Skip navigation

The current thinking is that enterprise software companies need to open source part of their technology as an effort in low-cost marketing. The theory being that marketing $ that aren't productive for customers (e.g. advertising) are put towards development, which does help customers (and yet still feeds the marketing engine with branding, leads, etc.). This is certainly why most companies do OSS, but it's a tricky process.Having gone through this with Wildfire, we have seen the challenges and have a better idea of what it takes to make an OSS project successful. I think open-sourcing software can reshape the way dollars are applied to projects, but it requires a very focused effort.


IMHO, some elements that are important:**


  1. Embrace professional services: Understand the service aspect of the model, and seek out the biggest users of your application to find opportunities for professional services. When these opportunities cease to be relevant to the majority of the community, find consultants to take on the work instead of core engineers -- this keeps the $ flowing efficiently.

  2. Create something great: What you create really has to blow people away. Just making it free is useless unless it has the potential to pull an avid community of users.

  3. Write a constitution & stick to it: Understand at the beginning of the process how the commercial offerings will differ from the free offerings philosophically. Communicate it and abide by it.

  4. Make it a complete solution: Whatever you open source needs to provide a complete solution in and of itself. It may not have all the bells and whistles, but it needs to get the job done well.

  5. Don't horde features: Arbitrarily pulling common features into a commercial edition will cause havoc with the community.

  6. Have up-sell opportunities: Given #4 above, make sure your commercial applications can add significant value to customers without denying them features from the core project.

  7. Be "open" with communication too: Stay in close contact with the community about the business as well: how you make money, partnerships, members of the team and more. Being tight-lipped about how you plan to make money doesn't foster trust and the community needs to understand how the project is funded.

  8. Get SaaS-y: It's a buzz-acronym, but software as a service has some merit -- specifically, tying performance of the application to the payment stream. In other words, being paid for doing a good job as opposed to licensing your software once and leaving. Whether you are responsible for hosting the applications or not, seek out ways to build accountability for performance with your commercial customers.

  9. Make support more appetizing: Support by itself is interesting to a certain percentage of customers, but finding ways to do proactive support (calling on customers periodically, sending updates that will affect them, helping them with performance, etc.) can expand the # of customers willing to purchase plans.

  10. Build on passion: If your engineers aren't passionate about it, don't do it. It will become abundantly clear to everyone that it's just another form of marketing and there will be no energy. No energy = no community. No community = no marketshare, profit or value.

Disclaimer: We don't do all of these....yet :).

Summary: OSS can change the way companies allocate $ towards more productive activities, but it is by no means a path to success. Without a massive commitment to being a "true" open source initiative (and the passion to make that happen), most of these projects will fall flat and cause more financial harm than good.

Yesterday marked a milestone for our growth as the temporary wall between our current office and our expanded office came down. The extra space is actually in the building next-door to us. The architects made a passageway into that building by opening one of our adjacent walls. Everyone here is super-excited to finally be in the same space again (some of us have been temporarily on a different floor). Not to mention having double the space we had before (bonus: cool exposed brick in the new space!).



Jive Software at JavaOne

Posted by bill Apr 26, 2006

[] [] This year, three of us will be making the trek to JavaOne in San Francisco. It's been three years since I was there so I'm definitely looking forward to it. The two pictures here are from the 2001 and 2003 shows.


We'd love to meet up with anyone interested. We'll be there starting Monday night and will go home Friday afternoon. If you're interested in getting in touch, please contact us and mention JavaOne. Or, please leave a comment on this entry and I'll get in touch.


I'm looking forward to hearing Josh Bloch's talks (always missed those in the past) as well as sitting in on some sessions about Mustang and Dolphin.


It's amazing how much we learn from our Customer Advisory Group conversations. One thing that was a bit of a shocker this go around was how many people not only don't have business metrics for their communities but that most of them "just set them up and let them do their thing." A few people even admitted to being "scared of their own community."


I guess its not a total surprise. I mean, how many companies out there do a good job engaging with their customers in order to really understand their needs? A handful? So its not a stretch to imagine that companies would avoid their community too.


That said, my opinion on this is changing as I continue to play with Googles Adwords Analytic tools.


[Googles Adwords Analytic tools|]


Theyre cool and fun to play with. The results quickly pull out important questions that shape our decisions. I now send them around since everyone asks to see them. The big "Aha" here is that it's no longer just a friend to Marketing. All of a sudden, the whole company cares about things that we didnt talk about before.


Between this and our conversations with customers, our eyes are very open to how we can improve our own products. Customers have said it's important to have this same level of "cool community buzz reporting." And our whole company is excited about the prospects. In fact, it's now one of the most prioritized projects going on in Engineering. Keep a lookout for some very cool stuff this summer.

The Power of Pubsub

Posted by matt Apr 18, 2006

In Wildfire2.6, we've introduced support for the Publish-Subscribe extension to XMPP (Jabber). A loose analogy is that it's like RSS on steroids, but for instant messaging. A slightly more technical take: it's a comprehensive system for publishing and consuming topic-based events.


