Contributor mp3phish

I know that the Ubuntu volunteers don't have any power over this, but I think that the parent company Canonical maybe could get some pull here. Maybe they could leverage the Dell partnership to force AMD's hands.

ATI Video card drivers are very poor. The thing just plain doesn't work, and I don't see this changing any time soon. This is probably the #1 thing holding Ubuntu back to the mass population where their laptops and desktops came with ATI motherboard chipsets and absolutely NO support for video, compiz, etc from AMD.

While I am not praising nvidia's effort (closed source only) at least their drivers work.

I propose that Canonical use their leverage in any way they feel possible to convince AMD to deliver high quality (and preferably open source) drivers for Ubuntu. They are not only holding back gaming on linux, but also adoption for every day users.

This idea ties together ideas #8033 and #7989 with some improvements.

Problems with USB Flash Unmounting: 1) When you pull the plug on a drive, the system does not detect it and take care of the forced unmount for you. User has to go to command line. 2) When you unplug a drive, even when the light isn't blinking, unflushed buffers can cause significant data loss for no apparent reason.

Solution:

1) the system should detect when the drive is unplugged and if it is still mounted, a forced unmount should occur. This will prevent the user from having to use the command line for an every day occurrence. Many newbies reboot the system to get rid of this error state.

2) There should be a timeout set that after X seconds of inactivity (I propose 5 seconds) the write buffer should be flushed.

3) A new .fslog (or similar) file should be installed in all usb flash drives upon connection. If its a journaled filesystem, then don't, but for fat32 and others, it should do this. It will have the following function:
a. for every flush of the write buffer due to the timeout in part (2), write an entry into the log at the end indicating all writes are complete. This can be called the "clean" state
b. when writing to the disk with an initially empty write buffer, check to see if the log indicates a clean or dirty state. If clean, change the state to dirty.
c. when mounting a flash drive the system should read the log file, detect whether it is dirty or clean, and do a fsck if dirty, warning the user to not unplug the drive when it is in use.
d. Due to (a) and (b), at maximum, a log write could only occur once after 5 seconds of inactivity if the buffer has data in it. So it would not reduce lifespan of the flash. And even if your worried about flushing after 5 seconds, it would only correspond to writes already on their way to the disk, negating any problem with limited writes and lifespan of the drive.


[....]

Hello,

Instead of marking bugs as "not an idea" I propose that the brainstorm developers export all the "non ideas" which are marked as bugs to be exported as a CSV file and FTP'd over to the Ubuntu bug tracker. Once there, the bug tracking database should import the data and be listed as a bug.

Since brainstorm and all other ubuntu logins will be universal soon, this will integrate nicely.

I just think its a lost cause to mark stuff as not an idea, and then do nothing with it. All that data just going to waste, and then telling the user to post it to the bug database. It just isn't efficient, and it is causing bugs to go unreported. One of the biggest problems with Ubuntu (and the Linux community in general) is that bug reporting is not taken seriously when reported by an end user. Most of the time it is because the end user reports it to the wrong place. This is why Brainstorm should allow this feedback from end users to flow freely to the bug tracker. Because brainstorm is a much friendlier feedback loop for end users than bug databases are.

We need to clean up the /home directory for the user from the cluttered hidden .configuration files: you go into /home/yourusername/ and reveal the hidden files. All you see is tons of junk, and when you need to find one, it is a total nightmare. If you want to backup your home directory, rummage through all the config directories created by the hundreds of applications and delete their temp files and caches prior to the backup. The dozens of .hidden files creates usability problems for everyone, power users and newbies.

Ubuntu should make a serious effort to work with upstream providers in an effort to get all major ubuntu applications to follow the freedesktop.org standard. That is to put all these files into ~/.config and all temp/cache directories under ~/.cache. I propose that this be done by setting two goals:

1) Short term goal: Start with the most common default applications. work on those that cause the most clutter.

