The Calendar extension’s status

Ever since Sunbird 0.3a1 was released, and even before, there has been a constant current of questions regarding the status of the Calendar extension for Firefox, Thunderbird, and Mozilla Suite/SeaMonkey. These questions have, for the most part, all had the general form “Are you going to update the Calendar extension as well?”

To begin with, throughout the development of Sunbird 0.3a1, all of the code for the Calendar extension was maintained, and changes and improvements made to Sunbird were also made to Calendar. At this point, anyone who wants to ought to be able to create an 0.3a1 equivalent version of the Calendar extension for their own use. You can follow the directions at the project page to do so.

The next logical question, of course, is “Why are there no new public versions of the Calendar extension available?” The short answer to this is ‘resources’. It should come as no surprise that the project simply does not have the same resources of say, Firefox. We have a very small set of build machines that are currently being forced to do double-duty, also building Lightning. Attempting to make these machines also build Calendar simply is not possible. It takes away too many build cycles.

The other main resource we’re lacking is manpower. Although the Calendar extension and Sunbird share a large majority of the code, there are still differences between them. Before we could even consider releasing an updated version of the Calendar extension, it would need to be thoroughly tested, on each of the 3 main platforms, in each of the 3 programs. (That’s 9 different, complex test sets.) After that, there would be the additional strain on the development team to triage bug reports that came in, to determine whether they were Calendar specific, or also applied to Sunbird (or to Sunbird and Lightning). Our development team is fairly small
so adding this additional burden would delay future releases of Sunbird and Lightning.

To conclude, development on the Calendar extension has continued alongside Sunbird development. Currently, if you don’t have the ability to build Mozilla projects on your own, Sunbird is your best option. Furthermore, it is important to note that the future of the Calendar extension really is Lightning. (Sunbird will, of course, be the future stand-alone.) In the short-term, we’re working hard to make Lightning compatible with Thunderbird 1.5. We do recognize that the current situation, having no updated Calendar extension for the 1.5 version of Thunderbird, leaves our users in a difficult position. Hopefully, this explanation has helped to elucidate why there really is no other option. Until such time as we can sort out the remaining obstacles for Lightning in TB 1.5, users who have upgraded from TB 1.0.7 are encouraged to use Sunbird.

13 comments

  1. Thunderbird will probably be at 2.0 by the time Lightening is ready! Maybe you should aim at that… the work really is painfully slow.

  2. I find this blog very informing, and it shows the progress much better than before. Keep up the good work!

  3. I love sunbird and have our entire web dev company using it for shared calendars via webdav and my entire family using it at home for the same purpose.
    work – 7 users
    family – 23 users
    Works great, combined with a an svn backup ever 5 minutes for data safety everyone is happy with the work. Keep it up.

  4. “Furthermore, it is important to note that the future of the Calendar extension really is Lightning.”
    I liked the Calendar extension a lot and liked having it integrated to my browser (suite), although I’ve had to switch to Sunbird to continue to use new builds of Seamonkey. Will Lightning support the Suite/Seamonkey, and will it be usable without having to use MailNews (I’m happily using Pegasus for my mail)?

  5. For those of us who don’t build our browsers from source code, but download released installers for our client platform, it seems like it wouldn’t be too much extra work to generate an xpi compatible with a release of firefox and stash it somewhere, unless I completely misunderstand the process here.

  6. I can appreciate your resource situation but I think you are making a big mistake here. All the people who embraced this work (and are now depending on it) are now screaming for a TB/FF 1.5 compatible version and I think you are going to lose most of them if you don’t at least put out a 1.5 compatible version of of the “old” Calendar extension.
    Do you really want to bite the hand that feeds you like this?

  7. Ricardo Palomares

    While I understand the current limitations, I’d bet for a 1/2 proportion for Calendar XPI builds, i.e., one Calendar XPI build for every two of Lightning.
    Besides that, I fail to see why managing calendar items should be done in a mail&news window instead of its own window, like it’s being done right now. It’s like trying to use a mail folder to give access to web browsing. Or maybe it’s just the “Let’s do it like Outlook” syndrome. :-/
    Don’t get me wrong, if there are people feeling comfortable with such UI concept, go ahead, but IMHO, getting the functionality working with a decent UI like the existing one (a really good looking one to me) should be the priority.
    Anyway, cheer up and keep the good work. I expect to announce the es-ES translation of Sunbird 0.3alfa1 in short.

  8. Dan, Michael, Ricardo,
    we can really understand your concerns here and you can believe us, that we would really like to release an updated extension for Seamonkey 1.0, Firefox 1.5 and Thunderbird 1.5, but our very limited ressource situation simply does not allow this.
    Releasing a new version is not just putting up an xpi of the extension and then be done with it.
    Those xpis would need to be thoroughly tested on SM/FF/TB on all three major operaing systems (windows, linux, mac), because we don’t want to release an xpi of low quality and because any bugs arising from lack of testing would hunt us later on.
    In my opinion this is really the time where our community can shine. It is entirely possible to build current trunk builds of SM/FF/TB together with the calendar extension and then put up the extension for public download.
    People who do not want to run the trunk nightly builds are advised to download Sunbird 0.3 alpha1.

  9. Raj, there are unofficial builds that can be integrated into SeaMonkey. You can find one at http://beta.aviary.pl/adrianer/2005.12.15/

  10. How about posting a Alpha/Beta XPI and let us test it for you?

  11. “How about posting a Alpha/Beta XPI and let us test it for you?”
    I second that.

  12. I want to produce at least a Calendar XPI for OS/2 that works on the 1.8 branch builds of SeaMonkey and possibly Firefox. But I am missing very basic information, like
    – Where is the SunBird/Calendar/Lightning development going on (trunk or branch)?
    – If trunk is the answer, how do I create an XPI for branch that works well? (The trunk XPI I made crashes SeaMonkey 1.0beta).
    – Was the Sunbird 0.3a1 release closer to trunk or branch?
    – Please document how to create the XPI. I just by chance today noticed (through bug 296390) that “make xpi” is wrong and that I should instead use “make -C calendar/xpi”. Are there more tricks I should know about this?

  13. Peter:
    The calendar development is currently mostly happening on the trunk. However, there is scheduled to be a branch-mirroring mechanism available shortly as per the new branch plan. Once that arrives, we will start mirroring all of our changes to the branch and supporting the branch as well.
    The 0.3a1 release was made from the trunk.
    Your best bet for getting more help on the XPI questions would probably be to ask in IRC (irc.mozilla.org, channel #calendar).