While TI did restructure on the OMAP side, the parts are still active, not EOL'd: http://www.ti.com/lsds/ti/omap-applications-processors/omap-4-processors-produ

While TI did restructure on the OMAP side, the parts are still active, not EOL'd: http://www.ti.com/lsds/ti/omap-applications-processors/omap-4-processors-products.page http://www.ti.com/lsds/ti/omap-applications-processors/omap-... The OMAP dev kernel is also quite active: http://git.kernel.org/cgit/linux/kernel/git/tmlind/linux-omap.git http://git.kernel.org/cgit/linux/kernel/git/tmlind/linux-oma... Look at the checkins in just the past few days. Here's what it looks like on the Google side as a contrast: https://android.googlesource.com/kernel/omap https://android.googlesource.com/kernel/omap While TI announced they wouldn't be pursuing new smartphone/tablet clients, they also announced they'd continue supporting existing customers. It's worth noting that Google is an active client - the Google Glasses they released this year are running w/ OMAP 4430s (and you can see the glass kernel branches are active on the Android side). Not to put too fine a point on it, but your post is completely unsupported by the (easily discoverable - it only took like 2min of searching to dig the above up) facts. I doubt that anyone from Google is going to step forward with their reasoning (although, in the spirit of things, I'd be happy to be proven wrong), but I'd assume it was a pretty straight forward cost/benefit analysis. There aren't going to be new OMAP-based phones/tablets, and the number of active GNex's running stock is probably miniscule (tens of thousands?), so it just wasn't worth doing.