Contributor jhoger

Often after doing an update the double arrow icon will appear indicating a need to reboot. But if you click it, no reason is given as to why reboot is needed.

My idea is that clicking on the icon (or a "Why?" link for more information) should indicate why the OS thinks a reboot is needed so I can make a decision about whether to do it now or later.

my other ideas

I don't see any intro/rules/FAQ on the main Brainstorm page.

I'd like to see something that clearly documents the various features available in Brainstorm, markup available/allowed in messages, a short list of common proscribed ideas or types that keep coming up that don't need to be posted again, etiquette, etc.

Imagine a desktop where the organizing principle for workflow becomes the "project" or "task." So I'm working on a coding project. Any research I do in the browser (history and bookmarks) automatically attaches to ("gets saved in") the task. Instant message or IRC sessions in that task are kept organized there. Documents I create associate with the task. Each task would feel a lot like virtual desktops today, but you could shut down your machine and pop up the same virtual desktop with your apps open in the same positions.

I'd like to see a task-oriented desktop realized (starting point ideas are in Plasma, Gimme). Basically take Virtual Desktop, "session" across reboots, and tag/metadata based file system, and application integration to permit work you do to be associated with different projects or tasks across applications.

You would be able to close a task and reopen it. Move a task to another machine. Switch between tasks as easily as virtual desktops. Easily post-hoc re-classify data events that occurred over a period of time to a given task in case they associated with the wrong one.

Also the window manager should do more of the window management, as IonWM, dwm do.

We're at 900,175 votes now. My idea is to heavily publicize the 1 millionth Brainstorm vote to build some more buzz around Brainstorm and Ubuntu.

This is a significant amount of direct user feedback on the distibution from all levels of users. It is an accomplishment.

I'd like to see the vote go along with a reason similar to Slashdot. It is a canonical summary of *why* you are giving the +1 or -1 that you are giving.

For example, I'm interested in promoting "innovative" ideas. With no ability to search on innovativeness, it's hard to find the ideas.

Being able to give/receive such reasons would give constructive criticism. It would be a shortcut for those who like to leave comment feedback.

Searching for LowHangingFruit would be cool for developers with some free time that want to do something impactful.

Some positive moderation would include:
Innovative - A great idea for a new way of doing things... shows "outside the box" thinking.
LowHangingFruit - Easy change that would make a big difference.
Obvious - Well, yeah why isn't this already done? Indicates a bit of surprise that it doesn't already work that way.
+0: DontCare,Funny

Negative moderation would include:
Unnecessary: There's already a good way to do that
Inconsiderate: would make things better for a few but worse for everybody else
WhatAreYouThinking: this idea would make things worse for everybody.
SciFi - that would be cool but it's not in reach of modern technology. "Intelligent Agents" for example.
TooHard: may or may not be a good idea but it would take so much blood and treasure that it deserves a -1.
Unintelligible: The submitter is much too clever to be understood
PureNoise: Annoying joke/sarcastic ideas
Abuse: not a legitimate idea, just an attempt to mess up the site.


[....]

Today ideas live forever. They generally keep whatever priority they get from users in the first day to week. This fails to take into account that the goodness (or badness) of an idea changes over time.

I recommend that if an idea has been in Brainstorm 1 year without being marked Done, then the original author (or author of a dupe) will be notified, and given a chance to automatically (clicking a button) resubmit it. If no one asks that it be kept alive, it will be moved into the BS Pile and some other users can sift through them and request revival. It would start its reincarnated life with 0 votes, but all original comments. The original author name would be unchanged.

This would clean out old crufty ideas. In particular this would deal with the fact that the userbase and "how BS works" changes over time. This will give ideas a fair shake in whatever the current "best BS ever" happens to mean.

I think it would be useful to be able to "beam" an open document running in a given application to another machine.

Say you have an OpenOffice document on your laptop, and you want to move it to your desktop. Open Nautilus, browse to a remote machine's share and drag the file in, go to the other machine, navigate to the file, and double click.

Imagine if you could just drag the running program window to the network "Place" representing the remote machine. The program state is moved to the other machine and the Window pops up there.

Just like a Star Trek "Transporter" for running programs.

One trick would be if the application you need on the other machine isn't installed. If so, it prompts you to install it at that time. This would be an intuitive way to install programs.

Note this is not the same as XMove or just setting the DISPLAY variable. The point is to actually beam the running instance.

I typically browse brainstorm by typing brainstorm.ubuntu.com into my address line.

But the default page that comes up is "Most popular today" which I don't like. I'd rather see "Latest ideas"

The idea is to add an option to the user profile that selects which brainstorm sort order is displayed when I browse brainstorm.ubuntu.com .

Occasionally you see comments on brainstorm that thus-and-so is an upstream issue, in a way to discount the importance of the idea here. I think this is irrelevant and a cop-out.

Users see Ubuntu as an integrated product, not a platform. That actually works in our favor: Windows is "just a platform." Ubuntu can be much more and thus provide a richer experience.

At the very least, for a set of applications considered "core", the distribution should take full responsibility for feature requests. A major value-add of Ubuntu is that it is not just a repository of well-made packages plus a release schedule. It endeavors to be a cohesive, usable whole.

So if a suggestion comes up for a feature in OpenOffice.org, I certainly think it is fair game.

So the idea is to push this as a cultural norm on Brainstorm: upstream issues for core applications are our issues, and that includes feature requests.

VMware and QEMU allow you to run Windows and Windows apps on Linux, but the app is confined to a complete virtualized desktop. This works well but in daily work it can be awkward.

Parallels for OSX has the "Coherence" feature that allows Windows applications to run as a window on the desktop. In X parlance, the application runs "rootless." Presumably the program acts as a Windows device driver or hooks GDI calls and translates them to X messages.

My idea is that this feature be added to QEMU for Ubuntu. This would permit installing Windows in a virtual machine, or utilizing the components of an existing installation of Windows on another partition to run Windows apps natively in a natural, familiar way.

For folks who absolutely require one or more proprietary apps, this would permit them to switch to an Ubuntu desktop.

EDIT: VirtualBox "Seamless Mode" looks like it does a good part of this. So maybe all that's needed is to opportunistically "find" the Windows partition during install, and do some "polish" things in terms of smooth integration.

Note the "Get (Paid) Support" button on SourceForge projects. This is the "give the program away, sell support" open source business model. However, most users will never find that button because they don't get their programs from SourceForge. Rather, they get their programs through their OS: Ubuntu.

Unfortunately there is no way that Canonical can provide expert support on every package in Universe or even expert support for every package in the core repositories.

My idea is that Canonical should put a certification program, standards/quality control, etc. in place for allowing upstream developers or other 3rd parties to do support under the Canonical/Ubuntu umbrella.

Canonical could charge a transaction fee, % gross sales, or a one-time yearly fixed marketing fee to facilitate connection with outside parties to provide support.

This would help Canonical get income that could be plowed back into further development and improvements. It would also help upstream to monetize their work. Ubuntu users get high quality, expert support. Win-win-win.

It would be important to make the steps, cost for initial establishment of communication with registered support people easy to find in the program they need support for. Ubuntu is in a good place since through the OS, Ubuntu has the final direct contact with the user (customer).

This idea was improperly moderated: Add a link to Debian homepage on http://www.ubuntu.com/ homepage as "already done" even though there is NO link on the Ubuntu home page (whether Ubuntu should do such a thing is a separate concern).

That stopped the voting cold. There is no way I can see to formally complain about this. It smacks of censorship since it stopped the voting process.

My idea is to add a formal review process that anyone can report into and follow look at. Maybe this was just an error... I certainly assume and hope so. If so, a review process would sort it out.

In general, the mods need to be careful especially when there's a possibility of the appearance of a "conflict of interest."

(Note: I'M NOT recommending BASIC here, read further...)

Way back when all home computers powered on instantly to an OK prompt where you could type in BASIC programs. This encouraged a lot of hobbyists to experiment and become programmers.

These days, all computers can still be programmed by the user but either you have to download a giant, intimidating dev environment, or you have to dig around to find the scripting language for your system and dig much further to find docs.

Since Ubuntu is based on programming by its users, I think it should encourage programming by installing, distributing, promoting a friendly scripting IDE. Probably Python since it has fairly simple syntax.

Pippy could be back-ported to Ubuntu:

http://wiki.laptop.org/go/Pippy

It should come with a friendly tutorial utilizing lots of cartoons (like the old Getting Started With Color BASIC for the TRS-80 CoCo).

Also, like the OLPC XO-1, you should be able to easily get to the source of any script based programs on your system. If it's written in Python, be able to right-click it and pop it up in the friendly IDE.

