New forums are ready!

Good news! Thanks to cbeard and the folks at Mozilla Labs, I now have placeholder forums where discussion can ensue in a public place, without requiring every post to be approved by me.

For starters, I created three fora: one for Thunderbird Planning, one for discussion of Internet Communications Innovations, and one for random MailCo topics if they come up. We’ll see if those are the right ones as things move along.

I look forward to reading everyone’s contributions there! Feel free to add comments on my blog posts or email me if appropriate, of course.

For people who are new to Thunderbird, you should also check out MozillaZine’s forums, several of which are about Thunderbird — they’re a good place to find day-to-day support and discussions.

Meta comments:

  • The labs forums site was chosen for three simple reasons: 1) it provides RSS feeds, which I see as a requirement, 2) it could be done very quickly with minimal coordination, 3) there’s something experimental about all of this! I’m happy to consider alternatives in the future.
  • I should explain as well why I’m using web-based forums rather than a mailing list or newsgroup. Basically, I believe that newsgroups are too obscure outside of the fairly insular “insider” community, and that mailing lists require too much commitment on the part of participants, whether that’s adding to their email load or requiring them to customize their mail flow to triage mailing list traffic. The market has spoken, and web forums seem to have won when it comes to “level” public discussion spaces (as opposed to blogs). RSS feeds are there for those of us who can’t take yet another destination and need integration.

    (If someone knows how to do a two-way integration between the SMF forums and a mailing list so that everyone can interact with the forums in their favorite way, I’m happy to see how hard that would be to add.)

PS: I have a hard time writing “forums” instead of “fora”, but I’m being good and fitting with the flow!


Here’s something interesting — Florian Queze and Quentin Castier just released Instantbird, which is a XUL application wrapped around libpurple, the library behind Pidgin (once GAIM), Adium, Meebo, and others. There are some licensing issues around doing the same in Thunderbird right now, but it’s nice to see that some people are connecting these various technologies in interesting ways.

“I just really want to know what’s going on with Thunderbird”

Someone called Rod asked a bunch of questions in reaction to a recent post:

Your developers left. Where did they go? How are they going to continue to participate in Thunderbird development? Who is going to be taking their place as the lead developers? If you don’t know yet, what’s your action plan for figuring that out?
What’s the release schedule for Thunderbird? Are we still aiming for a 3.0 this winter? Is it going to be pushed off indefinitely?
What about new features for Thunderbird? There have been a number of allusions to using MailCo as an opportunity to redefine email. Wonderful! What does that mean, what have you done in that regard?
That’s what I mean about the state of affairs… what’s the 10,000 foot view of the Thunderbird landscape? Where are you now, and where are you going with it, and why should we hang around and wait and continue to convert people to Thunderbird?

Btw, please note that I’m totally not trying to be a smartass or combative here. I just really want to know what’s going on with Thunderbird. In the past few weeks I’ve seen nothing except tight-lipped-ness from both Mozilla and the (former) Thunderbird developers, and vague hints that great things are coming. I’d just like some real info to back up those hints, that’s all.

I can’t answer all of those, but I’ll see what I can do.

First, any tight-lipedness on my part or the part of anyone else at Mozilla about Scott & David’s plans is simply because we can’t, and won’t, talk about their plans on their behalf. I don’t know what they’re planning on doing in any detail, and even if I did I wouldn’t say anything without their express permission. So for all questions about Scott & David, please talk to them.

Now, onto the things that I can talk about.

I just really want to know what’s going on with Thunderbird.

I don’t want to sound flippant, but I suspect the biggest difference this month compared to, say, June, is that there’s more talking with people outside of Mozilla about Thunderbird and the exciting possibilities that lay ahead. Many people come to me (the new guy, designated point man, bullseye, lightning rod, etc.) with ideas, plans, resumes, etc. That’s basically “what’s going on”.

At the bug & code level, as far as I can tell, it’s business as usual. For example, I noticed today that David Bienvenu was in the IRC channel, helping people with their patches. That’s not noteworthy, that’s just David staying involved like he said he would!

I want to emphasize that at this stage, it’s just talk. There have been no decisions about specific Thunderbird plans. Those take a long time to form, especially in a distributed, global, collaborative, multi-factorial system like Mozilla. Some people will, I’m sure, criticize me for that, wanting clear authoritative leadership, someone to take charge, etc. However, I don’t think that would fly very well with the dozens of people who have done a lot more to make Thunderbird what it is today than I have! I’m much more interested in getting to know them, understanding what their ideas are, and figuring out together a roadmap which enough people can align with, and which motivates others to join.

