Ubuntu Developing Its Own Calculator, Calendar, Etc

[GreatEmerald]

Second mouse hangs... Well I can tax my CPU 98% (video conversion) on 4 cores and still have a responsive desktop, I am sure there are cases where this may not hold up but I never have that issue. I remember taxing single core systems (even on Windows) and having the computer laugh at you when you tell it to do anything but those days are gone.... I hope. My phone has a dual core ffs.
If it is a ram issue, that maybe plausible. It's well known you don't want to use swap if you can avoid it though (on any operating system). I suppose if you have a SSD this maybe a non issue. Via your description sounds like it's a combination of swap usage and CPU taxing.

Yeap, mouse lag is squarely due to no free RAM left. It used to happen to me while converting 1080p videos from near-lossless to something more sane, and since the source files are humongous (only a single 1.5 minute video clip fitted in my entire RAM - now I could fit in twice that!), running out of RAM is not that uncommon. Doing something of an opposite - dumping raw video - can also cause that, as the data the CPU outputs cannot be committed back on the HDD fast enough, thus it stays in RAM until there is no more space (at which point the transcoder locks up or starts running super slowly). But, of course, that's not something usual people do often.

[Hamish Wilson]

Sorry, you're drawing lines in the sand which don't exist. Android uses the Linux Kernel. The fact they have their own patches which are not upstream doesn't change this.

I was under the impression it was now a completely separated code base with some interchange of code between the two projects. Still, what it takes to become a true fork is somewhat poorly defined.

Plenty of embedded devices ship with a minimal, customized kernel and minimal userland binaries. OpenWrt barely has a userland, is compiled with uClibc, and is still considered "Linux".

Again, the definitions are vague. Regardless, Ubuntu has utilized a lot more parts from other distro's than Android.

[jayrulez]

I was under the impression it was now a completely separated code base with some interchange of code between the two projects. Still, what it takes to become a true fork is somewhat poorly defined.

This seems to be very prevalent on phoronix. People make claims and statements based on feelings and things they think they know but don't really know rather than on facts.

Ubuntu has utilized a lot more parts from other distro's than Android.

No one said otherwise.

[Hamish Wilson]

This seems to be very prevalent on phoronix. People make claims and statements based on feelings and things they think they know but don't really know rather than on facts.

Well, in that case I am going to need to ask for a citation from you then.

[jayrulez]

Well, in that case I am going to need to ask for a citation from you then.

A citation for what?

[TheBlackCat]

Well, in that case I am going to need to ask for a citation from you then.

http://lwn.net/Articles/481661/

The android patches change a total of 7000 lines of code, compared to a total of 15 million. That is about 0.05%, and is as many lines of code changed in total as Red Hat has patches.

https://lwn.net/Articles/514901/

Android patches are being merged upstream. You can boot android with a regular Linux kernel as of 3.3, and a lot of other things have been merged since then. They are planning to and are on track to merge all of their changes.

[Hamish Wilson]

Alright, thanks for the links.

[kotakotakota]

Perhaps this is a bit of a silly argument, but am I the only one here who thinks that those new apps just look really ugly? :P

Joking aside, I usually appreciate it when a group tries to make a nice, uniform experience the way Canonical is trying right now. But, I don't trust they have the technical background for it. Their reasoning for doing things seems a little lackluster, almost spur of the moment. For instance, Mir... The developers even claimed that they don't fully understand Wayland, yet they dismissed it. Unity, Mir, Bazaar, Launchpad (though I appreciate this one), and countless other projects that are under the Canonical umbrella directly compete with existing, perfectly viable solutions. Mark Shuttleworth also loves to exaggerate and make up statements as well, and that definitely does not help their case in my opinion. And many of the decisions that Canonical has made have large potentially negative impacts to existing projects and communities. Besides, we already have enough trouble with drivers and such... Now graphics drivers will have to support both Wayland AND Mir... Though Canonical employee Christopher Rogers suggests that that isn't too much of a problem: https://plus.google.com/113883146362...ts/QwMqCgC7c9G

Back to the whole unified experience thing... While I know that the apps that they are writing are not very difficult to write, the state that they tend to release things is also worrisome to me. Remember when Unity was first released? The shock of that initial experience was enough to keep me from appreciating the interface today still. Perhaps it is just me, but I feel Canonical enjoys calling beta or sometimes even alpha quality software stable. Hope they don't do the same here... Also, they can do what they want with this but I don't really see much of a point, writing so much featureless software...

[smitty3268]

I never said it made sense. Just that I had no reason to care.

I think this makes sense, because obviously Ubuntu is going to want to create a polished user experience, and if everything they are doing is in QML then it probably makes sense for that reason to have all the basic desktop apps also written in QML. They can't go elsewhere for them - KDE apps will have their own look-n-feel, which won't match with Unity. And it sounds like these apps are going to be relatively simple - something i feel confident Canonical can handle well enough.

Mir is completely different, because there is no real goo reason for it - to the end user, it should be invisible, and they shouldn't be able to tell if the OS is running Wayland or Mir. The only issue is if 1 system actually has more capabilities of some sort, which it sounds like won't be the case. Also, i have no confidence that Canonical has the resources to successfully pull off a complicated low-level project like Mir. Certainly not within the short timeframe they've specified.