We Are *All* Making Ubuntu - Part 2

Yesterday, I introduced an opportunity for everyone who wants to help make Ubuntu's search feature the best in the world to participate in a specification hosted here:


(Note: You will need to login with your Ubuntu SSO, and also be a member of the ubuntu-etherpad[1] team on Launchpad. Join it here: https://launchpad.net/~ubuntu-etherpad

There are already some good ideas and constructive discussions on the Pad. Go ahead. Take a look, then add some more.

One thing missing though are mock-ups, wireframes, and prototypes. Do you have a background in designing interfaces? We need you!

I am not a programmer, nor do I play one on TV. But I do know that code does not look like this:

Really bad code that doesn't make Ubuntu better.Really bad code that doesn't make Ubuntu better.

Will the people that invest all that time and energy contributing to the bickering, name-calling, and dredging up the past take the time to participate in building a constructive future together?

Paul Ferson, Mohit Kumar, Emil Protalinski, Alex Williams, and others, I challenge you to use your words for good, not click-throughs. Join us here and become part of the solution: http://pad.ubuntu.com/4OnwYN3HVT


I'm waiting for approval to join the etherpad team, so I can't view the pad yet.

I'd love to see agreement on a solution, but I'm sceptical that it can be reached. As I see it, the problem solved by the Amazon search results is not "users can't search Amazon easily enough", it's "Canonical isn't making enough money". I would like Canonical to make enough money, because I want Ubuntu's future to be assured, but I'd like them to do it without compromising privacy.

The key issues for users are that it does remote search by default, and the easiest way to look for applications or files also performs a remote search. But making this any less obvious - even requiring one extra button to search online - would massively reduce how many users use it, and hence how much money it makes Canonical.


Hi Thomas, thank you for your comment.

No one that is in a position to know is saying that "Canonical isn't making enough money." A peanut gallery (of a handful of journalists) like to pretend they know or speculate as it makes for a juicy story and gets them page-views. Then there's all the competitors out there that would *love* for everyone to think that there's some kind of financial emergency. And even if there were, there's a contingency fund on standby: http://en.wikipedia.org/wiki/Ubuntu_Foundation

The reality is more likely that things are humming along and that people on the inside want to grow Ubuntu even faster. If adding revenue through search is helpful in that regard, then I'm standing behind it. The world needs a better solution than the current cartel ASAP, not in 10-20 years.

Opt-in vs. opt-out is a false dichotomy. A better way to imagine this is Opt-intelligently :) The initial search stance is largely irrelevant if it is super easy for people that enjoy Ubuntu to express their intent.


My thinking is that, if Canonical was already financially sound, they wouldn't have done this - or at least, they'd have spent longer getting it ready for prime time. They could have added more search providers, implemented proper preferences, let people test it in a PPA. The last-minute addition with only Amazon results, and the subsequent scramble to fix major problems before release, smacks of desperation. I don't want to believe this, but it's the only explanation that seems to fit.

A recurring theme of interface design over the years is that defaults matter, so I disagree that opt-in/opt-out is irrelevant. As a user, I want to be confident that when I search for files, Canonical is not listening in. Canonical seemingly wants the Amazon search results to be obvious to as many users as possible. Somewhere, there has to be some default that manages those aims - and Canonical gets to set it.

Your effort to make a positive conversation about this is commendable. Certainly, parts of the discussion have been far too pessimistic. But if we think, as I do, that there is something amiss, we shouldn't be shy about saying so.



I think its overly optimistic to believe that Canonical is going to change their mind since statements from employees at Canonical have already drawn a line in the sand suggesting no interest in budging on the issue.

"the simple answer is that the online searching capabilities of the dash are not going to be removed." - Jono Bacon (12/10/12)

" Canonical has made some changes based on feedback, but not all changes, and that responsibility lies with those who make the decisions around the policy." - Jono Bacon (12/10/12)

Now you do realize that Unity is a seperate upstream project (https://launchpad.net/unity) from Ubuntu right in fact so are the search lenses... So when you say "We all make Ubuntu" thats accurate but we do not all make "Unity" or "the Shopping Lens" both are Canonical projects (https://launchpad.net/unity-lens-shopping) and the Community are not decision makers in that development process and Canonical has been very clear so far that they are not going to budge.


Hi Ben, and thanks for your comment.

Times can and do change. And a truly compelling solution to this problem would be tough to beat down. (A bug report, or a really long bug report, or a hundred bug reports are not compelling solutions. A few nasty blog articles, or a manifesto from RMS, or Jono, or anyone else is also not a compelling solution. Re-hashing the controversy is also not a compelling solution.) Great software *would* be a compelling solution, and that requires great design and running code.

We can all make Lenses. We can all contribute patches, even upstream. We can help people imagine the unimaginable, even if they work for Canonical. We can even help those who would stop before they even try... ;)

Optimistic? Guilty as charged. Ubuntu was born from optimism. Have we lost the will to dream?


"A bug report, or a really long bug report, or a hundred bug reports are not compelling solutions."

Actually thats how Ubuntu Development's lifecycle works. (https://wiki.ubuntu.com/UbuntuDevelopment)

"We can all make Lenses. We can all contribute patches, even upstream."

Sure but ignoring the Ubuntu Development process is going to get patches ignored and creating more lenses does not address the privacy concerns in ones that already exist.

Mind you I have spoken not only with Legal, Design, Community Team and Core Devs its pretty clear Canonical is sticking with the feature as it is.

But I do hope you prove me wrong :) if that counts for anything.


Hi Ben,

We can design Lenses that work better than anything that has come before. Heck, we could even introduce something new to contain all these lenses.

One path I've seen work in Ubuntu is:

Concept --> Specification --> Blueprint --> Prototype --> Code --> Review --> QA --> Release

Legal, Design, Community Team, and Core Dev's might all be unable to visualize what the world that loves Ubuntu can visualize. We have a different "lens", pardon the pun. ;)


You are asking for ideological problems to be solved with code, merely as an attempt to deflect blame and play the victim. I've seen so much of this bullshit in the F/OSS community recently that, frankly, it's starting to bore the pants off of me.

Almost everybody criticising this has suggested methods of implementation, or non-implementation thereof. Canonical and Ubuntu have however, as usual, ignored consensus and continued along to further their own agenda.

When Ubuntu is ready to accept a patch to make external searches opt-in by default, I and many others will be quite happy to provide it. Until that time, however, please stop passing the buck on to some unnamed other to cover for Canonical's agenda. It's really embarrassing to watch.


There are other ways to solve this other than "opt-in by default."

Hostility is not getting us any closer to fixing the issue(s). I recommend that you check and contribute to the Pad.


There being more than one solution doesn't change the fact that a vast number of your users have spoken out against this change. It was not properly vetted, it was not properly consulted upon, it should be stopped. Period.

I have no interest in contributing to a project that requires me to agree to the CICLA, other than to contribute code which makes opt-in or removes the Amazon search from the base installation.


If you code the "opt-in" or "Amazon-removal" tools, then the community will have something that they can apply in the event there is no movement., or if things go really wrong Isn't that worth a try?


I think you are ignoring the point Chris and I have tried to make which is that there is almost no chance Canonical would accept such a patch. I know one person who is well respected in our development community who has already offered to address the bug and Canonical declined.

There is a process in Ubuntu Development and I really do think your ignoring the proper path for addressing bugs. You can create lens all you want... (we have over 100 lens on Launchpad already) but the reality is it will not ship default in Ubuntu without a greenlight from Canonical Product Strategy.


My bet: A patch that is *remarkable* would be accepted. And if not remarkable enough for Canonical Product Strategy, then it can be something that's applied after, from a PPA or similar. Thoughts?


So your saying that a former Canonical employee who works on security who pitch the patch with a solution just didnt offer a remarkable patch at all? Several pointed out that he made a excellent pitch but was just ignored.


I don't know the specifics, so it would be helpful if you (or he) could post a link to the patch in the Pad so the community can look at it. Thanks.



The Community is pretty much aware of it because he is also the community in fact one person on the community council said "He really made the case".


Thanks for trying to make this discussion positive, Randall!

Wish I had time to contribute to your pad, but it looks like it's working pretty nicely. That makes me happy :)


Thanks Dylan. Keep up the good work on all that you do for Ubuntu :)



A lot of people are just asking for opt-in remote searches. Why not just adding this option to the installer (unchecked by default) in the same way it is done for proprietary drivers for example?

This would make everybody happy, a lot of (IMHO revelant) critics would no longer be spread, and it would give Canonical a good real metric on how much people want this feature (if the packages for remote searches is to be downloaded on demand).



Thank you for your suggestion Philippe. Opt-in has advantages and disadvantages, and I'm sure a balance will be reached soon.


Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Images can be added to this post.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.

Tip Jar

Liked one of my articles? Please consider clicking the "Donate" button above

Namecoin (NMC) is also appreciated:

Thanks! Your support helps keep this site free and interesting.

An Ubuntu show right from Vancouver!

Real Local Community