Category Archives: work

cool stuff we do at work

Mir and Chromium

Ubuntu’s Display Server Mir is gaining more and more traction and the team is making good progress on the platforms that are at the core of Ubuntu.

Mir is proving itself everyday to be the exact technology that Ubuntu needs to power mobile devices. Mir’s features are on par with the requirements that we put out to run Ubuntu on mobile devices. A few recent highlights include the screencasting API, Sidestage and improved right edge navigation (in concert with Unity 8).

Continue reading

First Mir benchmarks

The first neutral (i.e. not published by us) benchmark of Mir is out. Michael over at Phoronix has a good write up of the current state of things and also mentions that the install was smoother than anticipated.

The results (~about 10% median, 15% average penalty in FPS, see below) are totally within the expected range for where Mir is at right now and doesn’t have us sweat. Continue reading

Running Mir

The documentation for Mir is growing and we also have instructions out how to get Mir running on your computer, I wanted to briefly summarize the necessary steps to get Mir up and running and how to go back.

Please be warned that while I tried to carefully document all necessary steps you might end up with a system that doesn’t function as expected and which will require further intervention, a reinstall or might even suffer data loss (most likely during the OS reinstall;). Also, don’t use Mir on a production or public system, as there is at least one security related bug.

Continue reading

the first 8 weeks

It has been 8 weeks since my first day in my new job at Canonical and I wanted to share some impressions with you guys.

Role
After too many years in QA I was looking for a new challenge, something that drives me to my limits and lets me push myself beyond these. Well, I guess I was heard…
In my current role at Canonical I am managing a stunning team of Software Engineers in Canonical’s Product Strategy division, which is headed by Mark Shuttleworth. We are focusing on the “plumbing” layer of Ubuntu’s Unity shell and everything related to touch and gesture technologies . We provide basic Unity & system services to the Desktop and make sure your browser scrolls smoothly across your favorite web pages. Right now I am spending a lot of energy in growing the team together with the rest of the Management team.

Team
Holy Guacamole, what a great team!!!… across all levels of the organisation! Canonical’s hiring credo is to hire the best people, regardless of their location – thus we all work from home.  This has mostly personal advantages (there is seldom a major traffic jam on the way from the bed room to the office downstairs) but also some drawbacks (you just can’t beat face time). However, I am astonished how efficiently the team collaborates together. In my (limited) view, this is mostly because of the motivation that every single one brings to the table to make it work. When people get together for a sprint to work on a project you can literally sense the energy that is in the air.  I haven’t felt so energized in years! (Disclaimer: Don’t get me wrong, I loved working with my old team very much, but the thrill of something new and exciting was missing – we have been routined & seasoned professionals)

Innovation
At UDS-P last week, Mark announced that by “… [April 2014] (14.4), Ubuntu will run on all personal computing devices, such as TVs, tablets, phones and will integrate them seamlessly to your desktop and cloud storage”. Admittedly, this is quite a big announcement and I have heard lots of doubts whether Canonical will make it or not. Let me tell you “we will”. The existing team that we are building on plus all the new rock stars that are lined up to start helping in that effort have me very positive about our ability to achieve our goals.
Being part of the Product Strategy division, we are working in an environment that not only allows for innovation but makes innovation one of your core responsibilities. Now, you might ask whether this type of “enforced innovation” produces any valuable outcome. The answer is “Yes!”. For one because it is not enforced but truly appreciated by the team and supported with the right resources. Second, the most important factor however is the team – whenever I walk into a room with engineers there is this vibrant feeling that something great is about to happen (and it does, I have seen it!).
My team and I carry our share in that effort as well and I am looking forward to soon blog more about some interesting stuff we are working on for 12.4.

Quality
Due to my background in Software Quality, I was asked to help build Quality as an integral part of the engineering organisation. I truly appreciate the focus on quality that everybody has. The biggest challenge will be to keep a very agile environment going while applying QA principles to drive higher quality. As this requires the dedicated focus of a person, I am more than happy to announce that Martin Mrazik will bring in his extensive Software Quality experience and help us to push our standards even further. Martin and I have been working together in the past and I am looking forward to working together again.

Canonical/Mark Shuttleworth
Prior to my start I have been asked frequently if I really want to join the dark side (Canonical) or if I really want to work for MS. Being ignorant as I sometimes am I, did not have a particular opinion on either question (plus Novell was and is also considered the “dark side” due to the M$ deal). A lot of the questions were targeting Canonical’s perceived practices in the open source world.

My personal opinion on where Linux is headed in the next couple of years is even more commercialization. More and more companies will be able to directly or indirectly leverage Linux to generate revenue and jobs. However, this cannot be based on the pristine idea of Free & Open Source Software. Commercialization brings along competition and in my books competing product features cannot be developed in public from day 1. Now, you might argue that one of the major benefits of FOSS (finding the best technical solution) will be lost and we end up with multiple competing, but half-baked technologies.

I am convinced that an evolutionary process will keep us all honest in how we do software and how we chose to work together. As a company in a commercial context you have 2 choices if you don’t develop in public from day 1:
1) you either make sure your new technology is rock solid and will be accepted by the community by the time you open your development (with the risk of having to go through a lengthy adoption phase and the possible outcome of still being rejected)
2) you acknowledge that you won’t be able to leverage the FOSS synergy of sharing code with the community and thus will have to carry a notable financial burden yourself (you can only do this so many times)

Investments typically mean risk (think of the Moblin, MeeGo, Tizen disaster) and opportunities (think of what Ubuntu has done to Linux adoption with the average user), some investment will utterly fail, some will just thrive and take of. Progress in my opinion is based on the willingness of taking a risk and corporate FOSS contributions will help accelerate this process. I have a hard time seeing this kind of risk or investment being translated into something the community as such can provide or is willing to shoulder.
This is where in my opinion companies like Canonical, Google, RedHat & SUSE come in. A lot of good things were done by the unnamed contributor but I believe that in order to take Linux further (200 million users…) investments are needed to drive and support innovation at that large scale. Investments typically want to be protected to provide some competitive advantage and I am not blaming anybody for trying to get a return of his investment. Even though this might contradict with the pristine idea of FOSS, I believe that we can all benefit if said investments end up back in the pool of open source software.

The opinion in this section represents mine and not the one of my employer – not speaking in particular about my boss (MS) or my employer (Canonical) is intentional 😉

Hiring
Canonical is hiring! If you want to be part of this exciting journey then check here for openings in the Product Strategy division. Feel free to drop me a mail (olli – at – olli-ries dot net) if you think you can make a dent into Linux’ personal computing experience.

Automation @ work

Last week I got confirmation that the automation rock star will join QA, I am excited. This is the long due step towards improved automated testing.

Automation unfortunately is the one area that is being cut first when schedules or resources run tight. Paradox enough, automation however is the only investment that will help you when resources are running tight – nothing beats the wage of an idling server that just waits to run your repetitive tasks for cheap.

With the yet to be announced automation rock star, I am going to introduce a couple of changes:
a) a dedicated manager/architect/project manager to drive the effort
a) a dedicated automation test team
b) a joint approach to improve our existing automation infrastrucure

This will be my focus in remaining FY09 and first half of FY10. I hope to not overwhelm the rock star – I started to collect a couple of features on my flight back to the US and got 70 high level features in about 90 minutes… my head is still full of ideas. The coolest highlight is a public API that I plan to provide to our partners & our community! Stay tuned…