Lessons from Google: Thunderbird as a Firefox extension!
June 1, 2010 – 11:39 amDuring my current trek to the Bay Area, I seem to be bombarded with news about Google. I was particularly interested in the different way that Google views its business model, and that got me thinking about how Thunderbird fits into Mozilla’s business model.
In an extensive article in Atlantic magazine (I always read the Atlantic while travelling), James Fallow describes efforts that Google is undertaking to try to revive the viability of news reporting as a professional activity. A lot of this revolves around different methods of bundling content with monetization schemes. In the traditional newspaper model, news is one of many parts of the bundle, with monetization provided by various advertising streams. The internet has provided many ways to unbundle the various pieces of the newspaper, with the result that the hard news content no longer has a sufficient bundling with the monetization schemes.
Google may be viewed as a new kind of bundle, with its many services (search, maps, email, news, etc.) all part of the bundle of services that are attached to the monetization scheme of search advertising. The Atlantic article notes that:
Virtually all of Google’s (enormous) revenue comes from a tiny handful of its activities: mainly the searches people conduct when they’re looking for something to buy. That money subsidizes all the other services the company offers …
That requires a new way of thinking about understanding the relationship of traditional “cost centers” to “revenue”. When Google’s Andy Rubin was asked to justify the expense of Android development in this interview in the San Jose Mercury News, he responded:
So all we have to focus on are those types of innovations that scale for large audiences, and … the revenue crank just turns.
So what is Mozilla’s bundle and monetization scheme, and how does Thunderbird contribute to that bundle?
Mozilla’s main bundle that is attached to a monetization scheme is the FireFox browser, which is monetized through people’s use of a particular search provider when connected through the bundle. Thunderbird (and its related Mozilla Messaging cost center) do not currently contribute anything directly to that bundle and its monetization scheme.
I don’t believe that Mozilla’s subsidization of Thunderbird can continue indefinitely. Perhaps that was possible for awhile, but Mozilla’s bundle can be expected to come under continual competitive attack, which will eventually force them to marshall the resources needed to defend the bundle. Mozilla Messaging and Thunderbird need to be part of the solution, and not just a cost drag, when the tough decisions are being made to defend the bundle.
If we think like Google, it’s pretty obvious what the step is that is needed for Thunderbird to become a major contributor to Mozilla’s main business of “Firefox the bundle”. We need to figure out ways to attract people to the bundle, and get them to linger as long as possible on the bundle while the “revenue crank just turns.” The obvious solution is that the underlying communications code that drives Thunderbird needs to be repackaged as a FireFox extension. (Since I call that codebase Skink, let’s temporarily use the code name SkinkFox to describe the Thunderbird-as-Firefox-extension product.) SkinkFox could have the same relationship to Thunderbird that Lightning has to Sunbird: the same product packaged both standalone, and as an extension.
“That’s just SeaMonkey!” you say. Well no it is not. If you think like an engineer and focus on features, then I guess SkinkFox is similar to SeaMonkey. But thinking like a marketer, the goal of this is to build the FireFox brandname and revenue stream, and SeaMonkey does not contribute to that while SkinkFox does.
Technically, great efforts are already being made to reduce the barriers that keep the mailnews (Skink) codebase separate from FireFox. I don’t think it would be a huge leap to figure out how to merge Thunderbird’s user interface and tab system into FireFox’s.
As Thunderbird enters a critical rethinking in the next few months in the aftermath of Thunderbird 3.0/3.1, SkinkFox is one of a variety of thoughts I will be giving on possible futures for messaging at Mozilla.
We could always rebrand SeaMonkey as “Netscape Communicator”. That’ll give us instant brand recognition for sure (all we need is to buy the trademarks and copyrights from AOL which mozilla could do for a pittance these days).
Ratty.
I think you’re starting out with the wrong questions… It’s not about competing with other “bundles” or how to generate revenue.
Rather, what’s the value for users, and how does it further Mozilla’s mission to the Web? It’s not clear to me that bundling Thunderbird really helps with either.
There is not just one starting place for questions, there are many dimensions to understanding a project. The issue of monetization is a perfectly valid view of a project. Which comes first, the monetization plan or the user value? Neither, you must have both.
“how does it further Mozilla’s mission to the Web” is an irrelevant question if Mozilla fails to generate enough revenue to sustain itself.
The idea that one of the highest paid open-source projects in history won’t make enough money to “sustain” itself seems a bit silly doesn’t it?
Is there ever a point when one has enough money?
I have not examined the financial plan for Mozilla, so I don’t know what their pressures are. But I can see that the hot browsers on the market are webkit based, and the Mozilla codebase is not moving forward quickly enough to keep up with them.
So while it may seem like Mozilla has a lot of money, their revenue is miniscule compared to their main competitors in this race, which are Google, Apple, and Microsoft. They need to invest more in their browser to keep up. So while you might claim that their income is enough for “one” (person?), it is not enough to both maintain a product edge, and to invest large sums of money in things that do not contribute to their monetization scheme, at least in the long run.
I think you’ll find the graphic in the “application architecture” section of the April 2003 Mozilla Development Roadmap very interesting: http://www-archive.mozilla.org/roadmap/roadmap-02-Apr-2003.html#application-architecture
Thanks for pointing that out. The vision from that document exactly matches technically what I am discussing here. Yet I should also point out that my main point here is not the technical proposal, but rather asking the question whether we should try to understand Mozilla as a bundle with a monetization scheme (like Google, or the traditional printed newspaper), and then make sure that whatever we do, it contributes to the bundle as well as to its other goals.
Would one way of thinking about this SkinkFox be, that just as gmail is a mail interface living inside a permanent tab in firefox, SkinkFox could be a mail interface living inside a permanent tab in firefox? Not to say that it had to work that way, but rather to say, that there already is a mail program, i.e. gmail, living inside firefox. So one way of seeing SkinkFox would be as “just” a imap-to-http filter.
The Thunderbird user interface it more than just a tab. It is really a three-level view of a tree. The SkinkFox extension would need to be more than just a tab.
:S
That does sound rather like a desperate measure to keep a dying application alive. Since Google launched Gmail (and the others played catch-up) very few see any point in using a stand-alone e-mail app. And for good reason.
While a web-based interface may meet your needs, it does not meet that of others. The vast majority of enterprises still believe that a standalone application has value.
Yet clearly Thunderbird is a little at odds with the rest of Mozilla’s culture, which is strongly focused on trying to make html-based applications to foxus of everything.
I believe the combo of Firefox and Thunderbird is already available as SeaMonkey by KaiRo and his team. Getting everything into the tabs is indeed the way to go, and SeaMonkey should be able to achieve this.
I addressed in the blog why SkinkFox is not SeaMonkey. Yet at the same time, I would agree that in many ways SkinkFox is more closely related to SeaMoneky than Thunderbird.
But there is a big difference as well. SeaMonkey’s limited monetization theme relies on the trademark difference between SeaMonkey and Firefox, while Thunderbird is a supported child of FireFox. So while technically it is more natural for SeaMonkey to adopt the SkinkFox concept, from a monetization perspective it is almost impossible for SeaMonkey to do so, yet it is natural for the MoMo team that produces Thunderbird.
Me thinks that the average person embedded in Mozilla culture does not appreciate how powerful the monetization issues are, since it is counter-cultural to openly discuss them.
And here was I hoping Mozilla would do something innovative with Thunderbird (like item centric interface…)… I guess we can forget about that if Thunderbird becomes a Firefox extension.
I don’t think it’s realistic though. Expecting Thunderbird to fare well as an extension, I mean. Maybe the best solution would be to integrate Thunderbird with Firefox, and vice versa, in a way that users using both of them would have benefits when compared with users using just one of them. Because, after all, Thunderbird is something completely different from Firefox, and they may and will run at the same time, on occasion, but not always.
I don’t really understand why an extension is inherently less innovative than a standalone application, other than that the effort to repackage takes away from time that might be spent innovating.
I’m not sure that innovation is the goal of Thunderbird either. Yes it is one of many possible goals, yet there are other pressures as well. I hope to discuss these issues in further posts.
A firefox extension is a crazy idea, at least until Firefox get multiprocess support, One of the reasons Seamonkey was split is that I do not want my email client to crash when my Browser crashes, we are talking about a lot of local content (POP3 and local folders), frequent local crashes in thunderbird are not as damaging as a crash on a browser checking GMail. Probably you are using the wrong name, and you want an XULRunner Thunderbird running on the same XULRunner used by Firefox, and instalable using Extension like features
No I am not using the wrong name, I really meant an extension under FireFox. An independent XUL application does not contribute to “turning the revenue crank”, and that is the perspective I am promoting in this post.
Isn’t that like going back to square one? (Mozilla Suite etc.)
While the Skink add-on would of course be optional, the performance hit on Firefox would probably similar as it was with having Mozilla Suite load all its components in addition to the browser. Additionally, I’d say that integration of email in current browsers ranges from bad (Seamonkey) to awful (Opera). Would the new add-on be that much better / more discoverable?
While the existing integrations may be awful, to me that is a critical issue in all of this that ought to be addressed. So much of email now has become an attention getter that points you to a website (Facebook friend request is an example). The integration is occurring anyway, and we need to make it work more cleanly.
Raindrop is Skinfox
Why does Mozilla feel the pressure to copy, copy, copy Google?
Your “copy, copy, copy” is my attempts to fight a “not invented here” mentality. Personally, I don’t find enough appreciation of what is happening with other vendors in the Mozilla community. I really don’t think that you are helping matters when you accuse people whoare being open-minded about the what is going on around them, of being nothing but a “copy” of someone else.
Innovation is usually 90% copying with a small additional insight.
I would say it depends on who you consider your competition. If your competition is the GMail, Twitter, and Facebooks of the world, an extension may make sense. For all of the talk of post-email technologies, well, Wave made a big splash but I haven’t heard anything much since; Twitter has an abysmal retention ratio, and Facebook appears to be hitting its peak like other social networking sites before it. The “fifteen minutes of fame” phenomenon is pretty strong on the internet, and you would have to expend a lot of energy to keep yourself well-known in an already desperately crowded field. The integration would have to be done extremely well, but you run into the problem of a bifurcated audience: people seem to either love or detest Web 2.0-y UI (note some of the TB UI, er, wars).
Another point I think might be worth bringing up. You do talk of emulating the bundling model, but it seems that the public opinion in recent weeks has swung to the anti-bundling side of things. I do think there appears to be an “uncanny valley” in effect for bundling: people want more bundling until they realize that the bundling just has so much data collected on you that it is scary. While FF is not exactly a service keeping track of all of your data, it does seem that people are pretty happy keeping their browser and their email client separate. No offense to the SeaMonkey developers, but it’s not like people are hopping to use SeaMonkey because it’s integrated data. Using FF’s brand recognizablity is probably not going to help a similarly-oriented extension.
If you consider your competition Outlook… such an extension is not going to buy you anything.
This post is not fundamentally about external competition, it is about understanding Mozilla as a system (or bundle) that works together to generate revenue. Mozilla is competing with itself by directing users away from its browser to Thunderbird, and it does not need to be doing that. From that perspective, Thunderbird is a “competitor” to FireFox, while Gmail is not.
“public opinion in recent weeks has swung to the anti-bundling side of things”
While that may be true, it is also true that in the last few weeks a master bundler (Apple) has now surpassed Microsoft in total market value. What people whine about and what they really want and need are not always the same thing. Nobody wants to have money extracted from them, but it is necessary for any organization to extract money from someone if they are to survive.
It would be better if Thunderbird.next revolutionized e-mail instead and made something that Gmail cannot offer. After all it’s a standalone app, i.e. with fewer limitations and better performance.
As a developer, I know it would be fun to work on revolutionary software. Yet Thunderbird’s user base for the foreseeable future are legacy users. I’ve been tempted to do a blog post “What went wrong with Thunderbird 3.0″ soon, and a major part of that is that the new Thunderbird team thought that their user base was more interested in innovation than they turned out to be.
Your post is a very airy roundabout way of stating the damn obvious: Thunderbird needs to find a way to make money. Has always and still does.
The solution you are suggesting is simply weird. There is no way to avoid the concept that you are asking for an email program’s chrome UI to be integrated into a browser. Very few people seem to want that and those who do are surely somewhat satisfied with SeaMonkey?
The argument that many people are happy with webmail personally doesn’t ring true for me. I use Gmail but I also prefer Eudora: probably the most feature-packed email-only application ever written. Alas Eudora is no longer supported by Qualcomm and the Penelope project consists of a grand total of one developer last time I heard (years ago).
Integrating en email application’s chrome UI into a browser is just weird. There must be better solutions. For example:
- Consider anonymously harvesting keywords from a user’s email and presenting inline ads based on those keywords. Ok it sounds like a privacy invasion but it’s very simple: Google does it and few seem to care. Why cannot Tbird be Gmail with the power of native applications?
- Why not reverse your proposal and keep user’s inside Tbird, searching from a Tbird-owned search box? Thunderbird already includes a html parser and a lot of browser-like code, why not (conceptually) ‘intercept’ searches that would otherwise take the user out to Firefox and send the revenue to Tbird instead?
- Make a better widget. If you really want to attract enough users to Tbird to make any monetization scheme worthwhile: improve Tbird! The application seems to have spent so many years receiving patches mainly to repair the buggy code inherited from way back in the Netscape 4 days. I am trying to avoid Lotus Notes at my workplace. It is horrible. I look up how to migrate from Notes to Tbird and it did not seem easy. Why not? I’m sure there’s whole host of pro-active features that could be added into a native email application that just have not even been considered.
Think a bit harder. Remember it is only in the last half of the internet’s modern life that the idea of funding a browser through the search box arose. Before that everyone thought the old user-pays model was the only option and that was butchered by MS monopoly.
“Thunderbird needs to find a way to make money”
I tried to bring up the example from the interview with Google’s Andy Rubin. He was asked this exact question with regards to the Android operating system. His response was that this is actually not the relevant question, and that is the point I was trying to make. The question really is, how does my product contribute to the entire monetization scheme of my organization’s bundle? As long as the product contributes significantly to the goal of keeping people engaged with the monetized bundle, then it is not necessary to directly relate “Android” to “search revenue” – or in the Mozilla analogy, “SkinkFox” to “FireFox”.
“Why not reverse your proposal and keep user’s inside Tbird, searching from a Tbird-owned search box”
Because the powerful bundle to build on is not Thunderbird, but FireFox.
[...] you’ve got suggestion drop them here; Lessons from Google: Thunderbird as a Firefox extension! Also Read Thunderbird 3 RC 2 Update AvailableGive Thunderbird 3 More Shine with the Silvermel [...]