Due to a critical issue that some Windows XP users are experiencing, we have decided to release Lightning 1.1.1, an intermediate release compatible with Thunderbird 9/Seamonkey 2.6. We recommend all users to upgrade, especially those who cannot see their calendar data and are getting an error console message "Failed to load native module at path ...\calbasecomps.dll".
You can get the builds on addons.mozilla.org, as always.
Now that Lightning 1.0 has been released, we can move forward and adapt to the rapid release schedule that Thunderbird and Firefox are using. In terms of development we will do this gradually but for localization we will do so right away. From now on, if you are already used to the Firefox/Thunderbird localizations you can think the same for Calendar.
Here is an overview of where to get the sources and see the translation progress
| en-US source | locale source | dashboard | current version until merge |
|---|---|---|---|
| comm-beta | l10n/mozilla-beta | cal_beta | Lighting 1.1 / Thunderbird 9 |
| comm-aurora | l10n/mozilla-aurora | cal_aurora | Lighting 1.2 / Thunderbird 10 |
| comm-central | l10n-central | cal_central | Lighting 1.3 / Thunderbird 11 |
Specifically for Lightning 1.1, we have managed to open a milestone for Lightning 1.1. I'd like to encourage you to sign off on this milestone until December 5th so we can be ready in time for the release on December 20th.
As a Lightning user I often wish that all events would arrive at my Inbox as invitations in the iCalendar format. This way I could just click the accept button and it would be added to my calendar. But they don't. Often people just write you a text-based email when they want to invite you somewhere. So you right-click the email, choose convert to event and a new event dialog opens with start date set to next full hour. Not quite correct unless you're really lucky.
This is why I'm working on an extension to Lightning which would fill in dates and times automatically from email content. The basic idea is that localizers can define most common ways dates and times are referred to in their language similarly to how they localize Lightning itself. Using these patterns event dialog can be much smarter about setting start and end times and in most cases you wouldn't need to correct them at all.
Since Lightning is localized into 38 languages and language variants I need your help. If you fluently speak a language other than English and would be willing to spend some time to localize the patterns and test this prototype on your emails then please leave a comment below. Be sure to include your email and language of choice so I can send you a customized pattern file and prototype. Do so even if you speak only English as you can still help with testing. Feedback on when it guesses correctly and when not is crucial to improving the extension. What works great in English may not work well in your language and of course everyone's email is different too. With your feedback I can improve my extension so that this functionality can be included into Lightning later on (the curious can follow bug 403222 where more technical information will be posted).
Merike Sell, Mozilla Calendar Team
It's been over a year since we last posted about the current state of Lightning usage. Since that day Thunderbird 5, 6, 7 and 8 have been released with their Lightning counterparts 1.0 beta4, 1.0 beta 5, 1.0 beta7 and finally Lightning 1.0.
Now it's time to look at things again and I'm happy to report, that Lightning is more popular as ever before with nearly 30% more daily users than one year before. Right now we average nearly 1,400,000 users each day during the workweek and about 775,000 users on the weekends, which brings to more than 1,200,000 users on average over the course of a week.
Most of our users (64.2%) are now on Thunderbird 8. Most of the rest still uses Thunderbird 3.1 (15.0%) and Thunderbird 7 (13.4%). The remainder (7.4%) is mostly distributed to Thunderbird 2 (still at 2.3%), Thunderbird 6 (2.6%), Thunderbird 5 (1.0%) and other installations (other Thunderbird releases, SeaMonkey and Firefox) at 1.5%.
So, this is it! My early birthday present, Lightning 1.0, has finally been released. For quite some time now we have had so-called "beta" releases, that were in no way different to our other releases. With Lightning 1.0 we can finally say good bye to the betas and hello to exciting new releases!
When planning for this release, we mainly focused on rounding off any sharp edges you may find in Lightning. Just as an example, you can now press Escape to cancel creating an event via dragging. Sounds like a simple, intuitive feature that should have been there all along? You bet! There is one exception though, Lightning now supports full offline editing using the cache feature you can find in the calendar properties. This great big feature has been in the works during the past Google Summer of Code and has now finally made it into the release.
As always, you can get the latest release from addons.mozilla.org. If you are using the Provider for Google Calendar, make sure to upgrade too. Please note that Lightning 1.0 is compatible to Thunderbird 8, which is also being released today.
So, whats happening next? I will be gathering some ideas during the upcoming EU MozCamp 2011 and will present you my plans shortly after. If you have any suggestions, please do use the newsgroups to get in touch.
I'd like to present you with Lightning 1.0rc2. You can find it at its new location on the addons.mozilla.org beta channel (If you still get 1.0rc1, then please try again in a few hours when the page cache has been updated).
There is a big new feature in this release which needs your testing. Thanks to the hard work of Mohit, who has joined us through the recent Google Summer of Code, we have extended the cache with full offline support, which means you can write your events while in offline mode. If you are using a CalDAV calendar, it will even retry events that failed.
If you encounter any issues, please file a bug at bugzilla.mozilla.org.
Note, if you are using the cache with the Provider for Google Calendar, please stay tuned. I will soon post a development version of the provider containing a patch that corrects offline support.
Its about time to give you an update on what is planned for the next release. As you may have noticed, we have had a lot of trouble calling our releases "beta". For one thing, some users and especially corporate users are cautious when they read beta and don't see that it has the same quality as all our other releases. Aside from that we also had trouble with uploading releases to addons.mozilla.org (they are automatically marked beta and require admin intervention to make them public for everyone) and we are constantly in need of new sub-version numbers. For example, if we want to release a matching build for Thunderbird 7.0b2, what do we call it? Lightning 1.0 beta 7 beta 1? Thats too much, do you agree?
To relieve that issue we're going to bite the bullet and call either our next release or the one after 1.0. Afterwards we can newly decide on version numbers and I'll tell you one thing: I'm not going to go into the version number trap again. If at some point the next available version number is 2.0, then that's the way it is.
We have taken great care to avoid patches with string changes for the next release. Therefore as it is right now, if you have all strings translated for Lightning 1.0b5, then you are good to go for the next release.
According to the l10n dashboard, there are 13 locales with missing strings. This information is quite old so a lot may have changed in between, but it would be splendid if all of the remaining locales could have their strings translated. I will be contacting the localizers personally to make sure things progress
Update: Likely the l10n builds are not working yet. I'll also post here and in mozilla.dev.l10n when that is fixed!
For you? Yes, for you! We need help testing Lightning to make sure there are no unpleasant surprises in the next two weeks. Now you'll surely want to know where to get a build compatible with Thunderbird 7.0 Betas. Unfortunately I must admit, this is not quite ready yet. Our goal is to automate the release process a bit more, otherwise at least a day of my time is taken just for producing a full set of release (candidate) builds. It may easily happen that a detail is forgotten, which will only make the process more painful.
So what to do in the meantime? Well you have two options. First of all you could test Thunderbird 8 Prerelease with Lightning 1.0b8pre. This may be a different Thunderbird version, but the Lightning code is the same between those versions.
Then we have the first set of test builds for Thunderbird 7 beta and Lightning 1.0b7. These builds might have their quirks since not all parts of release automation went well. Please stay tuned, I'll be posting an update when its advisable to move from testing with Thunderbird 8 to Thunderbird 7.
We need you now more than ever to make sure Lightning 1.0 is a success. This version will likely get more PR than the previous versions so it would be a pity if it weren't in good shape or needs to be skipped for an urgent fix. I'm counting on you!
Unfortunately, some major errors slipped through in 1.0b4. You might have noticed the alarm window disappearing and a lot of misplaced beeps from the alarm system. You might have also noticed, that some of your calendars are no longer working (specifically, webcal:// calendars).
To fix these issues we are releasing 1.0b5 in the next 1-2 days, which contains 10 bugfixes.
As another highlight, 1.0b5 will now be compatible with both Thunderbird 5 and Thunderbird 6 to ensure you can continue using Lightning without extra hassle. If you are using Thunderbird 6.0b1 it would be great if you could give these test builds a spin.
Update: Another issue slipped in, I've modified the packages to fix it. If you have already downloaded, please try the following links!
You can get the builds here. If you get a 404, its likely they have not been propagated to all mirrors yet, check back in a few hours.
If you are using the Provider for Google Calendar, I have marked the current version compatible with Thunderbird 6. You can get it as usual from addons.mozilla.org
Thank you for your support!
I would like to announce the release of Lightning 1.0b4, which is now available via addons.mozilla.org.
Since our last release, we have fixed 140 bugs. Notable improvements for this release are:
Lightning 1.0 beta 4 is available for Windows, Mac OS X and Linux in 37 different languages including English. Please read the release notes for Lightning 1.0 beta 4 before downloading.
As the release has been out since Tuesday, we have noticed that there is an issue with alarms and are working hard to fix this behavior. A new release will be ready in the next few weeks.
Thank you again to all our developers, contributors, localizers, testers, and supporters. You are an essential part of our success, so keep up the good work!
The Thunderbird 5 release is just around the corner so its about time to provide a release candidate for the Lightning extension for calendaring. The upcoming version of Lightning will be called 1.0b4 and will be compatible with Thunderbird 5.
Unfortunately, this round of testing will be very very very short. Thunderbird 5 is slated for this Tuesday, the 28th and its critical that we have our release ready for this date. To compensate for the short period, we will be ready to spin another release if something major is broken, so if you are reading this after the release, please do file bugs as quick as possible if you find some.
Candidate builds for Lightning 1.0 beta4 in 36 languages are available as of now for:
A corresponding build of the Provider for Google Calendar is also available at those locations.
Note you will need Thunderbird 5 (or a beta thereof) to try it out. You are encouraged to back up your profile before upgrading. To install these builds in Thunderbird 5, please follow these steps:
Remember, testing is very important. The earlier errors are found, the faster we can react and possibly post another release on short notice.
Please tell us what you think of these candidate builds and file bugs in Bugzilla as you go. Thank you very much for your support, we couldn't do it without you!
As you may have noticed if you are running a Thunderbird 3.3 alpha or nightly, Lightning has changed its version number from 1.1a1pre to 1.0b4pre. I'd like to take a moment and explain why we changed this and what effects it may have.
As mentioned on the Calendar Versions page, we need to follow the release cycle of Thunderbird. Users will (and have, in the past) massively complain if a new version of Thunderbird is released without a corresponding Lightning version. Although there is no final plan yet when Thunderbird 3.3 will be released, the ballpark figure is some time in April. This would mean we have something between 3 and 7 weeks to complete our next release.
If we continue our previous plan and release Lightning 1.0b3, which would work on Thunderbird 3.1 and 3.2, we would have just enough time to complete that release and would need a near to equal amount of time to prepare for a Thunderbird 3.3 compatible release. As you can imagine, this is not feasible.
Therefore we have decided to leave out 1.0b3 and concentrate on releasing 1.0b4, which will be compatible with Thunderbird 3.3.
For localizers, this means they will have to make sure l10n-central is correctly translated. Resulting l10n builds can be found here. Pike is working hard to make sure the l10n dashboard can keep up with our branch switching. As soon as he is done, we can officially announce string freeze and accept sign-offs. This of course doesn't mean you need to wait. We have checked in all bugs that require string changes, so go ahead and translate!
For Seamonkey, this means that Lightning 1.0b4 will support Seamonkey 2.1, which is currently in its beta phase.
In turn, we no longer have a need for 1.0b3pre nightly builds. This might be unfortunate for Thunderbird 3.1 users that want to stay up to date with the latest Lightning features. On the other hand, the list of bugs to be fixed for the next beta doesn't contain any more new features. We will concentrate on fixing bugs and regressions to make sure Lightning 1.0b4 doesn't cause any surprises for your daily work.
For those of you using a Thunderbird 3.3 Alpha version or later, we have great news! I have finally managed to fix bug 591744, which was the last of the many bugs that had to be fixed to recover from the major changes to the mozilla-central platform.
If you would like to give it a try, you will need either a nightly or alpha version of Thunderbird, together with the nightly version of Lightning. You can get these using the following links:
Update: As noted in the comments below, the trunk versions of Lightning also work with the following beta versions of SeaMonkey:
After a fair amount of duplicates (19 as of now!), we finally have the hardware set up so we can provide Linux 64 bit nightlys on Trunk. On a related note, we now provide mixed x86/mac64 nightlys for the Mac OSX Platform, in turn we have dropped support for PowerPC.
After Thunderbird started providing nightly 64 bit builds it was time to follow. Unfortunately, our Mac builder was a G5 with a PowerPC processor, which doesn't support OSX 10.6. Thanks to Mozilla Messaging, we have now received a new mac and linux builder with 64 bit support.
Also I'd like to thank gozer from Mozilla Messaging for setting things up for us. Thanks to him, we now have the builders running comm-central (Thunderbird 3.3 Alpha compatible) builds.
You can get the new builds here, they are compatible with the current Thunderbird nightlys from comm-central:
If you are looking for Thunderbird 3.1.x compatible nightly builds, you can get them as usual from here.
About one month ago we shared our download numbers and we announced the majority of the lightning users are on Thunderbird 3.x. The numbers showed that roughly 75% of our users now used a Thunderbird 3.x. During the last month some things have changed: Even more users have downloaded lightning for Thunderbird and the number of users who use Thunderbird 3.1 has grown.
In this chart you can see that most users are still using Thunderbird 3.0, but the number of users who use Thunderbird 3.1 is steadily growing and it will lead very soon. The number of Lightning users on Thunderbird 3.1 has been growing very fast in the last two months, which shows that Lightning is a great add-on and people just love it as much as we do. While the number of Thunderbird 3.1 users is growing, the number of Thunderbird 3.0 users has literally collapsed, although the number of Thunderbird 2.0 users is going down very slowly.
In this chart you can see the active daily users of Lightning on Thunderbird. You can see, that the number is still growing, but in the last few weeks the number has been falling a little bit.
That's all about Lightning - let us move on to Sunbird!
In this chart you can see the active daily Sunbird users. The average is still moving up, but earlier last month the number was moving down for some weeks. This chart explains itself pretty well, so let us move on again...
Sometimes people are still using an outdated version of Sunbird. Although the development of Sunbird has stopped, you are still able to update Sunbird. Now, you may ask how much the daily growth of updates is. In this chart you can clearly see that the numbers are really random...
So, let us finally compare the download growth of Lightning and Sunbird.
You can clearly see that we got more Lightning users on Thunderbird than Sunbird users. While the number of Lightning users is still growing, Sunbird is almost dying. In the 35-day moving average you can see that the download number was rising for a short while, but suddently just dropped again.
Now, what do these numbers mean for us, the calendar software and its future? To find this out, you need to have some different angles of view on the download numbers: The future of Lightning seems really good so far, while the future of Sunbird looks very dark. But what have we learned from this? Well, we learned lots of things, for example that some users don't want to use an application for mail and for calendar - they don't want to use more than one program, they want to use an add-on that brings the calendar-function to Thunderbird, an awesome mail client, while some users prefer using multiple applications. We also learned that people really like to have the latest version of software installed, although some people just want to stay on the old versions.
The future of our calendar software seems obvious, but even if you take a closer look at it you can't really predict what the numbers will look like...
As you may have noticed, that blog posts are not as common as they were before. This doesn't mean that development has slowed down, but as you may remember, Simon has reduced his work on the project to an absolute minimum, due to job and real life restrictions.
Nevertheless, we are working as hard as we can to get the next release out of the door. If you are currently testing a 1.9.2 branch nightly (labeled as 1.0b3pre), you will not have noticed any improvements over 1.0b2 yet. Our current work has happened on comm-central (labeled 1.1a1pre, this is not the final version number though), which uses the latest version of the Mozilla Platform. Mozilla has done loads of changes to the platform with far-reaching consequences for the Calendar Project. We've decided to follow these issues to reduce the work needed later on.
We've also been working on lots of new features and bugfixes though and will be porting our fixes to the 1.9.2 branch very soon. If you'd like to see the list of fixes beforehand, check out this link.
We haven't decided on a firm release date yet, but our goal is to release 1.0b3 towards the end of the year. Currently, there are 22 bugs left to fix for the release. 6 of them are in review and I'm confident that we'll be able to fix the others soon.
But what can you do to help us towards the release? Yes, exactly! Testing! There is no way we can make a stable release without the help of our testing community. If you'd like to help out, you have two options:
In any case, be sure to check out the Lightning Nightly Updater Extension to make sure you're getting the newest testing version(s) on a daily basis
.Thank you for your help and remember: We can't do it without you!
After reading about Thunderbird getting a new Try Server, I thought it would be nice if the same would be possible for Lightning. After just a few failed builds, I successfully built Lightning using the Try Server and had a lightning.xpi uploaded to the right target directory.
If you want to try this on your own, you need a hg account and this patch file. Note also that you should be using mercurial queues. This not only makes managing multiple patches easier, it also makes using the Try Server a piece of cake.
[alias] push-to-try = push -f ssh://hg.mozilla.org/try-comm-centralThis will allow you to simply start a try build by calling
hg push-to-tryhg qimport -n tryserver-calendar.diff https://bugzilla.mozilla.org/attachment.cgi?id=471566
hg qpush) any amount of patches that you'd like to test.hg push-to-try (yes, with all those patches applied to your tree!)Thats all! If you want to watch your progress, see the Thunderbird Try Tinderbox. You will get email when your builds have completed, together with a link to the ready built lightning.xpi. Thunderbird files will not be uploaded.
For more information on the Tryserver, especially on how to exclude certain platforms or provide extra mozconfig options, please read the Thunderbird TryServer Guide or the Firefox TryServer Guide.
Six month ago, I shared our usage statistics for Lightning with you, which showed that Lightning users on Thunderbird 3.x had surpassed Lightning users on Thunderbird 2.x for the first time.
Now it's time to look at things again and I'm happy to report, that as of yesterday roughly 75% of our users now use a Thunderbird 3.x build as can be seen in the chart below.
It's also important to note that nearly 60% are already using the latest Thunderbird 3.1.x builds, while users of the old Thunderbird 3.0.x series are decreasing fast. This is mostly due to the major update offers of Mozilla Messaging for users of Thunderbird 3.0.x.
Major update offers have recently (last week) also started for users of Thunderbird 2.x, but haven't really been unthrottled yet (throttling means, that currently only 1 in 5 users of TB2 gets a major update offer to Thunderbird 3.1.x). Once major update offers go out to every TB2 user, we expect their numbers to decrease much more heavily. The current state of things (showing the last six months) can be seen in the chart below:
As frequently noted in earlier articles, the ratio of Thunderbird 3.0.x and TB 3.1.x users is much higher on the weekends compared to the Thunderbird 2 users. The graphs show that very well.
For those interested: The "other" number contains users on older Thunderbird 1.5.x builds, SeaMonkey users and users, who have mistakenly tried to install Lightning into Firefox. Generally that number always fluctuates between 0.35% and 0.5% of our total active users.
I hope that you will find this as interesting as I do.
The Calendar Project is proud to report, that (finally) the 1.0 beta2 release of Lightning has been completed and is now available via addons.mozilla.org.
About 6 months after 1.0b1, we have managed to complete 86 further bugfixes and improvements for the benefit of our users. Notable improvements for this release are:
Lightning 1.0 beta2 is available for Windows, Mac OS X (universal builds) and Linux in 35 different languages including English. Please read the release notes for Lightning 1.0 beta2 before downloading.
Thank you again to all our developers, contributors, localizers, testers, and supporters. We would not be able to do this without your assistance!
This new release candidate contains fixes for a few locales, no further code changes. Since the Thunderbird 3.1 release is scheduled for this week, I assume this will be the final release candidate.
Candidate builds for Lightning 1.0 beta2 in 33 languages are available as of now for:
A corresponding build of the Provider for Google Calendar is also available at those locations and will be uploaded to addons.mozilla.org once the release cycle completes.
For install instructions, please see one of the earlier blog posts.
Please tell us what you think of these candidate builds and file bugs in Bugzilla as you go. The release is near, and we are glad we can finally concentrate on the next one!
A few people have asked me different compatibility questions around Lightning 1.0b2. I'd like to take a moment and tell you about what applications Lightning 1.0b2 will support.
What version(s) of Thunderbird will Lightning 1.0b2 support?
This version of Lightning will only support Thunderbird 3.1. To go a bit more into the technical details, Lightning 1.0b2 and Thunderbird 3.1 use the Mozilla Platform version 1.9.2. The previous versions (Lightning 1.0b1 / Thunderbird 3.0.x) use the Mozilla Platform version 1.9.1. Therefore we cannot provide a version of Lightning that is compatible with Thunderbird 3.0.x
Will Lightning 1.0b2 support Seamonkey?
As you can see in this blog post, Seamonkey does not support the Mozilla Platform version 1.9.2. Therefore, Lightning 1.0b2 cannot support Seamonkey 2.0 or 2.1. Note however, the next version of Lightning will most likely use the Mozilla 1.9.3 Platform, which means the next release (due in about 4 months) will be compatible with Seamonkey 2.1. Until then you can still use Lightning 1.0b1 together with Seamonkey 2.0.x.
Which version of the Provider for Google Calendar should I use?
Together with Lightning 1.0b2, I will release a new version of the Provider for Google Calendar. This version will be called 0.7 and will be available on addons.mozilla.org together with Lightning.
Why don't you use the latest Mozilla platform, verion 1.9.3?
The 1.9.3 platform contains the newest features and bugfixes to date, but this also means that some Mozilla components we use, as well as some UI features, have changed and require us to adapt our code. Changes in the platform can happen every day, which means it doesn't make sense to rely on it for a release. Just imagine something big changes between now and next week: this would postpone the release indefinitely, which is obviously bad. Another factor is that the newer platform has binary components that have a different interface. If we build Lightning with Mozilla 1.9.3, it won't be compatible with a Thunderbird on the Mozilla 1.9.2 platform. It makes sense to use the same Mozilla platform version that Thunderbird does, this causes the least amount of headaches.
Why are you releasing betas, and not 1.0 final?
There is a certain level of stability that people expect from such a major release such as 1.0. We have decided on a certain set of bugs that we think should be fixed before 1.0 final is released. Back when we released 1.0b1, we had a few possibilities on what we should name it. There was 0.10 (as in 0.8, 0.9, 0.10, ...), but people could think this is similarly mature as 0.1 was. There was 0.9.5, but we wanted to make clear that the Mozilla platform used has a different minor version (1.8.x vs 1.9.x). So we ended up with 1.0b1. Now we have no other choice than to continue with betas until we've fixed all the bugs we think make sense for 1.0 final.
I hope this gives a bit of insight, if you have any further questions, please comment!
With some delay, we are happy to announce Lightning 1.0b2 release candidate 2. With this release we've mainly fixed bug bug 570464, but also made sure we were using the right localization files.
Please re-test if Lightning works in your language! Back with rc1, I accidentally used the newest version of the localization files, but not the version that the localizers suggested. I did notice some further change to the localization files, so it would be great if you could double check.
Candidate builds for Lightning 1.0 beta2 in 33 languages are available as of now for:
A corresponding build of the Provider for Google Calendar is also available at those locations.
Note you will need Thunderbird 3.1 (or a release candidate thereof) to try it out. You are encouraged to back up your profile before upgrading. To install these builds in Thunderbird 3.1, please follow these steps:
We will be releasing the final version of Lightning together with Thunderbird 3.1. There is not much time left before this release, so I encourage you to do the testing rather sooner than later.
Please tell us what you think of these candidate builds and file bugs in Bugzilla as you go. We are really looking forward to this release, I hope you are too!
Due to (at least) bug 570394 and bug 570464, we need to respin for rc2. This will happen quite soon (in the next few days), since we need to keep up with the release dates for Thunderbird 3.1.
If you are a localizer and would like to do any further changes for 1.0b2, please file a bug and contact me as soon as possible.
This time around, we had a lot less trouble then last time when preparing 1.0b2 release candidate 1. The most trouble this time was again release automation. We managed to work around the issues by producing manual builds on the same machines as our usual releases are built on.
Candidate builds for Lightning 1.0 beta2 in 32 languages are available as of now for:
A corresponding build of the Provider for Google Calendar is also available at those locations.
Note you will need Thunderbird 3.1 (or a release candidate thereof) to try it out. You are encouraged to back up your profile before upgrading. To install these builds in Thunderbird 3.1, please follow these steps:
The candidate builds will be available for testing for up to 2 weeks. We want to release the final Lightning 1.0b2 together with Thunderbird 3.1, so we may choose to shorten this period by a few days.
Please tell us what you think of these candidate builds and file bugs in Bugzilla as you go. Thank you very much for your support, we couldn't do it without you!
Update: Windows link fixed, thanks for the note!
As announced nine days ago, the sign-off deadline for Lightning 1.0 beta2, the Lightning release that will accompany Thunderbird 3.1 is on May 24th at 23:59 pacific time.
Right now the picture looks like this:
I will contact every locale that has not yet signed off their locale individually as well, so that everyone is aware of the current status.
As announced by Philipp two days ago the strings for the next Lightning release (1.0 beta2) have been frozen and you can now start to sign-in for this release.
We will be accepting sign-in requests until May 24 @ 23:59 pacific day time, so you have a little less than two weeks time to get your locale in shape.
Lightning 1.0 beta2 localizations will be taken from the l10n-1.9.2 repositories, so please work from there.
If you have any questions, feel free to ask them in the mozilla l10n newsgroup or send me an e-mail.
I want to make those of you, who are actively translating Lightning aware, that we are now ready to freeze our strings for the approaching 1.0b2 release.
After the string freeze we are currently planning for two and a half additional weeks before we try to release our first release candidate on May 25th. We would really appreciate it if you could have your locales in shape for the first RC, as I do hope we will only need one release candidate. Please also note that we have to stay in sync with the Thunderbird schedule, so we might have to prepone our rc1 release to an earlier date.
That gives you 2-3 weeks to get your localizations in shape.
The 1.0b2 release will come from the comm-1.9.2 release branch, which means you should commit your work to the l10n-mozilla-1.9.2 repositories. If any of you will not be able to keep his or her locale updated for the 1.0b2 release, please let me know beforehand.
I am aware that we still don't have bug 346278 fixed, I am really sorry this is still the case. If rc1 is released and you notice any localization problems, please do let me know as quickly as possible so I can coordinate a second release candidate. I do hope to fix this bug in the next two weeks though!
Please let Simon Paquet or me know if any of you wants to join us for 1.0b2.
If you have any questions, please do not hesitate to ask them here
Thanks,
Philipp
Update: It seems the calendar10x dashboard is not correctly updated. Please use the latest version of the dashboard to find out how many strings are missing!
As we're moving closer towards our next release, Lightning 1.0 beta2, it's time to talk more about our recent accomplishments and bugfixes. We've been pretty active in the last six weeks and over 40 bugfixes have made it into our code repository.
Overall we have fixed an impressive 43 bugs since February 8. Here's the complete list of fixed bugs:
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
We (mostly our lead developer Philipp) have done some heavy lifting with our bug database in recent days. During the last month, we waded through hundreds of bug reports (some of those being several years old) to see whether those bugs had been fixed by all the work that we've put into Lightning 1.0 beta1.
The result of this exercise was a reduction of our total open bug count by nearly 18% or in absolute terms, we've reduced the number of open bugs by 276 from 1552 to 1276. This chart illustrates our total open bug count from the start of 2010 until mid-March:
It's important to note that not every bug in our database is a real program error. Our bug list also contains lot of enhancement request, project management issues and enhancement requests. Basically everything that comes up in our project that needs to be tracked somehow is entered in our bug database, so that it can be easily tracked.
Some of our users currently seem to experience problems with Lightning becoming incompatible when they update their Thunderbird installation to the 3.0.3 release.
The reason for this seems to lie in the fact that those users use a Lightning nightly build and update it on a daily basis with the Lightning Nightly Updater add-on or manually update to the latest available nightly build. But because we moved our Lightning nightly builds to only support the current Thunderbird 3.1 development builds, as reported earlier, this leads to the incompatibility.
The solution is simple though:
Just uninstall your current nightly build and install the Lightning 1.0 beta1 release. This release is identical to the latest available Lightning nightly builds for Thunderbird 3, but fully compatible with the Thunderbird 3 release. And to be on the safe side: Please backup your profile folder, before doing this!
Three weeks ago, I shared some usage statistics for Lightning with you.
One of the things that I looked at then was the adoption of Thunderbird among Lightning users. Back then Thunderbird 3 users were still lagging behind, but now I can report that on the last weekend, the ratio of Lightning users on Thunderbird 3 was higher than the ratio of Thunderbird 2 users for the first time. Here's the graph showing the steady increase of Thunderbird 3 usage since November 25th in 2009, when the first release candidate for Thunderbird 3 was released.
As already noted in the earlier article, the ratio of Thunderbird 3 users is much higher on the weekends compared to the Thunderbird 2 users. But even during the workweek we're catching up pretty fast, even faster than on the weekends.
Interested observers will notice that the graph in this post and the earlier graph differ slightly. The reason is that in the current chart I've also incorporated the numbers of pre-release versions, like 3.0 beta3 or 2.0.0.24pre.
And if someone is asking himself why I only list Thunderbird 3 and Thunderbird 2 ratios here, the answer is that all other versions, e.g. users of Thunderbird 3.1 nightly builds, Thunderbird 1.5 users, SeaMonkey users and people that installed Lightning into Firefox all taken together account for not more than 0.5% of our total active daily users.
I hope that you will find this as interesting as I do.
As you may have noticed, not much has happened since the release of Lightning 1.0 beta 1. One reason for this is that I've had 8 Exams in the last three weeks, all of which needed quite an amount of preparation. I haven't gotten grades for all of them yet, but those I received so far were pretty good.
Now that all of this is through, I have a lot more time for Calendar. I will be enjoying semester vacation until April, which means I can catch up on all the calendar work I have missed.
I'd like to give you a short overview of what my next steps are, so you know what to expect.
This should keep me busy for a while, I'll post again when I have new goals to share. Let me know what you think!
Philipp
It's now been thirty days since the release of Lightning 1.0 beta1 and so far it's been a blast.
I regularly take a look at our download numbers and wanted to share some recent developments with you. First of all, let's take a look at our daily download numbers since 2009-01-01:
What's easy to see here is the huge reception that Lightning has gotten from our users since the 1.0 beta1 release. You can see it with the small spike in mid-December 2009 (announcement of the 1.0 beta1 release candidate) and the large spike 30 days ago, that marks the final release of 1.0 beta1. The corresponding moving averages show even better how much our daily download numbers have increased. To give some hard numbers: We've gotten more than 310,000 unique downloads within the last 30 days. That's 50% more than we've gotten at any time since we measure our daily download numbers (which we do since September 2007).
I should add two things here:
Another good indicator are our number of active daily users, which we've been measuring since late May 2009 and which can be tracked on a daily basis at addon.mozilla.org.
Again, it's easy to see the steady upward trend that started around late August 2009. The downward spikes on every weekend are also easy to see, as are our low numbers around the year-end holidays, which seem to indicate, that Lightning is mostly used for business purposes and not so much for private use.
Lastly, let's take a look at what our users are actually using. How many people have already updated to Thunderbird 3?
The chart starts in late November with the first release candidate of Thunderbird 3 and it's easy to see the first large increase on the day of the Thunderbird 3.0 release (December 08 2009). The release date of Lightning 1.0 beta1 (January 12 2010) is harder to spot here, but the uptake is still noticeable. What is pretty clear is the steady increase of Thunderbird 3 users and the corresponding decline of users of Thunderbird 2. Interestingly, the ratio of Thunderbird 3 users is much higher on the weekends compared to the Thunderbird 2 users. One possible explanation would be, that normal users , which also use Lightning during the weekend, are faster with upgrading than business users, which only use Lightning during the work week.
One interesting additional fact, that cannot be gleamed form those charts is the fact that Lightning users seem to be much more eager to update to Thunderbird 3, than Thunderbird users which do not use Lightning.
The last available numbers speak of a total of 6 million active daily users of Thunderbird during the week of which 1.4 million are already using Thunderbird 3. That would put the ratio of Lightning users at 15.5% overall (calculated with a value of 930,000 active daily Lightning users during the work week). However if you take a closer look at the various versions of Thunderbird, you can see that the ratio of Lightning users among Thunderbird 3 users is much higher (22.8%), while it is much lower among users of earlier Thunderbird versions (11.4%).
I'd be interested in your thoughts on this.
Now that we're moving full-steam towards the Lightning 1.0 beta2 release, it's also time to make clean slate and show you in terms of fixed bugs, what has happened since the last status update back in late November 2009.
Overall we have fixed an impressive 55 bugs since then, keeping in mind the whole release process hassle, the holidays and Philipp's exam schedule. Here's the complete list of fixed bugs:
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
Some people have already noticed that we no longer release nightly builds that support Thunderbird 3.0 and SeaMonkey 2.0. As of January 31st, our nightly builds will no longer live in the latest-comm-1.9.1 folder, but in the latest-comm-1.9.2 directory.
The main difference between the 1.9.1 builds and the 1.9.2 builds is, that the latter will only support nightly builds of the Thunderbird 3.1 and the SeaMonkey 2.1 development branch. This also means that our next Lightning release (1.0 beta2) will not support Thunderbird 3.0.x and SeaMonkey 2.0.x and users of those releases will have to update to be able to get the latest Lightning release then. For now, users that want to test the latest Lightning fixes should either download Lanikai 3.1 alpha1 or a recent nightly build of Thunderbird (also called Shredder) or SeaMonkey.
The reason for this decision is (again) one of (developer) resources. Currently the following development areas/branches are available to us:
With our current resource situation we are not able to fully support all three areas/branches. We have therefore decided to only fully support the Gecko 1.9.2 branch and give a somewhat limited support to the trunk. Those of you who want to follow our work are encouraged to download the Lightning Nightly Updater add-on, which should make updating Lightning much easier for you.
Thanks for staying with us.
Yesterday, Mozilla Messaging released Thunderbird 3.0.1, its first security, stability and maintenance release after the initial Thunderbird 3 release.
While it's always a good idea to quickly update to such a maintenance release for security reasons, this release also contains one important fix for Lightning users, who like to use Lightning in combination with other popular Thunderbird extensions (Enigmail, Quick Folders, Buttons!, etc.). Without this fix, some of our users were affected by bug 398702, which caused the File, Edit and View menus to disappear.
This bug has now been fixed in Thunderbird 3.0.1, which is why we recommend this update to all our users of Lightning 1.0 beta1.
There seems to be some doubt in our community regarding the status of the recent Lightning 1.0 beta1 release in our release hierarchy.
So let me unequivocally state for the entire Calendar developers team:
The two most asked questions regarding the new Lightning release so far were:
I'm happy to report that both issues could be resolved successfully:
Our thanks go to the Calendar lead developer, Philipp Kewisch, who provided the x86-64 build and who maintains the GDATA add-on. Great work, Philipp!
The Calendar Project is proud to report, that (finally) the 1.0 beta1 release of Lightning has been completed and is now available via AMO.
Nearly 16 months after the 0.9, this release is more than overdue and we're more than happy to get the nearly 500 bugfixes and improvements into the hands of our users. Notable improvements of this release are:
Lightning 1.0 beta1 is available for Windows, Mac OS X (universal builds) and Linux in 37 different languages including English. Please read the release notes for Lightning 1.0 beta1 before downloading.
Thank you again to all our developers, contributors, localizers, testers, and supporters. We would not be able to do this without your assistance!
Since my announcement, that we've released a first release candidate of Lightning 1.0 beta1, there have been a few questions here in the comments, which I'd like to address:
I hope that answers most of your questions.
After all the trouble that we went through to prepare the Lightning and Sunbird builds for 1.0 beta1 (see Philipp's posts here and here, we have finally succeeded (at least on the Lightning front).
Candidate builds for Lightning 1.0 beta1 in 37 languages are available as of now for:
A corresponding build of the Google Calendar Provider is also available at those locations.
To install these builds in Thunderbird 3, please follow these steps:
What is missing right now are two things:
Please tell us what you think of these candidate builds and file bugs in Bugzilla as you go.
I'd like to give you guys another update of how our release process is going, especially since Thunderbird 3.0 is now released and we are not ready yet. To take the interesting details up front, I believe we will be able to release rc1 before Christmas. If no urgent issues show up, we will be able to release the 1.0b1 final for or shortly after Christmas. This release will be compatible with Thunderbird 3.0.
From the product point of view, we are ready. We have all the patches in that we think should belong in 1.0b1rc1. Therefore, the nightly builds should be of the same quality that the release candidate 1 will be.
If you cannot do without Thunderbird 3 and are brave enough to try pre-release grade builds, I'd suggest to take a look at our nightly builds. Be sure to do backups first though! Otherwise please bear with us until we have the beta released and stay with Thunderbird 2 in the meanwhile.
From the release engineering point of view, we are constantly running into Problems. I believe the biggest and most unsolvable problem is that I am in a different timezone than Mozilla Messaging's release engineering, which means there is only a small timeframe where we can communicate about build issues. All other problems are at least solvable, but need some time to identify and find a solution for. Each attempt for building the release candidate has a build number. Let me summarize what happened during our builds:
Another "problem" is that gozer needs to trigger and upload the Lightning builds manually, which also takes extra time once for each build. I do hope we can finish off the last problems soon. I'd like to again note that next time around this process will be much shorter. We don't need to fix the small build system issues, all we need to do is update the release config and trigger a build.
We are working hard on getting the first release candidate for Sunbird and Lightning 1.0 Beta 1 out and since our nightly testers have surely noticed that the latest nightlies have been named "1.0b1pre" I wanted to give a short update.
We have been working on releasing the beta since last Wednesday. To make releasing easier, we have decided to make use of the Thunderbird release automation scripts. This is also the reason the version number changed to 1.0b1pre. Since the directory structure of Calendar is a bit different to what Thunderbird and Firefox use, we had some complications that need to be mangled out first. Unfortunately, building is a lengthy process (it takes a while until we are able to find out what went wrong) and Mozilla Messaging's Release Engineering is in a different timezone (6 hours difference), things are taking a bit longer than I anticipated.
Nevertheless, I am confident that we will be able to publish our release candidate in the next week, and depending on if we find further blocking issues or not, release 1.0 Beta 1 in the next 2-3 weeks. Also, for the next release the process should be much quicker since we've now had the time to take care of the one-time issues (missing access, minor bugs in our build system, etc.)
Update: Linux and Windows builds of RC1 have succeeded! The fix for Mac is on its way, we now just need to take care of some branching issues and let the Mac box build.
Finally another status update. We're down to zero (0) blockers for our upcoming beta release. It's taken us some time to prepare all the necessary release bits for various reasons:
So please be patient and bear with us. The release is coming :-)
Since the last status update seven weeks ago, a number of bugs were fixed. Here's the list of the 35 bugs that were fixed since the last status update:
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
Finally! Since July, we have been working on bug 470430. After lots of testing, reviewing and frustration, we have finally checked in the bug's patch. This bug is of quite central importance since we want to make sure that calendar data is not broken on an upgrade between versions, which was sometimes the case.
I would appreciate if everyone could take a look at the current nightlys to see if anything fails. Especially if you have an old profile (i.e from Sunbird/Lightning 0.8 or earlier) around, it would be nice to find out if upgrading from that profile works flawlessly.
If you have ever used Sunbird, please also check if you have the following directory:
C:\Users\<your username>\AppData\Roaming\Mozilla\Extensions\{718e30fb-e89b-41dd-9da7-e25a45638b28}\calendar-timezones@mozilla.orgC:\Documents and Settings\<your username>\Application Data\Mozilla\Extensions\{718e30fb-e89b-41dd-9da7-e25a45638b28}\calendar-timezones@mozilla.org~/.mozilla/extensions/{718e30fb-e89b-41dd-9da7-e25a45638b28}/calendar-timezones@mozilla.org~/Library/Mozilla/Extensions/{718e30fb-e89b-41dd-9da7-e25a45638b28}/calendar-timezones@mozilla.orgIf you do have this directory and it is empty, please report on bug 470430. Having it causes the upgrade process to fail, since the timezones files can not be found. My hopes are that this directory was created in error due to something in the development build process.
Now all we have left for our beta release is bug 494140, which will hopefully make it into the tree a bit sooner. Also, some locales need a few more strings. When all that is taken care of, then we can finally bring out release candidates for this beta release!
I'm looking forward to hearing from you and am excited that the beta is now one giant step closer.
UPDATE: We had a small regression, bug 526264. Upgrading from an older profile should work again after tomorrow's nightlys.
Ok, I have seen all the comments declaring the project is dead. Let me just say that rumours about our death are slightly exaggerated. Over the last six weeks we have fixed 31 bugs, which are listed below.
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
Two weeks ago, I was interviewed by the Bavarian broadcasting station here in Germany for their weekly computer radio show. The interview went pretty well and we covered a lot of areas, including:
You can listen to the podcast of the show by downloading the mp3 file (the interview covers the last eight minutes). As this was a German radio station, the interview was conducted in German as well. Sorry to all our non-German-speaking folks about that.
Nevertheless, this gives us some much needed publicity, for which we are grateful. We are already discussing some followup actions with the radio folks. More about this, when things get more concrete.
More than one month has gone by without a status update. Obviously it's vacation time :-)) We're hard at work to finally get our next release out of the door for Lightning and Sunbird. More on that in a separate post. Over the last six weeks we have fixed 22 bugs, which are listed below.
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
In February 2008 I wasn't even a user of Lightning yet as it becomes obvious from looking at my home calendar. Four months later I was researching how to translate it so I could have Thunderbird and Lightning in the same language because an Estonian localization didn't exist. A year later being familiar with the localization side of Mozilla and using Lightning regularly the option to write tests for calendar looked really appealing when I noticed it in the wiki. In April I found myself participating in the testday for Mozmill 1.1. (There's another one coming where you could participate by the way!)
Now that I've written smoketests for bug #500469 I find Mozmill relatively similar to Selenium as I expected. Both of these tools test user interface functionality and therefore the basic three types of steps are common to both:
Element location in a Mozilla application can be done either by using Mozmill's inspector which gives you a ready to use locator or with the help of DOM inspector if you need to access an element that is further down in the hierarchy of UI elements than what Mozmill's inspector gives you. Locators in Lightning are quite long but mostly relatively easily accessible. There are exceptions to this though: bug #505336 being an example of an element that is fighting against automation.
As UI responsiveness is always slower than script speed pauses between test actions are essential to let UI load properly. Failing to do that can result in an otherwise fine-looking test not working and once in a while I'm still able to forget that despite of knowing it. There are also some inconsistencies between platforms support of Mozmill which made it impossible to stick to the Linux environment I mostly use but it also gives both the Mozmill and Calendar development version stress on both platforms which can only make these more bulletproof.
Result checking is the most important part of functional tests even if running a test without any asserts can reveal some of the bugs too. This is also the part of Mozmill code that recently reminded me of just how quickly Mozmill is developed. I was looking for a way to assert that a certain element is not hidden and found a assertPropertyNotExist function by scrolling through the available methods on Google Code. This seemed like a good way to make sure that there's no hidden attribute attached to an element. To my surprise the first call to this method caused a failure which stated that there's no such function. It had been added only a couple of hours earlier and the version installed didn't have it yet!
Active development of Mozmill and fantastic people on both the #qa and #calendar IRC channels make the difference between a person merely checking things out and a new participant in the project. That's what it takes to get a new person to participate.
It's been three months since I last posted a donations update. And it's more than time to update our community on the current state of things, because two exciting things have happened or are about to happen.
Back in April we had accumulated roughly 1150$ of donations and we were heavily contemplating what to do with it. At the same time, a long-term localization contributor, Merike Sell, offered to create an automated testing framework for the calendar applications as part of Google's Summer of Code. Unfortunately the project was not accepted (only a few of the dozens of worthy projects were accepted by Google), but the offer was too good to let it slip.
So we decided to use our money to fund Merike's project outside of Google's summer of code. Unfortunately our money was not enough to fund Merike for the three months period, but David Ascher (CEO of Mozilla Messaging) graciously agreed to cover the remaining 3350$.
Merike is already hard at work. You can watch her progress in Bug 500469. Her mentor, Clint Talbert, will post an introductory posting explaining the overall goal of her project. And hopefully Merike will post here as well with a short introductory post and a status update once in a while.
Let me close by aying "thank you" to everybody who has donated to the Calendar Project so far and to everyone who is planning on doing this in the future. As you can see, these donations can really do some good, so donate now!
Our friends over at Inverse are pleased to announce the immediate availability of Lightning "Inverse Edition" v0.9.5. This is a maintenance release of Mozilla Lightning based on our 0.9 release which focuses on stability and includes many bug fixes, several small enhancements and some new features of the upcoming Mozilla Lightning 1.0 extension.
The Inverse Edition of Mozilla Lightning is based on the last official release for Mozilla Thunderbird 2 from the Mozilla Calendar project - which is 0.9. Inverse's strong desire to continue supporting and improving Lightning for Thunderbird 2 consists on the following activities:
It is important to note that this is neither a fork nor an official release from the Mozilla Calendar Project. Also, keep in mind that the Inverse Edition of Lightning is not supported by the Calendar Project. Support is rather provided directly by Inverse.
The "Inverse Edition" of Mozilla Lightning can be downloaded from this page:
http://www.scalableogo.org/downloads/frontends.html
You can also download the sources by following the instructions on this page:
http://www.scalableogo.org/development/source_code.html
For any questions, please do contact Inverse directly by writing an email to: support@inverse.ca.
Continue reading "Announcing Lightning "Inverse Edition" 0.9.5" »
Three weeks after the last status update, it's time for another. We're hard at work to finally get our next release (1.0 beta) out of the door for Lightning and Sunbird. Over the last three weeks we have fixed 12 bugs, which are listed below.
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
It's time for another status update again. We're hard at work to finally get our next release (1.0 beta) out of the door for Lightning and Sunbird. Over the last three weeks we have fixed 31 bugs, which are listed below.
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
PS: We've read and noticed the repeated requests for a more high-level plan of our future goals (some call it a "roadmap"). Philipp is currently working on a blog post, that will hopefully satisfy those of you, who were kind enough to get in contact with us and ask us for such a high-level plan. Stay tuned!
It's been a long time, but it seems that we're finally getting near to a new release. Yay!
Philipp tells me, that we now have only five open bugs on the blocking list that are needed for the 1.0 beta release. All of those bugs have no localization impact, which means that we are ready for a string freeze.
We will therefore freeze all application strings by tomorrow, Wednesday, June 23 at 23:59 PST. That means that all strings will then be entirely frozen until we release Lightning and Sunbird 1.0beta. I will open up a separate opt-in thread for localizers in the mozilla.dev.l10n newsgroup soon.
One additional note to localizers:
Please take a close look at the status of your tinderbox (at http://tinderbox.mozilla.org/showbuilds.cgi?tree=Mozilla-l10n-locale where "locale" must be replaced with your locale code (e.g. es-ES or de). I've seen some locales having a red Windows tinderbox (and therefore broken Windows builds), because those locales didn't follow this change to our installer strings. Please check if this applies to you and make the necessary changes.
In an effort to better integrate into Thunderbird and also to clean up some very evil workarounds we have reworked the Menu Structure of Lightning. Work on this was completed yesterday in bug 456385.
Previously, we switched the contents of the whole menu bar when changing from mail mode to calendar or task mode. Aside from the fact that this forced us to use very heavy workarounds for certain platform-specific bugs, this only made sense for 0.9, since the modes were very dedicated. I believe this was also a nightmare with regard to accessibility.
To fit better with the new tabbing mechanism available in the Thunderbird 3 betas, we changed this mode switching logic to show each mode in a tab. It looked a bit awkward that the menus change, just because you switched to a different tab.
Therefore we aimed at an integrated solution that has a fixed menu for manipulating events and tasks, and sorted our other menu items into the existing menus. Aside from that, the idea is to provide buttons and other actionable controls right in context - the user shouldn't have to travel far to do what he wants to do.
Please take a look at our latest nightlies together with Thunderbird 3 beta 2 or later to get a feel for the new menus and to find out if there are any missing pieces.
Our regular nightly build testers may already have noticed. If you haven't here's a heads-up for you.
Bug 481685 was just fixed. The aim of that bug was to align the nightly build directory structure on the mozilla.org ftp server of Thunderbird, Sunbird and Lightning. Here's a n overview of the changes:
| Application (Gecko version) | Old Location | New Location |
|---|---|---|
| Sunbird 1.0pre (1.9.1) | latest-comm-central-sunbird | latest-comm-1.9.1 |
| Sunbird 1.0pre (1.9.1) localized | latest-comm-central-sunbird-l10n | latest-comm-1.9.1-l10n |
| Sunbird 0.6pre (1.9.0) localized | latest-trunk-l10n-cvs | Deleted |
| Directory | latest-trunk | Deleted |
| Application (Gecko version) | Old Location | New Location |
|---|---|---|
| Lightning 1.0pre (1.9.1) | latest-comm-central-calendar | latest-comm-1.9.1 |
| Directory | latest-comm-central | Deleted |
| Directory | latest-trunk | Deleted |
Sorry for the long hiatus here in the blog. We haven't forgotten you. Here are the 37 bugs that we fixed since the last status update in mid-April:
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
I currently have only received the numbers for the last two weeks of March (we started accepting donations on March 16 2009) and those numbers look quite promising. In the first five days after the initial announcement people donated $698.05 to Calendar using PayPal. For the remaining 10 days of March we took in another $437.05 bringing us to a total of $1135.10 for March through PayPal*.
We will now discuss internally on how to allocate this money in the most effective and efficient manner for the project. We will also be thinking hard about further fund raising possibilities.
Let me use this opportunity to thank everybody who has donated to the Calendar Project so far and to everyone who is planning on doing this in the future. I would also be interested in any ideas that you (our community) might have on how to raise our fund-raising capabilities.
* Please note that the total donation number should be higher than that since this doesn't include any donations this month (April) or any check donations.
Sorry for the 7-week hiatus. I was quite busy personally and the project is busy as well. That doesn't mean however, that Philipp and his crew aren't fixing bugs anymore. To the contrary, a sizable chunk of bugs was squashed in the last seven weeks.
Here's the total list:
As always, our thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
The original plan regarding Thunderbird 3 Beta 2 was to release a beta of Sunbird/Lightning that fits together well. Due to other obligations this goal has slipped quite a bit. We have tagged the list of blocking bugs by bugs needed for the beta and those we can move into the next beta.
Currently, there is only one bug left that we really need to solve for this beta. It has a preliminary patch and will require string changes. Unfortunately, the mentioned bug is not quite trivial, since a lot of UI changes need to be done and we haven't quite agreed on some aspects.
This means we will announce a string freeze as soon as the bug has been fixed and release 2 weeks afterwards, or sooner if all strings have been translated earlier.
Nevertheless, we'd like to provide a beta as soon as possible. To help make this possible what we need most at the moment is testing! Please download Thunderbird 3 Beta 2 at:
http://www.mozillamessaging.com/en-US/thunderbird/3.0b2/
And our latest nightly version of Lightning:
ftp://ftp.mozilla.org/pub/calendar/lightning/nightly/latest-comm-central/
Be sure to back up your data before using this version, since it upgrades your local storage database. Downgrading is not easily possible afterwards without creating a new profile.
IMPORTANT:Note however as we've announced in an earlier blog post this will be the last release for the time being that includes Sunbird. This means that now more than ever, testing is needed to make sure that Sunbird doesn't contain any critical bugs.
If you want future Sunbird releases to happen and either have experience with the Mozilla build architecture or are willing to learn, please do send me an Email and I'll send you some information to get you started.
We will make an effort towards releasing more beta versions of Lightning on the road towards 1.0 to make sure the next release will be as bug-free as possible and still contain the nice new features we have been working on. Remember that we need you to fulfill this goal! Without your valuable testing, we will surely be up for a surprise in case a critical bug has made it into the final version without notice.
I just wanted to let you know that our friends over at Scalable OpenGroupware.org (SOGo) pushed out their 1.0 release yesterday.
Congratulations fellas!
SOGo is a free scalable groupware server. It offers shared calendars, address books and emails through your favorite Web browser or by using a native client such as our Mozilla Thunderbird with Lightning. SOGo is standard-compliant and supports CalDAV, CardDAV, GroupDAV and reuses existing IMAP, SMTP and database servers.
The SOGo developers are good friends of ours and have helped us with developing Lightning today and in the past. So give their software a try! I'm sure they will appreciate it.
You can download their server software here. The source code is also available.
One of the most often made statements within the last few years from our users and supporters was that people would really like to support us by donating a few bucks to the Calendar Project.
That wasn't possible until now. But thanks to the Mozilla Foundation and the efforts of its employees, particularly David Boswell, it's now finally possible to support the Calendar Project by donating to our cause.
All the necessary details are on our donation web page. You can support us via:
If you are a regular user of our software, please consider donating to our cause.
Thanks in advance
The Calendar Project team
Since the start of February, the statistics capabilities of addons.mozilla.org have been upgraded. It is now possible to see the number of active users on a daily basis, where before it was only shown on a monthly basis. This shows some interesting facts about Lightning.
What is pretty apparent from the stats is that Lightning's usage is much higher during the week than on the weekend. This is a stark contrast to Mozilla's main application Firefox, which is usually used by more people on the weekend than during the week.
[UPDATE: As I was informed, Firefox's usage numbers are also much higher during the week than on the weekend. It seems that I mixed up Firefox's market share numbers (which are normally higher on the weekend than during the week) with its total usage numbers.]
On Windows the usage numbers during the week are about twice as high as on the weekends, while on the other operating systems (Linux, MacOS X, Solaris, *BSD, OS/2) the usage numbers are just about 30% - 40% lower on the weekends.
This seems to indicate that we have a lot of business users out there using Lightning and that most of these business users use Windows as their platform of choice.
What we also see is that roughly 87.5% - 89.1% of our users are on the Windows platform, 7.6% - 8.6% use Linux and 3.1% - 3.8% are MacOS X users. Users of other operating systems range between 0.13% and 0.16%.
Those of you, that want to play more with our download numbers, can do this on the Lightning Statistics Dashboard, where you can find much more data to play with.
PS: Don't ask me about the low numbers on five days in early February. These are clearly incorrect and probably a result of an addons.mozilla.org bug, that was fixed later on.
After all the bad news in the last few days, here are some good news at last: Our recent bugfixes. Everything considered, the last 28 days were pretty good in terms of the number of bugfixes. Hopefully we can keep this up as more people will step up and help us with bringing the Calendar Project forward.
Here's the list of the 47 bugfixes in the last 28 days:
Our next release will include Sunbird, but subsequent releases won't unless new contributors take on the work.and
This is really a tremendous opportunity for anyone who is interested and willing to help make a difference here for the project and its hundreds of thousands of users -- get involved!Things will happen if someone actually steps up and makes them happen. That is how it works in the open source space. Philipp is very important as the project's lead developer, but he's not anyone's boss and can't command anyone to do something. Interested people will need to step up and do something, then Sunbird can live on.
Our announcement that we're very short on actual developers has already raised quite a stir. I want to give some feedback to some of the questions that were already raised and also elaborate a little more on the situation that we're in. First the feedback part:
Various users were forthcoming with ideas about feature additions. To answer this let me use a quote from Philipp's blog post yesterday:
This is really a tremendous opportunity for anyone who is interested and willing to help make a difference here for the project and its hundreds of thousands of users -- get involved!
The bottom line is: With no full-time developers on board and only a handful of part-time contributors, there's no way that we're going to implement stuff like Exchange support, improved task management, device synchronization, far-reaching performance improvements, etc. in the foreseeable future.
This can only happen, if you make it happen. New features do not materialize out of thin air, they appear because someone is willing and committed enough to invest time and effort to make something happen. Or to use a famous (and slightly adjusted) quote from JFK: "Ask not what the Calendar project can do for you, ask what you can do for the Calendar project!"
So get in touch, either on IRC (#calendar channel on irc.mozilla.org), our developer newsgroup (mozilla.dev.apps.calendar) or via an e-mail to Philipp or me (our mail addresses can be found on our team page).
I hope to hear from you soon SimonThe Calendar project, which for several years has been working on the Lightning add-on to Thunderbird and the standalone Sunbird project, is at a critical juncture. We feel it's important to communicate this to our users and contributor community, as your input will determine how the project continues.
Recently, several contributors who were working on the project full-time have left the project or have shifted to free-time contributor status due to other obligations. This means their contributions will be limited to their spare time, which is quite sparse given a full-time job and family. This is a significant change from recent years, and if no new contributors come on board, it means that the rate of change will decrease dramatically. Our releases will necessarily have to become less frequent, and the amount of bug fixes and new features per release will decrease.
As a result of this, we have to take stock, and figure out how we're going to go forward.
First, as much as it pains us, we have decided to step back from Sunbird maintenance. Our next release will include Sunbird, but subsequent releases won't unless new contributors take on the work. Trying to support both takes too much time, so we had to make a painful prioritization decision.
Second, our next major goal is to have a version of Lightning that will work well with Thunderbird 3. Lightning won't be built-in to Thunderbird 3 for a variety of reasons (see this post by David Ascher for more on the topic), but we're still on track to have a release that gives users of Thunderbird 2 & Lightning a migration path. That version will be a significant upgrade from Lightning 0.9, including notable performance and usability improvements.
There's a lot of work to do to reach our goals, and as a result we're not yet planning much beyond that. This is where we need your input, both in terms of direction-setting -- given limited resources, where should we focus and why? and in terms of more directly useful help -- if you've been using Lightning for a while, and know a bit about mozilla technologies (any of XUL, JavaScript, CSS), or want to help test, write documentation, or do something else not listed here -- get in touch!
This is really a tremendous opportunity for anyone who is interested and willing to help make a difference here for the project and its hundreds of thousands of users -- get involved!
I hope to hear more from you soon!
Philipp Kewisch (Calendar Project - lead developer)
We're hard at work to fix many bugs in anticipation of our next release, so it's no wonder that 43 bugs were fixed within the last 16 days by our developers.
Of particular importance are 2 bugfixes, which have bothered our users for quite some time:
Here's the list of all the 43 bugs that were fixed within the last 16 days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
Happy new year to all our readers!
Here is the first installment of our status update in 2009. We fixed 30 bugs since Silvester 2008, many of those were new and updated holiday calendars, of which I wrote about extensively a week ago. Still a few important bugs were fixed, on which I will write about in more detail in the coming week.
Here's the list of the 30 bugs that were fixed within the last 12 days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
Hi,
almost two years have passed since I took over as calendar lead from mvl, and I think it's time for a change. So I am stepping back from this role. There are two main reasons why I think this is necessary:
So who's my successor?
I am grateful that Philipp Kewisch has volunteered to step in. He's both smart and a great guy to work with, and has already done a lot for the project. So he feels like the natural choice to me. Please support Philipp in his new role!
Finally, I'd like to thank the whole team for supporting me during this time. It's been a great time for me and I learned a lot. I am sure we continue to have great releases! Having less time does not mean I am leaving the project ;-) I will certainly do a few features/bugfixes/reviews. Stay tuned!
best regards,
Daniel
PS: Philipp is at the top of the "2008 bug fix ranking"!!
