Pixel C (opens in new tab)

It's really cause no one wrote any Windows apps, and a lot of the ones that were written have been removed because the respective companies don't want to maintain them. The Kindle reader for my Windows Mobile phone was sooooo buggy.

If it had the apps to back it, Windows would have been okay as a touch OS. But that whole thing totally failed.

Microsoft should have put the Courrier table into production. It was ready to go. That one move totally killed any chance at Microsoft making it into the tablet realm (and really set them back 2 ~ 3 years as a consumer tech company)

In Windows 10 tablet mode the Surface is a great tablet. Once I got the Surface pro 4 its completely displaced ipad usage for me.

What you describe is an "issue" I have with the Surface Pro 3: I always want to just use the physical keyboard. I eventually put Linux on it and I now use it as a normal laptop, using the touch input only occasionally (admittedly, support for touch within GNOME is worse than even Metro's efforts).

These hybrid touch devices that have "optional" keyboards can't seem to decide what they want to be, and I don't think this is actually a software problem in the case of Windows devices which seem capable enough in either "mode." Rather, I think if people are actually given the option of using a keyboard and a touchpad to do any non-trivial task, I expect most of them will do so. At which point... how is this tablet with a keyboard attached to it worth owning when one can get a small laptop?

Even at the height of my "Give Metro a Chance" phase with the SP3, relying exclusively on touch was at best a kind of awkward stopgap until I could position myself to use the physical keyboard. Sticking with touch only makes nothing easier, except in situations where your interactions with the device are limited by your physical surroundings.

> Chromebooks, and thus presumably the Pixel C, ship a fully functional ssh client.

The Pixel C is an Android tablet, not a Chromebook (unlike the Chromebook Pixel which is, as the name suggests, a Chromebook.)

Doing work over SSH isn't something I enjoy doing as developer.

I side with Xerox PARC in how a software development environment should look like.

I know many like to work like on a PDP-11, but for me that is only for time travel or server administration purposes.

I had my share of teletypes since the 80's, plus remote work on the go while traveling across Europe isn't something that I would recommend.

My personal definition of full blown OS is something that considers me enough of an adult that it allows a file explorer / remote shell based file sync rather than use of a specific sync app or MTP.

Not really? I'm not sure what would be ambiguous about it.

Thing is that Android apps, when properly made, should be able to go from phone to tablet without existing in different versions.

That at least was the idea back when they introduced the Fragments framework. there they took the existing activities concept and turned them into panes in a larger UI. Sadly the world quickly turned that into that "hamburger" bar.

Nor did it help that Apple, via the iPad launch, had gotten app devs to think about phones and tablets are separate spheres. Thus for a while there was "HD" apps popping up in the Play store that had the whole Fragments thing going, right alongside existing phone apps.

And then Google kinda abandoned the tablet world for a while, instead trying to pitch ChromeOS and Chromebooks as a corporate platform.

That didn't catch on so now they seem to be returning to Android tablets after doing some work on various things like multiple user accounts, and Android for Work that allows someone to have one app state for work and one for personal use on the same device.

Well, it doesn’t help that Google’s support libraries for developers are at best... bad looking. At worst, they are crashing with no way to find out what caused it.

1. AppCompat used to crash on device rotation with no clear error message on KitKat.

2. There are many third-party libraries that expand on what Google provides to allow devs to actually provide good-looking Material apps, from MaterialDialogs to MaterialDrawer to MaterialTabs.

- https://github.com/afollestad/material-dialogs

- https://github.com/mikepenz/MaterialDrawer

- https://github.com/florent37/MaterialViewPager

Many things, like a multi-pane preferences activity with AppBar, are not possible without hacks, reflection or access to internal com.android APIs either.

And when Google provides support for something, it’s often lacking features, for example: https://www.reddit.com/r/androiddev/comments/3vwpna/im_incen... or full of bugs http://stackoverflow.com/questions/30537413/textinputlayout-...

