Hacker News Reader

Comment by frik | Hacker News Reader

I imagine in near future, we will use smartphones as our personal computer in every situation. In the office, one can then interact with a Bluetooth keyboard, mouse/touch pad and use a big monitor. Probably, one of the next iOS will come with a macOS UI as well (under the hood the iOS is based on macOS anyway) - so if you connect it to a huge monitor, you have the desktop UI of macOS, and otherwise the iOS UI. The same will happen with Android/ChromeOS/Google Fuchsia, Ubuntu and MS has some ideas too (but they cannot execute it, forcing mobile UI on desktop failed big, now they bet on Win32 again to fill their empty AppStore).

View on HN

The Andromium guys are trying to do that right now. They had a Kickstarter for a dock for your phone that does this for you, but it failed a few years ago. Just a month ago they tried the same thing with a very streamlined "laptop-like shell", and that Kickstarter[2] succeeded. I am one of the backers and I am quite excited about it.

[2] Non-referral link: https://www.kickstarter.com/projects/andromium/the-superbook...

The idea that we'll want to plug our smartphones into a dock and use them as desktops reflects really poorly on modern operating systems. The driver for this seems to be so that settings and files are always applied and accessible in every context. Sharing data between machines on a local network requires a lot of setup, is easily broken and is never seamless, but I don't think is has to be like that.

Personally, I want the opposite of "settings and files ... accessible in every context"—I don't want my personal data or configuration touching a corporate network, or vice-versa. I want the "bring your own machine" philosophy to be taken to its fullest extent: where there's no such thing as a workstation any more, because everyone just owns a PC (their phone) that they bring to work, which maybe contains a sandbox/VM with their work stuff in it, or acts as a thin client for a corporate shell-server, but is ultimately their machine.

This enables you to e.g. set up your own VPN for host traffic, to securely access whatever you want during work hours, completely opaquely to packet-sniffing on the corporate network. The "work sandbox" can ignore that VPN and just use the corporate network (maybe through its own VPN) like normal. Whereas with a shared workstation + LAN, all your personal data and traffic and ability-to-use-software is completely at the behest of corporate IT policies.

Oddly enough, this is already pretty much what the experience is like for iOS with a MaaS360-managed profile installed and Exchange ActiveSync-based accounts set up. Translate that to a desktop UI while keeping the isolation metaphor and it'd be perfect.

The TDP limits of smartphones (no more than ~10W) make them unuseable for heavier workstation loads, and there will always be demand for a lot of local computational power.

Most people don't run workstation loads, let alone anything remotely resembling "heavy".

The parent comment was talking about "every situation", not "most people".

I feel (hope :-)) that in the future, network bandwidth will be large enough that large, high-power general computation services can be used to offload intense processing tasks from personal devices to work around things like this.

So, if you're on your phone and need to crunch a massive data-set, this gets offloaded to the big-datacruncher-in-the-cloud, and the result sent back to your device.

Bandwidth is only one part of the problem. Latency is another one, and it seems to be insurmountable with the modern network media.

I hope the closed iOS model with no option to choose between default apps doesn't become "desktop OS". I mean kudos for them for the HW achievement, but iOS is "dumbed" down as it gets.

I have been wondering on the same thing. 90% of people using the desktop or Laptop now are basically the same as they are doing on phone. Social Media, Youtube, Web Surfing etc. Even some heavy lifting like Video Editing can be done on a Phone. In terms of computational power, if you add two more Core ( Our current Desktop CPU gets 4 thread from 2 Core ), and double the memory, there is nothing much left apart from the UI difference.

How far are we from that? 2 - 3 years? Where will iPad fit into it?

As far as I know Ubuntu and Windows are already doing this. But as you said without much luck.

I also think this is the future. But not only for the switch to desktop. Also as car console, key store, tv remote, synthesizer (just put it in a midi device) and so on.

You personal key to everything. And your single attack vector...

My phone catches on fire whenever I open up chrome. How will it handle multiple IDEs? Would be interested to hear from a hardware person if cooling is a hard or soft bottleneck for something like this...

You have major phone producers removing 3.5mm jacks to save space, by the aame attitude I think it's safe to assume no one will add a "thick" cooling system.

Reply on news.ycombinator.com