"Being open source is killing Android"

[user1397]

From http://www.zdnet.com/article/being-open-source-is-killing-android/

Interesting article, what do you all think?

[RichardET]

From http://www.zdnet.com/article/being-open-source-is-killing-android/

Interesting article, what do you all think?

I have not had an android phone in several years because of these issues. Linux is better off the way it is.

[ventrical]

Using android 4.4.2 here. No problems whatsoever. Two of my tablets have ubuntu kernels (RCA Voyager). With the implementation of ubuntu snappy core any current problems in security will be solved. It's just a train that has to get rolling. :)

Regards..

[montag dp]

Android is based on the Linux kernel, which is GPL. How do you go about just making that proprietary?

[grahammechanical]

So, what is the big mess that Android is in?


That then results in both the fragmentation (OEMs loading the code onto any and every device form factor they can think of), and the problems with updates (Google can't push Android direct to devices because heavens knows what modifications and tinkering have been done, both cosmetic and structural, to the code).

That is not going to happen with Ubuntu mobile devices. Things are being done differently. All Google needs to do to avoid fragmentation and the delay in updating mobile devices with newer versions of the OS is to imitate Canonical.


Someone, somewhere has to be in charge, and put the interests of the platform over profit margins and market share.

And is Google going to be that "someone?" Does Apple put the interests of the platform over profit margins and market share?

It may seem like a nonsense question, but if Android had been proprietary from the beginning would it have now have such a massive market share? Or would Android devices be more on a level with Windows phones for popularity?

I am not saying that people brought Android phones because they believed in FOSS but I do think that some of the success of Android comes from it being FOSS.

Regards.

[Dragonbite]

From http://www.zdnet.com/article/being-open-source-is-killing-android/

Interesting article, what do you all think?

So much misinformation being fed by blind hatred.

I mean, look at Microsoft... they are proprietary and in charge of their few handsets and it's worked great for them! Hasn't it?

[ventrical]

So, what is the big mess that Android is in?



Yeah .. thats what I'm still trying to figure out.



I am not saying that people brought Android phones because they believed in FOSS but I do think that some of the success of Android comes from it being FOSS.

Regards.

All very well said. I would like to add also that 'linux' played a big part of it and Canonical will no doubt amass the lion's share with snappy.

regards..

[buzzingrobot]

Fragmentation and slow updates are not an open source problem.

And Android is not "all about being open". Android is all about making money.

Fragmentation happens because a zillion and one companies see a profitable market and jump in.

Besides, why are all those Android brands a problem?

Slow updates are a *vendor* problem. I have a phone and a tablet from Google and they both see regular monthly updates and the occasional "emergency" patch.

Folks who have been victimized by slow updates might consider *not* buying that brand again. I'm afraid, though, that most consumers aren't paying that much attention to updates on way or another.

In any case, the best response to this alleged problem is not for Google to take over and become the Apple of the Android world.

[ventrical]

In any case, the best response to this alleged problem is not for Google to take over and become the Apple of the Android world.

+1

Then Android would be part of the elitist clique. It is totally against the fundamental philosophy of FOSS. I have two RCA Voyager 7s. When I see Kernel Version 3.4.67 root@ubuntu-018#1, I really feel like I own my tablet, and that my tablet doesn't own me.

Regards..

[bearlake]

+1

Then Android would be part of the elitist clique. It is totally against the fundamental philosophy of FOSS. I have two RCA Voyager 7s. When I see Kernel Version 3.4.67 root@ubuntu-018#1, I really feel like I own my tablet, and that my tablet doesn't own me.

Regards..

So you are saying you rooted a few RCA Voyager and added ubuntu.

Is there a how to for this?

Regards

[ventrical]

So you are saying you rooted a few RCA Voyager and added ubuntu.

Is there a how to for this?

Regards

No. I am saying I bought them this way . The kernel was installed OEM. It is android running on an ubuntu kernel.

I did root them to experiment with ubuntu "vivid' touch but it was near impossible to get them to work as they were not recognized correctly with adb - but that is another story. However , there is an app called Linux Deploy which works with VNC to be able to run ubuntu desktops on top of android.

regards..

[bearlake]

No. I am saying I bought them this way . The kernel was installed OEM. It is android running on an ubuntu kernel.

I did root them to experiment with ubuntu "vivid' touch but it was near impossible to get them to work as they were not recognized correctly with adb - but that is another story. However , there is an app called Linux Deploy which works with VNC to be able to run ubuntu desktops on top of android.

regards..

Sounds better.

Many thanks for info.

Regards

[ventrical]

Sounds better.

Many thanks for info.

Regards

Just for the record ..

http://ubuntuforums.org/showthread.php?t=2254725&highlight=rca+voyager

Regards..

[montag dp]

Fragmentation and slow updates are not an open source problem.

And Android is not "all about being open". Android is all about making money.

Fragmentation happens because a zillion and one companies see a profitable market and jump in.

Besides, why are all those Android brands a problem?

Slow updates are a *vendor* problem. I have a phone and a tablet from Google and they both see regular monthly updates and the occasional "emergency" patch.

Folks who have been victimized by slow updates might consider *not* buying that brand again. I'm afraid, though, that most consumers aren't paying that much attention to updates on way or another.

In any case, the best response to this alleged problem is not for Google to take over and become the Apple of the Android world.I think the point is that many users wouldn't realize that the fault lies with the vendor instead of Google. Actually, most probably don't even realize they are running a modified version of Android. I agree with others that this is not really a big problem for Google, though. I think the open-source nature is actually very attractive for vendors for the very reason that they can customize it as desired, and that means more market share for Google.

[Warpnow]

Didn't android edge out windows this year and is now the most used operating system on any platform for any technology?

Yeah, its really struggling...

The weaknesses are also strengths. If google locked down android more, companies like samsung might try to make their own operating system instead of selling amazing android devices.

Updates also cause problems. Some users would rather not take the risk the new OS will break their phone. You also can't expect companies selling $20 android phones to offer a ton of support after sale. We have to be realistic about that. They'd rather release a new phone than make updates to a no longer profitable platform for them. People who want better support and updates have better, more expensive options. In addition, Huawei or whoever can release a crappy android phone and it doesn't really reflect on google. The same OS can power a $20 phone and a $600 phone without anyone questioning the value. Apple can't do this-- if they release a low cost product it affects their other product lines.

Google is simply much, much smarter than the person who wrote the article...

[T.J.]

From http://www.zdnet.com/article/being-open-source-is-killing-android/

Interesting article, what do you all think?

If I might speak plainly, after 20+ years in IT and development, I consider it bunkum. Everyone is so overly concerned with software fragmentation. The truth is that homogeneous environments are dangerous. Fragmentation is healthy. Fragmentation is good for users. It promotes competition to deliver a better product at a cheaper price.

Without framentation, computers are dangerous to businesses, to users, to mission critical systems, and so many other things. For proof, you need look no further than Windows. Because every copy of Windows is internally very much the same:

You have malware and viruses.
You have a thriving business that sells software to users that they do not need. Symantec (the company that pioneered the concept) admits AV is "dead." It offers very little in real safety for the costs of users. Professionals have known this for YEARS. We simply wipe the device, as it is far more efficient and guaranteed to destroy a virus - while AV cannot offer the same guarantee. Another example is registry cleaners. When you have the same environment, you have hucksters selling all manner of trash.
Homogeneous environments stifle change, even when it is to everyone's benefit. Example: in the 1990s Alpha RISC processors were more efficient than Intel's, yet Intel won out because it had Microsoft's help. Intel eventually got the Alpha patents and buried them in favor of Intel's design. Now RISC technology similar to Alpha powers the world's most powerful supercomputer - built in China after the U.S. government denied them access to Intel's market hardware.
It removes security. Homogeneous software like Windows allow for governments to exploit security weaknesses for their own benefit on a massive scale. The NSA does this all the time. So do hackers.


Hardware is a counter example as to why fragmentation is a good thing: the fact that PC hardware is so diverse is one of its strengths. Not only are replacement parts cheaper for users, but there are far fewer hardware attacks compared to software.


So how do you provide needed unity in the face of fragmentation? You depend on open standards. Data standards like VCard allow users to transfer data between systems. TCP/IP allowed for network interconnection that we now know as the Internet while using whatever hardware you had on hand that was cost effective.

Programmers have depended on standards for years. A real world example: C and POSIX are standardized and those tools allow for software to be ported to systems even if they are not the same. Furthermore, because they are standardized by international agreement, the costs developers pass on to customers are minimal when it comes time to port software to a new system. Linux has prospered because of open standards like C and POSIX. Because of open standards, Google was able to create Android in the first place.

No, I believe that fragmentation combined with open standards is a GREAT thing! This writer is little more than a standard-bearer for a way of doing business that leads to higher costs, less choice, and less security for users.

[ventrical]

If I might speak plainly, after 20+ years in IT and development, I consider it bunkum. Everyone is so overly concerned with software fragmentation. The truth is that homogeneous environments are dangerous. Fragmentation is healthy. Fragmentation is good for users. It promotes competition to deliver a better product at a cheaper price.

Without framentation, computers are dangerous to businesses, to users, to mission critical systems, and so many other things. For proof, you need look no further than Windows. Because every copy of Windows is internally very much the same:

You have malware and viruses.
You have a thriving business that sells software to users that they do not need. Symantec (the company that pioneered the concept) admits AV is "dead." It offers very little in real safety for the costs of users. Professionals have known this for YEARS. We simply wipe the device, as it is far more efficient and guaranteed to destroy a virus - while AV cannot offer the same guarantee. Another example is registry cleaners. When you have the same environment, you have hucksters selling all manner of trash.



Windows has always been the pitri dish to allow malware to grow. They have been the de facto service provider to facilitate security software suites and still is and will continue to provide the hive base for a plethora of crapware . But this is not all bad because it forces open source to become better!. In my own personal experience for over a 1/4 century it has caused me to come full circle, at one point testing Linus Torvalds' first bootable floppy fail to now , running firefox in an lxc container on xmir in the unity8 DE (development cycle). In between that time I marshalled most of my efforts cleaning crapware from other peoples PCs. If there is a hell within the matrix then that is it.

With open source there is a freedom like no other. Unfortunately bad habits die hard and detoxing from Windows & other commercial concepts can take a very long time. There is a paradigm shift taking place and ubuntu/opensource will come up in the light on the flip side.

Regards..

[grahammechanical]

If Linux was not open source then Google would not have been able to base its Android kernel on Linux. And Canonical would not have been able to use the Android kernel in Ubuntu phones.

That is not to say that Google did not have the money to pay enough software engineers to write their own proprietary OS. They just did not need to. They bought Android.inc instead. I wonder if the business managers at Google knew they were buying something covered by a FOSS licence? Or did they get an unpleasant surprise? :)

Regards.

[sam-c]

you have heard of convergence!? which is in the users interest so there is no difference

[T.J.]

In between that time I marshalled most of my efforts cleaning crapware from other peoples PCs. If there is a hell within the matrix then that is it..

My sympathies. I refuse to submit myself that form of torture any longer. If someone comes to me with an infected machine, it gets wiped and reinstalled from an image (if possible), I refuse to waste 2-6 hours worth of electricity and then have the scan fail.

Not to sound nasty, but I've better things to do than coddle Windows users, who do not understand that their OS is a badly designed cesspool without proper privilege separation. To be fair, I have an even lower opinion of Apple for other reasons.



There is a paradigm shift taking place and ubuntu/opensource will come up in the light on the flip side.


Even Microsoft realizes that. Project Roslyn made me laugh. Some said it would never happen, but Microsoft really has no choice now.

[ventrical]

My sympathies. I refuse to submit myself that form of torture any longer. If someone comes to me with an infected machine, it gets wiped and reinstalled from an image (if possible), I refuse to waste 2-6 hours worth of electricity and then have the scan fail.

Not to sound nasty, but I've better things to do than coddle Windows users, who do not understand that their OS is a badly designed cesspool without proper privilege separation. To be fair, I have an even lower opinion of Apple for other reasons.



Even Microsoft realizes that. Project Roslyn made me laugh. Some said it would never happen, but Microsoft really has no choice now.


I empathize with your honest anger here. Like second hand smoke, just hearing of some person infected with a Windows machine makes me physically ill but it is part of the Instructional Development mindset that I try to deploy as a migration assistant for 'spinners' who are detoxing from Windows. The mis-information and dis-information are like a plague. Some of the new stories I hear each day are epic and near make me weep. Having pity is no good and so I try to be firm in my pitch to instruct potential converts to linux/ubuntu or other form of linux-os. "You have to get with the program and learn the learning curve. "

Having said this; there will always be security vulnerabilities and persons trying to exploit them and we still (I still) have to be vigilant and realistic, keep my teeth and claws sharp so when I am trying to assist in a migration I can talk the malware crap talk with potential ubuntu noobs and follow up with them. With ubuntu , the old axiom holds true: "we have to give it away to keep it". eh ? New converts have to be detoxed from malware with a delicate patience. They have to be nurtured away from Windows-shock psyche with soft food, not coconut shells and acorns. They are trying to find their way back through the looking glass, to hop that free ride back to Kansas and we have to help them along like the good witch of the North:) 'There's no place like home.'

We can help where others cannot because we have been there, but , yes .. there is a certain saddle point that can break the camel's back and I've spent a lot of time in that desert.

Regards..

[Linuxratty]

Fragmentation and slow updates are not an open source problem.

And Android is not "all about being open". Android is all about making money.

Fragmentation happens because a zillion and one companies see a profitable market and jump in.
.

That's it in a nutshell. I've never used it.

[T.J.]

I empathize with your honest anger here...Having said this; there will always be security vulnerabilities and persons trying to exploit them and we still (I still) have to be vigilant and realistic, keep my teeth and claws sharp so when I am trying to assist in a migration I can talk the malware crap talk with potential ubuntu noobs and follow up with them. With ubuntu , the old axiom holds true: "we have to give it away to keep it". eh ? New converts have to be detoxed from malware with a delicate patience. They have to be nurtured away from Windows-shock psyche with soft food, not coconut shells and acorns. They are trying to find their way back through the looking glass, to hop that free ride back to Kansas and we have to help them along like the good witch of the North:) 'There's no place like home.'

We can help where others cannot because we have been there, but , yes .. there is a certain saddle point that can break the camel's back and I've spent a lot of time in that desert.

Regards..

Sorry about the rant...Some days are better than others as you well know. After so many years (fast approaching 30 now -- sheesh) in the computer racket - dealing with Microsoft, I just get tired of the whole mess, and your comments gave me an excuse to vent some frustration. Ironically most of the annoyance isn't users, but others fully brainwashed in the false history that Apple and Microsoft invented everything. The Linux crowd is not really different though. You have those who bash everything: Pulseaudio, Systemd, Mono, LSB etc. I doubt a single one of them has ever looked at the code, patched it or just spent the time to fix a problem without jumping ship, screaming like Chicken Little.

For many of us, it is very easy to get caught up in legitimate concerns. When Debian first adopted Systemd, I was definitely worried, especially with the talk about kdbus. I think most of the disaster never panned out, and kdbus still is not in the Linux kernel. For me, it served as an abject lesson on future happenings, an inoculation against the nonsense. Linux communities talk a lot, but the people who actually write the code are usually more reasonable.

The fragmentation fear that the press has always been spouting has never come to pass. Not with a single operating system that I have ever worked with has ever had a "fragmentation apocalypse." Not a single one - including Linux! I certainly don't know where they get their ideas. As long as the applications work, doom is averted. The applications are kept working by developers, who spend their time writing code. It's not a glory business, but someone has to do it.

You have the people who actually do the work and then those who just stand around and talk.

[mikodo]

"we have to give it away to keep it".

"Wherever I go, there I am."

"People who think they know it all are very irritating to those of us who do", always makes me laugh. :)

Short version

"Sit down, shut up and listen" :wink:

[CantankRus]

I empathize with your honest anger here. Like second hand smoke, just hearing of some person infected with a Windows machine makes me physically ill but it is part of the Instructional Development mindset that I try to deploy as a migration assistant for 'spinners' who are detoxing from Windows. The mis-information and dis-information are like a plague. Some of the new stories I hear each day are epic and near make me weep. Having pity is no good and so I try to be firm in my pitch to instruct potential converts to linux/ubuntu or other form of linux-os. "You have to get with the program and learn the learning curve. "

Having said this; there will always be security vulnerabilities and persons trying to exploit them and we still (I still) have to be vigilant and realistic, keep my teeth and claws sharp so when I am trying to assist in a migration I can talk the malware crap talk with potential ubuntu noobs and follow up with them. With ubuntu , the old axiom holds true: "we have to give it away to keep it". eh ? New converts have to be detoxed from malware with a delicate patience. They have to be nurtured away from Windows-shock psyche with soft food, not coconut shells and acorns. They are trying to find their way back through the looking glass, to hop that free ride back to Kansas and we have to help them along like the good witch of the North:) 'There's no place like home.'

We can help where others cannot because we have been there, but , yes .. there is a certain saddle point that can break the camel's back and I've spent a lot of time in that desert.

Regards..

I got lost in the hyperbole maze. :P

[ventrical]

I got lost in the hyperbole maze. :P

As I posted earlier on I have two RCA tablets that have android running on OEM ubuntu kernels and although the article referenced by OP suggests that Google may take over Android they cannot take over ubuntu/Canonical so I fail to see the hyperbole maze you reference :)

If you are referring to my content of experience, strength and hope, I kid you not , I do not embellish although I am not offended if somebody opines that to them it may appear to be psychobabble or pure conjecture. This , after all, is discussion thread and I like discussing things. It helps me think better and and clear my slates . And I like to read the feedback from others. Some of it inspires me to think harder. Some of it motivates me to brainstorm. Brainstorming helps me defragment my thoughts and in turn helps me to stay focused. I used to read zdnet articles all the time but most of the stuff on the net these days are mostly bifurcations of the truth.



Android is a practical demonstration of the sort of mess that Linux would have become if it had enjoyed widespread popularity with hardware OEMs. Someone, somewhere has to be in charge, and put the interests of the platform over profit margins and market share.


Obviously the author's research is not current otherwise they would have included that ubuntu/Canonical is providing the ubuntu kernels for several current form factors which android is running on, RCA to name just one. So the article has nothing to do with fragmentation. It is a rip at google and android and a veiled slam at Linux. I mean.. who cares if Android uses ubuntu kernels? Now that's hyperbole.

Regards..

[ventrical]

"People who think they know it all are very irritating to those of us who do", always makes me laugh. :)

Short version

"Sit down, shut up and listen" :wink:

+1 :)

[ventrical]

Sorry about the rant...Some days are better than others as you well know. After so many years (fast approaching 30 now -- sheesh) in the computer racket - dealing with Microsoft, I just get tired of the whole mess, and your comments gave me an excuse to vent some frustration. Ironically most of the annoyance isn't users, but others fully brainwashed in the false history that Apple and Microsoft invented everything. The Linux crowd is not really different though. You have those who bash everything: Pulseaudio, Systemd, Mono, LSB etc. I doubt a single one of them has ever looked at the code, patched it or just spent the time to fix a problem without jumping ship, screaming like Chicken Little.

For many of us, it is very easy to get caught up in legitimate concerns. When Debian first adopted Systemd, I was definitely worried, especially with the talk about kdbus. I think most of the disaster never panned out, and kdbus still is not in the Linux kernel. For me, it served as an abject lesson on future happenings, an inoculation against the nonsense. Linux communities talk a lot, but the people who actually write the code are usually more reasonable.

The fragmentation fear that the press has always been spouting has never come to pass. Not with a single operating system that I have ever worked with has ever had a "fragmentation apocalypse." Not a single one - including Linux! I certainly don't know where they get their ideas. As long as the applications work, doom is averted. The applications are kept working by developers, who spend their time writing code. It's not a glory business, but someone has to do it.

You have the people who actually do the work and then those who just stand around and talk.

..and that was a very intelligent and well informed rant that I thoroughly enjoyed.:)

Than you very much.

Regards..

[ventrical]

Sorry about the rant...Some days are better than others as you well know. After so many years (fast approaching 30 now -- sheesh) in the computer racket - dealing with Microsoft, I just get tired of the whole mess, and your comments gave me an excuse to vent some frustration. Ironically most of the annoyance isn't users, but others fully brainwashed in the false history that Apple and Microsoft invented everything. The Linux crowd is not really different though. You have those who bash everything: Pulseaudio, Systemd, Mono, LSB etc. I doubt a single one of them has ever looked at the code, patched it or just spent the time to fix a problem without jumping ship, screaming like Chicken Little.

For many of us, it is very easy to get caught up in legitimate concerns. When Debian first adopted Systemd, I was definitely worried, especially with the talk about kdbus. I think most of the disaster never panned out, and kdbus still is not in the Linux kernel. For me, it served as an abject lesson on future happenings, an inoculation against the nonsense. Linux communities talk a lot, but the people who actually write the code are usually more reasonable.

The fragmentation fear that the press has always been spouting has never come to pass. Not with a single operating system that I have ever worked with has ever had a "fragmentation apocalypse." Not a single one - including Linux! I certainly don't know where they get their ideas. As long as the applications work, doom is averted. The applications are kept working by developers, who spend their time writing code. It's not a glory business, but someone has to do it.

You have the people who actually do the work and then those who just stand around and talk.

...and so the sum of the parts.. those parts that are supposed to be fragmenting are actually converging to what many might think a much more secure linux (at least in the case for xmir and ubuntu snappy).

Sorry for the long quote , but,



In March 2013 Canonical Ltd. (https://en.wikipedia.org/wiki/Canonical_Ltd.) announced Mir as the replacement display server for the X.Org Server (https://en.wikipedia.org/wiki/X.Org_Server) in Ubuntu (https://en.wikipedia.org/wiki/Ubuntu_%28operating_system%29).[6] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-engadget-announcement-6) Previously, in 2010, it had announced that it would use Wayland (https://en.wikipedia.org/wiki/Wayland_%28display_server_protocol%29).[30] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-30) Canonical stated that it could not meet Ubuntu’s needs with Wayland.[15] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-ars_technica_2013-10-17-15) There were several posts made in objection or clarification, by people leading other similar or affected projects.[31] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-31)[32] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-32)[33] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-33)[34] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-34)
When originally announcing Mir, Canonical made various claims about Wayland's input system, which the Wayland developers quickly rebutted.[35] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-35)[36] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-NoteToCanonical-36) Official Canonical documentation in 2014 states, "our evaluation of the protocol definition revealed that the Wayland protocol does not meet our requirements. First, we are aiming for a more extensible input event handling that takes future developments like 3D input devices (e.g. Leap Motion (https://en.wikipedia.org/wiki/Leap_Motion)) into account...With respect to mobile use-cases, we think that the handling of input methods should be reflected in the display server protocol, too. As another example, we consider the shell integration parts of the protocol as privileged and we'd rather avoid having any sort of shell behavior defined in the client facing protocol."[37] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-MirWiki-37) In late 2015 Mir switched from a custom Android-derived input stack to Wayland’s libinput (https://en.wikipedia.org/wiki/Libinput).[38] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-38)[39] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-39)
Longtime Linux kernel (https://en.wikipedia.org/wiki/Linux_kernel) developer Matthew Garrett (https://en.wikipedia.org/wiki/Matthew_Garrett) criticized the choice of licensing for Canonical's software projects, particularly Mir. Unlike X.Org Server (https://en.wikipedia.org/wiki/X.Org_Server) and Wayland, both under the MIT License (https://en.wikipedia.org/wiki/MIT_License), Mir is licensed under GPLv3 (https://en.wikipedia.org/wiki/GPLv3) – "an odd [choice]" for "GPLv3-hostile markets" – but contributors are required to sign an agreement that "grants Canonical the right to relicense (https://en.wikipedia.org/wiki/Software_relicensing) your contribution under their choice of license. This means that, despite not being the sole copyright holder, Canonical are free to relicense your code under a proprietary license." He concludes that this creates asymmetry where "you end up with a situation that looks awfully like Canonical wanting to squash competition by making it impossible for anyone else to sell modified versions of Canonical's software in the same market".[2] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-MirCLA-2)[40] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-40)[41] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-41)[42] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-42) Garrett’s concerns were echoed by Bradley M. Kuhn (https://en.wikipedia.org/wiki/Bradley_M._Kuhn),[43] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-43)[44] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-44) Executive Director of the Software Freedom Conservancy (https://en.wikipedia.org/wiki/Software_Freedom_Conservancy).[45] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-sfc-ed-45) Richard Stallman (https://en.wikipedia.org/wiki/Richard_Stallman) of the Free Software Foundation (https://en.wikipedia.org/wiki/Free_Software_Foundation) has stated that he supports dual-licensing of GPL software, as long as there are no proprietary extensions or proprietary versions of the free program.[46] (https://en.wikipedia.org/wiki/Mir_%28software%29#cite_note-46)
In June 2013, Jonathan Riddell of Kubuntu (https://en.wikipedia.org/wiki/Kubuntu) announced that Kubuntu did not plan to switch to Mir. He stated "A few months ago Canonical announced their new graphics system for Ubuntu, Mir. It's a shame the Linux desktop market hasn't taken off as we all hoped at the turn of the millennium and they feel the need to follow a more Apple or Android style of approach making an OS which works in isolation rather than as part of a community development method. Here at Kubuntu we still want to work as part of the community development, taking the fine software from KDE and other upstream projects and putting it on computers worldwide. So when Ubuntu desktop gets switched to Mir we won't be following. We'll be staying with X on the images for our 13.10 release now in development and the 14.04 LTS release next year. After that we hope to switch to Wayland which is what KDE and every other Linux distro hopes to do"

In September 2013, an Intel developer removed XMir support from their video driver and wrote "We do not condone or support Canonical in the course of action they have chosen, and will not carry XMir patches upstream".

.. so despite the controversy, despite what could be called fragmentation actually made the road to convergence more practical and realisitic.

As you point out there are many who are just sitting on the sidelines, outside looking in, waiting to share the fruits of harvest. But that is what the ubuntu community is all about. Sharing. Some share back , those maybe being maintainers, triagers, developers and other contributors giving back in various different ways. No matter how large a contribution or how small, whether is be right or wrong it still builds the KB. The community is actually responsible for all the good things that have happened to ubuntu. It is a formerly proof of concept novel idea that has become a solid reality. The community design has been tested several times by several elements but it always seems to come out on top better than it was before. One would have to get by several well seasoned helmsmen to roll the sails up on this ship and IMHO the same applies to android development. The worst or best that can happen there is that it be replaced with an ubuntu/linux version or at least alternatives of this nature will be more available, ie; with snappy of course.

Regards..

[Wadim_Korneev]

In short, Android has become a mess, and the landscape has changed such that there's no reason for Google to keep on making an open source operating system, especially one that the competition can leverage.

[T.J.]

In short, Android has become a mess, and the landscape has changed such that there's no reason for Google to keep on making an open source operating system, especially one that the competition can leverage.


Do what the article suggests and Android will be dead; like Windows Phone is dead.

Without the open source nature of Android, the handset makers would go elsewhere because their costs will go up. They basically have a razor thin profit margin as it is. If they suddenly have higher costs, even in the form of changes to their existing production process, they will abandon Android in droves. They won't go to Apple - Apple is too strict and has no interest in licensing. Either they would use derivatives based on the existing open versions of Android, such as CyanogenMod; or they would jump to Tizen.

Google would be completely left out in the cold, and Alphabet's stock value would plummet. Worse for Google, close the source and the programmers who are contributors and bug hunters would disappear. The cost of producing Google Android would increase. Programmers refuse to work with companies that stab them in the back. CyanogenMod would probably take Android's place. Google's reputation would sink even lower than Oracle's. Remember what happened to OpenOffice when the programmers left and forked Libreoffice?

Contrary to what the mass media would have you believe, fragmentation is healthy for open source software, because useful changes can be folded back into the official versions at little to no cost to the primary developers. Close the source to prevent changes and the value of the source code decreases dramatically.

Windows 10 is closed source and its value as a commercial product is essentially zero in today's market. It has too many competitors that are free or zero cost.

[mikodo]

I am not trying to blow smoke up T.J. but, it is enlightening to understand how things work from the perspective of a developer like yourself.

Others here, do the same from time to time. I always enjoy reading how things have evolved to where we are at now and how factual history can be used to help others understand (like myself) the reality of these things.

Thanks.

[grahammechanical]

I found this. I was surprised at the number of companies that are building their own versions of a mobile OS but based on Android Open Source Project. (AOSP). The genie seems to be well and truly out of the bottle. If Goggle was to change the licensing of future releases of Android these companies could take the existing open source versions and build from there.

https://en.wikipedia.org/wiki/Mobile_operating_system

When does an OS become mature enough so that all that is needed is updated drivers for the newer hardware in the latest products?

Regards

[T.J.]

I am not trying to blow smoke up T.J. but, it is enlightening to understand how things work from the perspective of a developer like yourself.

Others here, do the same from time to time. I always enjoy reading how things have evolved to where we are at now and how factual history can be used to help others understand (like myself) the reality of these things.

Thanks.

Oh I know, and please never take anything I say as a condemnation of yourself or any other poster on the forum. My point in commenting is just to offer another prospective. I'm not claiming it is the correct one and it certainly isn't the only one.

License drives development when the software, such as an OS is a simple commodity. Take Windows 10 as an example.

Even if the software is free as in Windows 10, the fact that the license is very strict and the code is closed are major factors in its use and continued development. Windows 10 is eschewed professionally because a number of developers are bound by privacy laws. While this may or may not be a factor in the decision making process, the fact that there is an level of uncertainly regarding the privacy of Windows 10, as well is the closed nature of the code to conceal what is recorded actually harms adoption levels.

The fact that Windows 10 is essentially trying to create a closed ecosystem from what was once a more open one is a problem too. As a developer who has worked on Windows, I can tell you that neither I nor the companies I work for intend to use the Windows Store or develop for Windows 10. Why should we give Microsoft 30 percent of our profit margin to create an app that can be delivered on less than half of the existing Windows user base? Furthermore, Windows Phone is dead, and Surface is a tiny percentage of the market, so Windows "convergence" is basically worthless. Universal Apps are Windows 10 only, really only useful on Windows 10 PCs and the Xbox - because of closed source and license. If Microsoft provided an environment for Linux and Mac to run Universal Apps, it would be far more compelling.

License drives everything - or in the case of Windows 10 - ruins everything. It determines market position in the 21st Century. When you look at Windows today versus the 90s, even Microsoft admits they failed, and they can't figure out why. In the 90's Microsoft held virtually 50-90% of all computer devices. Now they barely hold on to 14% - mostly because of the traditional desktop machines. A closed source license is killing them and their "vision". If they want to save Windows, they must open the code. An opensource Windows has a good chance to catch up with Linux faster than any other system.

If Android were to take the same path as Windows 10 - suddenly close up - developers would abandon it, much as they have Windows 10. While I do not presently develop for Android, we have been in discussions over it for a few years now. I know I would lose interest if I can't reassure myself by inspecting code. I'd head over to CyanogenMod in a heartbeat.

[misfitpierce]

This will be remedied with Googles new hands on OEM Shaming board listing the OEMs that are slacking off on updates for that aspect. Other than that OEMs are scaling back how heavy they make their skins now but some people actually like the funtionality it brings and Android learns new ideas from it and the devs working hard on mods and roms... So personally I love that its open source. Android wear is not open source and google wants no hand in messing with stock wear atm. All in all I think it works for android and helps build it quicker, I mean if you look IOS is playing catch up to android now.

[ventrical]

It's all about trust. There wasn't any then and most likely isn't any now.




...Windows phone is dead,...




Robert X. Cringely...


"now the only way Microsoft can die is by suicide."


I think I mentioned earlier about a paradigm shift. I wrote an article (and I forget where I posted it) titled "The Emperor's New Blinds" in reference to the release of Windows Vista a long time ago. The point of the matter was that Microsoft would collapse by the bloat of it's own weight. The anti-pirating spyware and authentication routines built into XP , Vista and Windows 7 literally killed not only a beautiful platform but also a solid business model.

After reading some really good editorials here I have come to believe that if Microsoft continues it's outreach program they may once again claim a larger percentage of market share but it will be an uphill battle as Ubuntu/Linux continue in developing their Instructional Development knowledge base. The offer of fresh, free open installs to broken machines is just too attractive and the swap in , swap out feature is just too much of a bargain to turn away from. The end_user_space public at large is finally realizing it's OK to untie and uncuff themselves from their chairs and no longer do they have to be held in the stocks by anti-malware companies.

And just as a side note; I think Android is anything but fragmented, just more diversified, as it should be.

Regards..

[T.J.]

I think I mentioned earlier about a paradigm shift...

After reading some really good editorials here I have come to believe that if Microsoft continues it's outreach program they may once again claim a larger percentage of market share but it will be an uphill battle as Ubuntu/Linux continue in developing their Instructional Development knowledge base. The offer of fresh, free open installs to broken machines is just too attractive and the swap in , swap out feature is just too much of a bargain to turn away from. The end_user_space public at large is finally realizing it's OK to untie and uncuff themselves from their chairs and no longer do they have to be held in the stocks by anti-malware companies.


Regards..

If I might offer some speculation, I believe Microsoft's best shot at not only evening out the field, but seriously regaining and keeping ground - is to open the Windows and Windows RT source code. It doesn't have to be all of it. Just enough so that an open standard of some kind can be created with the Windows platform. Operating systems are a no profit commodity for users and developers. What Microsoft loses now, it regains tenfold later. The apps are where the money is. By restoring trust by opening the code and allowing others to extend the use of their platform by helping develop it, they sell more apps. Furthermore, they can certify hardware and software as Google does. Then they can create a proprietary version with all the open changes for their Enterprise customers at far less expense than the current situation and sell it at the same price for more profit.

They could even add "value components" such as DirectX support to "open Windows" and Linux, sell them in the Windows Store for a pittance/subscription and make a landfall profit in the near term while Vulkan is not fully developed. It would take the wind out of some sails for certain.

If they were brave enough to do it, I believe it would make Microsoft a majority player, especially in a world where Oracle sues over an API. So while some say being open is killing Android, I believe it would be Microsoft's salvation.

[ventrical]

T.J. wrote...



...Operating systems are a no profit commodity for users and developers. What Microsoft loses now, it regains tenfold later. The apps are where the money is.


They used QBasic on DOS.5.0 to write Windows XP. No matter how popular it became, was or still is the company gurus could not see fit to decouple Windows IExplorer. It became the back door and front door (as well as the lamp post) for every malicious attack imaginable. There are two core problems. One, is the way the OS was written in a spirit of exclusivity so that common folk, i.e. future developers, programmers and hacks were shut out even at the basic level. If you tried to innovate or experiment you would get slammed with the EULA or be accused of being a pirate. If you used a Linux version you were shut out even more. The marriage between Intel and Microsoft was a jealous one and ferociously defended. Microsoft kept Intel based hardware sales always at a high bar. This way they could make excuses to obsolete older processors - that is until Ubuntu came along and extended the longevity of several classes of legacy Pentium D dual core processors. This got everybody's attention and shook the bushes.

Two, is the way the depends are designed to lap-link with the constantly failing versions of IE. Because of this isolated homogeneity a true fragmentation exists. There is always razzle and dazzle on the promo laps at the local stores but once it gets one or two automatic updates it becomes injected with some sloth process the knocks both the startup and shutdown back several steps from it's original out of the box vigour.

I appreciate your speculations. The way I see it is that they have to dump Internet Explorer. It is the only way that they will be able to build some kind of trust and credibility with user_space at large. With open source there are always fresh innovations. Take for example Canonical's unpopular decision to develop their own compositor/server (xmir). Ubuntu is always moving forward. Windows is always trying to sell the same set up only with a few different themes here and there. Being able to examine MS code freely is something I really do not have a desire to do and would probably not reveal any surprises.


By restoring trust by opening the code and allowing others to extend the use of their platform by helping develop it, they sell more apps. Furthermore, they can certify hardware and software as Google does. Then they can create a proprietary version with all the open changes for their Enterprise customers at far less expense than the current situation and sell it at the same price for more profit.


That's a fair comment. I think it is also a fair vision that includes a larger legacy group of end users who feel very comfortable with their current platforms. It was really easy for Steve Jobs to spout off about how an end user should buy a new machine every year and throw out the old one. I mean that sales tactic of keeping "updated" and "current" with "the latest" only kept other peoples wallets fat and there was nothing wrong with the current machine. Simply put .. open source has a hawkeye on the economy . Their giving a platform to the voices of the future and embracing the little ones. It's the new bandwagon and Microsoft would do well to get with the program. Even if they kept their proprietary blobs, fine .. they should release an open source version if they want to reclaim their former market share standings otherwise they will become nothing less that a fragmentation.

Regards..

[T.J.]

They used QBasic on DOS.5.0 to write Windows XP.


I don't get it. Windows XP was written in C++ primarily when you consider the COM standard. If you are referring to the scripting host that comment might make more sense in context.




Being able to examine MS code freely is something I really do not have a desire to do and would probably not reveal any surprises.


I respectfully disagree. Do I think that open-sourced Windows code would help develop Linux? No, probably not. Linux is a monolithic kernel, whilst Windows is a hybrid kernel. That is literally a useless comparison, but it would facilitate better user software on Windows and Linux.

What it would also be good for is improving application standards across both systems. Windows and Linux have different application design problems that could be resolved for the majority of users by cooperation between the two; such as binary compatibility and interface stability. One of the greatest hangups that no one wants to admit is that Linux and Windows GUIs become increasingly unstable in recent years. Too much change without consideration of the consequences has a real world impact. Windows saw it with 8.1; Linux saw it with Gnome 3. They were different cases, but rooted in exactly the same errors. In both cases, the userbase retreated or refused to upgrade - which led to massive losses in marketshare or mindshare..

[ventrical]

http://ubuntuforums.org/images/ubuntu-VB4/misc/quote_icon.png Originally Posted by ventrical http://ubuntuforums.org/images/ubuntu-VB4/buttons/viewpost-right.png (http://ubuntuforums.org/showthread.php?p=13514898#post13514898)

They used QBasic on DOS.5.0 to write Windows XP.

Please disregard the above. I digress .. err about 12 years :) My bad. A wrong statement based on memory of one story I read.



I respectfully disagree. Do I think that open-sourced Windows code would help develop Linux? No, probably not. Linux is a monolithic kernel, whilst Windows is a hybrid kernel. That is literally a useless comparison, but it would facilitate better user software on Windows and Linux.

What it would also be good for is improving application standards across both systems. Windows and Linux have different application design problems that could be resolved for the majority of users by cooperation between the two; such as binary compatibility and interface stability. One of the greatest hangups that no one wants to admit is that Linux and Windows GUIs become increasingly unstable in recent years. Too much change without consideration of the consequences has a real world impact. Windows saw it with 8.1; Linux saw it with Gnome 3. They were different cases, but rooted in exactly the same errors. In both cases, the userbase retreated or refused to upgrade - which led to massive losses in marketshare or mindshare..



Yes .. major warts like gnome-disk-utility and starutup-disk-creator. What's good for the goose is good for the gander. The user base will have to pay a price for convergence. People are not actually willing to pay large money for hamper sized desktops any more unless they can get them dirt cheap. I mean there are developers that will work for free, but for how long? They have to put bread on the table, pay the bills .. ya know. With ubuntu heroics in restoring old metal they have widened their embrace of market share, at least in theory and on paper. On bump in the road, ie, like the upgrade mangler and that user base will quickly suffer themselves back to Windows or Apple. I've seen it happen. However the effervescence of their positive ubuntu experiences will count for something when Canonical rolls into the next generation of tablets and phones, the snap apps etc.. There will be bread and butter, milk and honey, plenty to go around. Yes .. I see the light .. there is a lot we can learn from microsoft code.

Regards..

[T.J.]

http://ubuntuforums.org/images/ubuntu-VB4/misc/quote_icon.png Originally Posted by ventrical http://ubuntuforums.org/images/ubuntu-VB4/buttons/viewpost-right.png (http://ubuntuforums.org/showthread.php?p=13514898#post13514898)

They used QBasic on DOS.5.0 to write Windows XP.

Please disregard the above. I digress .. err about 12 years :) My bad. A wrong statement based on memory of one story I read.


Oh no worries, it wasn't a criticism or sarcasm, merely confusion. I have read your comments and I have garnered a respect for your opinion, even when I do not agree.



The user base will have to pay a price for convergence


There is some truth to that, but the real question that no one has really asked is "Is convergence really desirable or necessary?" For myself, I say it is a 50/50 proposition. Convergence is only useful where it provides benefits to the user - not the developer. (If a developer needs convergence to write better code, it is a sign that he/she did a poor job writing it in the first place. You ALWAYS separate core program logic from the interface code.) Converging touch interfaces on static hardware is a pointless endeavor. It is only after a period of hardware refresh does convergence become valuable. I'd say that Gnome's philosophy (before Gnome 3) of incremental change over massive change was the correct one.



I mean there are developers that will work for free, but for how long?


Long term FOSS developers are either employed to work on FOSS or do so in their spare time. The truth is that it is either for money or for personal pleasure. In rare cases, it actually is altruism. Sometimes, you are just proud of your work and want to share it. FOSS programmers are good people, but they are also sometimes nasty, arrogant and just plain rude. As soon as everyone comes to grips with the reality, and stops trying to spread the delusion of a FOSS utopia, I think everyone will be better off.

FOSS is absolutely a better way to develop and deploy. It's not all unicorns and sparkly rainbows in a perfect world. I think a lot of people confuse FOSS ideals as incompatible with the fact it is also a business in which everyone has to make a living. As a programmer, I can tell you right now, that the idea the two are at odds is utter rubbish. FOSS ideals work very well in business as long as everyone keeps a clear head and thinks long term.



"On bump in the road, ie, like the upgrade mangler and that user base will quickly suffer themselves back to Windows or Apple. I've seen it happen.


That is Linux's greatest failing.


Being absolutely honest, the mission of FOSS is software. The point is the source code is public. We make no promises beyond that. We don't promise to support or hold anyone's hand. That said, there is value in treating users who are not developers with courtesy and respect. About 1/3rd to 1/2 of the community, in my opinion, is too self absorbed and does not care. It is also true that there are only so many hours in a day, and after a long day of coding, I am sometimes just too exhausted to care. There are no perfect answers.

It is important to note that there has been a fundamental shift in the Linux community from FOSS (free open source software) to FLOS (free linux operating system) in which portability of code and being a citizen of the wider community has been cast aside for a Linux only approach. In my opinion, that is sad. FOSS includes everyone Linux, Windows, Mac, and others - while FLOS cares only about Linux. Some have even advocated abandoning POSIX - which in my opinion is sheer madness. Sticking to standards is what has brought Linux this far.




There will be bread and butter, milk and honey, plenty to go around. Yes .. I see the light .. there is a lot we can learn from microsoft code.


I can't tell if you are being ironic, satirical, honest or sarcastic. I believe there is a lot we can learn. Microsoft has over the years, hired some very brilliant people. If the code were open, I do believe it would be of benefit. There is no point of view, no single approach, or software design that is so perfect an endeavor that something new cannot be learned.

[mikodo]

(If a developer needs convergence to write better code, it is a sign that he/she did a poor job writing it in the first place. You ALWAYS separate core program logic from the interface code.)

Does Ubuntu Snappy Core with Snap apps and multiple DE's fit this criteria?

[T.J.]

Does Ubuntu Snappy Core with Snap apps and multiple DE's fit this criteria?


Yes, it is true that Linux itself - being the kernel - separate from the DEs is helpful! =) You have good insights. Actually, I was referring to the design of the code itself rather than the overall packaging, such as Snappy.

Good program design requires that the parts of the code that do the work - save files, do calculations or whatever are not specific to what you see on the screen; that they be separated from the parts of the code that make up the interface on the screen. The reason is that it makes the program easier to move to another OS or a touch screen - where you might need a different interface for the same program. The second reason is that it makes it easier to find bugs. The third reason is that if you don't have to do a complete rewrite for each OS, it saves a lot of money and time.

Often when code is designed today, the code used to create window you see is often jammed together with the code that actually does the work such as saving files or whatever. Most of this is just laziness, but for a lot of younger programmers, it is just the way they were taught - especially on Windows. Some people want convergence so that the visible interfaces are all the same. In that way, they can keep writing code in a sloppy fashion and never have to correct their design mistakes.

The problem with that of course is that a person using a keyboard in an office has very different interface needs from a casual tablet user - even if they both use the same program.

[grahammechanical]

The problem with that of course is that a person using a keyboard in an office has very different interface needs from a casual tablet user - even if they both use the same program.

Which is why those people who think that the Ubuntu phone is Ubuntu on a phone are disappointed. They expect to be able to do the same customizations of the UI as they do on the desktop and all the desktop applications to be available. Some even ask if they can install an alternative DE.

Canonical is making progress with Mir + Unity 8 for phones & tablets. And app developers are given a lot of help to design for converged form factors right at the start but we have yet to see Mir + Unity 8 as a desktop environment that can replace Unity 7.

Regards.

[ventrical]

Oh no worries, it wasn't a criticism or sarcasm, merely confusion. I have read your comments and I have garnered a respect for your opinion, even when I do not agree.



There is some truth to that, but the real question that no one has really asked is "Is convergence really desirable or necessary?" For myself, I say it is a 50/50 proposition. Convergence is only useful where it provides benefits to the user - not the developer. (If a developer needs convergence to write better code, it is a sign that he/she did a poor job writing it in the first place. You ALWAYS separate core program logic from the interface code.) Converging touch interfaces on static hardware is a pointless endeavor. It is only after a period of hardware refresh does convergence become valuable. I'd say that Gnome's philosophy (before Gnome 3) of incremental change over massive change was the correct one.

I appreciate your point but strongly disagree. Convergence will not be mandatory, it will be an option. It gives more choices to both programmers/devs and end_users. We are not shut up in a box and thinking inside the box but are truley deploying an 'out of the box' path which to travel upon into the furture. No one gets shut out , the dinosaurs , the middle set and the new set. It is a mindset of inclusion, including the dinosaurs, the middle set and the new set.

I am not slamming commercial software houses but we have to be honest. Any company that has to resort to building and embedding code into their opertating system in the form of Product Activation Keys and VLKs has only perpetrated a legal method to install a malware into their clients form factors and we cannot forget the histories of these companies know matter how beautifully set they are with prodigy programmers. In the past tense they are accessories after the fact. So it goes back to trust again. What's that bisuness model bromide... 'The customers always right.' Ahh .. thats it isn't it?




Long term FOSS developers are either employed to work on FOSS or do so in their spare time. The truth is that it is either for money or for personal pleasure. In rare cases, it actually is altruism. Sometimes, you are just proud of your work and want to share it. FOSS programmers are good people, but they are also sometimes nasty, arrogant and just plain rude. As soon as everyone comes to grips with the reality, and stops trying to spread the delusion of a FOSS utopia, I think everyone will be better off.

FOSS is absolutely a better way to develop and deploy. It's not all unicorns and sparkly rainbows in a perfect world. I think a lot of people confuse FOSS ideals as incompatible with the fact it is also a business in which everyone has to make a living. As a programmer, I can tell you right now, that the idea the two are at odds is utter rubbish. FOSS ideals work very well in business as long as everyone keeps a clear head and thinks long term.

We just finished a cycle named Utopic Unicorn: It's a brilliant marketing idea. It includes the new set, the middle set and the dinosaurs.




Being absolutely honest, the mission of FOSS is software. The point is the source code is public. We make no promises beyond that. We don't promise to support or hold anyone's hand. That said, there is value in treating users who are not developers with courtesy and respect. About 1/3rd to 1/2 of the community, in my opinion, is too self absorbed and does not care. It is also true that there are only so many hours in a day, and after a long day of coding, I am sometimes just too exhausted to care. There are no perfect answers.

But with the ubuntu community the model is that the dinosaurs feed the middle set and the middle set feeds the new set and that creates a positive sand-box for instructional development. Yes, we can build sand castles and tear them down and rebuild them again - but - yes .. I admit that when the warts appear they are big and ugly.. but we still move upstream and avoid the big rocks.



It is important to note that there has been a fundamental shift in the Linux community from FOSS (free open source software) to FLOS (free linux operating system) in which portability of code and being a citizen of the wider community has been cast aside for a Linux only approach. In my opinion, that is sad. FOSS includes everyone Linux, Windows, Mac, and others - while FLOS cares only about Linux. Some have even advocated abandoning POSIX - which in my opinion is sheer madness. Sticking to standards is what has brought Linux this far.


If sticking to standards were so with Ubuntu then I would just have to go back to Windows. I respectfully disagree T.J. It is alsmost as if you are saying we have to sleep with the enemy.




I can't tell if you are being ironic, satirical, honest or sarcastic. I believe there is a lot we can learn. Microsoft has over the years, hired some very brilliant people. If the code were open, I do believe it would be of benefit. There is no point of view, no single approach, or software design that is so perfect an endeavor that something new cannot be learned.

How many people have tried to jump into the Microsoft sandbox and got kicked in the teeth. Don't forget what you said about 10 messages ago .. 'windows phone is dead' and the main reason for that is that all the new set are in the ubuntu/snappy sandbox.

As for your other question about ironic, satircal,honest or sarcastic... all of the above :) and I am not trying to be smart or disrespectful.

I worked with some Microsoft MVPs (geesh) and I pointed out to them at the time during Vista's release , while I was beta testing Facebook models, that there were possibilities of exploits and big malware that could get through and their reply was that they would send 45,000 of their hackers to fix it. And I got booted from the sandbox. So now I am in the ubuntu sandbox. I am not claiming sainthood. Ubuntu is not a monastic system. It is a community and whether we split hairs or not the community source concept has been extremely successful more so that the business model.

regards..

[T.J.]

If at any point someone wants to move this to different thread I certainty understand....


I appreciate your point but strongly disagree. Convergence will not be mandatory, it will be an option. It gives more choices to both programmers/devs and end_users. We are not shut up in a box and thinking inside the box but are truley deploying an 'out of the box' path which to travel upon into the furture. No one gets shut out , the dinosaurs , the middle set and the new set. It is a mindset of inclusion, including the dinosaurs, the middle set and the new set.

That's just it. The whole concept of "convergence" is poorly defined, at best. For you, it sounds like defining what is new versus what is old i.e, dinosaurs, middle and new.

I beg your indulgence while I expound a bit. For me, convergence is the attempt to merge the various forms of user interfaces into one single interface, and failing poorly. That they create is a user interface that is generally less productive in terms of time. The whole mindset of convergence assumes a few things:

a) Old is bad.
b) A merged interface is more efficient.
c) Every user case or condition is the same.

All three of these conditions fall apart in the real world. The first is a subjective opinion. The other two are simply untrue. I consider the history Gnome 3 and the massive abandonment by its userbase as an abject example. The closest example of a trusim that you can achieve is that similar designs give enough hints to help the user - but complete convergence is not practical. Ironically, pundits call that reality "fragmentation".


am not slamming commercial software houses but we have to be honest. Any company that has to resort to building and embedding code into their opertating system in the form of Product Activation Keys and VLKs...we cannot forget the histories of these companies know matter how beautifully set they are with prodigy programmers

I fail to see how that even enters into the discussion we have had so far. I know there is a lot of dislike for Windows, but if the code was open these things would be irrelevant. Open source is all about code. Personally, I don't have time to waste on grudges over past wrongs. However, how you feel is up to you. I can't change that, and wouldn't try. =)


But with the ubuntu community the model is that the dinosaurs feed the middle set and the middle set feeds the new set and that creates a positive sand-box for instructional development. Yes, we can build sand castles and tear them down and rebuild them again - but - yes .. I admit that when the warts appear they are big and ugly.. but we still move upstream and avoid the big rocks.

Somewhat. I've heard the best description of the Linux community - not Linux itself necessarily - from a BSD programmer. They described it as a chaotic undisciplined mess. The benefit of the chaos was that Linux might go in directions no one ever thought of - and that is a good thing. It also meant that a lot of things that came out of the process would be less than useful. It's been my experience that while Linux has much to commend it, it requires firm project leaders to keep it from descending into total stupidity or a single poorly planned agenda. Thank heavens that Linus Torvalds is very strict about poorly designed code in the kernel. On the other hand, systemd - while useful and the software itself has a lot of potential - is clearly a poorly planned and executed project.




If sticking to standards were so with Ubuntu then I would just have to go back to Windows. I respectfully disagree T.J. It is alsmost as if you are saying we have to sleep with the enemy.

When I refer to standards, I am referring to ISO/ECMA standards, such things as POSIX.

Serious programmers prefer ISO/ECMA standards over de-facto ones because compliance means things work and we waste less of our time. Without POSIX. some of the core software that Linux uses, such as X11 wouldn't work. Fortunately, Linux is mostly POSIX compliant. If the FLOS crowd has their way and Linux abandons POSIX in favor of creating its own standards, then I won't be using Linux anymore because I would either fork a new version or return to BSD. Without POSIX, porting Linux code would be as hard as porting from Windows.

You might think that de-facto community standards are better than ISO/ECMA ones. I don't.

Look at the mess they have created. Languages such as Python and Java - where versions and compilers are inherently incompatible with each other. Oracle was able to sue over Java because Java is a defacto language. Had it been ISO standardized as Sun had originally planned, that would have been impossible. Python has so many different licenses, it is ridiculous. You can't use Java, Rust, Python, Swift or other languages like them in complete security. You never know if some future lawsuit from their corporate sponsor is going to undermine you for using them.

ISO/ECMA standards have specific terms that limit or remove legal liability in virtually all use cases. I can't say the same for a lot of the newer "standards". I believe that there will almost certainly be a major schism between the older pro-POSIX/FOSS and the younger FLOS groups when Linus retires someday. I say that because while Linus manages the kernel, its POSIX functionality and binary interface are relatively stable. If that ever changes, I see the Linux community splitting down the middle like the Christians did during the Reformation period.

[user1397]

Look at the mess they have created. Languages such as Python and Java - where versions and compilers are inherently incompatible with each other. Oracle was able to sue over Java because Java is a defacto language. Had it been ISO standardized as Sun had originally planned, that would have been impossible. Python has so many different licenses, it is ridiculous. You can't use Java, Rust, Python, Swift or other languages like them in complete security. You never know if some future lawsuit from their corporate sponsor is going to undermine you for using them.
Just wanted you to clarify this statement. I completely understand it if you're talking about Java since from my understanding it is not fully open source (or else why would openjdk exist? Am I incorrect here?), and maybe Swift too since originally it was proprietary and owned by Apple (although after version 2.2 it is licensed under apache/MIT license so...), but I don't get how Rust or Python have this issue. They are both run by non-corporate entities from my understanding.

[T.J.]

Just wanted you to clarify this statement. I completely understand it if you're talking about Java since from my understanding it is not fully open source (or else why would openjdk exist? Am I incorrect here?), and maybe Swift too since originally it was proprietary and owned by Apple (although after version 2.2 it is licensed under apache/MIT license so...), but I don't get how Rust or Python have this issue. They are both run by non-corporate entities from my understanding.

I'll answer your questions about Java, but I have to explain the situation more clearly first. Most of the details are usually only known to programmers. There are two major divisions on the programming world: standardized languages and de-facto ones.

The standardized languages are officially recognized by the ISO/IEC (and usually the EU version called ECMA as well). As part of that recognition and standardization, the languages are defined in series of reference documents that everyone has to follow in order to be considered compliant with the standard. In other words, a compliant compiler will take the language as stated in the standard without complaint. This is one of the reasons that C/C++ is so portable between Linux, Windows and Mac. While each adds extra features to the language for their own use, the compiler has to adhere to the standard if they want to advertise ISO compliance. ISO standards have to be offered to EVERYONE by all parties involved under what is known as a RAND grant. A RAND grant is "reasonable and non-discriminatory" terms. That doesn't mean that they can't charge a fee for it, but it does mean that the technology covered by the standard and its terms have to be clearly stated and the same for everyone. In the vast majority of cases, ISO standards have no fees or royalties (even if there is patented material). In the absolute simplest terms, it is an international agreement so that everyone who uses the standard is guaranteed the same legal recognition. It's virtually impossible to sue someone protected under a RAND grant once all terms are met. Since the vast majority have no fees and open terms, that means if I decide one day to use an ISO standard - I am not only protected legally, but no one can stop me from doing so, regardless of wealth or market position.

The longest used programming languages in the world are all protected by ISO standards: C, C++, COBOL, and ADA are just a few examples. No one can go after you just for using those.

Then you have de-facto languages whose use is merely licensed. The standard of the language is what the community or the main sponsor say it is at any given time. Examples of these would be Java, Rust, Python, Perl, and PHP to name a few. To be absolutely fair and sterling from the outset, in most cases, the fact that it is de-facto rather than ISO is normally not a problem as long as you stick to one version on one OS. It does however present problems later on. First of which is that de-facto languages tend to be incompatible between versions. Python and Perl can be "poster children" for this fact. De-facto languages have no compliance standard that tools are expected to follow. Code on one compiler may react entirely differently on another - even though it is the same source code. Second, while community licenses: the CDDL, Apache, and even the GPL are generally sufficient protection, the licenses can be undermined piecemeal by local law, since they are only contracts - not international agreements. Third, the largest community sponsor may claim complete intellectual property over the de-facto standard and use it to eliminate or blackmail competition (as in the case of Java with Oracle vs Google).

So I generally consider de-facto languages as a bad choice, regardless of whether or not they have an open source license. I resist using them at all, period. If I am forced to, I only use ones that I feel have the least likelihood of being used as a legal weapon. As far as I am concerned, that pretty much eliminates the use of Java, Go, Rust and some others. Their sponsors - Apple, Google, and Oracle are fond of suing everyone in sight. Others like Mozilla, are simply high profile, and make easy targets. There is too much legal risk.

That does not mean that I will be sued personally, but it does mean that third party court cases could affect my development costs. Legal cases could lead to the retraction of the language entirely, changes in the language, re-licensing, or even new patent fees...The list of uncertainties is endless as long as one company or sponsor claims ownership and property rights to the language. That costs developers like myself money and time. Those are not something I have in abundance. In order to make a profit, I have to deliver on time and preferably under the budget.


Now then, OpenJDK is licensed under the GPL, so yes it is open. What you didn't know is that Oracle's Java is based on OpenJDK (but since Oracle is the sole owner of Java they can relicense their version as they see fit.) OpenJDK is considered the official reference version of the Java SE. However, there are three versions of Java: SE (standard edition/ desktop), EE (enterprise edition), and ME (micro edition used for embedded such as phones). Each is licensed under slightly different terms. Dalvik is Google's clone of Java ME, based on a "clean room" implementation, derived in part from Apache's Project Harmony. A "clean room" version means that they created their own "work-alike version" without looking at the original code. It's done under very strict conditions and is perfectly legal. The developers who created Dalvik were never allowed to see the original Java ME code.