Q: How are they [Scott & David] going to continue to participate in Thunderbird development?

A: Like any other contributor, especially as their privileged status as module owner doesn’t change just because their employment status changes.

Q: Who is going to be taking their place as the lead developers?

A: They’re still module owners, as per the cultural practices of the project. Clearly I need engineers on staff to work with them and the other contributors on the codebase. As to who that will be, I don’t know yet. See below.

If you don’t know yet, what’s your action plan for figuring that out?

A: I’m talking to other developers familiar with the code base, such as Seamonkey module owners, peers, etc., finding out if they or people they recommend would be good people to hire. I’ll be going through an interview process with the help of Mozilla engineers to identify the best people for the job. So far, I’m having interesting conversations. I look forward to being able to announce those hires!

Q: What’s the release schedule for Thunderbird? Are we still aiming for a 3.0 this winter? Is it going to be pushed off indefinitely?

A: I’d be a really bad software project manager (or whatever it is I am in this context) if I said what the release schedule was for Thunderbird before I had a good idea of scope, resources, rate of change, etc. However, given the current staffing levels as of mid-October, and my understanding that there are no major new features currently implemented on the trunk codebase from which a 3.0 build would come, I think that a 3.0 release this winter is unlikely.

Q: What about new features for Thunderbird? There have been a number of allusions to using MailCo as an opportunity to redefine email. Wonderful! What does that mean, what have you done in that regard?

A: I’m sure you don’t expect me to have a fully baked answer to what “redefining email” means after a couple of weeks on the job. Just to clarify — I wasn’t hired because I had a specific vision for Thunderbird. I was hired, I believe, because it was felt that I could help shepherd the community towards a new vision. That will take time and patience on everyone’s part.

Also — if this was a typical product, and I was the product manager, either I’d refuse to answer (if you don’t promise anything, you can’t disappoint — that’s why Apple keeps its secrets so well), or I’d answer based on a spec which I’d built over weeks, using estimates that the rest of the team had built over weeks. As it is, I haven’t even begun to go through the bug database, the mozillazine forums, or the newsgroups, to learn about the features that Thunderbird users have asked for in the past. So project management realities alone mean that you shouldn’t believe anything I’d say about specific new features, even if the choices were mine alone.

That said, I have a few general thoughts that I’ve been noodling on, based on my conversations and readings in the last few weeks, which might help with the current uncertainties. Again, they’re just my early thoughts, and don’t read much into them.

There are at least three timescales with which I’m thinking about Thunderbird planning.

  1. the very short term. Are there burning issues which need attention this week, or this month?

    I’m not ramped up enough yet to answer that question. Luckily, I don’t have to! The staff of Mozilla Corp, with Scott and David’s help, are keeping on keeping on with respect to emergency issues in the short term.

  2. the medium term (say, a year or so): what features and other changes should we be looking to add to the current system, that will make a significant difference in the lives of millions of users, but which are not so big or hard to implement that they can’t be ready for mass market use within a year.

    This is a big topic, but some high-level points include:

    • Calendaring and task management are generally desired features, that all major competitors provide, and which are supported by well-defined standards. I think it’s likely that Thunderbird will grow some related features, likely starting with Lightning, which is already well under way to being useful to masses.
    • Thunderbird needs to be as good a platform for extension writers as Firefox has become. There’s architectural, website, documentation, and community work to do there. I’m confident that if we do the foundational work that makes it easy for creative developers to experiment, amazing extension will emerge, which we can then consider for integration into the core product.
    • If there are ways to make Thunderbird interoperate with more mail providers than our current set (IMAP, POP, and special-cased Gmail) in such a way that the product integrity is preserved, then I think we should see about facilitating those use cases.
  3. the long term (say, five years from now): what major architectural shifts are needed to ensure that we’re relevant then, given the large shifts in the industry, whether that’s the maturation of the web as platform, the advent of much better mobile devices, the emergence of non-email channels, etc.
  4. I have very rough thoughts that fall into that bucket, but I’m not ready to discuss them yet — I’d rather spend my time figuring out who has informed opinions, code, people, or time to contribute to figuring out possible long-term roadmaps. (I tend to leak half-baked ideas though, so be a bit patient and you’ll probably get stuff to chew on).

I don’t think you’ll find much above to back up specific “great things that are coming”. That’s not because I don’t think great things won’t come, but because, even though no one believes us, we don’t know what those things will be. I still believe, maybe unreasonably, that there’s enough potential in a vibrant Thunderbird (which, by the way, is getting more investment than it ever has) that more people with better ideas will want to jump in, and we can then see where that leads us.

I hope this helps, Ron, and whoever else was thinking what Ron was. Feel free to ask me more specifics, and I’ll answer as best I can. I’m happy to answer questions as to what I’m doing, what I’m planning, what I’m thinking. Hopefully my blogging will make that clear over the weeks to come, and we can get over this patch of uncertainty and discomfort for some. Then if what I say sounds interesting, and you want to help, let me know.

CalDAV good times ahead?

While I wasn’t looking, a bunch of people have been making real progress in standards-based calendaring interop. I’ve been talking to a few of them, and it’s fascinating to see momentum building in that arena. There are clients getting mature (Thunderbird with Lightning, Apple’s iCal in Leopard, Chandler Desktop, Evolution, probably others I’m forgetting or don’t know about). There are servers getting there too (Chandler’s server, Apple’s CalendarServer, Bongo, a couple that I know about but don’t know if I can talk about publicly yet). As far as I can tell, the CalConnect consortium is working, making sure that people play the interop game, even though I’m sure it’s hard work. It would be nice if Google wasn’t so far out in left field, but you can’t have everything.

Nothing like being minority players to get competitors to cooperate on pushing a standard forward!

Open letter to the Thunderbird community

Thus far, my public comments about MailCo have been quite high-level, and spanning big fuzzy concepts like communication, organizational structure, etc.  Here I’d like to touch on Thunderbird specifically.

Both Scott McGregor and David Bienvenu have posted that they are leaving Mozilla Corp.  My understanding from chats with them weeks ago (I hope I’m not divulging anything that I shouldn’t) is that they have decided to start a new venture.  They’ve worked on Thunderbird and its predecessors within Mozilla and Nestcape for a long time, and I can certainly understand their desire to do something different.  After all, I’m doing something different myself.  At the same time, I know from talking to Scott and David that they care very deeply about Thunderbird, and I look forward to working with them in the future. 

That’s all I know about Scott and David’s plans, so I’ll talk instead about what we’re doing with the still-being-formed mail-focused subsidiary of the Mozilla Foundation, regarding Thunderbird planning in particular.  While Scott and David will I’m sure be forthcoming with advice and opinions, I’m hoping to solicit input from other Thunderbird community members.  

The bulk of the MailCo budget is expected to be spent on staff (as with all small software companies, and especially small open source software companies!), with most of that going to Thunderbird-focused staff for a while, I expect.  We’re recruiting experienced developers now to focus specifically on Thunderbird and more broadly on improving mail and communications in general. Everyone involved full-time in the development of Thunderbird has been offered a role and we’re moving forward as quickly as possible to hire additional developers.

I’m also going to be looking for non-developers to help in every other aspect of maintaining and evolving Thunderbird. If you’re interested in such a job, or know someone who I should talk to, please don’t hesitate to let me know.  People who care about the project and who want to make a different in the world of email will be key to Thunderbird’s continued success.  One of my tasks next week is to get a better picture of exactly which jobs I should fill — I’ll share the job descriptions as soon as they’re settled, but you don’t have to wait for that if you’re interested.

More broadly speaking, the intent of the new subsidiary is to gather together the Thunderbird community (including you) in a single “virtual” location, so that what happens or doesn’t happen in Firefox-land isn’t so relevant to Thunderbird anymore.  I’ll have more to say about the new community “hangouts” soon.  I’m (obviously!) convinced that Thunderbird is great, and can become even greater. There’s so much pent-up energy around the project, it feels as though it’s about to take over the world.  There’s of course a lot of work ahead to coordinate everyone’s enthusiasm, but I’m looking forward to it.  For the last couple of weeks, I’ve been gathering feedback and opinions from lots of directions.  The opinions of the core Thunderbird community are more important than many, so if you care about Thunderbird, please let me know what you think.  Now is a great time to influence the future of Thunderbird.

Thinking org-onomically

I’m cheerfully stealing the title from the current advertising campaign from TIAA-CREF, which I’m finding interesting in the context of this post — how can a corporation for the public benefit (my own non-legal term) make money?