2) Long term goal: Implement XGD on the rest of the apps. Start with the popular add on software most people use, and then on from there.

specification below (freedesktop.org):

$XDG_CONFIG_HOME defines the base directory relative to which user specific configuration files should be stored. If $XDG_CONFIG_HOME is either not set or empty, a default equal to $HOME/.config should be used.

$XDG_CACHE_HOME defines the base directory relative to which user specific non-essential data files should be stored. If $XDG_CACHE_HOME is either not set or empty, a default equal to $HOME/.cache should be used.

http://standards.freedesktop.org/basedir-spec/basedir-spec-0.6.html

[....]

I really like the front page displaying the most popular ideas of the day. However, if you know when the busy times of the day/week are, you can guarantee your idea to be a top idea for that day, week, and month because of timing. The votes/time sorting method is catering to timing, even though time is arbitrary.

I propose to modify the sorting method to either replace the current front page, or at least as an option to test out. Instead of the time bonus, I propose a "views" bonus. This would tend to bring ideas with fewer viewers to the front, and cycle through them. Also, ideas you voted on should be removed from the home page (continue to display them in other searches)

so logically: sorting value = [#Votes]/[#Views]

Once the idea gets front page exposure, unless it kept its vote to view ratio high, it would eventually slip its way down the pile and make room for other ideas who did not have such a good chance. It would behave as if a water pump were pumping water to the surface inside a large bucket of water. Only the popular ones would stay at the top. But even the bottom feeders would have their 5 seconds of fame, so long as they gained some amount of initial support.

You would want to make sure that each post kept track of its views by user. So if a user refreshes the page 20 times, it only gets 1 view. And anonymous views should not count, only logged in views. you might also want to make each page of results a little shorter so ideas at the bottom aren't given views which are never actually looked at by the user.

Thanks for listening.

Hi,

Being that all the 32 bit users are complaining about moving to i686 (from i386) and how it will break their old systems, why don't we just officially support the 64bit edition as the primary standard. All new processors are 64 bit now anyway, and you automatically get all the processor enhancements when switching to 64bit (guaranteed to support sse2, extended register sets, etc)

Problem is now that if you want to run 64bit, you lose compatibility with lots of stuff. Its not as well tested. It isn't the default installation being used. So you are either stuck with a low performance i386 distribution of an untested 64bit distribution.

Its tiring hearing about the i386 users who will be adversely affected if we move to i686. Well, lets just forget it all together and move straight to 64bit and leave them all with i386 as the backup. It should kill two birds with one stone (get us all on 64bit which is the future, and get us all on modern -march settings in gcc) and all the i386 lovers out there can still have their i386 compiled binaries.

When you click on the link "Most popular this month" it only has about 2 or 3 ideas on the first page that are not already on the front page of "Most popular this week". What this basically means is that this link is broken. There are two options to fix this problem:

1) Only show ideas greater than 1 week old
-or-
2) Give priority to the week 2, 3, and 4 ideas (ie: give week 1 ideas a 7-14 day sorting penalty).

Once this sorting issue is fixed, brainstorm will work much better. I filed a bug report on it, but it got no response and I think it is slipping through the cracks.

What do you think? Are there any other ways you can see to solve this problem?

I have noticed that the most popular (today, this week, this month) links are all date weighted so if you click on the "this week" link it starts with the most popular ideas from today for the first few pages before moving to anything new. So basically, the week and month links are useless because they just repost all the stuff from the today link and you are forced to scroll through pages and pages of ideas before seeing the requested content.

I don't know if this was intentional, or in error, but here is my proposal to fix it:

items in the "today" link should be items created less than 24 hours ago

items in the "this week" link should display items created less than 7 days ago, and more than 24 hours ago

items in "this month" link should display items created less than 31 days ago, and more than 7 days ago.

This will prevent the sorting by date/score weighted formula from cluttering the results with the "today" ideas and actually allow us to browse through the weekly and monthly ideas.

