Thunderbird’s Strategic Dilemma
July 15, 2010 – 4:58 pmIn a recent tb-planning post, neandr wrote:
With all respect for the people working at Mozilla/Thunderbird and fully understand the limitation they are faced with, I would like to see a more detailed mission statement for the products (TB/LG) and the future of it. Only expressing TB is for individual users, SOHO and not for the Enterprise is a very vague statement
I was going to respond to that in the thread, but I got wordy so I posted this blog entry instead.
At the recently completed Mozilla Summit, variations of this request were made by many people that are close to the Thunderbird project (including myself). But after listening to several days of Firefox people extolling the virtues of moving everything to the browser, and being “more like the web”, I have a new appreciation of how difficult the development of a vision statement is for the Thunderbird team.
The standard game plan that Mozilla projects are expected to follow is to develop an application with a significantly high market share so that they can use their market influence to fight for the rights of individual users. Mozilla is a fascinating organization as a hybrid commercial/public interest organization, and they take their values quite seriously.
Unfortunately Thunderbird, which is the only real product that MoMo currently has, hovers around 6 million users, which is much lower than the number they believe are necessary to have the influence they would like (I have heard 100 million users as a goal). Nobody currently has a concrete plan to develop a product with 100 million users. So the current strategy (as I see it) is to try a series of experiments to try to develop some concepts that might be used to specify the 100 million user product. I can’t resist naming things (my wife calls me “an Adam”), so let’s call this future product by the code name “Gigabird”.
One such experiment is Raindrop. Other experiments are going on in extensions to Thunderbird, which seem to be focusing on changes to the user interface. Right now, that is where the vast majority of the developer resources are focused at MoMo.
So if your real strategic mission is to develop Gigabird, what do you do with your legacy product Thunderbird? The big problem is that the “ordinary users” that are the primary focus of Firefox (and by implication also MoMo) are migrating away from email for many forms of messaging to other media – Twitter, Facebook, text messages, web forums, blog comments, etc. The hardcore users of email (who are likely to continue to use a desktop client) are sitting in office cubicles, yet going after these “enterprise” users is counter-cultural for Mozilla.
So what are the strategic choices available to MoMo?
the HailMary
The goal here is to try to come up with one or more really clever innovations that will form the basis of Gigabird. This is, after all, the way that some of the new messaging formats have occurred, with Twitter as the poster child. Using these innovations as a base, the basic plan for Gigabird will be formulated at some point in the future. This is the current MoMo strategy, at least as I see it. Given the existing Mozilla culture, I would probably do this as well. (Warning: should this strategy every become publicly revealed, the director will disavow any knowledge of these actions.)
the AboutFace
Here you notice that the values-driven direction that you felt so passionate about is actually not going to get you anywhere, so you make a major readjustment in values to allow you to pragmatically accept a new direction. Such moves have been done by Mozilla in the past, and are part of the standard corporate Myth propagated by Mitchell Baker (the story about how in the early days they were adamant about never shipping a binary). The application to MoMo could be to accept that what they have is an email product, and their future users are going to be sitting in cubicles. Users in cubicles should have rights too, so there could be a valid Mozilla Foundation purpose in fighting for the rights of these “enterprise users” and let Thunderbird develop into an enterprise product.
the SlowPlod
This is the direction that existing Thunderbird users are hoping for. The ultimate goal is to slowly improve Thunderbird until it is undeniably the best email client around. You fix any important bugs. You support all of the hot new messaging concepts. You spiff up the user interface, incrementally adding new features that provide small improvements to usability. You keep your power uses happy with lots of extensions for specialized purposes. It’s pretty clear that dmose does not believe that he has sufficient resources to pursue this strategy, nor is he likely to have them in the foreseeable future. The Thunderbird code base is also really hard to adapt to these new media (witness the struggles that I have had or jcranmer’s blog ). I think that the MoMo team wishes us well, but believes that the future lies elsewhere.
the VacuumTube
Just because you can’t change the direction of humanity does not mean that you have nothing. Vacuum tubes are long gone – yet the guitar player at my church proudly uses his fancy amp with glowing tubes showing through plexiglass. The company that bought one of my previous businesses had also previously purchased a manufacturer of vacuum tubes, which had morphed into specialized purposes like lamps for spectroscopy, and nuclear-warfare-resistant cathode ray tubes. Email clients will be with us forever, and in the hands of people who love them could have a useful future in various niches.
