Browser Calendaring Integration

One thing we didn’t do when posting about the status of the calendar extension XPI was explicitly say what our browser strategy would be. This has become more of an issue since the release of Firefox 1.5, as people who have upgraded from Firefox 1.0x no longer have the option of using calendar.xpi.
We do care about having a good calendaring integration story for the browser, but we’re still trying to figure out what we think that should be. One way of looking at it is that there are really a couple of pieces here: one involves supporting calendaring on the web (e.g. good hCalendar & ICS handling, other integration hooks for HTML calendaring software); the other is for full-fledged calendar management (a la Lightning or Sunbird).
One possible solution for the calendar management piece might be to support Lightning in Firefox. Another option would be simply to have good linkage and handoff to Sunbird (or whatever your system calendar app happens to be). Or perhaps the management/web split suggested above is not the best one, and there are other, better ways to handle these issues. Please feel free to comment with suggestions and thoughts.
In any case, we’ve started thinking about and discussing these things, but we probably won’t be able to put enough time into this to lay out a real roadmap until after Lightning 0.1 ships. In the meantime, we’d like to suggest that folks who have been using the calendar extension consider using Sunbird, either version 0.2 or version 0.3 alpha 1, depending on your level of risk tolerance. Additionally, if you are currently a Firefox Calendar user and like that setup so much that you’re not willing to switch to Sunbird, we’d like to hear about why that is. Please leave comments…

42 comments

  1. Here’s a use case:
    Someone who only has “user” rights and cannot install new software and who already has Firefox installed on their machine with the ability to install XPIs in it. Giving them an XPI gives them the ability to use the Calendar tool where they otherwise wouldn’t have it or would have to call an admin.

  2. My vote is for good linkage and handoff to Sunbird.

  3. I think calendar should be integrated with e-mail client. Sometimes I receive appointment information by e-mail. How can I add it to calendar if it stand alone app? I don�t want mess with files. People use e-mail clients more often than calendar. It�s just convenient to have calendar functionality in e-mail client.

  4. I’m going to take it off the deep end and suggest that the browser be added to the calendar.
    For a calendar to be effective, it needs to be in your face, not tucked away in another app you never think to launch.
    I’d love a calendar/todo centric browser mutation a la Flock that could place tasks in my menubar and pop up reminders when I open my browser. If there’s a flexible, simple calendar xpi that would let me do this in Firefox then I’m all set.
    I like Sunbird for advanced calendaring, but a lite xpi would be something I would use daily.

  5. I’m a strong believer in the PIM. Integration with Tbird makes most sense. Most professionals who would make use of calendar have email clients open all the time. Just about every office computer you’ll ever see has an email window open. Putting calendar integrated into it makes things easier.
    In the browser doesn’t make much sense, since not everyone’s job involves the web. But just about every job today involves email.

  6. I don’t use any of the Mozilla calendar tools currenly, but I do use calendars a lot on the web, mostly when booking train or airline tickets, but also when scheduling my classes.
    I would like to see a world where, for example, an airline booking calendar became “aware” of when my exams are without actually sending my private info to the airline site.
    Presumably the booking site would notice that my browser is calendar aware and when I open their timetable it would actually send the browser to an iCal page of flights to choose from. The browser would present this as a “choose date” widget with my personal calendar events superimposed on the available flight info. Choosing the date for the flight would send a POST back to the original page, or maybe just fill in a text field with the correct date.
    This sort of fantasy calendar experience is probably a long way off, but it would be nice to move in that direction.

  7. The number one reason for me to use a calendar is to keep track of birthdays and anniversaries of friends, family, and colleagues. Unfortunately, Regardless of where to put the hooks for the Mozilla Calendar, the program lacks functionality to make it a viable alternative:
    1. Sunbird doesn’t have an address book, forcing you to manually enter every contact’s birthday/anniversary.
    2. Thunderbird has an address book, but it _doesn’t_ have fields for date of birth or anniversaries.
    As a result, neither Thunderbird+Calendar, Firefox+Calendar, Sunbird alone, or Lightning, seem to be a fit for me. This is of course a shame, since I really love Mozilla products in general and would like to use your calendar as well.

  8. I agree with David. Anniversaries are most important in day to day living. To remember someone’s birthday just makes the day shine! A well deserved feature that lacks in Mozilla products.
    Integration with email client is also a definite. Easy way to add new appointments while still fresh in the memory!
    Thanks!

