Aart, you cited you all know that there is much that is new in EDA in terms of certainly accelerated growth, new customers, new customer requirements, new products and the like. But over time, EDA history has a way of repeating itself. And so, when we think about, for example, the thing that you started in your comments with AI -- and by the way, there’s a very interesting Synopsys presentation this morning at the ANSYS conference on AI. The question I have there is about historical precedence for earlier new tools. So, for instance, in the early days, our synthesis and implementation, eventually there were issues with respect to tool capacity, for example, in terms of synthesis blocks, for example, methodology. I’m sure you remember the great debate about flat and hierarchical and extensibility of the tools to new device types or applications. So, the question with respect to AI is how confident are you that the platform that you have is viable for 5 to 10 years in terms of capacity, performance and meeting the needs of a wider and wider set of device types, so that the kind of walls perhaps at some of the earlier tools eventually ran into, AI doesn’t necessarily run into. That’s question number one. For you, Trac, you raised the midpoint of your non-GAAP expense expectation for the year by $95 million. Could you talk about that in terms of headcount-driven growth behind that raise and guidance for OpEx? And if, for example, you were to fill every one of the more than 2,000 positions you now have opened, would you still be able to meet your margin objectives for the year?