I was hassled in Austin. During waiting an extra 30 minutes, a heckler TSA agent come by asking if I wanted the full massage treatment & if I enjoyed opting for the pat-down. No problems in SF though.
Mainframes failure is in handling the volatility of today's technology world. With nothing resell-able, cheaper costs are meaningless versus piecemeal purchasing of ever-better hardware.
Others have great ideas too, but how about GIT for everything. - Research in GIT:
Where's the data, how was it processed? - Politics in GIT:
Have a commitment? Put it in your TODO. Achieved it? Mark as done.
No one will vote for a claim/plan unless you register it to be held accountable. - Bills in GIT: If a law changes, who did it, how when? - Laws as Unit Tests: Unit tests show the behaviors & results. To start a business, follow the start_business unit test.
These tests could be "played with" to determine what is legal & illegal.
If most of Android became APK files (except kernel, driver, root-land & a few other pieces) then carriers would have far less work to push out these updates. The updates would happen automatically like for apps. This may require dependency logic: 'This app requires that you update libAndroid.apk". It works for Chrome.apk & could be used for nearly all Java libs.
Though not a complete solution, it would resemble part of the solution the Linux distros use. It would considerably reduce surface area to attack (an important goal). As a side benefit shared libs would make security easier for everyone: lib developers (as a first-class APK), app developers (look to lib developers to fix their bugs), carriers & users (less data to transfer).
I expect differences: - The GL part is the same, the renderer side is not. - The input subsystem is different. - Everything's Asynchronous by default - Daniel Stone testing Chrome startup showed that 497ms was due to just waiting for X responses (rendering & input). - Fewer context switches. Less message passing (since WM & rendering are 1 process). - Multiple GPUs for rendering are exposed to user-space
About Drivers: - With Android drivers supported, Games can run on GPUs they couldn't have before.
Watch that video. The only benefits lost are Motif compatibility & similar tools that no-one cares to update. Remoting will be faster with Wayland. It'll be rooted or rootless, user's choice.
I see "X11 => Wayland" like "x86 => ARM". Progress requires breaking compatibility sometimes when old assumptions are getting you down. It doesn't mean that everyone must change.
These aren't published formats. Everything is guesswork, so you can't be 100% compatible. And MS raises the bar instead of publishing anything helpful. So no, you'll never get MS Office XLS/X 100% compatible or MS Outlook 100% (what does that mean anyway, Exchange compatible?). Go with other standards like OpenDocument, PDF, etc and Groupware*(zimbra, open x-change).
To run a C++ (native) app on Android hardware (because that hardware is cheap to buy) supported by only closed drivers: - Using Android's SurfaceFlinger API means your apps use Dalvik, fail. - SurfaceFlinger is a bit basic to extend. - X11 isn't getting ported, Difficult? - Wayland shows more promise, (and sends fewer network packets with FreeRDP than X forwarding). - Ubuntu Mir is trying something similar.
Correction: They ported (wrote a new backend for) Wayland to sit atop Android drivers (which will sit atop Android hardware).
Your conclusions are right: Linux apps (on toolkits with Wayland ports) can now run on this. This has nothing to do with running Android apps. Though getting a great display manager on Android hardware is a good first step if that's where you're going. You'd still need the whole Android subsystem with adapters (AudioFlinger APIs in Alsa or whatever sound subsystem they use, SurfaceFlinger APIs on Wayland).
Good Multi-monitor support lets docked tablets equal usability to desktops: Big monitor, keyboard, mouse, always has power, etc. That's the big feature Android could get to start picking-up Desktop work applications. Ubuntu already has those desktop apps & multi-monitor UIs (not support yet), so when their port is done, a docked tablet will be a very Desktop-like experience.
There will always be a "best place to be seen" for inventors & developers, which will make that place a developing market. Desktops miss out on voice dictation since the energy to get it right went to the tablet area.
Cloud providers are shrinking Corporate IT. That makes the corporate market more closely resemble the public. The "bring your own device" era is arriving.
The source of the valuable ingredients control the product. In cars, you've got the Ford Motor Co. & other similar companies that wrap a car around the most valuable part that they built. Similarly, the best deal has always been products from the same company (like Samsung for Android phones).
Licensing their apps for use in the biggest phone OS (which they control) probably makes some money. Their GMail for business is taking off & so is their search appliance (for inside companies). Their data centers are cheaper per-seat than anyone. There's no lack of innovation. Businesses are warming up to things like Google Docs over MS Office + Sharepoint. They try many things (given for free), improve some, then work-out marketability (or axe). Selling mostly to businesses hides this income from public view.
Yes & (compression + drive tech) keeps up. But I save movies I will watch again because they're expertly-made or personally-made. My young kids broke 2 tablets already. It would have been worse if they were managing DVDs in a player w/o backup.
I get that they're showing both ends of the spectrum: a "Hard to get broadband" area, & a high-tech, high-saturation area. I just hope they don't need to dig much to install that fiber. Austin is on some solid stone & can take weeks to cut a hole big enough for a swimming pool.
Done wasting disk space, memory, copy time, & boot time for VMs? Push for LXC and get already-on "VMs" with software already installed. You're limited to no reboots & 1 kernel, but system administration happens for everyone by the system maintainer. Then "fatware" distros are a feature. You can skip virtualized filesystems with per-user home directories (and sensible browsing restrictions) if that fits your needs. It only requires hooks into bringup/shutdown since there's no live migration yet.
I'm using my last working CD/DVD drive ( & VHS with adapter) to put the media in digital formats. Then I make movie mixes (to tablets for the kids), and MP3 mixes easily. Sure, use legacy equipment if you have it & know it, but as it dies, replace it with modern solutions.
Slashdot Mirror
I was hassled in Austin. During waiting an extra 30 minutes, a heckler TSA agent come by asking if I wanted the full massage treatment & if I enjoyed opting for the pat-down. No problems in SF though.
Mainframes failure is in handling the volatility of today's technology world. With nothing resell-able, cheaper costs are meaningless versus piecemeal purchasing of ever-better hardware.
Except children's stuff. They can go on forever there because no tired old meme is so to a 4-year-old.
They're our representatives, not our leaders. This is a democracy, not militaristic rule.
& I was thinking "...on a tablet" patents.
Others have great ideas too, but how about GIT for everything.
- Research in GIT:
Where's the data, how was it processed?
- Politics in GIT:
Have a commitment? Put it in your TODO. Achieved it? Mark as done.
No one will vote for a claim/plan unless you register it to be held accountable.
- Bills in GIT: If a law changes, who did it, how when?
- Laws as Unit Tests: Unit tests show the behaviors & results. To start a business, follow the start_business unit test.
These tests could be "played with" to determine what is legal & illegal.
If most of Android became APK files (except kernel, driver, root-land & a few other pieces) then carriers would have far less work to push out these updates. The updates would happen automatically like for apps. This may require dependency logic: 'This app requires that you update libAndroid.apk". It works for Chrome.apk & could be used for nearly all Java libs.
Though not a complete solution, it would resemble part of the solution the Linux distros use. It would considerably reduce surface area to attack (an important goal). As a side benefit shared libs would make security easier for everyone: lib developers (as a first-class APK), app developers (look to lib developers to fix their bugs), carriers & users (less data to transfer).
I expect differences:
- The GL part is the same, the renderer side is not.
- The input subsystem is different.
- Everything's Asynchronous by default
- Daniel Stone testing Chrome startup showed that 497ms was due to just waiting for X responses (rendering & input).
- Fewer context switches. Less message passing (since WM & rendering are 1 process).
- Multiple GPUs for rendering are exposed to user-space
About Drivers:
- With Android drivers supported, Games can run on GPUs they couldn't have before.
Watch that video. The only benefits lost are Motif compatibility & similar tools that no-one cares to update. Remoting will be faster with Wayland. It'll be rooted or rootless, user's choice.
I see "X11 => Wayland" like "x86 => ARM". Progress requires breaking compatibility sometimes when old assumptions are getting you down. It doesn't mean that everyone must change.