My Prediction
MoMo will pursue the HailMary until they have enough ideas to formulate a real plan. At that point, they will want to devote all of their resources to Gigabird, and be looking for an honorable way to retreat from Thunderbird – which will be a variation of the VacuumTube. The likely retreat will probably be some sort of future custodianship by a conglomeration of companies that provide a freemium strategy. So if there was a basic, free Thunderbird product that could be enhanced with addons with commercial value (like my Exchange Web Services product, or Postbox as a Thunderbird addon), then MoMo could pursue their vision without abandoning their Thunderbird users, and let companies like MesQuilla and Postbox support Thunderbird.
I suspect that one of the reasons for Thunderbird’s relatively low market share are email services like Yahoo mail where you have to pay to use POP and there isn’t any IMAP support. If more people used services like Gmail or if there was a way to somehow make Thunderbird support web-mail services that don’t support IMAP or POP, increasing Thunderbird’s marketshare would be much easier.
“The big problem is that the “ordinary users” that are the primary focus of Firefox (and by implication also MoMo) are migrating away from email for many forms of messaging to other media…”
I’m really not sure this is actually true.
They are certainly adopting Twitter and other methods, but I have yet to find *anyone* who still does not rely on email “as well as” these other mediums.
I do understand the dilemma MoMo face but the *only* way Gigabird will get to 100million users is to get on the corporate desktop/laptop/tablet in the same way Firefox has…
“I have yet to find *anyone* who still does not rely on email “as well as” these other mediums.”
It depends on your definition of “rely on”. Email still has some irreplacable uses (if nothing more n to authenticate who I am when I sign up for Twitter!), so everyone has to have an email account. But you don’t need an email client for the occasional need.
But you might benefit from a “Messaging Client”, which Thunderbird is not but Gigabird certainly will be.
One important thing you only strafed lightly is that MoMo is a *commercial* entity and as such is desperately seeking for a way to earn money to sustain its mission – which is hard indeed, given that non-enterprise users are rarely willing to pay for an email program (if they even know such a thing exists – browsers are the predominant way of using the web these days)…
“non-enterprise users are rarely willing to pay” … and yet people need to eat. I have to admit that I am not a believer in “free as in beer” open source – though I am as guilty as the next guy in searching hard for a “free” choice to avoid paying a trivial $10 fee. Gigabird’s 100 million users don’t care about “free as in freedom” either, but they love free beer. Love it too much, IMHO.
I think the basic problem is the way messaging changed and still changes. Some points that I could think of:
- why should I use a desktop client, if a webmailer is sufficient?
- how can I check my email from different computers/devices? IMAP is fine for that, but you still need a client on every device which (probably in 99%) will already have a webbrowser installed -> use webmail
- why should I email at all, if all my friends communicate at facebook, twitter or any other social network?
- if you’re online all day long, why should you use some asynchronous medium such as email and not instant messaging?
Email has its strong points, especially if you get a large number of it or need search capabilities over all your mails from the last 10 years. But the number of people who have to rely on that is quite limited.
In my opinion, the basic plan behind such a thing as GigaBird could be to allow people to bundle all their messaging needs (email, twitter, social networks, rss, instant messaging, …) in one product where those parts are tightly interconnected. So instead of having to c&p a tweet to an mail or an instant message you got to facebook, you could just click on a button or menu entry to do it.
Of course it’s not feasible for MoMo to support all this things out of the box, but it should be possible to write extensions for GigaBird that can do it.
I believe having such a communication central in one application could be a way to cope with changing messaging needs, but of course it’s still possible that it’s completely wrong what I see from my restricted point of view
“basic plan behind such a thing as GigaBird could be to allow people to bundle all their messaging needs”
I call this a part of a larger issue of information fragmentation (see http://mesquilla.com/2009/01/27/information-fragmentation-its-the-enemy/) A lot of people talk about wanting to combine things together, but few people really describe the major advantages that such a service would give, sufficient to overcome its inherent disadvantages. (You have given a few). Aggregation is not a big enough idea to sustain Gigabird, but would be a necessary component.
I think your EWS and JCranmers WebForum Plugins are the right way for TB to go, but today it´s still too hard for average Extension developers to enhance TB with new Account Types. I think TB needs a much simpler and easy scriptable API for AccountTypes/Inbox/IncomingServers, so Twitter/Facebook/Youtube/whatever accounts will follow.
Before the forum, I was quite surprised that the MoMo team was not jumping all over this account work we were doing. There are several ways to look at it. First, we are part of the experiments that could eventually gell into Gigabird. But the prevailing wisdom is that what we are doing is too hard, and the likely future solution will be a direct injection of information into gloda rather than add something as a traditional mailnews account.
Michael, I doubt that. Yahoo’s webmail service is still years behind Gmail. If more people switch to Gmail the result will be just that – more people using Gmail. In the browser. Because Gmail’s web interface is really great and more than sufficient for most people. What are the compelling features that Thunderbird can offer which make it better than Gmail? For me those are:
1) Ability to take my mail offline – most people don’t need that, they are always online when they read mail.
2) Convenient RSS functionality – with the implementation being rather half-hearted this isn’t a real selling argument.
3) Ability to work with multiple accounts – most people don’t need that, they only have one email address.
4) Ability to sign/encrypt emails – way too complicated, almost nobody uses that.
5) Lightning – oops, not an actual reason because it is so broken and I have to run Sunbird instead (which is also broken but slightly less).
So I fully agree with the analysis here, Thunderbird really needs to decide on a direction. Corporate users are the market where it is most likely to have success in the current form. Getting significant market share among “regular” users isn’t something I can see happening, they don’t have any reason to migrate away from web interfaces. And morphing the application into something that is no longer a pure email client (Twitter? Facebook? Whatever new service is cool right now?) is something that *might* work provided good ideas and implementation.
Your “in the current form” as a desktop client is the key restriction here. Mozilla’s culture is very much “everything in the browser” so it would be natural to move Gigabird that way. (My post “Thunderbird as a Firefox Extension” at http://mesquilla.com/2010/06/01/lessons-from-google-thunderbird-as-a-firefox-extension/ got much better private feedback than it did public.) Another possible future that I did not mention would be for MoMo to be subsumed into a Mozilla Labs project that feeds Firefox. Mozilla Labs already has a “contacts in the browser” project, why not a “messaging in the browser” and “calendaring in the browser” as well?
As for “messaging in the browser” – there already is Snowl …
(
And actually, your blog software behaviour is a good way to demonstrate how discussions in the web often suck:
- “reply” jumps down to the textarea, but you lose context
- quoting is tedious
- you can’t efficiently follow threads, because you can’t tell what you already read
- and the mails I get about new comments just link to the top of the page
I hate web forums.
)
Yep, information fragmentation leads to lots of suboptimal implementations. I for one would use a really good integrated solution.
As for you Snowl comment, that’s a good example of fragmentation within Mozilla itself. Do we really benefit from Thunderbird, Snowl, SeaMonkey, Postdrop, and SpiceBird all pursuing their independent visions? I don’t believe that is in the user’s best interest, which is what MoFo is supposed to be all about.
I think Thunderbird needs to become a better e-mail client (your too corporate-focused “AboutFace”). Yes, this will need to be with corporate users also in mind. I don’t think imitating or integrating Twitter, Facebook, and the like will help much.
There are a few (incredulously neglected) bugs that would greatly improve Thunderbird for ALL users (corporate AND private):
- Bug 250539 “New paragraph (e.g. list) clears formatting, does not respect font prefs for HTML messages” (This bug *should* be a no-brainer)
- Better Address Book (more fields, sync)
- Make Lightning more usable (e.g., bug 455045, bug 151994, bug 137777)
Unfortunately, I see Thunderbird continuing to “rearrange the deckchairs on the Titanic” (Buttons-to-Header, neutered Quick-Search) for the foreseeable future.