Like RSS, pubsub offers a simple way to get notifications. But far beyond RSS, pubsub has rich publishing and permissions systems. As a more radical example than standard news syndication, a company could use pubsub to power a file sharing service; certain users would be allowed to publish files, while others could read them (with optional moderation). Users would be notified in real-time when files are added or modified, and could even filter notifications using keywords. Other IM twists to the pubsub protocol allow you to choose to only receive events when you're online, use your buddy list for permissions management, etc. The reason we're excited about pubsub is two-fold:

  1. It's much more comprehensive than the existing mainstream event protocols like RSS and Atom, which means you can do much cooler stuff. Of course, RSS and Atom should be thought of as complementary rather than competing technologies since they're for a different medium.

  2. If you believe like Jive and Google that an XMPP instant messaging client will be on every user's desktop, that means pubsub is a viable<span style="font-weight: bold"># platform</span># for building all sorts of services.

At this point, pubsub is just an interesting technology that remains to be proven. However, we'll be building some innovative services on top of it and I'm sure others will be too.


If you're interested in learning more about pubsub and how it can be applied, read our article on the topic.

Prediction Markets

Posted by djhersh Apr 13, 2006

While it's proven to be pretty good at predicting the payroll for Angelina Jolie on the Hollywood Stock Exchange, predictive markets are starting to generate some interesting buzz on the business side. Business 2.0 recently did an article on the Best Kept Secrets of the Best Companies and one of the pieces was on MSFT using it to predict release dates and #'s of bugs. Google has apparently been working on predictive markets as well, and will probably give the world something worth using. And the success of The Wisdom of Crowds has given the concept credibility.


It's worth it to read the Wikipedia entry, but as a quick background, prediction markets are essentially the equivalent of stock markets used for making predictions on outcomes. The theory is that people who buy at a low price are rewarded when that outcome goes up in value (i.e. is more likely), and that the collective wisdom is one of the most accurate methods of prediction.


We haven't spent much time looking into the applications, but I gather it would serve a lot of our clients really well -- faster, more accurate and more viral than surveys, predictive markets fit nicely into the community structure. Customers could invest in features that would make it into products, or even on how successful a product will be. Employees could invest in potential hires, growth plans, product features and more. As long as everyone shares the same goal, but has independent ideas on that goal, it would seem to be a no-brainer.


I'm obviously not an expert, but am interested to learn more from our customers about the use cases for their companies. I realize there are some companies out there with applications for managing these interactions, but we have never been asked about it, so I'm guessing that the company that makes it a) intuitive and b) viral, will be the clear winner.

How to know what people know

Posted by djhersh Apr 10, 2006

Expertise profiling comes up quite a bit with our customers. There is a lot of opportunity to streamline the painful email interactions (and over-the-cubicle-wall interactions) where people asking the same question many and/or don't know who to ask. Typically, our customers want to:

  1. Non-obtrusively route questions to the right people

  2. Notify relevant users of events in the system

  3. Quickly connect users with similar interests / skill sets

  4. Provide a level of confidence that an expert's responses are legitimate

To date, our ranking of expertise in Forumshas been based on what people say is their expertise and their points from answering questions (or providing assistance). Moving forward, there are a lot of opportunities to broaden the scope of expertise measurement.


To start with, calculating expertise can be done through implicit and explicit profiling. Explicit being the profile data that is actively managed by the user and/or company (e.g. "I am an expert in C# and a novice in Swing"). And implicit being the system's aggregate understanding of expertise based on the content and interactions provided by that user (e.g. Bill has 590 points in "Business Law" and 789 instances of "employment agreement").


How to Calculate


There are a variety of ways to calculate both explicit and implicit expertise, but the goal is to bring together the most relevant measures and provide an overall score that can serve as the basis for the intelligence described above (where to route questions, who is notified and when, etc.); however, the weighting of the different forms always depends on the use case.


The main areas are:

  1. Company Assessment: What does the sponsoring company say about the user's skill set (Explicit)?

  2. User Assessment: What does the user say about their own skill set (Explicit)?

  3. Certifications: What external accreditation does the user have (Explicit)?

  4. Interactions/Content Analysis: What does the system understand about the user's skill set based on interactions and content (Implicit)?

  5. Community Endorsement: What does the community say about the user's expertise (Explicit)?

  6. Points: How has the user scored in different areas of the community (Implicit)?



The most common use case is the routing of questions to people with the appropriate skill set based on the aggregate profile (a much more efficient mechanism for resolution than email). However, an understanding of expertise can be used in many ways, such as making social networking easier, using as the basis for incentives, improving personalization, etc. Right now, our system only deals with 2 and 6, but we are working on finding more advanced ways to calculate expertise as well as what to do with it when you know.

Awesome, I just came across this: Men's Journal: Portland best place to live. Here's a snip:

Broken down into a variety of categories to appeal to a broad range of interests, the "Best of the Best" picks on the list features winners for the perfect combination of adventure, attractiveness and affordability, with Portland No. 1, followed by Austin, Texas; Boulder, Colo.; San Diego; and Burlington, Vt.

We actually thought about moving the business to Boulder, CO a few years ago when we were based in NYC. Boulder is great but for a town of 90,000 people it's actually as expensive as Seattle. We opted for Portland based on that and a number of other reasons -- bigger city, affordable, good business environment, friendly people and low commuting times.

Filter Blog

By date: By tag: