Also, can someone speak to why the EF-S folks have a $1000 17-55 F/2.8 IS and EF folks are left wanting?
Is making such a lens for a crop that much more technically feasible or inexpensive?
Obviously - a fullframe equivalent would be 27-88mm. That reach is well covered for FF by the 24-70's and the 24-105, especially if you consider that the depth of field of the 17-55@2.8 roughly equals f/4 on FF. And if you'd really want 17mm, you'd be buying a 16-35 for your 2.8 or else the 17-40. The 24-70/2.8 I and the 16-35 I (and the 24-105) were both in the 17-55 price category, the mark II's outperform all of the above and are priced accordingly. Factor in build and weathersealing, and you'd almost think the 17-55 is the overpriced one

Enniehoo, the 24-70/4. Sub-$500 is not going to happen - this L IS-zoom is just no way going to be cheaper than the 2.8 IS primes. So 'cheap' would still mean it needs to compete with the 24-105 and 2nd-hand 24-70/2.8 I's. Doesn't make sense. A high resolution one with IS might make more sense. The only alternative I see is a parfocal STM lens for video, but I'm not sure they'd try to outfit an L with STM already, on the off chance it'll happily sit being a dud next to the DO lenses*. STM will need to prove itself first.
No, I say the new kid's going to perform
really good, and it's going to cost a bundle. And it'll probably sell like hotcakes to those who need the best, and even better to those who
want the best.
*) Yeahyeahyeah, I'm sure the DO's are good, even very good, they're just not the real deal, what with the weird bokeh