Good blog post though! I’m glad to see someone at least thinking/caring about Thunderbird.
The issue of neglected legacy bugs is an emotional issue for existing Thunderbird users, who as I mentioned generally would like what I call the SlowPlod strategy.
I hope you understand though that MoMo has been clear that SlowPlod has no reasonable change of success, if you define success as “getting large enough to impact the direction of messaging.” They don’t really have the resources for that, and today even a client that was much better than Outlook would have a hard time dislodging Microsoft from the market – because then you have to start concerning yourself with integration with Sharepoint, and Office, and Active Directory, and pretty soon you are doing AboutFace as well. We have a name for such a company that can do this, it’s called “Microsoft”.
Still Thunderbird exists, and has a following. What do you do with that? Maybe what you need is a separate company that could focus on Thunderbird the product, without being distracted by Firefox. Oops, perhaps you thought MoMo was that. Well it isn’t (though perhaps it could have been). They are a subsidiary of the Mozilla *Foundation*, and are bound to fulfill Foundation values. Thinking with Foundation values, there are evils that need squashing in messaging – siloing of user’s messaging data and relationships into separate entities like Twitter and Facebook that don’t represent the “open web” they are bound to promote, for example. MoMo wants to fix those issues, and they are not convinced that fixing bugs in Thunderbird gets them there.
The preferred solution is for volunteer open source developers to come forward to do this. But that is a hard sell. I could be one such person, but I am not convinced that I want to devote all of my precious time to that. Now, if you and nine like-minded friends were each willing to give me 10% of your salaries, perhaps I could be persuaded …
I donate 10$ each month to Thunderbird (and another 10$ to Calendar). If 399 other people did the same (i.e. 4,000 $/month), perhaps someone (you?) could make Thunderbird better (a’la “AboutFace”).
So, are there 399 people reading this blog…
http://www.mozilla.org/foundation/donate.html (why no Thunderbird here?)
There are 3 basic categories to the email software market:
(1) people using web-based email
(2) people using a simple desktop email client that don’t need calendering and other shared corporate stuff
(3) people using a desktop client that need calendaring and other shared corporate stuff
Category (1) is something that Thunderbird doesn’t play in, and in MSNHO should leave alone.
Category (2) Thunderbird dominates, although it could probably gain some market share here. This appears to be the current strategy, and it’s a reasonable approach.
Category (3) is the remaining big piece where TB could gain share, and it’s a pretty big piece. But it’s a complex piece. Outlook+Exchange is a powerful combination, and I can’t see TB having the resources to create such a combination.
But what it could do it start strategically identifying pieces of this market that could be targeted. Sunbird is an example of such a strategy. So would be developing an open-source Exchange connector.
“start strategically identifying pieces of this market that could be targeted” MoMo (and MoFo) don’t think this way. As a long-time small businessman, *I* do (my Exchange Web Services work is clearly aimed at the individual user, with a personal email account as well as a work account, and wants to check both on a single integrated client at home). But I have the very modest goal of being able to survive retirement without getting overwhelmed by health care expenses. MoFo (and MoMo) want to *change the world*! You don’t do that by picking off small edges of a much larger market. That’s what small businesses are good at.
I’m not disagreeing that Thunderbird might benefit from that kind of direction. I just know from my own experience of trying to convert my friends. I go to convert them only to find they use Yahoo mail and it won’t work with a mail client. They often have another account for school or work, which is often through another service that doesn’t support email clients. I think that Thunderbird offers something great that webmail clients don’t. You can easily access multiple accounts at once. The few friends that use Gmail and other IMAP supporting email have been very easy to convert. They can access their email easily and all at once. Even if they are not home, they can still directly use the webmail service to view their email. I admit that it’s possible that my experience has been outside of the norm.