So official OpenJDK is covered under the GPL, and Dalvik is covered by the Apache license. Both are open licenses. Neither could protect Google from being sued, because even though they are offered under open licenses, Java remains the property of Oracle, licensed solely on its terms. Had Java in general been covered under an ISO specification as Sun had originally planned, Oracle would not have been able to claim that it had been harmed by Google's use of Java's design, because Java would have been offered under a perpetual RAND grant - allowing everyone to use it for any purpose.

Python, Rust and so on - are also not covered under an international agreement with a RAND grant, but only a contracted license (even though they are open) - thus they have the same problem, just different details.

I am sorry it is so long-winded, but hopefully that helps clear things up.

[grahammechanical]

For me, convergence is the attempt to merge the various forms of user interfaces into one single interface,

I do not see Canonical's vision for convergence to be that strict. As I understand the design philosophy it is to write a UI that is a Phone UI when installed on a phone device, a tablet UI when installed on a tablet device and a desktop UI when installed on a desktop/laptop machine. There will be visual similarities but function will differ according to the screen capabilities of the device. The goal for the Ubuntu converged phone or tablet device is for it to switch modes when it is connected to an external screen, keyboard and mouse.

And application developers are being encouraged to write code that will allow their apps to morph from a phone app to a tablet app and to a desktop app depending on the mode the device is running in. That surely requires a well thought out code structure. And an SDK and APIs that make the developers task easier rather than harder.

Consider this actual situation. I run a certain Android app on my Ubuntu desktop. The developers provide versions for IOS, Android & Windows phone but not for Linux. I use Google Arc Welder to get this app running on Ubuntu desktop. But it still remains what it is designed to be and that is a tablet application. For example, when I switch to using an other application the Android app greys out. This is necessary on a mobile device to save battery power but not necessary on a mains powered desktop machine. It makes it difficult to read the information in the app. And as this app is a research library I am having to squint to transfer information from the Android app into a LibreOffice document.

The app designers did not think of this use case. How could they know that I would come along and start using it on a Linux desktop OS? I do not know how to do it but I do think that it is possible to package this android app as a Ubuntu snap. It is after all open source. But it would still be a tablet app which is useful for running on a Ubuntu tablet and it would also run on Unity 7 & Unity 8 but it would not be a converged app unless there is a re-write.

Regards
Regards.

[T.J.]

I do not see Canonical's vision for convergence to be that strict. As I understand the design philosophy it is to write a UI that is a Phone UI when installed on a phone device, a tablet UI when installed on a tablet device and a desktop UI when installed on a desktop/laptop machine. There will be visual similarities but function will differ according to the screen capabilities of the device. The goal for the Ubuntu converged phone or tablet device is for it to switch modes when it is connected to an external screen, keyboard and mouse.

Fair enough.


And application developers are being encouraged to write code that will allow their apps to morph from a phone app to a tablet app and to a desktop app depending on the mode the device is running in. That surely requires a well thought out code structure. And an SDK and APIs that make the developers task easier rather than harder.

You would think so, although I will comment that what is great in theory does not always work well in practice. Code design is far too often dependent on the programmer to exercise restraint, which was my earlier point. Windows in particular, encourages bad judgement by giving less experienced programmers tools that encourage them to hedge code into event driven blocks based on an interface. Frankly, it would be fine if they exercised some discipline when designing, but they don't.


Consider this actual situation

Just a heads up for future reference, when speaking of code, it's not "actual" to a programmer unless you have "actual" code or an working app to demonstrate objectively. Please understand that I don't doubt your word, but like most sciences, when we throw around "actual situation" in computer science, we like third party observation and confirmation. =)


The app designers did not think of this use case. How could they know that I would come along and start using it on a Linux desktop OS?

Naturally, they didn't. You are not running it in the environment it was intended for. Nor would it have helped if they had. When designing software, you generally design it for a specific operating environment. Just because it is possible for someone to run it in emulation, does not make it a valid use case. If it did, the world might be a happier place. WINE would certainly work better.


I do not know how to do it but I do think that it is possible to package this android app as a Ubuntu snap. It is after all open source..

I have my doubts, but I have been wrong before.


But it would still be a tablet app which is useful for running on a Ubuntu tablet and it would also run on Unity 7 & Unity 8 but it would not be a converged app unless there is a re-write

Which is exactly my point. What good is the "convergence" concept if you have to do a rewrite for each system anyway? In the real world, it is likely that you will have different versions for the sake of different hardware (mouse versus touch) even on Ubuntu.

Now the best possible ideal would be something that abstracts the mouse/keyboard/touch but in practice you run into certain issues, such as display scaling. In the best methods, you design separate interfaces and reuse as much backend code as reasonable. It's easy enough to change the frontend on the fly, while keeping the rest of the program held in RAM - but only if you separate the logic from the interface - as I mentioned previously. You also have to depend on the operating system or some form of threading in the SDK to make the necessary real time determinations as to which interface is being used (should you unplug the keyboard and "go touch" for example).

This doesn't mean that you can't have what you see as convergence, but it does give us something we both want. You want the right interface, I want an end to certain kinds of "sloppy" programming.

[vasa1]

..., I only use ones that I feel have the least likelihood of being used as a legal weapon. As far as I am concerned, that pretty much eliminates the use of Java, Go, Rust and some others. Their sponsors - Apple, Google, and Oracle are fond of suing everyone in sight. ...
Hi, can you link to a few cases in which Google has sued someone in the context you describe? Thanks!

[grahammechanical]

but only if you separate the logic from the interface - as I mentioned previously.

I am not a programmer. I cannot speak as a programmer but I do see your logic. We should not need convergence, however we think of it, to force programmers to write structured code (keeping code interfacing the OS distinct from code interfacing the user, as an example). The programmer should be writing structured code anyway. Following Best Practice often has practical benefits. I imagine that an app developer who is a sloppy programmer will have an up hill struggle porting his apps to other platforms.

Regards.

[T.J.]

Hi, can you link to a few cases in which Google has sued someone in the context you describe? Thanks!

It's not a lawsuit per se, but how about forcing Acer to cancel its plans to use Aliyun OS? Google used contractual obligations; threatening to boot them out of the Open Handset Alliance for using a what Google considers a fork of Android. All OHA members are forbidden from using unauthorised versions of Android. Google's lording over Android handset makers is one of the reasons Samsung has developed Tizen as an alternative.

http://arstechnica.com/gadgets/2013/10/googles-iron-grip-on-android-controlling-open-source-by-any-means-necessary/

Android's code is open, but the Android platform is not. It doesn't matter of you can run the app and the version you are using is a legitimately purchased device. It is not legitimate unless Google says it is.

Who does that remind you of? Personally, it reminds me of Microsoft.

[vasa1]

It's not a lawsuit per se, ...
I took your words literally. My bad.

[user1397]

It's not a lawsuit per se, but how about forcing Acer to cancel its plans to use Aliyun OS? Google used contractual obligations; threatening to boot them out of the Open Handset Alliance for using a what Google considers a fork of Android. All OHA members are forbidden from using unauthorised versions of Android. Google's lording over Android handset makers is one of the reasons Samsung has developed Tizen as an alternative.

http://arstechnica.com/gadgets/2013/10/googles-iron-grip-on-android-controlling-open-source-by-any-means-necessary/

Android's code is open, but the Android platform is not. It doesn't matter of you can run the app and the version you are using is a legitimately purchased device. It is not legitimate unless Google says it is.

Who does that remind you of? Personally, it reminds me of Microsoft.
Thanks for your long clarification earlier (and no need to apologize, the more thorough the better!)

I totally agree on this point you made, that android's open source but the platform isn't (or namely, google apps (gapps)) which means google basically does what it wants with android and no one bothers to use AOSP as their main OS because compared to google's android it is quite unusable.

In fact, I think that was one of the main points of the original article I posted in this thread, that google was right in making android an open platform at first to sort of get it quickly out there in the world to compete with iOS. But now given the fact that android is so established and in fact dominant, google is trying to close off more and more parts of android so as to have more and more control. In fact, they have been steadily doing this as more AOSP apps are abandoned in favor of gapps. And yes I agree, this is very Microsoft-esque, but hey what can you expect from a multi billion dollar company.





Now back to your other point, I fail to understand why you are so resolute in discarding de-facto languages. Although everything you said makes sense, and there really is no ultimate protection from using any of the de-facto languages as opposed to standardized ones, you can also clearly see in the world around us how thousands or millions of people are using these languages and for the most part there has not been a single incident of the type which you describe.

I get what you mean by the backwards incompatibility of python and perl, and how changing to the newer versions may have cost time and money to certain developers, but for the most part if a company depended on say python 2 and saw no need to change, then they are still using python 2 and won't stop until it is unsupported. But I don't see the python foundation or the rust foundation (which I think was set up by mozilla since rust originated there?) either pushing people around or getting pushed around themselves.

The only de facto language that I see has huge legal implications is Java because of its complex situation, the others I don't see why anything could go wrong.

But please by all means, if you want to elaborate or answer any of this please continue, I find this subject fascinating :)

[mikodo]

^^ If you look back in your thread, you can see the why to your questions.

In a broadly defined sense, he is advocating for openness, compatibleness, standardization, while applying discipline to his professional work and advocating for the same. He states very important reasons for these needs in his examples. There is a saying by George Santayana that is seemingly apt here. “Those who fail to learn from history are doomed to repeat it”.

[T.J.]

Thanks for your long clarification earlier (and no need to apologize, the more thorough the better!)

You're welcome.





Now back to your other point, I fail to understand why you are so resolute in discarding de-facto languages. Although everything you said makes sense, and there really is no ultimate protection from using any of the de-facto languages as opposed to standardized ones,

If I understand what you mean, you equate the risk as equal. I disagree. There is far better protection with ISO standards. That is why ISO standardization requires a RAND grant. There is also the fact that the ISO is composed of experienced delegates from member countries and the general weight of other international agreements. Suing over an international agreement is much harder than a simple contract - which is ultimately all that a FOSS license is. I understand the confusion. I could say "trust me", but (and I mean this with the greatest respect) the truth is that there are some things you have to experience for yourself before you understand why things are done a certain way.


You can also clearly see in the world around us how thousands or millions of people are using these languages and for the most part there has not been a single incident of the type which you describe.

That was what Google vs Oracle was all about. Now that U.S. federal courts have upheld the right of owners to copyright APIs, using de-facto languages leave you even more vulnerable, legally speaking. ISO specifications insulate users from frivolous lawsuits regarding the language. FOSS licenses in general do not. It is true that we have seen very few lawsuits of this type, but I do not believe for one minute that we will not be seeing more of them. If you have done this as long as I have, you start assessing what risks are worth taking long term, because if things change you might have to embark on a very costly rewrite to stay on the right side of the law. The simple truth is that I have no way of knowing if it will happen - that is very true - but is also true that if I elect to use a language covered by the ISO, then I don't have to worry about it.

This is not the first time things like this have happened.


I get what you mean by the backwards incompatibility of python and perl, and how changing to the newer versions may have cost time and money to certain developers, but for the most part if a company depended on say python 2 and saw no need to change, then they are still using python 2 and won't stop until it is unsupported. But I don't see the python foundation or the rust foundation (which I think was set up by mozilla since rust originated there?) either pushing people around or getting pushed around themselves.

Probably not, but without the protection that the ISO provides, the potential is there. Patent trolls and copyright lawyers will not usually go after new software. They wait until they are sure they can extract the largest monetary benefit from the most people.


The only de facto language that I see has huge legal implications is Java because of its complex situation, the others I don't see why anything could go wrong.

