Thoughts in the margins and a concise translation of Ubuntu Open Week materials
I’ll transfer my comments on this topic to my blog. And we'll see.
Brief background. Aiming to be closer to the people, Canonical hosts UbuntuOpenWeek in a chat format of system developers with interested people. After reading, I tried to give concisely the main thoughts that were raised in the live communication of developers with the community. Naturally, this is not a literal translation.
The process of developing ubunts is described in an interesting way. Due to the fact that the ubuntu-next git tree was created in Intrepid, we got 2.6.27 and not .26, as it should have been according to the ubuntu development cycle. In a normal cycle, they do not begin work on the distribution kernel while it is under active development. But the .27go advantages outweighed and we all got an improved suspend, wifi, hw support. Among the shortcomings of this approach, the facilitator singled out an exception to the main tree support for sparc and ppc platforms. Soon their fate will be considered and perhaps they will be included again. There will be .28 in Jaunty, but this has not yet been decided.
One of Jaunty's priorities is kernel loading speed. Specific goals are not standing in seconds, the core idea and scope for optimization activities is kernel modules.
They also work closely with XOrg and really want to get a working kernel-mode-setting. For all of this core entertainment, Canonical has just 6 full-time nuclear developers.
Finally, a question was asked about how soon (or far) in the future btrfs awaits us as the main fs. An answer was received that not even 50-50, but 40-40. Where did another 20% go? The answer was witty: either ext5 will be out by then, or Reiser will be released ...
And this is only one meeting at ubuntu open week. and there is also packaging 101, reporting & fixing bugs, kernel bugs, ubuntu vs debian, intrepid & servers, ubuntu & money.
recommend!
This hot English guy is involved in the MOTU (Masters of the Universe) project. The goal of the project is to support the Universe and Multiverse repositories . In parallel, he participates in Debian's projects. And here is what he told the community as part of ubuntu open week:
Debian is a great distribution that many people work on, including and Shuttleworth. When Mark created Ubuntu, he certainly looked at Debian, trying to fix several things in him. The differences affected org. moments, membership in the development team and, of course, in some technical aspects. The difference in the goals of both projects made it possible to concentrate resources on supporting a small number of popular architectures, thereby strengthening other areas of development. After all, support for 11 (well, or how many official platforms does Debian have?) Requires tremendous efforts and slows down the release cycles. The trend in Ubuntu is to choose the best solution for a particular task that really works well, and in Debian, they often try and try many solutions. The efforts, of course, are both true,
The author considers the most important difference that becoming a member of the Debian development team is much more difficult, and for a non-developer it is almost impossible. Perhaps that is why Debian vegetates in terms of "non-programmer" tasks.
To the question - “Are the Ubuntoids so severe that with the Pulseaudio vs Alsa alternative there is only one subsystem left?”, The answer was “yes” :) But with the caveats that this is not a dogma and does not always work, for example, for kde, gnome, xfce - no, but for MTA - yes.
Then they quickly hushed up a rather stupid question about why Canonical is not eager to support third-party derivatives from Ubuntu and moved on to the problem of choice. So, what is better for everyday work - A or B?
The answer is not easy. Many people recommend Ubuntu to beginners because it puts a lot of effort in to working “out of the box”. But it’s best to download the live-cd of both systems and try to choose it yourself.
From the point of view of stability, Debian has advantages due to a longer 18-month cycle, with stable but outdated packages. Here is such a compromise. Ubuntu is in a hurry to bake brand new packages with its 6-month cycles, but for lovers of stability they also have an answer. His name is LTS ...
What does it mean to be based on Debian for Ubuntu? Have their paths diverged since the first historic fork? Of course not! The start of each new release is accompanied by the so-called "merging", the merger of the Debian and Ubuntu branches. This process brings to Ubuntu both new improvements from Debian and errors.
Between distributions there is a mutual exchange of improvements and fixes. Debian receives stable packages, and Ubuntu stops accompanying them and only imports them from Debian back, because they do not have resources for all 15,000 packages.
What does Ubuntu do to refute the prevailing belief that they give little to Debian? Answer. 2 things: give patches to Debian and make the process of smuggling noticeable. Debian Stable practically does not leave anything and everything goes to Unstable, which, by the way, may be a visible factor for prosecutors who claim that Ubuntu takes a lot and gives little. In fact, a system for tagging bug reports and patches from Ubuntu has been developed, and anyone who wants to can see the statistics in the Debian bug tracker. We will take a word for what they give :)
The question of life and death - what will happen if Debian suddenly dies? The answer was not very clear, so the question was asked again and even more insistently. Will Ubuntu have the strength of developers to start autonomous navigation having, for example, 8.10 and that's it. Or everything is bad and the project is so dependent that it cannot be developed, at least in terms of security fixes. Answer: It takes a lot of developers to release the release, and there it will be seen.
Then they praise Launchpad, praise each other for bug fixes and ask to send fixes not only to Ubuntu, but also directly to the authors of the programs. There is also a big request - to monitor bugs both in the Ubuntu bugtracker and in Debianovsky. They have a vision of the issue that linking reports between systems is very important. We wish them good luck on this, I hope the topic of the coexistence of Ubuntu and Debian has opened up for us as well.
I do not think that many have read here, but if so, thank you, leave comments and help transfer to the section on Ubuntu.
Brief background. Aiming to be closer to the people, Canonical hosts UbuntuOpenWeek in a chat format of system developers with interested people. After reading, I tried to give concisely the main thoughts that were raised in the live communication of developers with the community. Naturally, this is not a literal translation.
Noise and din in the “class” Kernel . Speaker: Ben Collins .
The process of developing ubunts is described in an interesting way. Due to the fact that the ubuntu-next git tree was created in Intrepid, we got 2.6.27 and not .26, as it should have been according to the ubuntu development cycle. In a normal cycle, they do not begin work on the distribution kernel while it is under active development. But the .27go advantages outweighed and we all got an improved suspend, wifi, hw support. Among the shortcomings of this approach, the facilitator singled out an exception to the main tree support for sparc and ppc platforms. Soon their fate will be considered and perhaps they will be included again. There will be .28 in Jaunty, but this has not yet been decided.
One of Jaunty's priorities is kernel loading speed. Specific goals are not standing in seconds, the core idea and scope for optimization activities is kernel modules.
They also work closely with XOrg and really want to get a working kernel-mode-setting. For all of this core entertainment, Canonical has just 6 full-time nuclear developers.
Finally, a question was asked about how soon (or far) in the future btrfs awaits us as the main fs. An answer was received that not even 50-50, but 40-40. Where did another 20% go? The answer was witty: either ext5 will be out by then, or Reiser will be released ...
And this is only one meeting at ubuntu open week. and there is also packaging 101, reporting & fixing bugs, kernel bugs, ubuntu vs debian, intrepid & servers, ubuntu & money.
recommend!
About the relationship of Debian and Ubuntu . Speaker: James Westby .
This hot English guy is involved in the MOTU (Masters of the Universe) project. The goal of the project is to support the Universe and Multiverse repositories . In parallel, he participates in Debian's projects. And here is what he told the community as part of ubuntu open week:
Debian is a great distribution that many people work on, including and Shuttleworth. When Mark created Ubuntu, he certainly looked at Debian, trying to fix several things in him. The differences affected org. moments, membership in the development team and, of course, in some technical aspects. The difference in the goals of both projects made it possible to concentrate resources on supporting a small number of popular architectures, thereby strengthening other areas of development. After all, support for 11 (well, or how many official platforms does Debian have?) Requires tremendous efforts and slows down the release cycles. The trend in Ubuntu is to choose the best solution for a particular task that really works well, and in Debian, they often try and try many solutions. The efforts, of course, are both true,
The author considers the most important difference that becoming a member of the Debian development team is much more difficult, and for a non-developer it is almost impossible. Perhaps that is why Debian vegetates in terms of "non-programmer" tasks.
To the question - “Are the Ubuntoids so severe that with the Pulseaudio vs Alsa alternative there is only one subsystem left?”, The answer was “yes” :) But with the caveats that this is not a dogma and does not always work, for example, for kde, gnome, xfce - no, but for MTA - yes.
Then they quickly hushed up a rather stupid question about why Canonical is not eager to support third-party derivatives from Ubuntu and moved on to the problem of choice. So, what is better for everyday work - A or B?
The answer is not easy. Many people recommend Ubuntu to beginners because it puts a lot of effort in to working “out of the box”. But it’s best to download the live-cd of both systems and try to choose it yourself.
From the point of view of stability, Debian has advantages due to a longer 18-month cycle, with stable but outdated packages. Here is such a compromise. Ubuntu is in a hurry to bake brand new packages with its 6-month cycles, but for lovers of stability they also have an answer. His name is LTS ...
What does it mean to be based on Debian for Ubuntu? Have their paths diverged since the first historic fork? Of course not! The start of each new release is accompanied by the so-called "merging", the merger of the Debian and Ubuntu branches. This process brings to Ubuntu both new improvements from Debian and errors.
Between distributions there is a mutual exchange of improvements and fixes. Debian receives stable packages, and Ubuntu stops accompanying them and only imports them from Debian back, because they do not have resources for all 15,000 packages.
What does Ubuntu do to refute the prevailing belief that they give little to Debian? Answer. 2 things: give patches to Debian and make the process of smuggling noticeable. Debian Stable practically does not leave anything and everything goes to Unstable, which, by the way, may be a visible factor for prosecutors who claim that Ubuntu takes a lot and gives little. In fact, a system for tagging bug reports and patches from Ubuntu has been developed, and anyone who wants to can see the statistics in the Debian bug tracker. We will take a word for what they give :)
The question of life and death - what will happen if Debian suddenly dies? The answer was not very clear, so the question was asked again and even more insistently. Will Ubuntu have the strength of developers to start autonomous navigation having, for example, 8.10 and that's it. Or everything is bad and the project is so dependent that it cannot be developed, at least in terms of security fixes. Answer: It takes a lot of developers to release the release, and there it will be seen.
Then they praise Launchpad, praise each other for bug fixes and ask to send fixes not only to Ubuntu, but also directly to the authors of the programs. There is also a big request - to monitor bugs both in the Ubuntu bugtracker and in Debianovsky. They have a vision of the issue that linking reports between systems is very important. We wish them good luck on this, I hope the topic of the coexistence of Ubuntu and Debian has opened up for us as well.
I do not think that many have read here, but if so, thank you, leave comments and help transfer to the section on Ubuntu.