As a bit of background — the premise of the creation of MailCo is that Mozilla will provide $3 million in seed funding, to build the initial organization and fund it for an initial period. One of my mandates, however, is to work towards self-reliance. Currently, there is no set deadline by which we must be “cash-flow neutral”. The Foundation board has made it clear through Mitchell’s comments both public and private that it’s possible that the foundation would provide more funding in the future, if MailCo is seen to make progress on a variety of fronts. It’s also clear that the preference is for MailCo to find other sources of funding to at the least complement Mozilla’s contribution.

I think that encouraging me to figure out other sources of income is critical to ensuring that MailCo does what it’s supposed to do, which is to help people communicate over the internet. If, after years of work, we haven’t found anyone who cares enough to allow for some workable funding mechanism, then something’s not right. There is so much money in the software and communications arena right now, and we have the potential to reach so many people, that, based on no analysis but a pure gut check, I believe there’s got to be a funding model out there. In fact I’m sure there are a few, and that the biggest challenge will be picking the right ones. It won’t be easy. But it could be fun.

One unfortunate bit about business deals is that there is often value in keeping the details private, and very little value in talking about them in public. So I don’t think I’ll be blogging extensively about potential deals, negotiations in progress, and the like. Luckily, there’s nothing yet to worry about “leaking”, as I’m just ramping up and have only had the most exploratory of conversations so far.

In trying to come up with funding models, it’s sometimes useful to come at the problem from a variety of perspectives: market segmentation, understanding the industry players, identifying possible customer types, fitting into a competitive landscape, building a strategic “plays”, identifying “customer pain”, trendspotting, forecasting industry shifts, etc. Each one can lead to possibilities. it’s also clear that different people, because of their different backgrounds, tend to come up with different ideas as well.

If you’re used to selling “enterprise software” (a category name that I don’t particularly like), then the “obvious” thing for MailCo to do is to tackle the Outlook/Exchange market. I don’t know the exact bookings from Exchange licenses, but it’s a billion-dollar business (if anyone knows the exact number, let me know). It’s important to realize that Microsoft has an incredibly strong hold on that market, as many who have tried to tackle it have found. At the same time, it wouldn’t take a large percentage of that annual business to make a big difference for MailCo’s budget. If we had a scalable “enterprise support” offering, it wouldn’t take that many customers to allow us to break even! Please don’t read too much into this paragraph — I’ll have more to write about Outlook and Exchange in the future, I’m sure. At this point, I’m just using it to point out the incredible amounts of money being spent on email systems today, even though many email clients (and servers) are free.

Alternatively, if you’re used to selling to teenagers, you might think about making something pink and furry with a USB stick and some built-in, branded communication software built-in. With a bit of luck and design talent, those would sell well, and the pink would become unfashionable after three months, making it easy to sell the better, orange version. Or something.

Alernatively, if you’re used to selling “eyeballs”, or “users”, or “subscribers”, then the details of how the people associated with those eyeballs, etc., interact with other people over the internet is a potentially valuable asset. People seem on the whole willing to trade some exposure to advertising for free software and services. At this point, I don’t think the fit between something like Thunderbird and standard web advertising is a very good one, but a) no one said the income had to come from Thunderbird, b) there are lots of very interesting variations on “advertising”, and some companies will see value in advertising-like things which most of us wouldn’t even recognize as advertising-related.

Alternatively, if you’re a central organization for a possibly niche or clearly segmented part of the market, you might want to pay for to ensure that versions of software exist that are particularly well suited to your community/market/members/customers. Mainstream proprietary vendors routinely ignore “small” markets because they’re not “big” enough. But small for them might be plenty big enough for us.

Alternatively, … Well, you get the idea. At this point, I’m happy to talk to anyone who has ideas as to how we could make money someday. I’m happy to talk to businesses about possible relationships. But I’m going to be concentrating, for the first little while at least, on building the organization, building the community, improving the product, enabling innovation beyond the product, and improving our ability to reach users. If we have great products used by more and more millions of people, then, with some care and feeding, the economics will sort themselves out.

The Mozilla Organizational Hack

It’s been discussed before, but given some of the comments I got about my post discussing MailCo’s legal structure, I feel I need to try again and explain the whole for-profit/non-profit structure, and why it’s not only the logical thing to do but also a brilliant organizational hack.

First, the usual caveat — I’m very much not a lawyer, and I may get some of the details below wrong. I think I have the fundamentals right, though.

A constant: Mozilla assets belong, either directly or indirectly, to the Mozilla Foundation, which is a “California not-for-profit corporation dedicated to the public benefit”.

One of the significant consequences of this is that the full power of the US Internal Revenue Service (the friendly folks who took down Capone, remember?) is available to make sure that the public benefits from the Mozilla Foundation’s activities. This is a fundamental bit of legal architecture which has a real impact on the trust that can build between Mozilla and its users, developers, partners, and the public at large. People generally trust the IRS to make sure that, in exchange for being tax exempt, non-profits are monitored so that no swindling of the public can occur.

The downside is that the IRS ends up defining the rules of operation of such an organization, and those rules are so focused on funds transfers (in and out) that many non-profits end up being effectively glorified (and occasionally glorious) financial houses, raising funds and disbursing funds. Which is great in some fields, but doesn’t tend to produce great software, or allow for the kinds of relationships with companies which are needed to change an industry.

On the other end of the spectrum you have “Corporations”, which are taxable entities, and in exchange have more freedom and autonomy from these IRS regulations, as their social purpose is not direct public benefit but private benefit (from which, at least theoretically, the public should benefit in aggregate).

As any manager or executive is well aware, if a company has shareholders, the courts have, since 1916 (we’ll get back to that date), ruled that companies must act in the best interest of their shareholders, and they’ve defined that interest as fiduciary interest — meaning, how much money the shareholders can make.

While this standard corporate structure has led to awesome world changing businesses that employ hundreds of millions of people and have generated wealth for many, it’s also true that the need to look out for shareholders first and foremost sometimes leads companies to do things that hurt the general public. The sad thing is that company directors and managers legally don’t have a choice — they are bound by law to look out for the interests of the shareholders, beyond all other interests. I’m grossly oversimplifying, and there are lots of great debates to have on this topic, but the main point is valid.

(I do recommend both the movie version and the book version of “The Corporation” to people who are interested in that topic. As an example, starting on page 35 of the hardcover edition, you’ll find the story of the fight between Henry Ford and the Dodge brothers which led to the landmark 1916 ruling mentioned above).

So it would seem that corporations are legally bound to be “amoral”, while non-profits, while “moral”, are stuck in a mess of paperwork. If you want to work in a software shop that wants to make the world a better place, that would be a depressing conclusion.

We now get to the the organizational scheme that Mozilla picked and why it’s such a wonderful hack. By making a wholly owned subsidiary of the foundation (which means making a corporation with a single shareholder), it’s possible for the corporation to do what’s in the best interest of the shareholder, using whatever metric the shareholder wants to specify. And given that the shareholder’s interest must be the public interest, it means that while the corporation is taxable, there are safeguards in place to ensure that it stays aligned with the mission of the foundation. The only “downside” is that the subsidiary must pay taxes. Personally, I’m a big believer in taxes as a key mechanism for the improvement of humanity, so that doesn’t seem like a significant bug.

As an aside: Mozilla isn’t the only organization to use this setup — it’s just the only one I know in the software field, and probably the first to have such a big impact on an industry of “hard core” for-profits.

For the conspiracy theorists out there, I realize it’s a bit disappointing. There cannot be an IPO or acquisition. No one can buy or sell stock in any of these organizations.

Which leads to the next question I wanted to clarify, viz. how can MailCo make enough money to become more self-sufficient? Tune in tomorrow for the answer, or a least a hand-waving version of an answer.

In the Bay Area for 24 hours

I’m in the bay area for 24 hours, just long enough to meet the folks at mozilla headquarters, have a few meetings, and participate in what should be an interesting new experience: a live interview with Mitchell Baker and Asa Dotzler on Air Mozilla, mozilla’s internet video outlet. In the future, I’ll try and make my visits longer and mention them earlier, so that I can schedule a broader range of conversations, be they in offices, homes, or organic fair trade coffee shops! It’s clear from the state of my inboxen* and the comments on this blog and Mitchell’s that a lot of people have opinions and ideas to contribute, ranging from perspectives on the problems to possible solutions to possible business relationships to employment discussions. In the first few hours, I’ve heard from friends old and new from as far away from home as China, Singapore and even Redmond! This is going to be a fascinating adventure.

*: yes, that’s an old vax joke which the kids won’t get, but I’m tired and punchy, and I get to make 2 email jokes a month now.