Geek stuff from a french geek and photographer
Recent Tweets @lhirlimann
Posts tagged "quality"

We just released the second beta of Thunderbird 31. Please help us improve Thunderbird quality by uncovering bugs now in Thunderbird 31 beta so that developers have time to fix them.

There are two ways you can help

- Use Thunderbird 31 beta in your daily activities. For problems that you find, file a bug report that blocks our tracking bug 1008543.

- Use Thunderbird 31 beta to do formal testing.  Use the moztrap testing system to tests : choose run test - find the Thunderbird product and choose 31 test run.

Visit for additional information, and to post your testing questions and results.

Thanks for contributing and helping!

Ludo for the QA team

Updated links

We just released the first beta of Thunderbird 30. There will be two betas for 30 and probably 2 or more for 31. We need to start uncovering bugs nows so that developers have time to fix things.

Now is the time to get the betas and use them as you do with the current release  and file bugs. Makes these bugs block our tracking bug : 1008543.

For the next beta we will need more people to do formal testing - we will use moztrap and eventbrite to track this. The more participants to this (and other during the 31 beta period), the higher the quality. Follow this blog or subscribe to the Thunderbird-tester mailing list if you wish to make 31 a great release.

Ludo for the QA team

Thunderbird 24 beta is out. This is the first beta of the next Major release of Thunderbird - with many, many core improvements and many small bug fixes (we are missing locales on windows). Good news is Lightning was build at the same time , so calendar users we’ll be able to download their favorite extension.

Starting now we are running a Test week , in order to catch any major regression that didn’t get caught. There’s tow ways to achieve this  and both are difficult.

1) We need a few more tens of thousands of users (so Thunderbird get’s tested with AVs, Anti-spam solutions etc …) so raise our low rate of 0.43% of beta users compared to mainstream users. People tend to be afraid of running betas for email while not for things like a web browser. I had the same thinking before testing Thunderbird became my main Activity. Since February 2009 - I’ve been using various betas/alphas and nightlights of Thunderbird, and didn’t loose a single email on either imap nor pop. Your mileage may vary but It’s been pretty stable for me.

2) We need people with more desktops setting to helps us during our first test week - which is starting today. That’s easier than 1 as you can just use a test account with no “important” emails in it. Then you’ll need to run the beta on that account and spend sometime testing ( that means reading a test - trying to do it in Thunderbird and report if it worked or not). This is less stressful about the mails you care. All you need is some time, the capacity to read English , and try things in Thunderbird. The tests are at , and the quality team is available on on #tb-qa  (just click that link to talk to us). Bugs found need to be reported in bugzilla and marked blocking of (if that’s chinese to you just add :usul on the cc list).

3) if you don’t have time this coming week - consider joigning our other efforts outlined at

4) the beta is at

Ludovic for the quality team.

Thunderbird - the email client from Mozilla is about to have version 24 reach the beta channel. Thunderbird has been running version 17.0.x since November 2012. It’s been a year of core development that we are about to catch up , this means changes in the networking layer, the javascript engine, gecko. It also means that all the changes (over 600) we made in Thunderbird code need to be tested. Depending on who you are it’s going to matter differently.

If you are a user, you’ll want to enjoy the same feature set, the same extensions and the same comfort for reading email, newsgroups and RSS feeds. To make sure we didn’t break important features that matter to you, switch today to the beta[1] and report issues to us. Also we’ll have a testday/testweek[2] come and participate and help make Thunderbird just better.

You are an extension developer - your extension works great in Thunderbird 17. It’s now time to try the beta series[1] and see if your extension still works in version 24, if not time to fix it and submit it to AMO. We’ll probably be able to help you in #maildev on

You are an IT administrator and are using Thunderbird in your corporate network - make sure major things don’t break before 24 comes out. For Those wondering Thunderbird ESR[3] is currently exactly what Thunderbird 17.x is except for the update mechanism, we don’t differ by a patch on both branches. So testing 24 is the right way to go.

You are an integrator and use Thunderbird as your mail client in your solution (say OpenXchange or Blue-mind) - well now is the time to figure out if the next major version of Thunderbird is still working the way you are used to.

I know that currently the beta version available for download is version 23 - but that will change to 24 soon - and in most cases running the version for more than a few hours is way more useful, so running 23 nos is the way to go.

For those wondering why I’m asking for more beta usage I’m going to give you some “stats”, they aren’t precise or anything but it’s numbers and they still mean things we have roughly 8 Million daily users, and around 35 Thousand beta users, that’s 0.43%. It’s too low and has bitten us in the past when we discover that for example a major AV vendor doesn’t work with latest release and that it affects 10% of our users, just because none of our beta testers were using it. 

Ludovic - for the Thunderbird Quality team.

[1] available at .

[2] To be announce a  bit later , probably here or on the various Thunderbird mailing lists .

[3] see .

We’ve just released Thunderbird 22b and lightning a corresponding version of lightning. Our next major release is due in a few weeks with version 24. In order to catch anything that might have gone wrong since the release of version 17.This week we need our users to spend a few hours testing the
applications, so that anything that might have gone wrong in improvements and fixes since the last big release of version 17 can be caught. There are two ways to test, a formal way and a less formal way.

Let me start with the formal way - as it’s the best way for us to know
the level of quality we’ve achieved with the products.

The informal way of testing is just to switch your daily usage of Thunderbird to the beta version above and make sure that : you send crash reports and you file bugs in bugzilla as described above.

During the testing week we will be available for chat on #tb-qa (on to help you the best we can - please be patient as we might
be there but not answer so quickly.

Ludovic for the quality team.

To try it out with

  • your favorite anti-virus
  • your favorites extensions
  • your email provider

Let us know if you have issues with it and how serious the issues are.

Thunderbird 17 beta can be found at

Some popular extensions can be found on AMO and the very popular Lightning and enigmail , here and here (choose the beta branch) respectively.

Right now I’m pretty sure a Teacher in a Tech school is teaching it’s student something on filling bugs and on using bugzilla, mozilla’s bug reporting system. At some point in the lesson they are handed a paper with a workshop instructions.

How do I know that ? Well I follow a number of things in bugzilla and we get :

  • 10 bugs created in a spawn of ten minutes
  • these 10 bugs are almost identical
  • once the bug is created a field is changed (always the same)
  • the email addresses form those bugs are very similar.

Now that I’ve explained what’s going on (and this goes one like maybe once a day on the course of the last few weeks).

I love that people are teaching how to use bugzilla and giving them real hand try and experience on the product but it’s bad.

Why is that bad ?

On a system like that is almost at 800000 bugs, you have a lot of users, a lot of things are happening. When you create a useless bug (e.g. an school exercise) you are wasting a lot of time of many people. When a bug is created emails are sent to a huge group of people and a good portion of them will spend a few minutes reading it - and will loose all this time instead of working on issues that affect the mozilla codebase.

When the same thing happens on landfill, it won’t affect as many people. So consider using it instead.

There’s a bugzilla test installation, it’s called landfill, and it’s here:

As some of you may know I’m the QA lead for mozilla’s email client Thunderbird. I’m responsible for the quality of the product that is used by millions of users. There is no way for me during a release cycle to test all the servers that do POP3, IMAP, SMTP, LDAP, NNTP, that are running on the Internet. A good portion of our quality process relies on crowd testing.

This means we rely on having enough users using betas and alphas and filing bug so that we can fix them before we release to a broader public. I’ve been running various alphas and betas and nighlies for the last 3 years on a daily basis and never experienced any data loss (either on pop or imap (and yes it’s easier to save data using imap)). On beta with have 0.28% of our user. This is low and we really need to get more users using beta so we can find issue before we release instead of after. Running  beta is easy just download it and run it. And when you find something that doesn’t look right, file a bug.

A very nice battle report video channel for the Warhammer Fantasy battle game. Video quality isn’t very high, but the story lines are, and they are improving all the time. If you like the game, you’ll love this youtube channel.