The question you should ask yourself is do you feel comfortable with it. If you do, then by all means proceed as you think best. I don't feel comfortable with it. I like the legal guarantees that ISO standardization provides, because I have seen how people try to work the system for their own benefit.

All that it would take for a developer to lose money is for a patent or copyright suit to be filed against the projects, and everyone would have to re-engineer their code. It wouldn't matter necessarily if they win or lose, just the uncertainty is enough to damage a community, and make development costly. If you do not believe me, look at the history of BSD. BSD was here before long before Linux. Had BSD/i386 not been under legal dispute in the early 90's - it is quite likely you would be using BSD right now, and Linux would have never existed. (Consequently, the BSD license would have shaped the FOSS movement rather than the GPL,) Ultimately, the legal problems were solved, but BSD missed its chance. Linux had beat it to the punch.

[user1397]

snip
Fair enough T.J. I understand your points and although I may not totally agree I see why you're saying what you're saying.

Now as the old saying goes, "beggars can't be choosers" which I think applies to this if you start thinking about what real companies are looking for. Many companies look for Java and Python (and other de-facto languages) developers and most people will use whatever tool is most in demand so that they can make a living. If you can avoid all these major languages and still make a living (not saying specifically you, just mean in general) then kudos to you but many people can't avoid them.

Also, I hate how so little meaning gets conveyed through text and writing. No vocal tonality, no body language. I'm sitting here typing this and reading what I'm saying and can't help but think I sound like a jerk in certain parts, but I swear I'm really not! Haha :)

[grahammechanical]

I hate how so little meaning gets conveyed through text and writing. No vocal tonality, no body language.

Ah! Yes. With spoken speech a person does not need to be clever with words to be rude or insulting. It is all there in the tone & the body language. With written speech a cleverly constructed insult could be taken as a compliment and an awkwardly phrased compliment taken as an insult.

It is often a matter of the difference between "imply" & "infer" that causes arguments.

Regards

[T.J.]

Now as the old saying goes, "beggars can't be choosers" which I think applies to this if you start thinking about what real companies are looking for. Many companies look for Java and Python (and other de-facto languages) developers and most people will use whatever tool is most in demand so that they can make a living.

Oh absolutely. At no point did I mean to imply otherwise. I meant that if I have a choice, I avoid them. Fortunately, as a contractor, I usually have that option, but not always. We've all had our jobs where we had to take what we were given and make the most of it. I am no exception to that rule of life, my friend.

There is no right or wrong here, and I am certainly not trying to persuade you that de-facto languages are "bad" or that everyone has bad intent. I do think that it is very important that the ramifications of the legal system and the licenses we use, be discussed openly. I am advocate for the ISO primarily because it offers the best protection for everyone equally - high or low. Companies or sponsored foundations who control de-facto languages are not "evil", but they can and will serve their own interests ahead of everything else, even at the detriment of their "community" processes or even the licenses they use. Imagine how Google vs Oracle is going to affect Java, and the community; even though it is GPL. I'd say at the very least people are asking if the GPL is enough to protect them. Sadly, in the U.S, it isn't anymore.

There is something to be said for looking at the situation with reasonable caution and skepticism. All I am suggesting is to look at the situation objectively and provide yourself with the greatest latitude for long term investment, software maintenance and frankly: career safety. That is something that you (and only you) can judge for yourself. When it boils down to it, if the legal "crap does hit the fan" the man (or woman =) ) who chose the tools in the design process takes the heat. If that happens to be you, I sincerely hope that the discussion is useful someday.


If you can avoid all these major languages and still make a living (not saying specifically you, just mean in general) then kudos to you but many people can't avoid them.

All too true, and I have used most of them. If I might say so, I've found most of the "de-facto" language standards to be terrible. Of course, that isn't saying much. The ISO C++ 2011 and 2014 standards are two of the most awful, disgusting, convoluted messes I have ever met. That is an opinion, not a fact. I'm "old school" and come from the days when C was considered mandatory. If you were into UNIX, you knew C. I consider C to be a clean and well designed language for its intended purpose: replacing assembly. I find that far too many languages are designed without long term use or a clear purpose in mind, so you end up with constant compatibility problems when the next version is released. That in itself doubles my burdens because someone has to maintain the software for it to be useful.



Also, I hate how so little meaning gets conveyed through text and writing. No vocal tonality, no body language. I'm sitting here typing this and reading what I'm saying and can't help but think I sound like a jerk in certain parts, but I swear I'm really not! Haha :)[

OMGosh, yes! I hate that myself. I try to be very precise as a counter-balance, but the truth is that lack of sleep sometimes works against me, as it does this morning.

[ventrical]

I am not a programmer. I cannot speak as a programmer but I do see your logic. We should not need convergence, however we think of it, to force programmers to write structured code (keeping code interfacing the OS distinct from code interfacing the user, as an example). The programmer should be writing structured code anyway. Following Best Practice often has practical benefits. I imagine that an app developer who is a sloppy programmer will have an up hill struggle porting his apps to other platforms.

Regards.

@graham

We both know that convergence is needed and that is the set track that ubuntu is on. If you really want to test this out then just try the android-x86_64-6.0-rc1.iso from http://www.fosshub.com/Android-x86.html on an amd64 non-touch desktop. There is some movablility in Intel Xeon5thGen machines and the installer greeter works and it will install on Intel based graphics but there is no network or tweak mouse gestures. So why are they advertising it for desktop use without a real network (like unity8 has).

and try running that on your nVidia device. Then when you look at how far unity8 has come (and Canonical's willingness to fix these code_tag bugs) and the option to fall back to unity7 set .. I hope anyone would see the clear and conventional wisdom, especially with Libertine containers. Canonical (ubuntu) is lightyears ahead of this game. Getting into a code spraying contest is futile at this point. I don't see the reason for it. But, after some research, I will concede that perhaps android does possess some of the hallmarks or 'fragmentation' but I can't really comment in a precise way if open source is 'killing' Android. One can't just "crystal ball' or conjure up a smack comment considering all the data and history involved... it is actually more complex than that... but to each their own.

edit:

actually reported there are cases where it works on nVidia graphics. Touche'

Regards..

[mikodo]

^^ Good!

I am reminded by this thread of a post Geoffrey_Arndt once made.

http://ubuntuforums.org/showthread.php?t=2297282&page=9&p=13378142#post13378142

[grahammechanical]

If you really want to test this out then just try the android-x86_64-6.0-rc1.iso

@ventrical

I did. I was looking for a way to run that Android app I mentioned earlier.But I could not get android-x86 to install. By the time I found it needed an existing ext3 partition I had grown bored by the failures. And then I discovered Google Chrome Arc Welder. That works with but one complication. Any upgrade of Chrome breaks the relationship between Arc Welder and the Android app. I have yet to find a way to correct this except the bulldozer of a fresh install of 16.04/16.10. For the sake of my sanity I have disabled the Google PPA.

I accept your point that running a mobile OS on a desktop machine is a pointless exercise. We have both experimented with personal_x86 which was Mir+Unity 8 on desktop hardware. I do think that Canonical is going about convergence the right way. They are doing more than squeezing a desktop OS into a mobile device or expanding a mobile OS to fill up a desktop screen. But just they same I cannot help but think of Canonical engineers as being like a person walking in thick fog and groping their way forward with arms out stretched to avoid walking into something. And I am old enough to have lived at the time when there were foggy days in London town. To quote a very old song.

Regards.

[ventrical]

@ventrical

I did. I was looking for a way to run that Android app I mentioned earlier.But I could not get android-x86 to install. By the time I found it needed an existing ext3 partition I had grown bored by the failures. And then I discovered Google Chrome Arc Welder. That works with but one complication. Any upgrade of Chrome breaks the relationship between Arc Welder and the Android app. I have yet to find a way to correct this except the bulldozer of a fresh install of 16.04/16.10. For the sake of my sanity I have disabled the Google PPA.

I accept your point that running a mobile OS on a desktop machine is a pointless exercise. We have both experimented with personal_x86 which was Mir+Unity 8 on desktop hardware. I do think that Canonical is going about convergence the right way. They are doing more than squeezing a desktop OS into a mobile device or expanding a mobile OS to fill up a desktop screen. But just they same I cannot help but think of Canonical engineers as being like a person walking in thick fog and groping their way forward with arms out stretched to avoid walking into something. And I am old enough to have lived at the time when there were foggy days in London town. To quote a very old song.

Regards.


If it seems like I am frantically boosting convergence and unity8 is it because I am doing just that! and it is not because of snappy(but snappy will play a part) - it is because of lxc containers (now libertine and xapps). You had posted a link in U+1 from M.Hall and his dogfooding experiences with unity8. I took up that challange and have had some success with how Libertine Containers can work on Unity8. I recall the time we had spent experimenting with lxd/lxc and you mentioned that you would be happy if you could just get firefox to run on one container (which we were using snappy-core and lxd on that). I could get graphical htop and midnight commander to run in containers but not standalone firefox. However, I say again, that I was able to run two successful sessions of firefox from two completely different containers while in the unity8 desktop so I am concurrently running unity8 on xmir and also xapps on xserver in mutiple containers. So Mark was right about this little hypervisor. I was drawn to this like a moth drawn to a candle flame. I mean where else would a beta tester who tests and breaks machines want to be?

Now as for snappy and snappy personal .. I am still holding out for good news. I watch mail from the snapcraft mailing list and they are full steam ahead there. and I think it was Mark who said at one of his keynotes that developers would be also able to convert click into snaps and so it may be that many of those android apps may be available in 16.10 . Think about that :) I am. And as for snappy and xmir being proprietary , the benevolence of ubuntu/canonical/community has far outweighed anything they have ever extracted and perhaps this is a small price to pay for the security that the snappy repos will offer.

Regards..

[ventrical]

^^ Good!

I am reminded by this thread of a post Geoffrey_Arndt once made.

http://ubuntuforums.org/showthread.php?t=2297282&page=9&p=13378142#post13378142

Thanks for that . Microsoft has a lot of good people. They have worked a successful business model. I personally have had nothing but angst with their former security models. For almost 20 years I have had to backstep my own programming career to focus on malware removal for microsoft products. Then I found Ubuntu and it , well , set me free from malware removal. So now I am moving forward with the convergence. I like it. It is fresh and I know it will work. Taking backsteps into that other genre' only causes me anxiety and trepidation. I am moving forward where angels fear to tread :)

regards..

[user1397]

Wow, just found out about this project today, extremely relevant to this thread: https://microg.org/

Says it's a "free-as-in-freedom re-implementation of Google’s proprietary Android user space apps and libraries."

About:
The linux-based open-source mobile operating system Android is not only the most popular mobile operating system in the world, it’s also on the way to become a proprietary operating system. How is that?While the core operating system is still released as part of the Android Open Source Project, the majority of core apps is not. It even got worse: More and more libraries and APIs are only available on phones that run various Google apps pre-installed, effectively locking third-party apps to the Google ecosystem. For these reasons Android is called (http://arstechnica.com/gadgets/2013/10/googles-iron-grip-on-android-controlling-open-source-by-any-means-necessary/4/) a “look but don’t touch” kind of open.
By now, several popular open-source applications like the secure messenger Signal (https://github.com/WhisperSystems/Signal-Android) already require some of Google’s proprietary libraries to be installed. Increasing demand in the free software community in addition to severe problems in Google’s proprietary software discovered by the Android modding community led to the development of a free software clone of Google’s proprietary core libraries and applications - the microG Project was born.
Although most microG components are far from complete, users are amazed by the results. Free software users got extended application support, privacy-caring users can reduce or monitor data that is sent to Google and especially older phones can expect some battery life improvements. microG is not only used on real devices, but also replaces Google tools in test emulators and is even used in virtual mobile infrastructure.

[bapoumba]

Test ...

[grahammechanical]

I doubt this:


More and more libraries and APIs are only available on phones that run various Google apps pre-installed, effectively locking third-party apps to the Google ecosystem.

