February 2012 Meeting Minutes
2012-02-23 19:45 UTC Fezile, #meeting
Proceedings =========== * Meeting started at 19:45 UTC by Aleksey Bragin.
* Point 1: Release status: 0.3.15 --------------------------------- * Aleksey Bragin proposed that we prepare trunk and release 0.3.15, as it is now, before anything major happens. Several members then wondered about the status of USB and the "mshtml bug". Jerome Gardou explained a bit about how the mshtml bug is far from completely being fixed, because if you look at testbot's results, there are still some problems with ASSERTs hit and some bad pagefaults happening when paging out. * A lengthy discussion occurred, about how ready trunk is, whether it's in a better state than 0.3.14, the state of the theme to be bundled with ros, the plan for CLT, whether we should go 0.4<something> or 0.3.15... This was settled through a voting: "Do you agree to release 0.3.15 with current trunk features, before CLT?". * The total number of votes was 21, with 11 votes as Yes, 5 as No, and 5 abstentions. As a result, we will be having a 0.3.15 release with CLT being the deadline.
* Point 2: New website status and migration plans ------------------------------------------------- * Amine Khaldi gave a quick summary on the state of the website revamp: - Quite some progress has been made in the theming department, and it's visible from the playground. It's starting to look a lot like the current one, which means it's almost done (only some issues are left). - Maciej Bialas has been investigating how to import users from the RosCMS. We will most likely need to compromise: members will have to reenter passwords when we migrate, but we'll see. - Amine Khaldi then mentioned that we need help from web devs familiar with Drupal. Alexander Rechitskiy mentioned that he has a guy that wants to be part of the web team, and he passes his info to Amine to establish the contact. - He also mentioned that Maciej is very busy, so the progress will slow down, and that the current short term plan is to continue the theming work, provide a way to import users, import the rest of data from wiki/bugzilla...etc and finally add the phpbb bridge. * He then summarized the summary by saying: We needs skillful drupal guys, we're almost there, and that's it.
* Point 3: CMake migration and finally abandoning RBuild -------------------------------------------------------- * Amine Khaldi suggested that it's time to ditch rbuild, as he suspects most of our members have migrated already. Agreement on that, from ours members, was unprecedented as *everyone* were on favor of this :) * We agreed on doing the necessary commits/infrastructure changes ASAP, ie "tonight" (relative to the meeting of course), but that didn't happen just yet so it's just a matter of time. Amine Khaldi wants to do the honors (the commit that will remove rbuild and its related files). * Amine Khaldi also mentioned that he's preparing a nice little surprise for major build performance boost, on many levels, so stay tuned guys ;) * Jerome Gardou explained that he'll be handling the PCH support to get it to a much better state than it is right now.
* Point 4: Added per Amine Khaldi's request: GSoC preparations -------------------------------------------------------------- * Amine Khaldi explained that he created a Google Doc and invited the interested members. He asked the members to collaborate and shape up an excellent ideas list, projects that fit in GSoC at both the time and complexity levels. * Colin suggested building up on the last year in terms of informative content (wiki pages from last year), covering questions about how we deal with students, our application form... etc.
* Meeting closed at 21:45 UTC by Aleksey Bragin. * Minutes written by Amine Khaldi.
Good! Let's proceed with 0.3.15... a faster release cycle is better, and will help interest in ROS as people will be able to see clearly that progress is being made (as opposed to appearing to be stagnant to those who do not look at trunk). Plus, I want a "stable" release to try some new features ;P
On 24 February 2012 11:49, Amine Khaldi amine.khaldi@reactos.org wrote:
February 2012 Meeting Minutes
2012-02-23 19:45 UTC Fezile, #meeting
Proceedings
Meeting started at 19:45 UTC by Aleksey Bragin.
Point 1: Release status: 0.3.15
- Aleksey Bragin proposed that we prepare trunk and release 0.3.15, as it is
now, before anything major happens. Several members then wondered about the status of USB and the "mshtml bug". Jerome Gardou explained a bit about how the mshtml bug is far from completely being fixed, because if you look at testbot's results, there are still some problems with ASSERTs hit and some bad pagefaults happening when paging out.
- A lengthy discussion occurred, about how ready trunk is, whether it's in a
better state than 0.3.14, the state of the theme to be bundled with ros, the plan for CLT, whether we should go 0.4<something> or 0.3.15... This was settled through a voting: "Do you agree to release 0.3.15 with current trunk features, before CLT?".
- The total number of votes was 21, with 11 votes as Yes, 5 as No, and 5
abstentions. As a result, we will be having a 0.3.15 release with CLT being the deadline.
- Point 2: New website status and migration plans
- Amine Khaldi gave a quick summary on the state of the website revamp:
- Quite some progress has been made in the theming department, and it's visible from the playground. It's starting to look a lot like the current one, which means it's almost done (only some issues are left). - Maciej Bialas has been investigating how to import users from the RosCMS. We will most likely need to compromise: members will have to reenter passwords when we migrate, but we'll see. - Amine Khaldi then mentioned that we need help from web devs familiar with Drupal. Alexander Rechitskiy mentioned that he has a guy that wants to be part of the web team, and he passes his info to Amine to establish the contact. - He also mentioned that Maciej is very busy, so the progress will slow down, and that the current short term plan is to continue the theming work, provide a way to import users, import the rest of data from wiki/bugzilla...etc and finally add the phpbb bridge.
- He then summarized the summary by saying: We needs skillful drupal guys,
we're almost there, and that's it.
- Point 3: CMake migration and finally abandoning RBuild
- Amine Khaldi suggested that it's time to ditch rbuild, as he suspects most
of our members have migrated already. Agreement on that, from ours members, was unprecedented as *everyone* were on favor of this :)
- We agreed on doing the necessary commits/infrastructure changes ASAP, ie
"tonight" (relative to the meeting of course), but that didn't happen just yet so it's just a matter of time. Amine Khaldi wants to do the honors (the commit that will remove rbuild and its related files).
- Amine Khaldi also mentioned that he's preparing a nice little surprise for
major build performance boost, on many levels, so stay tuned guys ;)
- Jerome Gardou explained that he'll be handling the PCH support to get it
to a much better state than it is right now.
- Point 4: Added per Amine Khaldi's request: GSoC preparations
- Amine Khaldi explained that he created a Google Doc and invited the
interested members. He asked the members to collaborate and shape up an excellent ideas list, projects that fit in GSoC at both the time and complexity levels.
- Colin suggested building up on the last year in terms of informative
content (wiki pages from last year), covering questions about how we deal with students, our application form... etc.
- Meeting closed at 21:45 UTC by Aleksey Bragin.
- Minutes written by Amine Khaldi.
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
A note of caution, regardless of the vote for 0.3.15, it is entirely possible, even likely from the perspective of several of us, that the state of ROS trunk will preclude the ability to roll out a 0.3.15. We're talking about a 3 week, more likely 2 week, turnaround, which is absurdly if we wanted to do genuine regression testing and fixing.
On Fri, Feb 24, 2012 at 10:31 AM, Andrew Faulds ajfweb@googlemail.com wrote:
Good! Let's proceed with 0.3.15... a faster release cycle is better, and will help interest in ROS as people will be able to see clearly that progress is being made (as opposed to appearing to be stagnant to those who do not look at trunk). Plus, I want a "stable" release to try some new features ;P
On 24 February 2012 11:49, Amine Khaldi amine.khaldi@reactos.org wrote:
February 2012 Meeting Minutes
2012-02-23 19:45 UTC Fezile, #meeting
Proceedings
Meeting started at 19:45 UTC by Aleksey Bragin.
Point 1: Release status: 0.3.15
- Aleksey Bragin proposed that we prepare trunk and release 0.3.15, as it is
now, before anything major happens. Several members then wondered about the status of USB and the "mshtml bug". Jerome Gardou explained a bit about how the mshtml bug is far from completely being fixed, because if you look at testbot's results, there are still some problems with ASSERTs hit and some bad pagefaults happening when paging out.
- A lengthy discussion occurred, about how ready trunk is, whether it's in a
better state than 0.3.14, the state of the theme to be bundled with ros, the plan for CLT, whether we should go 0.4<something> or 0.3.15... This was settled through a voting: "Do you agree to release 0.3.15 with current trunk features, before CLT?".
- The total number of votes was 21, with 11 votes as Yes, 5 as No, and 5
abstentions. As a result, we will be having a 0.3.15 release with CLT being the deadline.
- Point 2: New website status and migration plans
- Amine Khaldi gave a quick summary on the state of the website revamp:
- Quite some progress has been made in the theming department, and it's visible from the playground. It's starting to look a lot like the current one, which means it's almost done (only some issues are left). - Maciej Bialas has been investigating how to import users from the RosCMS. We will most likely need to compromise: members will have to reenter passwords when we migrate, but we'll see. - Amine Khaldi then mentioned that we need help from web devs familiar with Drupal. Alexander Rechitskiy mentioned that he has a guy that wants to be part of the web team, and he passes his info to Amine to establish the contact. - He also mentioned that Maciej is very busy, so the progress will slow down, and that the current short term plan is to continue the theming work, provide a way to import users, import the rest of data from wiki/bugzilla...etc and finally add the phpbb bridge.
- He then summarized the summary by saying: We needs skillful drupal guys,
we're almost there, and that's it.
- Point 3: CMake migration and finally abandoning RBuild
- Amine Khaldi suggested that it's time to ditch rbuild, as he suspects most
of our members have migrated already. Agreement on that, from ours members, was unprecedented as *everyone* were on favor of this :)
- We agreed on doing the necessary commits/infrastructure changes ASAP, ie
"tonight" (relative to the meeting of course), but that didn't happen just yet so it's just a matter of time. Amine Khaldi wants to do the honors (the commit that will remove rbuild and its related files).
- Amine Khaldi also mentioned that he's preparing a nice little surprise for
major build performance boost, on many levels, so stay tuned guys ;)
- Jerome Gardou explained that he'll be handling the PCH support to get it
to a much better state than it is right now.
- Point 4: Added per Amine Khaldi's request: GSoC preparations
- Amine Khaldi explained that he created a Google Doc and invited the
interested members. He asked the members to collaborate and shape up an excellent ideas list, projects that fit in GSoC at both the time and complexity levels.
- Colin suggested building up on the last year in terms of informative
content (wiki pages from last year), covering questions about how we deal with students, our application form... etc.
- Meeting closed at 21:45 UTC by Aleksey Bragin.
- Minutes written by Amine Khaldi.
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
-- Andrew Faulds (AJF) http://ajf.me/
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
Yes, that is a very quick turnaround. But if we could get a faster release cycle (not THIS fast), it would be better for ROS.
On 24 February 2012 17:18, Zachary Gorden drakekaizer666@gmail.com wrote:
A note of caution, regardless of the vote for 0.3.15, it is entirely possible, even likely from the perspective of several of us, that the state of ROS trunk will preclude the ability to roll out a 0.3.15. We're talking about a 3 week, more likely 2 week, turnaround, which is absurdly if we wanted to do genuine regression testing and fixing.
On Fri, Feb 24, 2012 at 10:31 AM, Andrew Faulds ajfweb@googlemail.com wrote:
Good! Let's proceed with 0.3.15... a faster release cycle is better, and will help interest in ROS as people will be able to see clearly that progress is being made (as opposed to appearing to be stagnant to those who do not look at trunk). Plus, I want a "stable" release to try some new features ;P
On 24 February 2012 11:49, Amine Khaldi amine.khaldi@reactos.org wrote:
February 2012 Meeting Minutes
2012-02-23 19:45 UTC Fezile, #meeting
Proceedings
Meeting started at 19:45 UTC by Aleksey Bragin.
Point 1: Release status: 0.3.15
- Aleksey Bragin proposed that we prepare trunk and release 0.3.15, as it is
now, before anything major happens. Several members then wondered about the status of USB and the "mshtml bug". Jerome Gardou explained a bit about how the mshtml bug is far from completely being fixed, because if you look at testbot's results, there are still some problems with ASSERTs hit and some bad pagefaults happening when paging out.
- A lengthy discussion occurred, about how ready trunk is, whether it's in a
better state than 0.3.14, the state of the theme to be bundled with ros, the plan for CLT, whether we should go 0.4<something> or 0.3.15... This was settled through a voting: "Do you agree to release 0.3.15 with current trunk features, before CLT?".
- The total number of votes was 21, with 11 votes as Yes, 5 as No, and 5
abstentions. As a result, we will be having a 0.3.15 release with CLT being the deadline.
- Point 2: New website status and migration plans
- Amine Khaldi gave a quick summary on the state of the website revamp:
- Quite some progress has been made in the theming department, and it's visible from the playground. It's starting to look a lot like the current one, which means it's almost done (only some issues are left). - Maciej Bialas has been investigating how to import users from the RosCMS. We will most likely need to compromise: members will have to reenter passwords when we migrate, but we'll see. - Amine Khaldi then mentioned that we need help from web devs familiar with Drupal. Alexander Rechitskiy mentioned that he has a guy that wants to be part of the web team, and he passes his info to Amine to establish the contact. - He also mentioned that Maciej is very busy, so the progress will slow down, and that the current short term plan is to continue the theming work, provide a way to import users, import the rest of data from wiki/bugzilla...etc and finally add the phpbb bridge.
- He then summarized the summary by saying: We needs skillful drupal guys,
we're almost there, and that's it.
- Point 3: CMake migration and finally abandoning RBuild
- Amine Khaldi suggested that it's time to ditch rbuild, as he suspects most
of our members have migrated already. Agreement on that, from ours members, was unprecedented as *everyone* were on favor of this :)
- We agreed on doing the necessary commits/infrastructure changes ASAP, ie
"tonight" (relative to the meeting of course), but that didn't happen just yet so it's just a matter of time. Amine Khaldi wants to do the honors (the commit that will remove rbuild and its related files).
- Amine Khaldi also mentioned that he's preparing a nice little surprise for
major build performance boost, on many levels, so stay tuned guys ;)
- Jerome Gardou explained that he'll be handling the PCH support to get it
to a much better state than it is right now.
- Point 4: Added per Amine Khaldi's request: GSoC preparations
- Amine Khaldi explained that he created a Google Doc and invited the
interested members. He asked the members to collaborate and shape up an excellent ideas list, projects that fit in GSoC at both the time and complexity levels.
- Colin suggested building up on the last year in terms of informative
content (wiki pages from last year), covering questions about how we deal with students, our application form... etc.
- Meeting closed at 21:45 UTC by Aleksey Bragin.
- Minutes written by Amine Khaldi.
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
-- Andrew Faulds (AJF) http://ajf.me/
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
"Faster release cycle is better" is what a bunch of idiots at Mozilla Corporation said. And now their browser has begun the process of becoming more bloated, much slower, and contains more and more useless rubbish and "features" with every release, while the competition (including Internet Explorer) is excelling on simple and useful things, such as a UI that doesn't randomly freeze solid every five minutes.
I strongly oppose this "faster release cycle" unless it means simply tagging a TRUNK build that "just works" as a release and releasing without doing anything further, apart from maybe the occasional bootscreen/login screen change to reflect the version number.
If you want to try new features, use a trunk build.
Quick release may actually *reduce* the interest in ReactOS when people try the "stable" builds only to find that it is a lot buggier because of the absurdly limiting deadlines, and so few final fixes could be made.
On Fri, 24 Feb 2012 16:31:52 +0000 Andrew Faulds ajfweb@googlemail.com wrote:
Good! Let's proceed with 0.3.15... a faster release cycle is better, and will help interest in ROS as people will be able to see clearly that progress is being made (as opposed to appearing to be stagnant to those who do not look at trunk). Plus, I want a "stable" release to try some new features ;P
On 24 February 2012 11:49, Amine Khaldi amine.khaldi@reactos.org wrote:
February 2012 Meeting Minutes
2012-02-23 19:45 UTC Fezile, #meeting
Proceedings
Meeting started at 19:45 UTC by Aleksey Bragin.
Point 1: Release status: 0.3.15
- Aleksey Bragin proposed that we prepare trunk and release 0.3.15,
as it is now, before anything major happens. Several members then wondered about the status of USB and the "mshtml bug". Jerome Gardou explained a bit about how the mshtml bug is far from completely being fixed, because if you look at testbot's results, there are still some problems with ASSERTs hit and some bad pagefaults happening when paging out.
- A lengthy discussion occurred, about how ready trunk is, whether
it's in a better state than 0.3.14, the state of the theme to be bundled with ros, the plan for CLT, whether we should go 0.4<something> or 0.3.15... This was settled through a voting: "Do you agree to release 0.3.15 with current trunk features, before CLT?".
- The total number of votes was 21, with 11 votes as Yes, 5 as No,
and 5 abstentions. As a result, we will be having a 0.3.15 release with CLT being the deadline.
- Point 2: New website status and migration plans
- Amine Khaldi gave a quick summary on the state of the website
revamp: - Quite some progress has been made in the theming department, and it's visible from the playground. It's starting to look a lot like the current one, which means it's almost done (only some issues are left). - Maciej Bialas has been investigating how to import users from the RosCMS. We will most likely need to compromise: members will have to reenter passwords when we migrate, but we'll see. - Amine Khaldi then mentioned that we need help from web devs familiar with Drupal. Alexander Rechitskiy mentioned that he has a guy that wants to be part of the web team, and he passes his info to Amine to establish the contact. - He also mentioned that Maciej is very busy, so the progress will slow down, and that the current short term plan is to continue the theming work, provide a way to import users, import the rest of data from wiki/bugzilla...etc and finally add the phpbb bridge.
- He then summarized the summary by saying: We needs skillful
drupal guys, we're almost there, and that's it.
- Point 3: CMake migration and finally abandoning RBuild
- Amine Khaldi suggested that it's time to ditch rbuild, as he
suspects most of our members have migrated already. Agreement on that, from ours members, was unprecedented as *everyone* were on favor of this :)
- We agreed on doing the necessary commits/infrastructure changes
ASAP, ie "tonight" (relative to the meeting of course), but that didn't happen just yet so it's just a matter of time. Amine Khaldi wants to do the honors (the commit that will remove rbuild and its related files).
- Amine Khaldi also mentioned that he's preparing a nice little
surprise for major build performance boost, on many levels, so stay tuned guys ;)
- Jerome Gardou explained that he'll be handling the PCH support to
get it to a much better state than it is right now.
- Point 4: Added per Amine Khaldi's request: GSoC preparations
- Amine Khaldi explained that he created a Google Doc and invited
the interested members. He asked the members to collaborate and shape up an excellent ideas list, projects that fit in GSoC at both the time and complexity levels.
- Colin suggested building up on the last year in terms of
informative content (wiki pages from last year), covering questions about how we deal with students, our application form... etc.
- Meeting closed at 21:45 UTC by Aleksey Bragin.
- Minutes written by Amine Khaldi.
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
On Sat, Feb 25, 2012 at 5:00 AM, Adam geekdundee@gmail.com wrote:
"Faster release cycle is better" is what a bunch of idiots at Mozilla Corporation said. And now their browser has begun the process of becoming more bloated, much slower, and contains more and more useless rubbish and
You're being ironic aren't you ? Firefox has never been so slim and less memory hungry than it is these days.
Yes. Sorry. You're absolutely right. Firefox really is slimmer and less memory hungry than ever these days. On another note, eating McDonald's is good for you so you should do this everyday.
Did I mention that I am the King of England?
On Sat, 25 Feb 2012 05:52:09 +0100 Christoph Thompson cjsthompson@gmail.com wrote:
You're being ironic aren't you ? Firefox has never been so slim and less memory hungry than it is these days.
On 25.02.2012 09:38, Adam wrote:
Yes. Sorry. You're absolutely right. Firefox really is slimmer and less memory hungry than ever these days. On another note, eating McDonald's is good for you so you should do this everyday.
Did I mention that I am the King of England?
Ehm... Firefox IS less memory hungry in its current version than before. I noticed this on my smaller Windows system, where Firefox used to have around 400 MB with around 20 tabs open and now it's down to 200 MB. That this is still far from optimal is a different topic ;)
[On Linux it seems to still be the same memory hog as ever...]
Regards, Sven
That's interesting. On all my Windows systems Firefox 9 and 10 hogs up almost 1GB of memory (and uses up the pagefile!) for only 5-6 tabs open. I even reduced the memory cache settings. Not even Firefox 3.x used to hog so much RAM on any of my systems. Approx. 100-200MB at most with 15 tabs open. And memory cache enabled.
Note that I am talking about visiting the same sites I normally do here.
On a test now (x64 system) Firefox 9.0.1 hogs up over 170MB RAM with only 1 tab open. Not even Firefox 4 was this bad; I've noticed usage getting worse since they switched to this "quick release" scheme and turned their entire user base into beta testers, without their consent.
I rest my case.
On Sat, 25 Feb 2012 11:15:23 +0100 Sven Barth pascaldragon@googlemail.com wrote:
On 25.02.2012 09:38, Adam wrote:
Yes. Sorry. You're absolutely right. Firefox really is slimmer and less memory hungry than ever these days. On another note, eating McDonald's is good for you so you should do this everyday.
Did I mention that I am the King of England?
Ehm... Firefox IS less memory hungry in its current version than before. I noticed this on my smaller Windows system, where Firefox used to have around 400 MB with around 20 tabs open and now it's down to 200 MB. That this is still far from optimal is a different topic ;)
[On Linux it seems to still be the same memory hog as ever...]
Regards, Sven
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
On Sat, Feb 25, 2012 at 11:15 AM, Sven Barth pascaldragon@googlemail.comwrote:
Ehm... Firefox IS less memory hungry in its current version than before. I noticed this on my smaller Windows system, where Firefox used to have around 400 MB with around 20 tabs open and now it's down to 200 MB. That this is still far from optimal is a different topic ;)
[On Linux it seems to still be the same memory hog as ever...]
Regards, Sven
Actually, it's been severely less memory hungry since version 7 (down by 20 to 30%) and it keeps improving with each new version since then cause Mozilla has made it one of their top priorities to reduce memory usage. And there's no difference for Linux. Looks like the king of england must have forgotten to update ^^
I'm sorry, have you tried version 9 or version 10 yet? It's usage is increasingly high since around version 4. I was hoping it would improve as well but I'm not seeing it.
And in any case, did any of you read the other bits of my post? What about the buggy useless features? The UI that keeps freezing solid?
Bear in mind that I am not the only person with these complaints.
I was using this as evidence for how silly the idea of "fast release" is. This bumping the version numbers is all rubbish.
On Sat, 25 Feb 2012 12:30:55 +0100 Christoph Thompson cjsthompson@gmail.com wrote:
On Sat, Feb 25, 2012 at 11:15 AM, Sven Barth pascaldragon@googlemail.comwrote:
Ehm... Firefox IS less memory hungry in its current version than before. I noticed this on my smaller Windows system, where Firefox used to have around 400 MB with around 20 tabs open and now it's down to 200 MB. That this is still far from optimal is a different topic ;)
[On Linux it seems to still be the same memory hog as ever...]
Regards, Sven
Actually, it's been severely less memory hungry since version 7 (down by 20 to 30%) and it keeps improving with each new version since then cause Mozilla has made it one of their top priorities to reduce memory usage. And there's no difference for Linux. Looks like the king of england must have forgotten to update ^^
Quicker releases have the benefit that it forces us to fix lingering regressions and release. Otherwise regressions and significant bugs pile up until release and we have to spend months fixing them.
On 25 February 2012 04:00, Adam geekdundee@gmail.com wrote:
"Faster release cycle is better" is what a bunch of idiots at Mozilla Corporation said. And now their browser has begun the process of becoming more bloated, much slower, and contains more and more useless rubbish and "features" with every release, while the competition (including Internet Explorer) is excelling on simple and useful things, such as a UI that doesn't randomly freeze solid every five minutes.
I strongly oppose this "faster release cycle" unless it means simply tagging a TRUNK build that "just works" as a release and releasing without doing anything further, apart from maybe the occasional bootscreen/login screen change to reflect the version number.
If you want to try new features, use a trunk build.
Quick release may actually *reduce* the interest in ReactOS when people try the "stable" builds only to find that it is a lot buggier because of the absurdly limiting deadlines, and so few final fixes could be made.
On Fri, 24 Feb 2012 16:31:52 +0000 Andrew Faulds ajfweb@googlemail.com wrote:
Good! Let's proceed with 0.3.15... a faster release cycle is better, and will help interest in ROS as people will be able to see clearly that progress is being made (as opposed to appearing to be stagnant to those who do not look at trunk). Plus, I want a "stable" release to try some new features ;P
On 24 February 2012 11:49, Amine Khaldi amine.khaldi@reactos.org wrote:
February 2012 Meeting Minutes
2012-02-23 19:45 UTC Fezile, #meeting
Proceedings
Meeting started at 19:45 UTC by Aleksey Bragin.
Point 1: Release status: 0.3.15
- Aleksey Bragin proposed that we prepare trunk and release 0.3.15,
as it is now, before anything major happens. Several members then wondered about the status of USB and the "mshtml bug". Jerome Gardou explained a bit about how the mshtml bug is far from completely being fixed, because if you look at testbot's results, there are still some problems with ASSERTs hit and some bad pagefaults happening when paging out.
- A lengthy discussion occurred, about how ready trunk is, whether
it's in a better state than 0.3.14, the state of the theme to be bundled with ros, the plan for CLT, whether we should go 0.4<something> or 0.3.15... This was settled through a voting: "Do you agree to release 0.3.15 with current trunk features, before CLT?".
- The total number of votes was 21, with 11 votes as Yes, 5 as No,
and 5 abstentions. As a result, we will be having a 0.3.15 release with CLT being the deadline.
- Point 2: New website status and migration plans
- Amine Khaldi gave a quick summary on the state of the website
revamp: - Quite some progress has been made in the theming department, and it's visible from the playground. It's starting to look a lot like the current one, which means it's almost done (only some issues are left). - Maciej Bialas has been investigating how to import users from the RosCMS. We will most likely need to compromise: members will have to reenter passwords when we migrate, but we'll see. - Amine Khaldi then mentioned that we need help from web devs familiar with Drupal. Alexander Rechitskiy mentioned that he has a guy that wants to be part of the web team, and he passes his info to Amine to establish the contact. - He also mentioned that Maciej is very busy, so the progress will slow down, and that the current short term plan is to continue the theming work, provide a way to import users, import the rest of data from wiki/bugzilla...etc and finally add the phpbb bridge.
- He then summarized the summary by saying: We needs skillful
drupal guys, we're almost there, and that's it.
- Point 3: CMake migration and finally abandoning RBuild
- Amine Khaldi suggested that it's time to ditch rbuild, as he
suspects most of our members have migrated already. Agreement on that, from ours members, was unprecedented as *everyone* were on favor of this :)
- We agreed on doing the necessary commits/infrastructure changes
ASAP, ie "tonight" (relative to the meeting of course), but that didn't happen just yet so it's just a matter of time. Amine Khaldi wants to do the honors (the commit that will remove rbuild and its related files).
- Amine Khaldi also mentioned that he's preparing a nice little
surprise for major build performance boost, on many levels, so stay tuned guys ;)
- Jerome Gardou explained that he'll be handling the PCH support to
get it to a much better state than it is right now.
- Point 4: Added per Amine Khaldi's request: GSoC preparations
- Amine Khaldi explained that he created a Google Doc and invited
the interested members. He asked the members to collaborate and shape up an excellent ideas list, projects that fit in GSoC at both the time and complexity levels.
- Colin suggested building up on the last year in terms of
informative content (wiki pages from last year), covering questions about how we deal with students, our application form... etc.
- Meeting closed at 21:45 UTC by Aleksey Bragin.
- Minutes written by Amine Khaldi.
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
The only time I've ever seen this work is with OpenBSD - but that's because they do a strict six-monthly release schedule, with set periods of time hacking away, set periods of time testing things, and set periods of time finalizing the release. Any new features that doesn't make it into the (I think it's 4 months?) hacking period just doesn't make it. In other words, auditing is very strict there; that's hardly what I could say for other projects. And fair enough; it's a huge job!
And some of the big rewrites aren't even enabled/in the main tree until years later!
Quick releases work the way you are talking about *in theory* - while in practice (at least, here in ReactOS that is) I have noticed a lot of "Oh we're waaaay past the deadline dude. Let's like, you know, leave this big regression till the next release, man." - in other words, great in theory, but rarely practical and benefits are either negative, or negligible.
The MM rewrite was an example. Now that's been pushed to the next release. (correct me if I am wrong) - it was going to be released with 0.PI before.
On Sat, 25 Feb 2012 11:57:57 +0000 Andrew Faulds ajfweb@googlemail.com wrote:
Quicker releases have the benefit that it forces us to fix lingering regressions and release. Otherwise regressions and significant bugs pile up until release and we have to spend months fixing them.
On 25 February 2012 04:00, Adam geekdundee@gmail.com wrote:
"Faster release cycle is better" is what a bunch of idiots at Mozilla Corporation said. And now their browser has begun the process of becoming more bloated, much slower, and contains more and more useless rubbish and "features" with every release, while the competition (including Internet Explorer) is excelling on simple and useful things, such as a UI that doesn't randomly freeze solid every five minutes.
I strongly oppose this "faster release cycle" unless it means simply tagging a TRUNK build that "just works" as a release and releasing without doing anything further, apart from maybe the occasional bootscreen/login screen change to reflect the version number.
If you want to try new features, use a trunk build.
Quick release may actually *reduce* the interest in ReactOS when people try the "stable" builds only to find that it is a lot buggier because of the absurdly limiting deadlines, and so few final fixes could be made.
On Fri, 24 Feb 2012 16:31:52 +0000 Andrew Faulds ajfweb@googlemail.com wrote:
Good! Let's proceed with 0.3.15... a faster release cycle is better, and will help interest in ROS as people will be able to see clearly that progress is being made (as opposed to appearing to be stagnant to those who do not look at trunk). Plus, I want a "stable" release to try some new features ;P
On 24 February 2012 11:49, Amine Khaldi amine.khaldi@reactos.org wrote:
February 2012 Meeting Minutes
2012-02-23 19:45 UTC Fezile, #meeting
Proceedings
Meeting started at 19:45 UTC by Aleksey Bragin.
Point 1: Release status: 0.3.15
- Aleksey Bragin proposed that we prepare trunk and release
0.3.15, as it is now, before anything major happens. Several members then wondered about the status of USB and the "mshtml bug". Jerome Gardou explained a bit about how the mshtml bug is far from completely being fixed, because if you look at testbot's results, there are still some problems with ASSERTs hit and some bad pagefaults happening when paging out.
- A lengthy discussion occurred, about how ready trunk is,
whether it's in a better state than 0.3.14, the state of the theme to be bundled with ros, the plan for CLT, whether we should go 0.4<something> or 0.3.15... This was settled through a voting: "Do you agree to release 0.3.15 with current trunk features, before CLT?".
- The total number of votes was 21, with 11 votes as Yes, 5 as
No, and 5 abstentions. As a result, we will be having a 0.3.15 release with CLT being the deadline.
- Point 2: New website status and migration plans
- Amine Khaldi gave a quick summary on the state of the website
revamp: - Quite some progress has been made in the theming department, and it's visible from the playground. It's starting to look a lot like the current one, which means it's almost done (only some issues are left). - Maciej Bialas has been investigating how to import users from the RosCMS. We will most likely need to compromise: members will have to reenter passwords when we migrate, but we'll see. - Amine Khaldi then mentioned that we need help from web devs familiar with Drupal. Alexander Rechitskiy mentioned that he has a guy that wants to be part of the web team, and he passes his info to Amine to establish the contact. - He also mentioned that Maciej is very busy, so the progress will slow down, and that the current short term plan is to continue the theming work, provide a way to import users, import the rest of data from wiki/bugzilla...etc and finally add the phpbb bridge.
- He then summarized the summary by saying: We needs skillful
drupal guys, we're almost there, and that's it.
- Point 3: CMake migration and finally abandoning RBuild
- Amine Khaldi suggested that it's time to ditch rbuild, as he
suspects most of our members have migrated already. Agreement on that, from ours members, was unprecedented as *everyone* were on favor of this :)
- We agreed on doing the necessary commits/infrastructure changes
ASAP, ie "tonight" (relative to the meeting of course), but that didn't happen just yet so it's just a matter of time. Amine Khaldi wants to do the honors (the commit that will remove rbuild and its related files).
- Amine Khaldi also mentioned that he's preparing a nice little
surprise for major build performance boost, on many levels, so stay tuned guys ;)
- Jerome Gardou explained that he'll be handling the PCH support
to get it to a much better state than it is right now.
- Point 4: Added per Amine Khaldi's request: GSoC preparations
- Amine Khaldi explained that he created a Google Doc and invited
the interested members. He asked the members to collaborate and shape up an excellent ideas list, projects that fit in GSoC at both the time and complexity levels.
- Colin suggested building up on the last year in terms of
informative content (wiki pages from last year), covering questions about how we deal with students, our application form... etc.
- Meeting closed at 21:45 UTC by Aleksey Bragin.
- Minutes written by Amine Khaldi.
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
My opinion about this:
There is a difference between a project that is in use (like Firefox) and a proejct that has no use except showing how far it is. For a project that is in use, there is a benefit from a shorter release cycle (as long as it's done thoroughly), due to the fact that users demand bugfixes, to improve their "working experience".
A project like reactos can only show "look, this is what we have so far, it still doesn't work". A too short release cycle can result in too few actual changes, and people don't really see many improvements. Also release can become boring (yet another release without major improvements...) Another final disadvantage is that heavy changes are really hard to get in in a tight schedule. After a release devs might be tempted to rush their changes in as soon as possible, possibly causing multiple major changes being comitted at once.
Therefore I would prefer releases in a multiple month cycle. Every month or every 2 doesn't seem to be reasonable for our project at the moment.
Just my 2 cents, Timo
A real interesting side-note.0.3.14 has had a greater and more positive impact than previous ReactOS releases.After reading tons of Forums, criticisms, etc, the positive feedback can be summed up in 2 ideas:1) "ReactOS is more compatible!I can feel its improved stability and now more apps are working"Well, really, this is a fact of not releasing after one whole year. They have been comparing 0.3.13 to previous ones, released in a 5-6months cycle. Thanks to a delayed release, they have felt the evolution of a one year release (2-3 normal old releases 'till now).If they would have just compared 0.3.13 with 0.3.11-0.3.10 they would have felt exactly the same "jump". 2)"New features"All the work that has been done since long ago has met into a release (GSOC projects, rewrites, etc). This has boosted the general release impact.
For these 2 reasons, I'd prefer to stick to a 4 or 5 month release after 0.4 is launched. If we release too late, we can drop the interest of our followers(but it will be recovered when releasing, as 0.3.14 shows). Twitter,Facebook, and media will help to recover them if we post regularly through them.If we release too soon, they won't see "changes" or "new features" so they won't have the need of updating and retesting. Btw, some stats: ReactOS 0.3.14: 370.000*.ReactOS 0.3.13: 217.000*. *Downloads after the 3 first weeks.
On 02/25/2012 05:27 PM, Timo Kreuzer wrote:
There is a difference between a project that is in use (like Firefox)
and a proejct that has no use except showing how far it is.
It is true that one of the major arguments for "release early, release often" is to improve the contact between the users and the developers. It improves user feedback and with that the chances that you are working on what users actually want. It's arguable whether ReactOS could benefit from that.
There are, however, other benefits. For example, making QA easier is an oft cited benefit. Similar to what Andrew Faulds noted.
A project like reactos can only show "look, this is what we have so far, it still doesn't work". A too short release cycle can result in too few actual changes, and people don't really see many improvements. Also release can become boring (yet another release without major improvements...)
Too long without hearing anything is also boring. If you want to keep ReactOS interesting, you need to launch great features on a regular basis. Waiting with a new release, until an awesome feature arrives, is not more interesting than releasing all the time without exciting features.
Another final disadvantage is that heavy changes are really hard to get in in a tight schedule. After a release devs might be tempted to rush their changes in as soon as possible, possibly causing multiple major changes being comitted at once.
Fast release projects usually have several branches simultaneously. Firefox, for example, develops features in branches. If it doesn't look like it will be ready for the planned version the feature would land in, the targeted Firefox released is simply moved. When it's ready, the feature is merged into master. When it is time for a release, master moves to Aurora channel, the next release it moves to Beta, until finally it moves into 'the real firefox' three releases later.
Therefore I would prefer releases in a multiple month cycle. Every month or every 2 doesn't seem to be reasonable for our project at the moment.
Matching ReactOS's release schedule to Firefox' doesn't make sense. It's a completely different product, with different people and different needs. The general principle of release early, release often still stands, but what is "fast" and "often" can get a different meaning compared to Firefox.
What do you think about developing multiple channels simultaneously? After 2 months, release alpha, after 2 months, release beta, after another 2 months, released. That way, new features appear 'on the radar' every 2 months, but they don't need to be completely polished for 6 months. Users would expect Beta to be generally usable, even if there are some bugs or regressions, and you'd do your best to keep Alpha usable, but don't guarantee anything.
Emanuel
well, i think thats being done, in some way. Lets see:
"alpha channel" may be the regular daily builds, which can be totally broken, completely unstable... or not. "use them at your own risk" (tm)
"beta channel" are releases done right after branching (with the -RC tag)
and "stable releases", considering those are not actually stables, are the so-called 0.3.13, 0.3.14, 0.3.12, 0.3.10.... and so
On Mon, Feb 27, 2012 at 3:07 PM, Emanuel Rietveld codehotter@gmail.comwrote:
What do you think about developing multiple channels simultaneously? After 2 months, release alpha, after 2 months, release beta, after another 2 months, released. That way, new features appear 'on the radar' every 2 months, but they don't need to be completely polished for 6 months. Users would expect Beta to be generally usable, even if there are some bugs or regressions, and you'd do your best to keep Alpha usable, but don't guarantee anything.
Emanuel
______________________________**_________________ Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/**mailman/listinfo/ros-devhttp://www.reactos.org/mailman/listinfo/ros-dev
Excuse me being so harsh, but
On 25.02.2012 8:00, Adam wrote:
"Faster release cycle is better" is what a bunch of idiots at Mozilla Corporation said.
... and another idiot is trying to argue.
Moral of the story is that one needs to have some background knowledge and experience to state something or to teach others. I understand when Olaf states his concerns - he actually makes huge amount of work related to releasing, and he actually understand the underlying problems.
There are not much large open source projects out there, but still we are not the first and only large open source project. Ignoring others experience is stupid.
WBR, Aleksey Bragin.
On February 25, 2012 at 5:05 PM Aleksey Bragin aleksey@reactos.org wrote:
There are not much large open source projects out there, but still we are not the first and only large open source project. Ignoring others experience is stupid.
Moreover, arguing anything by one data point (if they are using short or long release cycles) while ignoring all others, technical, social, political pressure, money and so and coming to a any one conclusion, is un-scientific.
best regards, Jakob
This.
Also, drawing conclusion based on one's own anecdotal evidence is plain stupid.
Op 27-2-2012 16:33, Jakob Eriksson schreef:
On February 25, 2012 at 5:05 PM Aleksey Braginaleksey@reactos.org wrote:
There are not much large open source projects out there, but still we are not the first and only large open source project. Ignoring others experience is stupid.
Moreover, arguing anything by one data point (if they are using short or long release cycles) while ignoring all others, technical, social, political pressure, money and so and coming to a any one conclusion, is un-scientific.
best regards, Jakob
Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
-
Adam -
Aleksey Bragin -
Ameen Ross -
Amine Khaldi -
Andrew Faulds -
Christoph Thompson -
Emanuel Rietveld -
Jakob Eriksson -
Javier Agustìn Fernàndez Arroyo -
Sven Barth -
Timo Kreuzer -
victor martinez -
Zachary Gorden