I have a hard time feeling sympathy for users who are so cheap, that they want everything for free, and are unwilling to pay Yahoo a little to upgrade their accounts. Should I dedicate my precious time to enabling them? Should MoFo? Nobody is willing to subsidize their habit.
Their is a well-known small businessman joke, that “I’m going to lose money on every sale, but make it up with volume!”. The “everything has changed” euphoria of the internet boom tried to pooh-pooh those ideas, but 1 + 1 still equals 2, and understanding the revenue stream is still critical to success. “I’m going to accept donations and subsidies from foundations” is a valid revenue stream, but you have to be doing something those that those foundations want. Making the world safe for cheapskates is not such a value.
You won’t get any argument from me there. I’m a capitalist and I know there’s no such thing as a free lunch. Let’s look at it the way economists do for a moment where everything is assigned a dollar value. People may not be willing to pay much money because it doesn’t buy much utility for them. They might be willing to spend tiny amounts of money a month, but they’d rather not bother with the effort. On the other hand, they may be willing to accept things like an advertisement or two on the Thunderbird start page because if they aren’t annoying and are sometimes helpful it would be a form of payment, but it wouldn’t involve effort on there part. I’m not saying that Thunderbird should necessarily have advertisements, but it is one of many possibilities.
“People may not be willing to pay much money because it doesn’t buy much utility for them” There is another issue to consider as well. A common issue in the developing world, is that if someone comes in and provides something for free, but it is not sustainable, that actually does more harm than good. The free solution drives away anyone who might try to actually run a business, and when the unsustainable solution ultimately fails, then everyone is left worse off than if the free solution never happened in the first place. Unsustainable open source projects need to consider this issue, because their very existence can prevent people from paying for something that does provide value to them.
I think Thunderbird/Lightning is one product that would seriously win if it targeted women. It doesn’t make sense to take the same user model as Firefox, since that user will probably just use mail in Firefox.
You’re right, the enterprise is one area where people still really need a mail client, but that is due to their business, and more importantly their organizational needs.
As a busy, and technically inclined mom, I needed to find a good calendar application to keep myself straight and also to keep the husband in the loop. Sunbird was the *only* choice (and IMO is still the only good choice, at least on windows).
I have two kids and myself to keep track of, and now that we’re no longer together I also have the added complexity of custody. There’s ballet, sports, appointments, school, holidays. What weekends are mine? When do we need to switch weekends because of travel?
Then (bear with me) there’s my period. For some reason keeping track of this is really important to women. Even more so to families that are trying to get pregnant, but still important to families who are not! I held on to ICQ 95 for years because the timer function was perfect for reminding me to take my pill, as well as allowing me to snooze it if I didn’t take it right away. I use my phone for this now, but it’ll only snooze for an hour, so if I miss it, I miss it.
So yeah right now I have a lot more use for a calendar than an integrated product, but what happens if my kids’ schools start learning how to use email/calendar invites instead of paper newsletters? How awesome would that be.
But you’re right, the people who use the products right now are sitting in cubicles, and people tend to want to use the same products at home as they do at work. So I dunno, maybe the way to break into the home market is still through the enterprise. Or maybe it’s time for a whole new grassroots movement to reach a whole new user base with a product that could change their lives.
The demise of Sunbird, IMHO, is an example of the failure of open source to fully recognize how important monetization is to success. Without a plan for that, there was no real long-term plan to keep it alive. Open source as a movement of volunteers is really hard to sustain.
Messaging naturally drifts into the broader area of personal information management, of which calendaring and task management is a key component (along with contacts and various kinds of notes). From my perspective, my life has been disorganized ever since I quite using ECCO (which my wife still relies on, running on Windows 7, even though the last release of the product was over a decade ago). For me, lack of messaging integration makes it unusable.
Unfortunately the commercial failure of ECCO and similar products have given an aura of “avoid this” to the whole area of personal information management. Meanwhile, as you observe, there are really no good products available, and our lives remain disorganized. I think that the dream of, one day, coming up with a viable personal information manager is part of what keeps me motivated to stay involced in messaging, as that would need to be a critical component of any successful product.