These aren't published formats. Everything is guesswork, so you can't be 100% compatible. And MS raises the bar instead of publishing anything helpful. So no, you'll never get MS Office XLS/X 100% compatible or MS Outlook 100% (what does that mean anyway, Exchange compatible?). Go with other standards like OpenDocument, PDF, etc and Groupware*(zimbra, open x-change).
*from osalt.com
To run a C++ (native) app on Android hardware (because that hardware is cheap to buy) supported by only closed drivers:
- Using Android's SurfaceFlinger API means your apps use Dalvik, fail.
- SurfaceFlinger is a bit basic to extend.
- X11 isn't getting ported, Difficult?
- Wayland shows more promise, (and sends fewer network packets with FreeRDP than X forwarding).
- Ubuntu Mir is trying something similar.
Sailfish doesn't run Android apps. It runs Qt-based apps (KDE apps). It now can run them on Android hardware.
Correction:
They ported (wrote a new backend for) Wayland to sit atop Android drivers (which will sit atop Android hardware).
Your conclusions are right: Linux apps (on toolkits with Wayland ports) can now run on this.
This has nothing to do with running Android apps. Though getting a great display manager on Android hardware is a good first step if that's where you're going. You'd still need the whole Android subsystem with adapters (AudioFlinger APIs in Alsa or whatever sound subsystem they use, SurfaceFlinger APIs on Wayland).
Good Multi-monitor support lets docked tablets equal usability to desktops: Big monitor, keyboard, mouse, always has power, etc. That's the big feature Android could get to start picking-up Desktop work applications. Ubuntu already has those desktop apps & multi-monitor UIs (not support yet), so when their port is done, a docked tablet will be a very Desktop-like experience.
There will always be a "best place to be seen" for inventors & developers, which will make that place a developing market. Desktops miss out on voice dictation since the energy to get it right went to the tablet area.
Cloud providers are shrinking Corporate IT. That makes the corporate market more closely resemble the public. The "bring your own device" era is arriving.
So one of the best encryption methods to avoid snooping is turning a message into a large captcha? That's easy enough.
The source of the valuable ingredients control the product. In cars, you've got the Ford Motor Co. & other similar companies that wrap a car around the most valuable part that they built. Similarly, the best deal has always been products from the same company (like Samsung for Android phones).
"they .. benefit the most from a broken system"
So they will defend attempts to change it. And they'll be able to fund such defense.
Licensing their apps for use in the biggest phone OS (which they control) probably makes some money.
Their GMail for business is taking off & so is their search appliance (for inside companies).
Their data centers are cheaper per-seat than anyone. There's no lack of innovation.
Businesses are warming up to things like Google Docs over MS Office + Sharepoint.
They try many things (given for free), improve some, then work-out marketability (or axe). Selling mostly to businesses hides this income from public view.
I'm very interested to know how "moralism" as a government/society form would work. I think they thought they were getting that with communism.
Yes & (compression + drive tech) keeps up. But I save movies I will watch again because they're expertly-made or personally-made.
My young kids broke 2 tablets already. It would have been worse if they were managing DVDs in a player w/o backup.
I get that they're showing both ends of the spectrum: a "Hard to get broadband" area, & a high-tech, high-saturation area.
I just hope they don't need to dig much to install that fiber. Austin is on some solid stone & can take weeks to cut a hole big enough for a swimming pool.
Done wasting disk space, memory, copy time, & boot time for VMs?
Push for LXC and get already-on "VMs" with software already installed. You're limited to no reboots & 1 kernel, but system administration happens for everyone by the system maintainer. Then "fatware" distros are a feature.
You can skip virtualized filesystems with per-user home directories (and sensible browsing restrictions) if that fits your needs.
It only requires hooks into bringup/shutdown since there's no live migration yet.
I'm using my last working CD/DVD drive ( & VHS with adapter) to put the media in digital formats. Then I make movie mixes (to tablets for the kids), and MP3 mixes easily. Sure, use legacy equipment if you have it & know it, but as it dies, replace it with modern solutions.