This is a weird one...

The problem is with the vote merging code. Since I didn't program it, I might not be all accurate, so bear with me. But the idea still stands:

When you merge two ideas, say one idea has a good, well thought out description and title and has 500 votes. Then other people post the same idea with crappy descriptions, or in inflammatory way, and it gets voted down. Then the bad idea is marked as a duplicate of the original idea, and all the negative votes will be detrimental to the good idea.

This is bad because people could maliciously vote an idea down by using this technique, and also because the bad character of other's (or bad description of other's ideas) should not cause a drain on a perfectly good idea.

My proposed fix is simple. if an idea is in the positive, and an other idea is merged in as a duplicate, it can only add to the vote (and not subtract from it). I don't mean only count positive votes, just that the net positive gain must be greater or equal to the net negative gain during the merge.

Other cases where this might be a problem is if the same people voted both ideas up, but the down votes are from different people. This could cause the "net positive" to be low and the "net negative" to be high, causing the end result to be a net drain on the master idea.

please tell me what you think about this idea, implementation issues, or other comments if you have experienced this problem in any of your ideas.

In the dashboard, there is "my ideas" and "ideas I promoted" etc. But the sorting method for each tab isn't really that good.

For example, "my ideas" sorts the ideas in order of popularity. But that order isn't optimal, because I often want to sort by date posted, so I can keep track of my newest ideas. But there might be times I want to sort by popularity as well. Same thing for ideas I promoted/demoted/commented on. There should be a button or sorting link at the top of each page allowing you to sort by:

date posted, date last commented, date last voted, most popular, unread new comments, etc...

Those are just some of the sorting ideas, there might be more or less to put in there. But it would be a great service for us because currently, it is not very easy to browse through your current ideas because they are usually somewhere in the middle of the pack and constantly changing position so it is not easy to find them easily.

Thank you

Hi,

If you have 2 pages worth of ideas under the "my ideas" link, page 2 comes up blank and you cannot view the rest of your ideas.

Please correct this behavior

The hard drive problem with laptop mode is a big issue, but this idea isn't the same as that problem.

What I propose is that the OS decide whether to spin down the drive not only because of time between reads, etc.. but also whether or not the user is using the computer.

Also, it should look ahead in the crontab to see if any scheduled tasks will be running in the next XXX minutes.

Also, the system should have two (or two sections) in the crontab so there can be "critical" crons and non critical ones. If the hard drive is spun down, it has the option to delay non-critical crons for xxx hours and then run them all at once.

The system by default has too many crons running that prevent spinning down of a hard drive to be efficient.

Please check out the dohickey client here: http://www.dohickey-project.com/client/about.shtml

It is an online database of user ratings for hardware. The user runs the program, it detects the hardware IDs of each device, and then the user can submit a 1-5 star rating of how the hardware works, comments, and suggestions to other users running Linux on the same hardware.

This program accomplishes about 10x more than the past 10 years of driver development on Linux simply because it connects users of the same hardware together and allows developers to see specific problems and solutions the vast majority of users are seeing in the real world. Something that does not happen in the traditional Linux driver development circles.

Virtualbox was easy to install and configure in my brand new installation of Ubuntu. However, the network was another story. I came to find out that it runs behind some sort of ipmasq behind the host machine's network configuration. It also has performance problems.

I propose that the network connection between a virtual machine and the host be bridged by default (or at minimum, the ability to bridge the connections by flipping a switch) so when you want to run VPN, or other specialty network apps from windows it works. Not to mention that if you are in a home network it behaves as a double NAT and we all know how bad that works. The #1 reason people want to run a VM with windows is to get the compatible with XXX legacy windows app that they have to have for work/school/you name it. And a crappy network install by default just makes things that much worse.

Steve Jobs and Apple have really started a whirlwind around the idea of switching to a mac and having windows installed inside a virtual machine for those "pesky windows apps" that the user just can't live without. In the educational sector (hint: where you can find lots of young users to switch) it is an every day occurrence where people will pay good money to buy VMWare or Parallels, a new mac, and a retail copy of windows xp home to install on the system.

We need this to be a prominent feature of Ubuntu. The first thing I did to show off a new Ubuntu install the other day was put Virtualbox with XP installed and stick it on one of the faces of the compiz cube. Boy was everyone impressed. Half the people thought it was a mac, the rest thought "wow, windows runs in a window and if it gets a virus I just zap it!?"

The imagination of things to come are limitless with VM's, but just to start out, can we have a sane way to run windows in a VM without having to do tons of behind the scenes linux network scripting and configuration?

The title says most of it. I think that in our user area, specifically the My Ideas link, we should be able to see just our active (or "New") ideas in one area, and all our inactive (or Duplicate/Done/Working) ideas under second link.

Also, when you have more than 1 page of ideas in your area, page2 does not work.

Man, brainstorm needs more intaction. I can't really get to have a conversation with anyone because there is no real place to have the discussion other than inside an idea, which is subject to change all day.

Is there an IRC channel, or forums, for brainstorm chat?

Hello,

This is intentionally similar to http://brainstorm.ubuntu.com/idea/5743/ but it is not a duplicate. The first part of idea #5743 was implemented, however, the second part of the comment was not.

Thanks for correcting the behavior of the "Most popular this week" link. However, I also posted the same problem about the month link, and while posts from the past 24 hours are removed, it still is being flooded out with bias towards the posts from the "week" link. Most of the ideas on the first few pages of the "month" link are the same ones from the "week" link.

I believe this is because of the way the votes/time since creation bonus is overpowering the other ideas.

Even if you don't want to remove the past 7 days worth of posts from the "month" link, maybe you can take away their time since creation bonus in a way that will prevent them from flooding the results.

Specifically, I propose that for the "month" link:

1) For posts created in the past 7 days, add 14 days to their "time since creation" for purposes of the weighted sorting calculation. This would effectively normalize their time bonus (to being in the same range of the "week 3" ideas). Remember, they are already given front page attention on the "most popular this week" link.

2) For posts created between 7 and 31 days ago, treat them as you currently do for sorting purposes.

There are a lot of posts about Kubuntu or KDE specifically, and also about xfce. Also there are a lot of posts requesting specific features in specific GNOME panels.

I propose we create either separate brainstorm categories for each one, or create at least one category called "desktop system" so we can file our ideas into there. Either one would be fine for me.

Right now, there is no obvious category to put Kubuntu, KDE, Xubuntu, Xfce, or Gnome ideas into. I would say most are filed under "System" but there are others under "Graphics", "Look and Feel", and others.

Thanks

I have been using Kubuntu for a while and it just is not up to par with other KDE distributions. I love Ubuntu and Kubuntu, but KDE in Kubuntu just isn't up to par with Ubuntu, or the other distributions shipping with KDE.

KDE4 is out now, and 4.1 will be out soon which will be the first point release of 4 intended for the masses. I propose that Canonical focus their attention on making the next Kubuntu release to standardize on KDE4, and polish it and make it shine. If this version isn't the best because 4.0 is new, then wait till the next version. But whatever you do, please make it polished and make it work well.

It is not good that the last 7.10 release was shipped with the most stable KDE version to date, but did not work proplery with wireless, DHCP, and adept package management tools.

I think it would be a good idea to have an option to sort the comments by the most divided ideas. What I mean is that a comment could be rated fairly neutral, but might have thousands of up and down votes on it, while another idea might have thousands of up votes and not many down votes. This makes it look like the second idea is has more interest, when in fact, the first idea has more voters and more people passionate about it. There is a clear distinction on these ideas and I think it would be worthwhile to be able to determine which ideas are very controversial, and for which ideas a consensus has formed.