Ok, 34 days without a status update :-( Some of you might think, that there's nothing to report... which is wrong. We're still there, we're still fixing bugs, we're still working hard to on the next release. In fact, even with the Christmas holidays, the Christmas shopping, the year-end preparations and celebrations we still fixed 46 bugs in the last 34 days.
But before I come to that, let's look at 2008 as a whole. 2008 was the year, where we fixed 866 bugs. The highest count since the creation of the project eight years ago. Our leading bug-fixer that year was Philipp Kewisch (Fallen) with 190 bugs fixed. Right on his heels was Daniel Boelzle (dbo) with 176 fixed bugs, followed by Berend Cornelius (berend) with 96 bugfixes on third place. Congratulations guys!
One thing is bothering me though. As I already said yesterday we're way to dependent on very few precious souls, which do most of the development work in the project. If you look at the numbers 53,35% of all fixed bugs in 2008 were fixed by the top3 bugfixers. 63,39% were fixed by the top5 and nearly 77% (76,91% to be exact) were fixed by our top10 development contributors. Hopefully in 2009 we can broaden our base some more while still retaining the extraordinary talent that we have in our current developer base.
That said, here's the list of the 46 bugs fixed since the last status update:
Speaking on behalf of the Mozilla Calendar project, I want to say a big thank you to everyone, who has contributed to our project in 2008, be it by providing bug fixes, doing QA work, writing documentation, writing bug reports or just telling some friends about our software. Thank you very much! I hope you stay with us in 2009. Happy new year to everyone!
We're just two days away from 2009, so this is probably a good time to look back on 2008 and some of the good and some of the not so good stuff that happened in that year from my own perspective.
First the good stuff:
Now the not so good stuff:
That is my perspective on 2008. I would be interested in hearing your take on the Up's and Down's of our project in 2008. Until that I wish everyone of our blog readers a happy new year 2009!
We need great t-shirts!
Dear designers, we now have a great new community store, where anyone can upload designs, and t-shirts come out the other end.
We would really appreciate it, if you could help us with some cool designs for Sunbird-wear or Lightning-wear.
22 days with no update here :-(
Some of you may think that this is, because we're not making much progress, but this couldn't be farer from the truth. Actually we have fixed 57 bugs in the last three weeks, which is is really an impressive feat.
Because of the large amount of changes, we're asking everybody to go and download a nightly build and test it intensively and report all the bugs that you find in bugzilla.
As a reference, here's the list of the 57 bugs that were fixed within the last 22 days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that have made this possible.
Two weeks have gone by and the Calendar developers are fixing bugs like crazy, as evidenced by the 44 bugs that we could fix in the last 14 days.
Because of the large amount of changes, we're asking everybody to go and download a nightly build and test it intensively and report all the bugs that you find in bugzilla.
As a reference, here's the list of the 44 bugs that were fixed within the last 14 days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
Another week has gone by and we are making good progress with bugfixing and driving the integration into Thunderbird forward. Here's the list of the 19 bugs that were fixed within the last 8 days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
It's been a while since my last status update on September 19 and I want to give you an update on what we've been up to in the last four weeks.
But before I do that, I want to move your attention to the fact that all calendar code development has been moved away from the MOZILLA_1_8_BRANCH to our new repository comm-central, that we share with the Thunderbird and SeaMonkey developers.
This benefits us in various ways:
There is one major downside as well:
As announced earlier on various occasions, Lightning 0.9 was the last release for Thunderbird 2.0.0.x releases and Sunbird 0.9 was the last release for Windows 98, Windows ME and MacOS 10.2 and 10.3.
The reason for this is, that with the very small development team that we are having, we can only support one platform in an optimal way. For us this is Thunderbird 3, which we hope will feature a fully integrated Lightning extension in its default install. Everything else has to come after this longstanding goal of our project. There will still be Sunbird releases of course and we are planning to release Sunbird 1.0 together with Thunderbird 3
With that being said, it's time to mention the tremendous work of our developers in the last four weeks, where 63 bugs where fixed. Here's the complete list:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
A few days ago, a new Calendar contributor, Fred Jendrzejewski, fixed bug 455728 and bug 455733. I believe that what he experienced here might be beneficial for other would-be contributors, so I asked him to write about his experience in the calendar codebase. Here is his report:
Some introducing things
Few weeks ago I started to work on the mozilla calendar code. My first interesting patches became available in the nightly builds last week and I would like to report about my experience. As this is my first blog post for the mozilla community, I will introduce myself shortly. I am a German 22-year-old physic student in his last year. Since two years I have the great pleasure to live in Paris and to get to know the French way of life (I love it).
A few months ago, I started to work on the Spicebird project, but the problems in the calendar code also catched my attention. I began working on the task / event lists and didn't like that everything was written twice. The idea was that tasks and events are very similar. This property is used almost everywhere in the code( by the calIItemBase object), except those trees. So I decided, highly motivated, to rewrite the two trees and to work on the simpler code.
So, I asked Philipp and Berend for there opinions and they liked it, but over time, the tree code became so complex that it was almost impossible to review the patches nicely. I had to step back and simplify the code first. Following Berends proposition, I started to sort out the hole mechanism for sorting and filtering items in the trees.
The new way of sorting and filtering
So I sorted out functions, modified the code and abstracted it. If you want to sort a item array now, you don't need to write this function completely by your own: you can use a preset of sort functions in the calUtils.js file. Even more practical is the solution for the filtering. We now have a calFilter object, which can handle the most common filtering mechanisms for you. We now have built-in solutions for property filtering and date filters.
First, you can associate a text field for text search to the filter object, just by passing the id as a string to the text-field to the filter. Another possibility is to set a property filter out of the property filter bag (a predefined set of filter functions) in the calFilter object. This code is just one string away from usage, as the text-field filter. If you don't like any of the built-in functions, you also can write your own filter function and set it as the property filter of the calFilter object.
Here again, the date filter can use a predefined set of times (today, this week...). But you can also write your own date filter. To filter out the next 8 days you have to pass the string
Why is it nicer? I think that these are some nice functions that developers of the core code and for extensions can use. If they find other very common filter functions the filter bags can be extended easily. So actually I have cleaned up the trees now and we have created a first common set of filter and sort functions in the calendar code, that can be used as a standard implementation.
What will come?
Now I started to work on a unification of observers and tree internal mechanisms. It seems possible to have only one tree implementation with only minor modifications for the event or task tree. Wait and see...
Thanks to gozer, Mozilla Messaging's build guru, we now have a working l10n dashboard for the calendar localizations. Localizers, who also do l10n work for Firefox or Thunderbird should already be familiar with the dashboard, for the others here's a quick explanation of the dashboard table:
I hope this helps our localizers in working even more effectively than they have already done in the past (if that's even possible). Thanks again, gozer!
To answer all the questions with a single post:
Yes, we are aware that every time you startup Thunderbird, it will try to upgrade Lightning to 0.9 but will fail with a "Download error -288". The reason for this is bug 456710, which Mozilla IT is working on. We hope that this will be resolved soon.
Until then, if you want to upgrade to Lightning 0.9, please uninstall Lightning 0.8 in the addon manager, download Lightning 0.9 and install it. Your calendar data will be preserved. We are sorry for the inconvenience.
The Calendar Project is proud to report, that the 0.9 release of Lightning and Sunbird has been completed and is now available via AMO and the Calendar Project website.
Five and a half months after the 0.8 release we move one step closer to our 1.0 release, currently planned for early 2009. Notable improvements of this release are:
Lightning 0.9 and Sunbird 0.9 are available for Windows, Mac OS X (universal builds), Linux, Solaris (Intel) and Solaris (Sparc) in 31 different languages including English.
Please use the following links to download the release:
Please read the release notes for Lightning 0.9 and Sunbird 0.9 before downloading.
Thank you again to all our developers, contributors, localizers, testers, and supporters. We would not be able to do this without your assistance!
The second release candidate (RC2) of our 0.9 release is available! We really think that this is it. If nothing serious shows up during the next few days, this RC will become the 0.9 final release. Here are the links:
The Lightning builds contain all available locales (30). To download a localized Sunbird build, you'll need adjust the link and replace the "en-US" string with the locale code of your locale, e.g. "ko" for Korean.
Here are the fixes that went into this 2nd RC:
Now that 0.9 RC1 is out, I thought some of you might be interested in the latest news on what patches made it into the RC1. We fixed 34 bugs in the last two weeks and most of the bugs in the list below made it into the RC1 (I'm not 100% sure, which bugs didn't make it, sorry). So that is really good news and with you guys out there testing the hell out of those builds, I'm sure we can make it to the RC2 and the final release in a very short timeframe.
Here are the 34 bugs fixed within in the last two weeks:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
Finally! The first release candidate (RC!) of our 0.9 release is available! I know you've been waiting a long time for it, so grab yourself a build. Here are the links:
The Lightning builds contain all available locales (30). To download a localized Sunbird build, you'll need adjust the link and replace the "en-US" string with the locale code of your locale, e.g. "ko" for Korean.
We really need your help to complete the rest of the L10N Checks and the Update testing, as you can see from our Test Plan.
The Update Testing is of special concern to us, because historically this has always been where we have found critical issues in our release candidate builds. So, please take a look. Feel free to update the test plan wiki or leave a comment on this post with what you tested, so we can track what has been done versus what is still remaining to do.
If you have problems or questions, please drop by #calendar-qa on IRC.
Happy Testing!
I recently fixed some issues around the event boxes in the calendar views (see calendar-multiday-view.xml and calendar-view-core.xml). They are implemented by means of XBL and although they are certainly a very sophisticated use case I found that the implemented bindings do not always utilize all features of this markup language. Probably some of these features are also missing in many other bindings around the world and so I thought this is worth a blog entry :)
Basically an XBL binding contains several elements: One 'content' element where the structure of the binding is defined, one 'implementation' element, that controls the behavior and one 'handlers' element that listens to events.
For the maintenance of an arbitrary XBL binding it is imperative to keep as much information as possible within its 'content' element. This way it is far easier for the developer to perceive the structure and logic of the binding and it also helps to avoid redundant information that may lead to bugs during the development process. If the developer should decide to modify the structure of a binding binding he will most likely run into problems when he then has to adjust the 'implementation' code at many places, too. In this context the 'attribute inheritance' feature executed by the 'xbl:inherits' attribute proves to be very helpful. This attribute takes care that modifications are forwarded down to the appropriate child elements. When it is done like this, attributes may be directly set at the binding without any internal knowledge about it.
In some cases anonymous elements were added unnecessarily to the bindings. For example quite often one box element can be removed just by extending a binding to a 'xul:box' or whatever. 'Spacer' elements may be exchanged by setting the 'pack' or 'align' attribute to the parental element.
I observed that there is also some redundancy in the code. I remember that this has not been resolved with regards to the possible performance loss. From all I have read and heard about XBL this argument is to be taken seriously, but I believe that the use of the 'children' tag within a binding could be an elegant solution, because it allows to add various anonymous content at a defined insertion point.
XBL goes hand in hand with cascading stylesheets - whereupon one should pay close attention to move all style settings to '.css' files. In my struggle with the calendar views I was frequently confronted with completely unexpected problems and side effects caused by seemingly harmless style attributes. For example I could see that setting attributes at certain elements seem to force a distinctive relayout and consequently trash the whole view because of too many other 'compact' widgets involved. This and some other side-effects have made me occasionally become anxious to destroy the layout by merely looking at the .css files!
Last but not least I certainly don't get around to admit that the views have always run quite reliable and offer a vast functionality with an appealing design! As we all know a lot of blood, sweat and tears have accompanied their implementation...
I hope that we will find some time to give the code its finishing touch at the beginning of the next release cycle so that we have time to react to the inescapable regressions that will come up.
I guess it's time to give you an update on our efforts to finally nail down the 0.9 release.
The major newsitem is, that we have decided on our last developer conference call that we will cut CalDAV scheduling from the list of 0.9 features. The reason behind this is, that we have found that this feature is still too immature for wide-spread consumption and the list of regressions would likely put the release date a few weeks (or even months) in the future. Therefore CalDAV scheduling will be disabled by default in 0.9, but interested users can enable it, by setting a hidden preference via about:config.
