft.com/For...5-6acc20e23e92
“We're considering disabling _SECURE_SCL (well, its VC10 equivalent, _ITERATOR_DEBUG_LEVEL) by default - I am personally in favor of this - but we have not yet decided one way or the other.”
We have had a lot of feedback on this feature (particularly on the default setting for this feature), which is great as it tells us what we need to be thinking about. We always want to make the best decisions we can, which often means waiting until we finish implementation of a feature so we can understand/quantify it’s true benefits/cost in multiple scenarios (as always, we prefer not to have to guess.) What we do know is that the feature, albeit with a different name/switch as Stephan says above, will (still) be in VC++. Ale, Stephan and I speak often on what the default should be, in fact Ale was just in my office and we spoke about this post/reply, and once we decision on the default we will of course let you know and again listen to feedback. The decision will not happen for a while yet however (well after the Slow Chat finishes – so best you stay looking at our VC Blog for this.)
Click Here to Expand Forum to Full Width