Ubuntu Wiki

Ubuntu Open Week - Ask Mark - Mark Shuttleworth - Wed, Apr 25, 2007

see also Tuesday session.

<Hobbsee> can you ask what the status of the commercial repos are, and what canonical's plans for them are

• the commercial repos are expanding as companies realise that it's worth them going to the effort of putitng together proper, supported packages. many companies that come from Windows or UNIX don't appreciate how cool and important packaging is. hopefully, once they have a taste of that, they move towards a better understanding of free software in general and have more information to base their own decisions about license freedom around

<fql> QUESTION: Several ubuntu pages and also bug #1 in launchpad note that ubuntu is entirely free (as in free speech), however in the default install this is not true (drivers etc.), are there plans on making this more transparant to the user?

• we should always qualify that as "only free applications". at least, i fix it that way in the wiki when i run into it. we have had the driver exception in place since the beginning, so it's not as though this has changed. hopefully, the new flavour will also carry the "pure free" flag

<sampbar> QUESTION: sabdfl - Mark, what new features would you like to see implemented in ubuntu in the next couple of years? and are there any features you would like to see improved?

• of course :-). there are two separate lines of attack, i think. one is "the race to parity". we need to get our category-specific apps to be as good as the current best of breed proprietary app in that category. we need a word processor that is as good as word. a spreadsheet as good as excel. we already have a better browser :-). then there is the second race, which is "the thought leader", which is all about innovation. we saw with firefox, when it hit parity, that the extension mechanism was a HUGE driver of creativity and innovation. this is why i fund bazaar - because i think innovation comes from the periphery of the community as much as from the center and we need revision control which is cross-platform, robust, and distributed. i would like to see us reach parity in more desktop apps and of course, i'd like people to start to see free software as the place where the genuinely groundbreaking innovation happens. this is also why i think it's so important to get compiz/beryl into ubuntu, because they are fertile grounds for creativity. the proprietary guys are not far ahead. if we give developers the tools and an audience, they will do amazing things

<dee> QUESTION: Today I read (German: http://www.heise.de/newsticker/meldung/88800) that Jane Silber has said that Firefox 2 might come for Dapper Drake? Is this true? And do you want to upgrade all firefox-dependent packages or should it be just additionally to Firefox 1.5? Doesn't this stand in a strong contrast to the LTS?

• **sabdfl scrambles to find a translation of that article

  i can't find jane either, so not sure of the context. my gut feel is its unlikely we would make radical changes to LTS releases without a long -proposed test period. my other gut feel is that jane doesn't speak unless she's certain of the facts, and probably knows more about that than me. so i can't really add anything useful now, and will just ask for the next question!

<McKinney> QUESTION: i read an interview with andreas barth criticizing ubuntu. what do you think, in which way can ubuntu and debian profit from each other in future? http://www.golem.de/showhigh2.php?file=/0704/51531.html&wort[]=andreas&wort[]=barth

• again, hard for me to respond, i can't read the german. however, i think debian and ubuntu already benefit hugely from one another. we have brought debian millions of new users, in a category that it was never previously taken seriously in, the desktop. in addition, many new DD's come through Ubuntu, or first discover debian through Ubuntu. we are expanding the debian universe, which is very good for debian. we lead a lot of very useful work, which eventually gets included in debian. they are considering upstart, for example and also considering our live cd infrastructure. perhaps even the ubuntu installer. they largely depend on work we do on the toolchain, on python, on java etc. the flip side is also true. we benefit hugely from debian's depth and breadth. we consider ourselves to be part of the debian family. it's sad for me, that when ubuntu releases there are 2 messages about it on debian planet - one of celebration from a person who contributes to both, and one a bug report, but i'm happy that, when debian releases, there are tons of congratulations to debian on planet ubuntu. i would like to see better collaboration. many dd's routinely read patches that we automatically mail to them when a package is modified in ubuntu. others just ignore them. it would be nice to have debian recognise the contribution ubuntu makes. we get twice the volume of bug reports now, not because we are more buggy, but because we reach a wider audience. debian would benefit if they took an interest in ensuring that their packages are getting that wider exposure

<zorglu_> QUESTION: what is the status with cnr ? any idea of when it gonna be usable for ubuntu ?

• i know the linspire folks are working on that, and i hope it will be ready soon! but i don't have an eta

<popey_> QUESTION: how do you feel about the way you are portrayed by the online comic strip Everybody Loves Eric Raymond?

