Discliaimer: I work for Google, but don't know about real strategy decisions about this.
I think it's all about establishing more demand for web apps.
Already I can do everything I need to in Chrome except for a nice development environment, though Cloud9 is getting closer, and music apps like Logic. Those are quite doable though, especially with the Chrome App APIs, someone just needs to get the ball rolling.
The current crop of Chromebooks were really inexpensive and nice for the money, but they weren't going to appeal to the cutting edge tech users who will loudly demand and use better web apps. Chrome needs something that inspires, and I think the Pixel will do that. I've used it for a while now and the hardware is incredibly nice. It'll also create space between the $200-300 Chromebooks and the Pixel for some decent mid-range devices.
I can't imagine there are that many people who want to spend $1,300 to use web apps. People use GMail, Google Docs, etc, because they're cheap, not because they're good. That's why the $250 Chromebooks sell any units--because they're cheap. This thing takes the "cheap" aspect out of the equation, which seems like a mind-blowingly stupid decision.
I use gmail and google docs because they're good. Why else would I use them? Cheap? There's plenty of free email programs out there. None compare to gmail. I'll agree that when I need to get really down and dirty with a spreadsheet, I whip out Excel, not google spreadsheets, but for most of my "lists that need some numbers to add/multiply together", google spreadsheets work great and are always available from wherever I am.
Which is not to say I think you're wrong. No one's going to spend this much money on something so limiting. If we were 10 years in the future and we all had gigabit network speeds, and there were enough awesome web apps to replace the hard working desktop apps (photoshop, IDEs, CAD, etc)... then yes.... but I would still hope the specs would be better for the price. 4GB of RAM in a $1300 machine? What is this, 2005?
IMHO, all it needs for a decent developer environment is semi-sanctioned shell access and a package manager. I've been using the age-old SSH+tmux combo fine, all that's really hurting me is latency. I'm kind of surprised Google hasn't taken up the task of building their own web-app IDE (but not that surprised, if what I keep hearing about Google not allowing code checked out on laptops is correct it's hard to improve much on SSH).
I want this machine to work, I just don't see how it can without being self-hosting.
Well said, that's my thought as well. If you look at the complaints in this thread (lack of VMs, lack of Photoshop equivalents, etc.) won't get created until there is compelling hardware capable of running them. We're already seeing an explosion of web apps to fill the need of regular Chromebook users (stuff like RDP clients are a hot item in the Web Store these days), but it will take something more capable before we get, for example, good WebGL games.
I think it's all about establishing more demand for web apps.
Already I can do everything I need to in Chrome except for a nice development environment, though Cloud9 is getting closer, and music apps like Logic. Those are quite doable though, especially with the Chrome App APIs, someone just needs to get the ball rolling.
The current crop of Chromebooks were really inexpensive and nice for the money, but they weren't going to appeal to the cutting edge tech users who will loudly demand and use better web apps. Chrome needs something that inspires, and I think the Pixel will do that. I've used it for a while now and the hardware is incredibly nice. It'll also create space between the $200-300 Chromebooks and the Pixel for some decent mid-range devices.