> Worse, even if Google fixes the problems in the OS, there are (still!) virtually no 3rd-party apps that support tablet Android natively; they're all just blown-up phone apps. Not only does that ruin the experience but-- like Microsoft and their stalled phone ambitions-- it's not a problem Google can fix no matter how good their developers are.

Actually, with sufficient number of sufficient-quality developers, its a problem Google can solve fairly directly, by developing their own, compelling, Android tablet apps.

OTOH, I think that this is a problem that is overstated; Sure, there's lost of Android apps that are just "blown up" phone apps, but there's a lot of apps where there's no compelling need for something more, and "blown up" to tablet size improves interaction when you have a tablet available.

Its also not really like like Microsoft's problem.

> Not only does that ruin the experience but-- like Microsoft and their stalled phone ambitions-- it's not a problem Google can fix no matter how good their developers are. All the software patches in the world can't fix the issue of apathetic 3rd-party devs.

Hell, Google isn't even doing that much. Half of their apps don't have proper tablet UIs. How can third parties bother to do decent tablet apps if even the company who develops the platform doesn't care enough to do so? Google is in a leadership position, which means they have a responsibility to lead, and not only did they drop the ball, they threw an interception straight to Microsoft.

It gets worse: Google's newest, most Material apps, like Inbox, have no tablet layouts. They are "blown up phone apps." Especially ironic (in an Alanis way, maybe?) because GMail is an exemplary app that shifts layout decisions based on screen size and orientation.

That said, Drive Docs, Slides, and Sheets are very nice on larger tablets. Still, Google needs to raise awareness that Android's APIs have very nice support for larger screens.

What's the actual real world experience of using Linux on an Android tablet though? I've looked into it briefly and you can do things in a chroot environment which has some limitations, or run a more vanilla Linux distribution that very likely has none of the drivers, power optimizations, etc. needed to make a usable mobile machine. It just seems like a horrible experience but I'd love to be proven wrong.

Also note that the 6.0.1 update for Android introduced a slightly different tablet UI again.

Since 4.3-4.4, tablets have sported a bottom bar with 3 buttons centered. Now two of them (back and home) are found in the bottom left corner, while the "switcher" button has landed in the bottom right.

While this seem to be nice when holding a tablet with both hands, it also works well with "Fitts' law" as corners are easy to hit with a mouse.

Lack of software. Or at least lack of high-quality software for getting real work done. I had an Android tablet for a while. It was nicely portable, had good battery life, could do a lot of things well, but I just couldn't find any apps I liked for two things I do often: ssh and creating presentations. One ssh app could do tunnels, but got all confused if I tried to open multiple terminal sessions. Another handled multiple sessions OK, but couldn't do tunnels. Then the two would interfere with one another if I tried to run both simultaneously. For presentations, many couldn't do something as basic as keep track of where my cursor was, and would add text at the wrong place. Others at least worked, but couldn't do anything more complex than a bullet list. A real email app would have been nice, but I had to use both home and work webmail (which are very different) instead. From what I can tell, all of this has only gotten a tiny bit better in the last couple of years. It would be great if there were more Android developers coming from the desktop space instead of the phone space, producing more desktop-like apps, but that doesn't seem to be the case. As long as that continues to be the case, using Android on one's primary machine will continue to be frustrating.

ChromeOS, on the other hand, seems to be in better shape. I'm seriously debating with myself about whether to get a regular Pixel (not Pixel C) as my next main computer.

Can't speak for GP, but it's an embarassment that in 2015, Android still doesn't have built-in multi-window support. For crying out loud, even Apple beat Google to this, and they're typically behind when it comes to these things.

Here: http://imgur.com/a/rpQqP

Yeah, also the original IBM PC which inspired a lot of complaints, replacements, and a change when the IBM PC AT arrived.

Gnome was exploring touch too. For my use case, though, a window manager is superfluous, as I'd use this machine as my portable emacs. At 150 bucks it's way cheaper than any other option. All I need is Emacs and Firefox anyway.