There is a class of ideas on Brainstorm that come up again and again. They are the "Yeah, So What?" ideas. For example, "Make boot faster."

My idea is that we should have a ceiling on # votes at least for the front page. Once an idea reaches, say, 50 votes in a day it no longer gets to be on the brainstorm front page. If it has 50 votes, it's pretty clear that it is a "good thing." This would give other ideas a chance to get vetted. No one wants to sift through 20 variants of "+90 Make boot faster." The ceiling would have to be adjusted over time, I'm not sure what the formula should be.

Also, ideas that have a large second derivative of +1 votes over time should automatically be suspect to the admins as "Yeah So What?" kind of idea. They should quickly be scrutinized as possible duplicates, and removed from the front page for that reason.

I'm not saying that the idea can't be voted on any more. I'm just saying that the front page is special and ideas need to be nurtured at the very least for the first day.

Often Brainstormers credit another user with an idea. Also comments are often prefaced with a user's name you are replying to.

My idea is that these user names (just like appear on ideas) should be clickable.

So whenever a user is credited, browsing to their ideas is a one-click affair.

To make it work, either allow simple markup ala ##jhoger## in the idea, or just match the text against the username database and ask the user on submit if they would like to make them clickable. The second approach may be to processor intensive on the server side, so markup may be the best that can be done.

Add comment threads and moderation to brainstorm. This would allow better organization of the comments and the ability to see if someone has responded to one of your comments.

It would also allow the average BSer to moderate abusive comments or highlight insightful comments. As a flat list of messages it is hard to separate the wheat from the chaff.

This would allow better discussion.

Brainstorm Search is really, really, really slow. I find it completely unusable. It is no surprise that folks aren't finding dupes when search is this broken.

My idea is that a Google site search box should be added to Brainstorm until the Search engine can be fixed. (NOTE:I am not suggesting that Brainstorm search should be dropped. I am suggesting a workaround until Search can be fixed.)

Till then, BSers: try
site:brainstorm.ubuntu.com (then search terms)

in your Google search to find things exclusively on Brainstorm.

It would be nice to be able to look for a given user's other idea submissions. Or, if I want to share all my ideas with someone else, to be able to send a simple URI that displays them. The reason is that someone with one good idea may have others. Also I may want to promote my ideas by sending a single link to a friend.

On the User Menu there is a link http://brainstorm.ubuntu.com/my_ideas

But ideally we would be able to do

http://brainstorm.ubuntu.com/jhoger/ideas

EDIT: figured out this is possible already:

http://brainstorm.ubuntu.com/contributor/jhoger/
Maybe we just need to add the ability to search by user in the Search ideas page so that this fact is more easily discoverable?

I keep reading in the press that Canonical is not yet profitable. Yet Ubuntu has quickly become one of the trusted "brands" in the Linux community.

It is imperative that Canonical become profitable since it is the thought engine that drives Ubuntu beyond just being another Debian derivative.

My idea is that Canonical should leverage their brand and become profitable using techniques that are not in conflict with the free software ideals.

a) Charge for support
b) Charge contracting/consulting fees for adaptation/customization
c) Advertising/marketing agreement fees for connecting downstream customers with upstream or 3rd party developers
d) Publish books
e) Get into the "trade show"/conference hosting business
f) Offer web hosting on Ubuntu servers
g) Offer paid-for certification on Ubuntu products
h) Sell Ubuntu foo-foo merchandise (hats, shirts, ...)
i) Sell Ubuntu branded computers (desktops, laptops, servers)... partner with an existing OEM to make these Ubuntu branded lines.

etc.

Probably Canonical needs to pursue a combination of these. Their long-term competitive advantage is their brand. If they guard it and build it up then they can sell all kinds of things with their name and logo on it.

Votes should be secret but right now they are not.

Why should votes be secret? For the same reason that votes and surveys responses are usually anonymous in real life: concern about retribution or other blowback.

Say another BS'er sees that I voted against his pet idea and he jumps over to a forum and asks his buddies to come vote against all my ideas.

Now I generally DO comment even on things I vote against. But it's my choice. If the voter wants to reveal why he voted such a way he can put a comment (or Brainstorm devs can add a "vote reason").

If it's left this way I think it's important to let the users know that this is the way it is. I didn't know it till I started playing with the new "My dashboard."