  9. My favorite is “handoff”, we did go for the “separate app” path for browsing and mail already. My problem would be the OS integration part here. Are there OS APIs for stuff like “get my calendar(s) for this week”, or Addresses of “Mr Hecht” or something like that?
    Dispatching URLs seems much simpler than that.
    I would like to see things like auto-fill in of dates in airline travel forms, or pop-up of calendar for dates in a web page or mail.
    Given that webmail and mail get interchangable, the integration of calendaring should be as good for the browser as for mail.

  10. I was using Sunbird for a while (part of a long sucession of PIM and calendar apps) but I’m reluctantly back to using Outlook for the moment. As I’m using Thunderbird for e-mail and Firefox for browsing, I’d love to dump Outlook completely but I’m not convinced, despite worthy efforts on the part of some, that the various Mozilla calendar projects will eventually provide an adequate alternative before other more compelling alternatives appear. The Mozilla projects seem to lack a clear vision, are under-staffed, seem to be an afterthought to other Mozilla projects, and seem to be running behind other Open Source projects, notably Chandler (http://chandler.osafoundation.org/) and Evolution for Windows (http://evolution-win32.sourceforge.net/). It may be a year or more away, but I can see myself moving to a Chandler with Firefox combination and dumping both Outlook (for Calendar) and Thunderbird (for e-mail). Sorry to be so negative but I think Mozilla needs to get serious about calendar integration with e-mail or cut bait.

  11. I don’t see any use for a calendar in Firefox.
    Even in the “calendar-aware” browsing experience described by Ami Ganguli (great idea btw) I think the Calendar UI would not live in the browser.
    So… since resources are limited… I would say focus on standalone calendar and TBird integration… and forget Firefox !

  12. How many times should I repeat? — Think of Firefox as a PLATFORM (sort of like Eclipse) for hosting all sorts of web related applications/extensions. The future and success lay with “platforms”. Forget about plain browsers. Plain browsers are antiquities. From the “platform” point of view even the question itself about the use cases is ridiculous. Firefox as a platform must have a calendar extension. Nothing more to discuss.

  13. I see more point in attaching the Calendar XPI to the email component than to the browser component, despite the cool ideas posted above.
    Anyway, as a loyal SeaMonkey user, what I really care about is not Calendar running on Firefox or Thunderbird, but in SeaMonkey (and, thus, mantain its ability to run on the other two, giving each one of the three the best possible integration).
    BTW, I can’t really see the point in managing calendar as folders inside an email window. Despite Outlook perversion, most people don’t browse from the email window, nor use (local) email through the browser window so, why do calendaring in other than a calendar window?

  14. I’d love an excuse to abandon Outlook, but right now the integration between tasks, appointments, and email is better than anything else I’m aware of (on Windows, anyway).
    I’m using the Outlook add-on by David Allen, which makes associating the three object “types” significantly easier. Even if you’re not a David Allen evangelist, you’d be well advised to see how Netcentrics extended the default functionality of Outlook.
    I’d also really like to publish my calendar to select people, or share a centrally available calendar without resorting to Exchange.

  15. Firefox as a platform…
    Yes , but a platform will only support good PIM functions if there is a strong design philosophy. I think this is what makes Chandler compelling–they’ve really thought a lot about how to design a platform for personal information managment. Sure you can add a calandar to Firefox or Thunderbird but I don’t see a coherent underlying vision/design/philosophy thing going on that will result in software that will take us somewhere new and better. It’s same old, same old.

  16. Yeah, I always thought the calendar as an extension was overblowned. Maybe something like a contact list(for webmail), or popup calendar(I use Windows date program currently) or todo list(need it everywhere) that shares the same fileformat/database with the big app.

  17. Thinking more about my use case mentioned earlier…
    I think there should be very basic calendar functionality in the browser, with links to richer functionality provided by your choice of outside application.
    A web developer should be able to:
    1. create a link (similar in spirit to a mailto link) that adds an event to the user’s private calendar (with permission, of course)
    2. pop-up a calendar date selection widget that combines private calendar info with web developer’s info
    3. provide an rss feed associated with a calendar event, so that users who have added the event to their personal calendar have the option of receiving updates pertaining to the event.
    The user should have a rich set of choices for where personal calendar information comes from. Obviously the Mozilla calendar solution should be one choice, but other local applications should work too, as well as web-based calendars. It should even be possible to combine several different sources.
    The browser should then at least support displaying the calendar widget (since I doubt seamless integration is possible otherwise). The widget cannot come from the web site because it doesn’t (and shouldn’t) know about the user’s private calendar events. It probably can’t come from third party applications because the interface would be complicated and probably too slow and jarring to provide a decent user experience.
    Actually adding info to the private calendar(s) doesn’t need to be supported within the browser, but there needs to be a seamless and secure way for the browser to call an external application (either local or a web app) to perform the needed operations.
    The WhatWG has put some thought into this. I think a good direction would be to think about what their calendar spec could look like if the browser had more information about the user’s calendar:
    http://www.whatwg.org/specs/web-apps/current-work/#scs-calendars

  18. This discussion on the integration of calendar seems pretty useless for me. The main problem of the mozilla calendar, in my point of view, is that it did not make any progress for the USER for months.
    The calendar has been available for firefox and thunderbird for years and the reality is that the XPI has not been updated for about a year. The known bugs has not been corrected and the most premium RFE are not implemented (I totally agree with David on the field for the address book and also could think to the necessity of PDA synchronisation).
    It is also true that some alternative exist.
    Sunbird has (had?) some regression compared to the thunderbird XPI : no email alarms and also no printing. Moreover, it will never be integrated with the address book.
    The first shot of Lightning was due for last June. When I wanted to try it in September, I discovered that I had to use a trunk build and unfortunately the lightning was only giving a lot of errors in the JS console. I tried again in november without more success. So lighting is not still there and It could be a good idea to develop it on the 1.5 branch instead of the trunk to meet more users, testers and potential developpers.
    I just now want to conclude by saying that the user can not live with the hope of the killer application for tomorrow. He needs an application with no regression compared to the one he was using yesterday.
    If the calendar application does not provide a working XPI for firefox 1.5 and thunderbird 1.5, it will loose all its actual users and I don’t think it could be good for it. Myself I already switched to my palm desktop, but I hope to be able to use the XPI calendar again in the future (and may be also develop the features I will then need).

  19. Eric, development for the areas that you like will only happen if you make it happen.
    This is a typical chicken-egg problem. You are waiting for a release on which you can start developing and we are waiting for developers who can make such a release happen.
    Personally I have never much liked the Calendar extension, I always preferred Sunbird so I certainly won’t put my energy into anything benefitting the extension. But we certainly could use someone who would put his energy into the Calendar extension.
    Make it happen! …or die trying :-))

  20. Mostafa Hosseini

    Well, the last hours of 2005 are passing by and I’ve managed to build the calendar xpi from trunk on a Firefox 1.5 release branch for windows. In a hurry I may add.
    The result: Not bad. It installs on a pristine Firefox 1.5 and adding an event works. No time to test it further though.
    What this means, I’m not sure.
    If others start using it and the regressions are not that much, xpi support for firefox might become a reality again.
    I’m sure of one thing though: The calendar development team has done a great job trying to stay compatible with extension support while developing the stand-alone version and doing a huge back-end rewrite.Thanks guys.
    And here is the xpi:
    http://ftp.mozilla.org/pub/mozilla.org/calendar/xpi/windows/calendar_windows_20051231.xpi
    Happy new year!

  21. J.B. Nicholson-Owens

    I’m not seeing convincing arguments for a web browser to have a calendar here.
    Not being able to install programs is not a use case for a calendar browser extension. If one can run Firefox from a directory, one can run a calendar program from a directory too. The running Firefox can pass a message to either start up a calendar app or signal a running calendar app and tell it to do something with a date or appointment. Passing messages is how this ought to work because many other programs can be modified to add such messaging.
    The kind of calendar integration I’d like to see is OS integration�I’d like to choose which app handles calendar tasks (so, the calendar programs should all list themselves with whatever interface the OS provides to let me pick from them). I’d like to see apps signal a calendar app with a protocol I can use in my own programs (so, this requires an unencumbered and fully documented protocol).
    Perhaps the place for a browser extension is something that scans the text of a page for something that “looks” like a date and (in a non-clobbering fashion) lets me easily add that date to a calendar of mine.

  22. I like the idea of good linkage and handoff to Sunburd

  23. I have been using the calendar extension in firefox to make it easy to share event calendars. Users can click a webcal:// link on a web page and automatically be subscribed without having to know the complete uri of the link.
    This also allows users to view the event calendar without installing additional calendaring or email software.
    I will be trying the plug in posted above – failing that, I would be forced to compile my own version of firefox with calendar functionality or avoid upgrading firefox.

  24. Simon, I already spent a lot of my spare time on the calendar extension development and I can not find enough time to make it happen again. Sorry.
    At this time, we had the extension as the only development . When the decision was made to concentrate do the Sunbird development, I was not so happy because of:
    – the fact that I did not see the main usefulness of a separate application. The integration of the extension in the mail application like what was proposed (done?) in Lightning had my preferences.
    – the lack of human ressources which would surely lead to the end of extension support.
    During the months that I contributed to calendar, I was almost the only one to propose patches. Those patches were at the XUL+JS level. I didn’t have any way to develop the backend. So, I was not able to contribute to the Sunbird backend platform. My idea on the Sunbird application was that it would lead to some regressions. Those regression are the only reason why I did not use the first Sunbird build. I hate regressions.
    I was calendar contributor because I was a calendar USER.
    Simon, you wrote “Personally I have never much liked the Calendar extension, I always preferred Sunbird”. For me, it is not really a matter of preferences but a matter of usefulness: I needed to print my calendars; I needed to send email alarms.
    I don’t know what is currently the situation but I can imagine that it is impossible for you to develop: the extensions + Sunbird + Lightning. I understand that you have to make choices. The only problem is to make those choices clear and to propose some builds to increase the number of users.
    It is not a “typical chicken-egg problem”: you will have more contributors if you have more users.

  25. Eric, nobody wants to give up the extension. But most of the current developers care more for Lightning (because of its better integration into Thunderbird) or for Sunbird (probably because they like standalone apps) than for the extension.
    To survive the extension needs someone who maintains and tests the extension regularly. This shouldn’t be too hard, since 95% percent of the code is shared between Sunbird and the extension and with the move to the new views (see a few articles before this one) now most of the code between Lightning and the extension is also shared.
    We need someone who wants to maintain the remaining 5% and who regularly releases new XPIs to keep the extension alive.
    This shouldn’t be too hard as Mostafah has shown with his new XPI for Firefox 1.5, which is a little rough on the edges but at least provides some basic functionality.
    If someone can keep the extension alive then we’ll all be happy and then you can contribute to a project where you are no longer the only active developer.

  26. Daniel Neveceral

    I think calendar should be integrated with Thunderbird

  27. My recommendation is to focus on integrating the calendar with the Thunderbird mail client and of course, standalone Sunbird. But as I said above, none of the solutions would fit me as long as it’s not possible to keep track of birthdays and anniversaries of the people in your address book. I don’t see that happening soon enough to wait for it, so I’ll probably look elsewhere for a solution.

  28. I think the calendar sould be reached easily from Tbird and Mozilla (with an XPI component that anyone can add depending on his needs).
    This XPI could be a small client of the Calendar DB without all the functionalities of the full app (for lighter XPI). I think the most useful functions could be: alarm popup, read today calendar, maybe add an event (no search, no heavy GUI). A button can then open the rest of the full app.
    Another function linked to Tbird integration, would be to manage a link with e-mails. For example: you receive an invitation for a birthday, then you add an event on your calendar and you create a link with the e-mail. This way the day before the birthday, the popup reminds you of the birthday, and with the link you access to the original email that contained the map to go to the birthday party.

  29. I was happier when Sunbird came out. It works better for us than it did as a FB extension and we (those of us sharing calendars) have no compelling reason to want it to be part of either browser or mail client.
    YMMV obviously

  30. Simon, I think that I was not clear enough when I replied to you. My english is obviously too poor to make my writings clear. I am trying again.
    You wrote: “Eric, nobody wants to give up the extension”. Why ?
    Obviously, the ressources are not currently available to maintain it (sorry to not being able to provide those 5% effort). Moreover, The Lightning project want to better integrate the calendar with thunderbird and then he should provide the solution.
    Suppose now, that you decide to give up the extension. What would you have to do:
    – completely revamp the web site to: make the extension a secondary application instead of the primary one ; add the reference of the Lightning application.
    – try to modify the Lightning application to make it run on thunderbird 1.5.
    – finish the Sunbird 0.3 release and propose it to the users. A release candidate would be a minimum, I think.

  31. I’d love for the History view to make good use of a calendar widget, so we could easily select the particular day/date we are interested in, rather than go by “1 day ago”, “2 days ago,”… or “more than 6 days ago”.

  32. James, what you are suggesting is currently in the works for the ‘places’ implementation in Firefox, which aims to replace the current bookmarks and history system in Firefox 2.

  33. My thought is that the extension for Firefox should be maintained because Firefox has the most distribution and it’s a lot easier to get somebody to put an extension into a program they already have than it is to get them to install a new program (not always the most rational decision, but one that is fairly prevelant in my experience at least). The short of it is, Firefox is already on the machines in my company, and it’s a whole lot easier to get the extension through than a separate program (especially when Sunbird is still at version 0.3, that makes installing Sunbird an even harder sell)

  34. In my case (and I believe I’m not alone) I’m waiting for one thing only:
    A stable Lightning for TB (hopefully 1.5).
    Secondly sync abilities offcourse.
    I see the point of using extension with FF, but in the end I believe calendaring belongs and relates to TB and emailing. Personal communication, scheduling and contacts are within this app. (There’s probably a reason why MS created Outlook instead of “IE professional”).
    Today I use Sunbird 0.3a1, but why launch another app, when TB is already running:)
    Keep up the good work.

  35. I vote for good linkage and handoff to Sunbird.
    Firefox knows how to deal with mailto links. It will make the same for webcal ones.
    I think that the 3 apps (ff,tb & sunbird) must interact rather than being built one on another.
    I’ll be very happy when I’ll be able to automatically add a webcal:// link received by mail to sunbird by clicking on it in thunderbird.

  36. Integration of Calendar into Thunderbird seems to be best for me. However, I had XPI extension installed into my Firefox until 1.5 version because of some problems running TB and Cal at once.
    I use Calendar extension for TB from upgate to 1.5 and the most important change is the need to logon every time when I visit my web email account. Another thing which I miss is that WeatherFox extension0 is not funnig on background as it was before when calendar was part of firefox with this extension installed.

  37. As a user of Firefox, not Thunderbird, not Sunbird, I liked the smaller download (dialup), and being able to quickly pop open the Mozilla Calendar extension from Firefox (no extra app). And that it could also be lauched as if it were standalone from a shortcut or command line with -cal switch, if I did not already have Firefox going. I was dissapointed when it wouldn’t work with 1.5 as were others. I’ve noticed some others saying they won’t upgrade to Firefox 1.5 or are going back to 1.0.x also, till it gets fixed. Right or wrong I don’t really need or want to get the standalone calander. I’m happy to see that someone is willing to work on Calendar extension for Firefox. Maybe now I can stop using using Firefox 1.0 and the old profile to load Calendar. It’s more to look after but I can see reasons why all 3 (in Firefox, in Thunderbird, and as Sunbird) are good to have available, and let user decide which best fit for them. After all isn’t that part of the reason for the different versions? Someone didn’t like where it was and decided to change it. My download is done, I’m going to see how it works. Thank you Mostafa and all.

  38. Personally, I am fine with a calendar just in Thunderbird. However, you have to remember that most people use their browser way more than they use their email. I think the reason that calendar is more commonly integrated into email is because the kind of people with the skills to implement a calendar are in high demand, and thus get a lot of email and spend more time checking their email than browsing. This doesn’t apply to the other 99% of the population though. Not sure, just a theory I thought I’d put out there.

  39. Hi, I have the following experience with my friends and fellow university students. The younger non-working people a dependent on high mobility. They are using webmail, no mail client, and webbased calendars like yahoo. So their integration is mainly a matter of exporting features and compatibility with mobile devices. All my professional friends (including myself) do use mail clients all day long, especially when coordinating meetings (compare to Outlook) this is where the calendar belongs to, i guess.

  40. The reason I like the calendar as an extension of the browser is because I always have my browser open, whereas with standalone calendar apps I constantly had to remind myself that there might be some reason that I neede it open, which, of course, is one of the reasons I need a calendar app in the first place!
    Right now, I have gmail as my email, so I get to have my email client right as a tab in my Firefox. Similarly, I have the calculator extension, which I use alot, and like having that be able to be a tab in my Firefox. If I get the calendar app to work as a tab in my Firefox, than I will, with some few exceptions, really need to only keep my Firefox, whatever program I’m using to listen to news or music, and my Word processor open on my computer.

  41. I agree mostly with Robert Accettura’s comments – having to open Sunbird etc. separately means that i have on occasion missed appointments, whereas TB is almost always open, so I don’t need to remember to open another prorgam!

  42. I think the least useful option is to maintain a standalone application (Sunbird), from both an end usage perspective and management of project perspective (despite what devs like to do, for understandable reasons ;) I would add). The Calendar ‘function’ is a Big App – not whether it runs in its own address space.
    Architecturally, there is a UI + rules engine, and a data store (webdav or local .ics, but well known). A UI unintegrated with either email or browsing make little sense, imho. No reason not to have .xpi in both TB and FF, whose code will overlap very much, publishing to various places.
    I use the very lite Rainlendar as both master entry and master data store; its big advantage is fast fingertip access. FF is sized so that mouseover on a small desktop space brings up 6 months of the calendar, and 70% of time is spend in FF as opposed to email.