• i love it! did you see today's? http://geekz.co.uk/lovesraymond/archive/taking-freedom-further. as they say in the movie business "did ya spell my name right? there's no such thing as bad publicity"

<[doctor]> QUESTION: Mark, what about "Ubuntu Control Center" as Yast or Mandriva Control Center ... A lot of newbies ask for it

• first we'd need to have good packages, so if those users want to start work on that we can get them into universe with MOTU help, and then consider them for main

<Moniker42> question: mark, what do you think of Charles Simonyi's claim to be "the first nerd in space"

• hey, i know a bunch of russians and americans who are just as nerdy as me and him, and went up before either of us. nonetheless, i think it's great that he flew, and hope he had a smooth flight. it's a privilege to get up there, and i'm envious of anybody who goes without me!

<popey_> QUESTION: What are you doing to counter the claims that launchpad is evil because it is closed source?

• funny, i had an email about that the other day, from a gmail address :-). first, i would like to have LP published under the GPL. i have a plan to get there, in a way that will (a) not fragment the eyeballs, and (b) allow us to continue to pay the developers. i don't see the point in releasing it until we can get those two things right. the main issue right now is that LP is a centralised design, when really we need a federated standard that works for bugzilla, roundup, sourceforge, collab, launchpad etc. if we released it, and 10 people set up their own instances, then the work required to keep track of everything LP keeps track of would go up by 10x. alternatively, the quality of info in LP would go down by 10x! so, for example, right now if a bug is filed in zope and the zope guys think it's also in the ubuntu zope packages. they can get us to look at it just by updating LP and if they think its in gentoo too, they can link to their bug tracker and they only do that once per community. if there were multiple LP's, they would lose that

<spr0k3t> QUESTION: I've seen some of the estimated numbers for *buntu, but I'm curious... how many unique IP addresses did the canonical servers see on th 19th of April?

• wow, not sure i have that number, but there were 53 mirrors before we announced, and 130 by the end of the day, that we know of. so hits on canonical.com are a small fraction of total. i think we were serving 12 gigabits / second from ourselves and top 5 mirrors :-). probably 20 gbits/s in total. 3 cd's per second for 12 hours. pretty amazing. if you were in #ubuntu-release-party, you know it was quite a rush

<gumpa> QUESTION Does Canonical have a relationship with the OLPC folks?

• not officially, no. i would like to, i think their work is amazing, and ubuntu is a good fit, but we've never been asked to participate

<deniz_ogut> sabdfl QUESTION: How do you evaluate the chances/challenges for Ubuntu based business models. Can Canonical try to push governing bodies in proper ways to spread Ubuntu in addition to local voluntary efforts?

• good question, i guess we will have to find out! we are already seeing that many governments like ubuntu, and base their own efforts on it. we have official relationships with some of them, and it works very well, but it's not universal. free software is still very new in government terms - they take a LOOOONG time to think things through

<pixelpapst> QUESTION: I am leading the development of a highly customized CDD, and am considering switching to Ubuntu partially as the base. Would I be expected to track all of our work in Launchpad ? I am concerned with keeping security-bugs secret (and cannot pay Canonical lots of money for this, sadly).

• you would never have to use LP, use whatever works best for you. we will have a security bug framework that lets you keep those confidential without being a subscriber (but only for the designated security team) and if your project is community driven and non-commercial, we will in many cases give you free access to LP advanced features

<darich> question: are there any plans for large scale marketing campaigns a la Firefox for Ubuntu(full page adds etc..)

• you mean bigger than the billboards? that was a fun campaign. all in good time. first we have to crack the corporate market, and be sustainable. we are taking a very unconventional approach, but we are on good track. next week you will see two big announcements, one of which will probably dominate the media, but both are really nice steps towards sustainability for the project. our goal is to be sustainable and completely free. we are neither right now, but we are moving towards that goal very steadily

<sampbar> QUESTION: Mark, do you feel that open office is in the right direction to becoming as good as word and excel?

• tough question. i'm getting worried about their developer approach. it still feels like a closed community. i know they WANT to fix that. the head of that group at Sun has the right ideas, but it's hard to move a big ship. oo.o has been a huge boost for free software, because it is cross platform. i had lunch with an interesting lawyer today, and he was telling me that he installed oo.o and was "really impressed that it worked so well". he was amazed that it just opened word docs nicely, and he could find his way around easily. great! but it doesn't have the rapid evolution that we see in other Gnome or KDE projects. i think AbiWord, and KOffice would be slicker, faster, lighter if they had more developer time, but because OO.o is the 800 pound gorilla, it's hard to get a lot of developers on those others. 63,000 unique IP's hit releases.ubuntu.com on 19 april, not bad for half a day's work :-). so i would like to see oo.o get "more open." it took mozilla long time to gestate and produce firefox. we need the same thing from oo.o

<grail> QUESTION: You have written a quite interesting piece about DRM on your blog, so what would be your recommendation for the music industry. How - in your opinion - should they position themselves?

• i don't give free business advice outside of my blog

<McKinney> sabdfl QUESTION: in an interview you said that maybe gutsy+1 could be the next LTS-release. is it already concluded which release will be the next lts?

• tech board needs to discuss this, and canonical needs to take a view on resources too. there are lots of questions
  • (a) which will be the next LTS (b) will we follow the same process as we did for Dapper, or do it differently? and many others
  i think it will be gutsy+1 or gutsy+2

<Riddell> QUESTION: some developers are wondering why UDS Sevilla specs from canonical employees are being organised on a private wiki and not publicly on launchpad

• **sabdfl goes to look. i don't see anything in the proposed list that looks that way. there are some specs i know of which have not been submitted, because they relate to unannounced partnerships (but hey, there's only a week in which we have to announce them). everything at the UDS's happens transparently. *sometimes* we'll have private meetings with partners there, but the sessions are generally open to all participants. we once had a funny situation where we had a private meeting with a very big partner and then we realised that one of the folks, who we thought worked for them, they thought worked for us, and was actually a community participant who worked for a media company

<welshbyte> QUESTION: which areas of the ubuntu community do you think need the most focus towards their expansion at the moment in order to cater for the huge amount of bug reports and new users that Ubuntu is seeing?

• developers and bug triagers. there's a nice bug triage community forming. they have a fantastic impact, because they help to distinguish between bugs which can be reproduced and those which cannot. they help the developers to focus on the items that are most important

<Daviey> QUESTION: Why is 'ubuntu' a trademark of Canonical and not the Ubuntu foundation?

• Canonical is the funder of the brand. The Ubuntu Foundation is a trust that exists in the event that Canonical can no longer support Ubuntu. so, when we made a long term support commitment, i figured out how much money we needed in the bank to meet that commitment and setup the Ubuntu foundation and gave them that amount of money. no matter what happens to Canonical, the money is there to meet the support commitment. you can safely deploy Dapper and you will not suddenly find yourself unsupported, but we don't touch that money, it's an insurance balance. so, in that light, it makes more sense to have Canonical manage the brands. it invests in the marketing, and also makes decisions about the ways in which people can use the brand. so far, it's working well. http://www.ubuntu.com/aboutus/trademarkpolicy

<pwnedd> Question: For a large number of people, the number one reason for getting a pc is for gaming. Right now however there are few mainstream games written for or even compatable with linux. Do you think there is anything the ubuntu team can do to change that, and open up linux to a larger population?

• i agree this is a big, important question. gaming developers are like other ISV's. they write software to sell software, so they want to see (a) how many people use a platform, and (b) what percentage of them WOULD PAY to use their software on that platform. that gives them a market size. they then compare that to the cost of the port and the lost time in making the port (while their competition works on new products). and then take a decision

  <popey_> QUESTION: Please buy transgaming, and package cedega for us :( they key is not who owns cedega, the key is the business model that supports it. no point in buying it and changing the model if that means it becomes unsustainable

<flossgeek_> QUESTION: Do you not think it would be a good idea to push the sun java stack as part of a solution in the enterprise and create a good partnership with SUN

• i'm cautious of the word "push". we try to choose the best free software components and make those instantly available and have everything else accessible from the network package repos i do think the SUN implementation of Java is the best and can't wait for it to be free software. most of the core should be ready for gutsy. some components will not make it because of third-party license issues, but it's a big step forward. i do like to partner with the best companies in each sector and for Java stacks, that would have to be SUN

<maccam94> QUESTION: how would you convince a school or company that ubuntu is ready for the desktop? there are very few other local deployments, so the impression is that nobody else is using it...

• schools and companies are different. for education, free software has huge advantages. the tremendous breadth of free software makes it possible to teach EVERYTHING using computer tools: art, science, maths, computer science, music, you name it. i don't know a school that has all of the windows apps to teach those, but it's easy to do so with Edubuntu or K12-LTSP. schools and companies do share the same support / ecosystem concerns and there its just a question of showing them where they can buy support and training. it will take time

<Armagon> QUESTION: What steps would you advise an experienced programmer (who lacks F/OSS development experience) to take in order to quickly become a valuable contributor to the community?

• find something you are personally really interested in! preferably a smaller component. that's written in a language you know. then climb in! a good community will welcome you. if you don't feel welcome, look elsewhere. so WELCOME!

<McKinney> QUESTION: lets suppose you're on the way to a public discussion with a debian enthusiast (not the debian "evangelist" himself, but quite close). what would your strategy be to come out of it as the "winner"?

• hey, we've already won with him, we're on the same side, thanks everybody! phew