Every OS has libraries and APIs that apps make use of. It means the app as it was written can only be run on that OS. That is not lock in. There is nothing stopping third party app developers from writing apps that run on other platforms. Or, from porting their apps to other platforms.

Android apps do not natively run on Linux. You can call that lock in, if you want. I call it, different operating systems. If an app developer wants to make use of a Google customer service and that limits the app to Android devices that have that service activated, then that is the choice of the developer. And it is also the choice of the user to activate these services and install to the app.

Regards

[bapoumba]

Test

[user1397]

@grahammechanical

I get what you're saying, but we can't act like operating systems are on a level playing field. Android and iOS but especially Android have hegemony in mobile (at least world wide; it's more 50/50 in the US), so it's not like a developer will opt to not release an Android version of his/her app just because they don't like a Google library or API. I mean, sure they of course can, but most won't because...money.

[T.J.]

@grahammechanical

I get what you're saying, but we can't act like operating systems are on a level playing field. Android and iOS but especially Android have hegemony in mobile (at least world wide; it's more 50/50 in the US), so it's not like a developer will opt to not release an Android version of his/her app just because they don't like a Google library or API. I mean, sure they of course can, but most won't because...money.


I'm waiting for the day that the U.S. government threatens to sue Google under anti-competition statutes.

Under the Google's agreements, handset makers cannot use derivatives of Android with Google apps - which is clearly an anti-competition measure designed to keep phone makers under Google's thumb. If they leave the Open Handset Alliance and use an Android derivative, Google can use its position as the main supplier of applications to force them back into line. Fortunately for Google, Amazon has enough sales diversity in its own ecosystem to ignore Google and host their own store. That in turn keeps the lawyers off of Google, but I suspect the day will come soon when Google pushes too hard.

I expect then that the current Android ecosystem will shatter when handset makers and developers start pushing back. Samsung has been developing its own OS, Tizen, so that it does not have to be beholden to Google.

I don't think that will be a bad thing because an open API for the applications has already been established. No one cares who makes the apps as long as they work. As far as I am concerned, provide the API for everyone and let them use what they want. It's actually a derivative of Java anyway.

[user1397]

@T.J.

Speaking of which, did you see my other post about the new project microG?

https://ubuntuforums.org/showthread.php?t=2328518&page=7&p=13516968#post13516968

[T.J.]

@T.J.

Speaking of which, did you see my other post about the new project microG?

https://ubuntuforums.org/showthread.php?t=2328518&page=7&p=13516968#post13516968

Recent developments in legal cases, such as Google vs Oracle - have established ownership of an API. Given the current status quo, it is likely that microG is going to "whither on the vine" before it ever reaches fruition. All significant handset makers have commercial interests and are likely not to touch it for fear of Google sending them a "cease and desist" letter as soon as they put their money behind a commercial prototype that uses microG. Worse, Google could revoke their OHA license for Android in retaliation for using an unauthorized version of Android.

It will take a shakeup of Android's legal status before real progress can be made in adopting more open alternatives. There is just too much money invested in the status quo. This is exactly what Google intended from the start. Because the code is open and a few makers go in a different direction, Google can claim it hasn't a monopoly - but because of the OHA agreements it does - for all intents and purposes.

That is why Samsung's Tizen is a completely different system. Google can't punish Samsung for developing, using it, or for licensing it to others. Conversely, Samsung is licensing Tizen just as poorly. Tizen is built with FOSS, but the SDK is proprietary. Sounds familiar doesn't it?

I'd say at this juncture, the only FOSS phone/tablet you will have is one you hack yourself, using a few proprietary blobs as required - just as you do with your Linux PC.

Personally, I think the best way to get Google is to get a partnership of handset makers to start a own "App Stores" around a common API/VM that can be used on any of their phones, and then license that as FOSS forever. Once developers realize they could sell apps to all users, away from the walled gardens with a less restrictive license, and get higher profits, I think Google would reconsider their position. Then handset makers can compete by offering better hardware in much the same way that PC makers once did.

[ventrical]

I doubt this:



Every OS has libraries and APIs that apps make use of. It means the app as it was written can only be run on that OS. That is not lock in. There is nothing stopping third party app developers from writing apps that run on other platforms. Or, from porting their apps to other platforms.

Android apps do not natively run on Linux. You can call that lock in, if you want. I call it, different operating systems. If an app developer wants to make use of a Google customer service and that limits the app to Android devices that have that service activated, then that is the choice of the developer. And it is also the choice of the user to activate these services and install to the app.

Regards

And then there is aptoide.

Which I have slates that are non-google locked.

https://www.aptoide.com/?lang=en

The other alternatives will bend Google Corp. to toe the line. Bill Gates, Steve Jobs (RIP) nor Lawrence Ellison own the internet. The internet belongs to the people that use it. All else are service providers and they better get used to that handle.

Regards..

[jasonstathan022]

In my opinion nothing is killing Android. On the contrary, Android is thriving and continues to eat iOS and Windows market share.

[sp40140]

Android is doing just fine. It could be doing much better, but that's the price google agreed to in the beginning.
Apple strong armed the wireless industry when they released the iPhone (they were in position to do so because of the popularity and built up excitement and demand). Apple set the terms.
Now when google wanted to get into this market, they had to give concessions to the service providers and OEMs so that they use android. I don't think android / google had any choice if they wanted to compete. (That's how MS beat Apple years ago).
But these concessions caused the fragmentation. It's not google's fault.
If it wasn't for google / android all of us would be using iPhone and paying ridiculous money for it.
So the problem exists, but it was a bargain. Google is doing best it can.
Also, they are not charity, they deserve to make money from their efforts. Nothing wrong with it.

[Kris_M]

I don't see a problem with Android on phones - Have had many of them - all custom rom'ed. Windows phone once a long time ago. I am currently running LineageOS (used to be "CM")(close to 2 million users I believe) on my Moto G3 (2015) - we get monthly security updates and have Nougat 7.1.2 . I flash weekly to help the dev path and to have fun. The things folks have filed bugs on, for the most part, don't affect me. Again, it is for fun. I ran 6.0 stock on that phone for a year with nary an update. Didn't really expect one. Just got bored.

On the other hand, if Google were to take it over in such a way that any phone could get monthly updates, and subsequent releases, I'd be all for it. The catch is the release changes - sometimes it is hard to run the latest release on a 2 year old phone. So would they give us security patches forever (unlikely). So plan on replacing your phone every 1 1/2 years?

It would certainly change the landscape and probably spawn a lot of non-google android-like op-sys-es. There would be thousands of devs out there with nothing to do. And probably a ton of various FoxCom phones by then to run it all.

[zer010]

Jeez. I read about 7 pages of this thread, happily so, until I realized this was a year old! :p
I was wondering why there were so many references to Ubuntu phone. I could've sworn that and Unity
were nixed a while back... unfortunately.

[simplydt]

From my experience Android is only a problem on low spec phones! But as far as I've seen Android continues to grow and grow so despite any problems it may be having, overall it seems to be more than ok.

[lisati]

From my experience Android is only a problem on low spec phones!

True. We have two android phones that we had to retire. They were fine when manufactured, but a year or so after we bought ours, Google made some changes which, by hogging a sizable chunk of the available space in the phone's internal storage, effectively made it impossible to update apps or install new ones. Thankfully our home's current crop of smart phones cope a lot better.

[simplydt]

True. We have two android phones that we had to retire. They were fine when manufactured, but a year or so after we bought ours, Google made some changes which, by hogging a sizable chunk of the available space in the phone's internal storage, effectively made it impossible to update apps or install new ones. Thankfully our home's current crop of smart phones cope a lot better.

Yeah its ironic isn't it? A phone whose specs would perfectly handle all software just a couple of years ago is now a major pain because today's software just eats more and more resources. Just another way to force us to spend more money upgrading our phones. After all who can handle the frustration of an unresponsive device. I know I can't!

[T.J.]

Yeah its ironic isn't it? A phone whose specs would perfectly handle all software just a couple of years ago is now a major pain because today's software just eats more and more resources. Just another way to force us to spend more money upgrading our phones. After all who can handle the frustration of an unresponsive device. I know I can't!

Agreed. The phone makers add more crap with each new generation because they need to force people into buying new and faster hardware to keep their profit margins and market share. The ownership is pretty much saturated, so the only way to keep people upgrading is to tweak the software.


From a programming standpoint, what do you expect from an Android ecosystem that uses a Java derivative? IMHO, OOP and byte-coded VM languages programming languages are just plain resource hogs. The more you ask them to do, the more grandiose the objects in the software get to maintain and the slower they become. What is really needed is a phone OS that imposes a standardized C API that is lightweight but familiar, such as a subset of POSIX.

[exhile]

Being closed source is killing iOS from being the mobile operating system with the largest user base. Not to mention that Apple charges a premium price for their devices which increases the value of Apple shares. As of May of this year, Android has over 2 billion active devices worldwide so I'd say open source is not killing Android.

Market competition between iOS and Android is a benefit for consumers. The latest innovations and technologies by each mobile operating system will compete for your dollar. If one mobile O/S upgrades to a new version, the other competitive mobile O/S will release a new version. Variation between each mobile O/S product offerings allow customers to decide which product is right for them. Mobile payments, facial recognition, button-less screens, wireless charging, smart watch integration, blue-tooth earphone connections, 4K video recording, etc.... are all improvements in each yearly smart phone generation.

I can't wait until blue-tooth connected Augmented Reality glasses get offered.

[KD5NRH]

I don't see a problem with Android on phones - Have had many of them - all custom rom'ed.

Therein lies the problem, and most of it isn't Android itself; it's OEMs doing stupid things (like pointing /SDCARD to the internal memory so many apps fill that up while ignoring your 64G MicroSD) and then locking it down so the typical end user can't fix the stupid.

For other devices it's not so bad; I don't need a custom kernel to install...well, anything I've found myself needing or wanting. But if I want to put a few apps with massive locally stored data (podcasts, etc.) on my phone and it's making those apps think they're saving to my SD card while redirecting that to internal memory, now I've got to root the phone and start messing with the guts of the OS. Nor do I need to do anything special to get rid of all the useless crap installed with Ubuntu other than just apt-get remove, while there are apps clogging up memory and burning bandwidth on updates on my 3 year old phone that I've only opened by mistake (PITA to root, so I do whatever I can on my CyanogenMod equipped Kindle Fire.) and yet I can't get rid of them without extraordinary measures.

[ulysses77]

Being an Android dev, I can safely say that Android's open source nature with the backing from grand overlord Google makes it insanely powerful. If desktop Linux had a powerful direct backer like Google in the past, all of Linux would have enjoyed the prominence that Android Linux has. Could you imagine? That would be an awesome world.

Saying Android would be better off closed source opposed to open is like saying "kiling net neutrality is a good thing!"

Ha, nope.

[ulysses77]

My man! Custom ROMs! You know what's up! I miss the days of CyanogenMod, I sunk so much time into that. I haven't looked into LineageOS, is it good? Another thing that's good to do, may be a little more expensive outright, but to buy an unlocked phone or a Google direct phone (Pixel, Nexus). No bloat or company crap on those. A lot of custom ROMs have more security and privacy features though.

On the other front, yeah, custom ROMs can greatly extend the life of your phone that's for sure. After 2 years usually the hardware is too underpowered to support the new stuff. That's the problem with older hardware in the mobile world. Mobile tech is evolving like mad so a lot gets left in the dust. I usually find a purpose for older phones though, they make awesome media players.

[jee.smith]

In fact, Android is not an opensource OS. Today most of all devices work on custom OS build upon Android and they are not opensource even possible to use without a device.

[faraco2]

What is your opinion of a free alternative Android operating system? https://www.replicant.us/

[kurja]

In fact, Android is not an opensource OS. Today most of all devices work on custom OS build upon Android and they are not opensource even possible to use without a device.

Huh? Android is very explicitly an open source project - which means that OEMs are free to modify and use it in their products.