With that out of the way, we are now down to nine release blockers, where only four of the nine blockers will require code changes and the other five covering tasks to be done during the release preparation process by our build and documentation crew.
So hopefully, we can fix the four remaining code issues during the next week and give you first release candidate (0.9 RC1) shortly thereafter. We also have fixed some other bugs (15) during the last week. Here's a list of last week's bug fixes:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
When I arrived at the Mozilla Firefox+ Summit 2008 on the 28th, I was surprised to hear, that Google announced the release of CalDAV support for Google Calendar while I was up in the air. At first I thought Simon was joking when he told me, but he was not!
I had very mixed feelings on this step. On one side this is a really great step. Supporting CalDAV means that more applications will be able to natively read and write to Google Calendar, without the need for third party applications that proxy the write process through the web-based API. In the long run this means less work for me since at some point the Provider for Google Calendar will not be needed anymore. Instead of fixing the Provider for Google Calendar, I can focus on CalDAV bugs, which benefits both Google users and those of other servers. Also this is something I've been hoping for a while now and have asked Google employees about in the past.
On the other side, I was of course a bit disappointed, that all the work and pain I've put into the Provider for Google Calendar is now being made obsolete. This didn't last long though, since thats just the way it goes with software and I was always aware that at some point something like this would happen. Nevertheless I have no regrets since I think I've done a great deal for users of the Provider and the feedback I've gotten was overwhelming (and mostly positive). Writing the Provider also taught me a great deal about the Mozilla codebase and how the calendar application works, so it was definitely worth it!
Since the above sounds much like I'm going to drop work on the Provider right away and suggest to use CalDAV, let me tell you that it is quite the contrary. As noted in Google's blog post, the release of CalDAV support is developer-focused. There are a number of known issues and application support is currently quite minimal. At the time of release, as far as I know only Apple iCal actually works. Therefore, I'd still suggest to use the Provider for Google Calendar until Google's CalDAV server is more mature.
Using a few workarounds introduced in bug 448561 and bug 451455 I was able to make Sunbird/Lightning's CalDAV work with Google. If you are interested in testing, please feel free to take a look at the latest nightlies (Sunbird / Lightning).
Philipp
We're getting nearer and nearer to our first 0.9 release candidate. With 31 bugs fixed in the last 10 days, many of them release blockers, we're now down to only 11 release blockers at the time of writing. 8 of those 11 bugs currently still need a patch, so if anyone of our readers wants to help out, now would be a great time.
On the localization front it looks like the Macedonian and Turkish localizations will not be able to make it, but Slovenian will, so we'll still be in better shape l10n-wise than we were during the time of the 0.8 release.
For anyone interested, here are the 31 bugs fixed within in the last 10 days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
One of the things we're excited about for the 0.9 release is the addition of CalDAV scheduling support. CalDAV scheduling is an emerging standard; the first widespread use of it was in Mac OS 10.5, where it allows iCal users not only to store their calendars on a shared server, but also check each other's free/busy schedules, send each other meeting invitations, respond to meeting invitations, and so forth.
With 0.9, people using Sunbird and Lightning will be able to do all these things, using not only Mac OS 10.5 servers but also a growing number of other CalDAV servers, both open-source and otherwise. And they'll be able to do so regardless of whether the other calendar users they are interacting with are using Sunbird, Lightning, iCal - or some other CalDAV client we haven't even heard of yet.
CalDAV scheduling support has just started to appear in nighlies over the last few days. It will be filled out further over the next few, but what it really needs most right now is TESTING! We're really hoping the community will give this new code a serious workout and help us find whatever bugs remain so that it will be rock-solid for the 0.9 release. As Daniel has noted, testing nightly builds can be dangerous to both your profile and your data - so be careful but please do bang at it and let us know what you find.
There has been some trouble with an accidental SQLite database schema change I've backed out on monday (see bug 446303), which has forced people (which have updated to that specific nightly) to manually restore their storage.sdb.
I'd like to remind everybody that the nightly builds of Sunbird and Lightning are development versions that might break your profile and data. Even though the recent case didn't bring any dataloss (at least I am not aware of any) and a workaround is available, please take care and do backups before updating.
To prevent SQLite database schema trouble in the future, we've decided to require an additional second review on schema changes, and may add a bug keyword to tag those important changes.
I hope you understand...
We're getting nearer and nearer to our first 0.9 release candidate. With 34 bugs fixed in the last two weeks, many of them release blockers, we're now down to only 18 release blockers at the time of writing. 10 of those 18 bugs currently still need a patch, so if anyone of our readers wants to help out, now would be a great time.
On the localization front, 30 of our 33 are ready for the release (hopefully Macedonian, Slovenian and Turkish will still be able to make it), so we'll likely be in better shape l10n-wise than we were during the time of the 0.8 release.
For anyone interested, here are the 34 bugs fixed within in the last two weeks:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
Sunbird and Lightning both have a feature called the "unifinder". It's the box at the top if you click the toolbar "Find Events". Next to the search text field you can select different time ranges to be scanned matching your search (or just present all matching the range if you leave it blank). The drop down box offers several time ranges, such as "All Events", "Today's Events", "Next 7 Days", ...
While the ranges look sensible, they cause me some headache, more specifically the unbounded ranges "All Events" and "All Future Events" w.r.t. infinite recurring events. If you select one of the unbounded ranges (e.g. "All Events"), the current implementation will only show the master items of recurring events, because there's no finite range to expand into. It shows the events in their condensed (unexpanded) form. While this approach misaligns the UI a bit (the other filters show recurring events in expanded form), it looks like a feasible way at first glance (modulo bugs around it). But there's yet a further problem with this approach: You could override´single occurrences of a recurring series, e.g. modify the start and end of an occurrence. Those are currently left out.
The latter and the inconsistent UI let me doubt that showing event series in condensed form is really what we want here, moreover because IMO showing expanded occurrences is far more intuitive. Solving the problem, I can imagine two solutions:
For the moment, until we have a good solution, I'd vote to remove the unbounded ranges.
I hope there are more options that we could consider, and would like to hear your feedback on this topic...
So I'm back from the Mozilla summit and hope to get out of my jet lag within the next few days. It was really a great adventure with bears, a power outage and a rockslide which got me first floatplane flight ever.
While many people seem to believe that Microsoft is behind all of this, I believe that this was all carefully planned out by Dan Portillo and John Lilly of MoCo :) My hat goes off to dan, who did a perfect job of organizing this whole event for over 400 people.
The summit was great in every aspect. I met lots of people, who I had only known online for years like Gary from the Rumbling Edge, Robert Kaiser (KaiRo) of SeaMonkey fame, Wayne Mery (Thunderbird QA), Axel Hecht (Pike), Mitchell Baker, the hopefully soon to be appointed MoFo Executive Director Mark Surman and many more who I can't possibly all name, because then this post would get a few pages longer.
The sessions were great (I went to most of the Calendar/Thunderbird related ones) and I even got to hold one session myself, where I could introduce myself to a lot of localizers in my new role as Thunderbird localization coordinator.
Besides the sessions, I also got the chance to talk with some people more in depth about some issues, most notably with Mitchell Baker, David Boswell and Mark Surman about sorting out the issues of getting the Calendar Project into the Mozilla Foundation directed giving program, the Mozilla Foundation vision and more ways of cooperating with the Mozilla Foundation and leveraging its assets to move the Calendar Project forward. We also did some product planning regarding the necessary steps for enabling Lightning in Thunderbird.
There's much more, that I could talk about, but I will just close this post by saying that this was a really great event, that many people will likely be talking about for years to come. Thanks Mozilla for organizing this!
Recently I was thinking about how we could show to people the progress in terms of activity, feature work and bug fixing that the Calendar Project has seen in the last few years. I came up with a number of bugzilla reports showing the number of bugs fixed per quarter since the inception of the Calendar Project.
I put all those numbers into a spreadsheet and created a graph from that and here it is:
While it's important to note that not all bugs are equal (A very complex thing like "Support device sync" would be one bug as would be something trivial as "Change window title from 'Edit event' to 'Open event'"), I think it still gives a good sense on the kind of increased activity that one could notice within the last few years in the Calendar project.
If anyone is wondering why we had such a huge drop of fixed bugs in Q2 2007, that was due to one of our key contributors being away for a few weeks in that time frame and the other developers waiting for him fixing a particular bug that was blocking the 0.5 release that was supposed to be released in that time frame.
As promised here is a new status update.
As you can see, the last 16 days have brought us a huge load of 45 bug fixes in anticipation of the upcoming 0.9 release. Unfortunately, due to the high amount of remaining blocking bugs (28 at the time of writing) we had to postpone the originally planned release date of 0.9 RC1 to a later date. I'll keep you posted on this.
Here's the list of all the 45 bug fixes:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
I'll be leaving today for the Mozilla summit in Whistler, Canada. Daniel, Philipp and Martin are also coming from the Calendar team as are hundreds of other key contributors from all major mozilla projects and products.
I hope to have some good discussions on the future of Mail & Calendar, localization and other issues that I've been pursuing. This will be a good time to finally meet in person many of the people that I've only been in contact with via newsgroups, mail and IRC.
You guys can also expect a new status update within the next few days. Until then...
With all the developers talking here now, it probably makes sense to also show you, what we've done in terms of real bug-fixing in the last few weeks. So within the last three weeks 28 bugs were fixed. Here's the list of all these changes:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
String freeze is coming up so I've decided to give an overview of what's on my radar and what has been. Aside from calendar work my semester is about to end but that unfortunately doesn't mean I have more time.
I have put a fair amount of work into bug 353492 in hopes of supporting multiple alarms and different alarm types for the provider for Google Calendar. Due to the large amount of code and the high possibility for regressions I don't think this will make it into the next release. There is still an open issue regarding repeating alarms which has kept me from requesting review on that bug. If you are interested in helping out I have summed up the problems in the bug quite well. Nevertheless I think its reasonable to already check in the strings before the string freeze in case we come to a solution. If you are interested in testing (as an end user) I have a linux xpi attached to the bug. Be sure to back up your profile though since a database upgrade takes place.
Something mentioned in our roadmap for the next release is "better interoperability". Of course this is a very wide topic and I'd like to tell you a bit about what I have planned. Something that was (almost violently) discussed after the last release was a number of problems with accepting invitations. I'm looking to improve the situation a bit also in hopes that I can fix a few Provider for Google Calendar bugs regarding invitations too.
<rant>
Note however that this motivation does not come from the newsgroup discussions. I must say these were quite annoying and it aggravated me to see that users were insisting on us developers dropping everything and fixing the bug that annoys them most. Of course its annoying if something that you were used to broke in the next version but the way things were demanded was just wrong. The way things were discussed made me think for quite some time that I should not do a da** thing for invitations, but there are other reasons I'd like to go forward. I don't want to offend anyone but next time something doesn't work as you'd like it to, please be gentle and understanding instead of trying to force me into things. That works better and usually lets you reach your goals quicker.
</rant>
Recently Daniel fixed a bug that allows me to display invitations from Google Calendar in the invitations dialog. This patch was also checked in on that bug so it would be nice if you guys could give it a whirl to see if it works. Please be sure to enable "display invitations in your calendar" on calendar.google.com.
The next thing that would be nice for the Provider for Google Calendar w.r.t interoperability is managing sending invitations automatically when saving events. Currently this is globally disabled via an advanced preference option. It would be better to instruct the server to send invitations only when needed.
I hope this gives you a better understanding of whats going on in my (calendar) world.
Until next time
Philipp
I recently dedicated a good portion of my time improve the calendar views. Initially I wanted to solve a nasty regression bug dealing with overlapping events, but as I delved into the quite monolithic code I got entangled by other issues that caught my attention.
For example I found it not so easy to keep track of all the shadow elements of the event boxes. Also I noticed – certainly like other users too - that the shadows sometimes seem to be sort of tattered at the corners. For this reason I was not too annoyed when Christian (our UI expert) told me that he already had in mind to change the appearance of the shadows anyway. So our common idea was remove the current shadow elements and only denote a slight shadow with a 1Pixel border around the events. I am currently working on this.
As we have discussed on our last F2F meeting in Hamburg we wanted to improve the look and feel of the calendar views in various respects. From the design point of view the decorated view header has the greatest backlog. Christian and me are working on this (Bug 444292). Christian also is currently heavily engaged to modernize the styling of the calendar views that spread a flair of the last century.
So I can promise you that you will hear and see more about the calendar views in the coming weeks even though we are approaching the final round of our 0.9 roadmap. One last note about the monolithic code: I won't change too much about this shortly, although it is certainly better for the project in the long run if we do this. But with the lack of a compiler I have developed a certain respect for such challenges...
Yesterday we had a pretty interesting developer conference call, which was basically about the issue that only a very small team is currently developing on Lightning and Sunbird (basically just three people), that it is hard to reach community expectations with regards to bugfixes and feature work with so small a team and what would be necessary to attract more outside developers.
From my point of view one of our shortcomings in the past has been on the communication front. With that I don't mean the communication between different people on certain bug characteristics or code issues (that works great), but the communication between the calendar project and our community.
Currently our channels of communication are:
The result - at least as far as I see it - is that
So what can we do to improve our situation, to make the project more transparent, to raise the interest in the project and by that to gain more outside contributors? A few things come to my mind:
Two weeks have gone by and lots of fixes have happened since the last installment of the status update. Thanks to our developers and reviewers 41 bugs have been fixed in the last two weeks, which is really amazing. Nothing is really noteworthy. The fixes contain mostly backend fixes, cleanup work and (unfortunately) regression fixes.
Here's the list of all the 41 changes within the last three days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that make this possible.
Only three days after the last status update, it's already time for another as the developers have fixed lots of bugs (19) in the last three days.
The most notable fixes are in the area of iMIP/iTIP (sending and receiving e-mail invites) with the fixes for bug 377761 (better compatibility with earlier Outlook releases) and bug 429938 (support for meeting cancellation).
Here's the list of all the 19 changes within the last three days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that made and make this possible.
I apologize for the long delay between status updates. It is certainly not due to a low number of bugfixes in recent day. In fact quite a large number of bugs (32) have been fixed within the last 18 days, showing that the developers have been quite active.
Three bugs are particular worth mentioning:
Here's the list of all the 32 changes within the last 18 days:
Many thanks go to all developers, contributors, localizers, testers, and supporters that made and make this possible.
Two weeks ago, we had a great face-2-face meeting, which only now I am able to post about. Among the many interesting talks and discussions particular question we've discussed was "How do we proceed with Thunderbird 3 and trunk?". I'll try to sum up the outcome of this discussion here.
There are a ton of good reasons to move on to the trunk ASAP and help Thunderbird 3 get into shape: calendar will be a core feature of it, trunk has a couple of platform improvements that make our live easier and it will lessen the load on developers and localizers, that we no longer have to think about both the 1.8 branch and the trunk.
On the other hand, we have existing users on Thunderbird 2 waiting for fixes and features. Looking at our resource situation, it was pretty clear to everyone that we can't support and develop both branch and trunk longer term in a reasonable fashion. So where and when do we make the cut?
We've agreed to do one further release on branch in the coming 4 months, namely 0.9. We will focus on the following topics:
More details can be found on http://wiki.mozilla.org/Calendar:Lightning09.
We hope that this release will be sufficient to be used for a longer period of time until the next Thunderbird releaseis out. There will definitely be usable betas (perhaps even alphas) of Thunderbird with integrated calendar, but we also know that some users reject to use those and stay on Thunderbird 2.
We had some further discussion about whether the naming of this last branch release shouldn't be 1.0, rounding up our development for Thunderbird 2. This seems obvious from a marketing POV, but doesn't fit into the mozilla way of promoting major releases. So we agreed that we have a thorough look at what we've reached both w.r.t. features and quality when 0.9 is almost done, postponing that decision.
To answer some obvious questions:
It's been nearly a month since my last status update and nearly two weeks since 0.8 was released, so I guess it's time to show you what we've been working on in the last days. Since 0.8 came out, we've already fixed 29 bugs. So grab yourself a nightly and help us test those fixes.
On a sidenote: Many of the calendar developers will meet in Hamburg (Germany) to discuss future plans, meet up with some of the Thunderbird developers to discuss the future integration of Lightning into Thunderbird and perhaps also do some hacking. We're hoping that this will strengthen the team spirit and give us a better vision of what we wanna do with 0.9, 1.0 and onwards.
Here's the list of bugs fixed since 0.8 was released:
Many thanks go to all developers, contributors, localizers, testers, and supporters that made and make this possible.
Since the release of 0.8 many people have asked for builds on additional platforms and we're happy to comply as our community has supplied us with builds for Linux x86-64, Solaris (Intel), Solaris (SPARC) and OS/2.
Here are the links:
Many thanks to Philipp Kewisch for the Linux x86-64 builds, to Sun Microsystems for the Solaris builds and to Walter Meinl and Peter Weilbacher for the OS/2 builds. Enjoy!
I am happy to report, that the 0.8 release of Lightning and Sunbird has been completed and is now available via AMO and the Calendar Project website.
Five months after the 0.7 release we are again one step closer to our 1.0 release, which we hope to release at the end of 2008.
Notable improvements of this release are written down:
Lightning 0.8 and Sunbird 0.8 are available for Windows, Mac OS X (universal builds) and Linux in 30 different languages including English.
Please use the following links to download the release:
Please read the release notes for Lightning 0.8 and Sunbird 0.8 before downloading.
Thank you again to all our developers, contributors, localizers, testers, and supporters. We would not be able to do this without your assistance!
After all the previous hassles we're happy to report, that the second release candidate for our 0.8 release (0.8 RC2) has been released. We know you've been waiting for it, so grab yourself a build. Here are the links:
We would really appreciate it, if you test those builds heavily, as we plan to push out the final release, if nothing serious comes up within the next 2-3 days. We're especially calling out to our Mac OS X community, to test these builds, as there were no recent nightly builds available for you for the last two weeks and we want to make sure, that everything looks fine on the Mac as well.
If you find a bug, that you think is serious enough to block the final 0.8 release, then please file it over in Bugzilla and make some noise in the bug report.
If you have problems or questions, please drop by in #calendar-qa on irc.mozilla.org or in the mozilla.dev.apps.calendar newsgroup available via Google.
Thanks and happy testing!
Now that the new release of AMO has gone live, a much improved statistics solution is also available. I've marked our download statistics page as public, so everyone can take a look at it. Here's the link:
Public Lightning download statistics
A good explanation of the possibilities of the new statistics dashboard can be found on Justin Scott's blog.
Good news, guys!
The Camino guys have been helping us out with a spare machine of theirs. The machine is up and running and if everything works out, we'll hopefully push out the 0.8 RC2 on Friday.
Please join me, when I say a big "Thank you!" to the Camino developers. We love you guys!
I have some bad news:
Yesterday the Apple Xserve machine producing the Mac OS X builds of Sunbird and Lightning went down and it hasn't come up since. According to Mozilla's IT department, it's a hardware failure and the machine needs to be sent to Apple for repairs, which will take a while. You can find the gory details in bug 423816 (Please don't comment in the bug unless you have something really useful to add. We don't want to piss off Mozilla IT. These guys are great!)
For us this means, that the RC2 is stuck, since we can't release without one of our major platforms. But you can still help us by testing the current nightly builds (just pretend that they are the RC2) on Linux and Windows. This will help us to shorten the time from the RC2 release (once it's out) to the final release.
Here are the links:
Thanks for staying with us!
Somewhere on Sunday evening, we passed the 4-million-downloads barrier for Sunbird and Lightning downloads in total, three and a half months after we reached the 3 million downloads mark.
At the moment we have 2,215,000 unique Sunbird downloads and 1,814,000 unique Lightning downloads. These numbers do not incorporate the downloads of people who updated Lightning via addons.mozilla.org or via the automatic update service in Sunbird.
Some additional information:
Furthermore 218,000 installations of Sunbird 0.5 have been updated to 0.7, so the total numbers of Sunbird 0.7 downloads is at 820,000 at the moment.
We currently do not have any information about the number of different Lightning releases in the wild.
Let me use the opportunity to say "Thank you" to all our users and supporters. We couldn't have done this without you. We hope to bring you a great 0.8 release soon (hopefully with even more downloads).
We're getting nearer and nearer to the final 0.8 release. We've fixed 19 bugs in the last 13 days, are now down to zero (0) blocking bugs and will release a second release candidate (RC2) within the next days.
Here's the list of bugs fixed within the last 13 days:
Special thanks again go to all developers, contributors, localizers, testers, and supporters that made this possible.
The 0.8 RC1 is available! I know you've been waiting a long time for it, so grab your self a build. Here are the links:
We really need your help to complete the rest of the L10N Checks and the Update testing, as you can see from our Test Plan.
The Update Testing is of special concern to us, because historically this has always been where we have found critical issues in our release candidate builds. So, please take a look. Feel free to update the test plan wiki or leave a comment on this post with what you tested, so we can track what has been done versus what is still remaining to do.
The test day today has been a great success, but we need to keep up the intensity and keep pounding on this build. It'd be great if we can make our next RC the final build, but we can only do that if we're sure we found all the issues with this RC.
If you have problems or questions, please drop by #calendar-qa on IRC.
Happy Testing!
We're getting nearer and nearer to the final 0.8 release. We've fixed 26 bugs in the last 11 days and we'll probably push out a first release candidate in the next few days.
The RC1 will come with two blocker bugs unfixed, which we will fix until the final release:
We know, that releasing a release candidate with known blockers unfixed does not fit perfectly into the definition of a "release candidate", but we feel that it is the right thing to do, to get our code out to more testers, which hopefully will not find more blocker bugs.
On the positive side are, as mentioned above, 26 bug fixes within the last 11 days. A lot of the fixes were in the area of .csv file import and export, which should be greatly improved over what we had in 0.7. Here's the list of bugs fixed within the last 11 days in alphabetical order:
Special thanks again go to all developers, contributors, localizers, testers, and supporters that made this possible.
I believe our localization community and our non-English speaking users may be happy about the fact, that since yesterday Lightning nightly builds now contain the strings for all the currently available 30 languages.
So localizers can now test their changes nearly immediately by downloading a tinderbox test build. And many of our users from abroad can now start testing Lightning nightly builds in their native language.
I guess some of you are asking themselves how we are doing with respect to our next major release. So here's a short update:
At the moment we are down to seven blocker bugs, which will need to be fixed before we push out the first release candidate. The status on those bugs is:
I'm sorry that we're not fixing those bugs faster, but two of our main developers are away for a few weeks and such a 33% drop in manpower is hard to handle for such a small project as ours.
But there's also something positive to mention:
A few days ago, the last outstanding localization completed the translation of newly added strings. That means that Sunbird builds for all 29 languages (including English) are now available for testing purposes. For Lightning we'll provide localized builds when the first release candidate (RC1) comes out. But all our non-English testers are encouraged to give builds in their native language a try.
English test builds for Sunbird are available here.
All other 28 languages can bei downloaded from here.
One more week has gone by and some 24 bugs got fixed. The most notable fix this week was probably:
Here is the list of all the 24 bugs which were fixed within the last week:
Special thanks again go to all developers, contributors, localizers, testers, and supporters that made this possible.
As mentioned in yesterday's status report a lot of last week's bugfixes fix important or longstanding issues and deserve special attention. This is the first part (of two) describing those important fixes in more details. So let's get down to business:
WOW! Just nine days have passed since the last status update and 44 bugs have been fixed in that time period. A lot of of those bug fixes fix important and/or longstanding issues and deserve special attention, but if I put everything into this posting you'll probably need a few hours to read through this post :).
Therefore I'll post one or more additional posts within the next few days dedicated to those important bugfixes.
So without much further ado here is the list of all the 44 bugs which were fixed within the last nine days: