Skype Technologies S.A. is failing its independent software developers.
What must Skype do to develop a successful developer community? How would you score them? How do you approach such a problem?
Developing an effective developer community requires more than just the desire, it require a model, something that will stretch the internal team and inspire developers. It must be so simple that the parties get it. So effective that key dialogs can start.
Eight key dimensions drive the success and behaviors necessary to nurture an effective developer community. This is not just about words. It is also having the right types of personalities and roles involved to make it happen. Too often a developer community is viewed with systems focus. I'd offer up that it is about people. People in all these roles
Judge for yourself: How does Skype's management of the Skype API and developer program score on each of these factors? Where could they improve? How would this map versus Microsoft or Java or...... Rate them "A" to "F" on each of these. Rate them today, then rate where you think they will be in six months. Are they on tract to be the ultimate partner for developers? What new ground must they break to get there?
Architecture Evangelism: Are the systems and documentation for developing your product on the Skype platform clear and comprehensive? Are short-term feature release timetables published? Are road maps disclosed and updated? Are contacts easy to find? Do you know who to talk to? Is access managed and measured? Is the developer education program diverse (accommodating many kinds of programmers), dispersed (geographically and across time zones), stepped (from beginners to gurus, from generalists to specialists), affordable, and comprehensive?
Score: D-
In Six Months? This is totally dependent on Lenn Pryor. C maybe.
Creative Opportunities: Does the API expose many features? Can they be combined to do novel and interesting things? Do they provoke innovative and competitive products and services? What unique opportunities does the API offer? Can solutions bridge APIs etc? Does the Skype developer program provide tools for experimenting and testing a developer's work in progress?
Score: B.
In Six Months? This will be a C- unless they expose more
User Experience: Does Skype help developers create "star" products and services? Toolkits? Is there effective brand synergy and marketing impact? Are third party tools seamlessly blended into the Skype user experience? Best practices: Apple's UI standards.
Score: F.
In Six Months? We'll see whether they become developer friendly C-.
Supportive Team: How effectively does the ecosystem work as a team, as a community? How free and productive is the exchange of ideas? How effective is Skype's communication and updates to the community? What is the opportunity for co-development with Skype? What are the risks of Skype obsoleting third party products through surprise changes to the API? When and how does Skype compete directly with developers and other partners? How well is Skype staffing to support the developer community? Best practices: Microsoft Developer Network.
Score: F.
In Six Months? Unlikely to see a roadmap in less than six months. Could still be an F although a frustrated and trying F.
Legal Agreements and Public Policy: Are contracts and deals between Skype and developers effective? Are they fair? Do they reflect the realities of how programmers develop software and how users use software? Is the legal language clear? Are accurate translations easily available? Are the license terms and conditions best-in-industry? How much does Skype protect developer rights and interests? How well does Skype protect developers by protecting end user privacy? How actively does Skype advocate to governments and industry for personal data privacy, the right to connect, and against hostile regulation?
Score: F.
In Six Months? Complete lack of action or general obstruction. This needs to be an A if Skype is to win.
Business Exchange: Working on Developer Time: Do you respect the ISV developer's time? How many minutes does it take for a developer to get a technical answer? to apply to the developer program?
Does the Skype application enable a two way information exchange? Is data flow through the API one way or two-way or even multi-way? How much is static vs dynamic? Exchanges with the client, exchange of information with the user? Security of information? Privacy management, user rights protection? Can developers build on information exchange to create commerce transactions?
Score F:
In Six Months? I truly wish for improvement. C.
Value Creation: How's the money managed? Integrated? Can payment to the Skype ID be made? Can withdrawals or payments be made? Who pays for services or products? Does Skype offer download "options" for certified software add-ons? Is there a river of monetary opportunity?
Score: D. Although other free services are creating value on Skype's back.
In Six Months? D expect little change.
Investors & Peers: As a developer can you sit at the table? Do you coexist like eBay resellers, integrated into the financial ecosystem? Is Skype a positive facilitator or are their always barriers? Are investors willing to put money in? Who bankrolls the opportunity? Are business cases relatively easy? Do you travel first class or coach?
Score: F:
In Six Months? D at best on current trajectory.
My conclusion.
Now many will say I am very harsh. You are probably right. Most developers would say "Skype is doing their best." "They are a young company." "They basically get it." What developers want is more access and functionality in the API. Communication is a big deal. They also don't want to be screwed by changing Skype road map. I gave the highest score to the API. Without it the only developers interested in Skype would be those applying for a job.
I thought long and hard about posting this blogpost. I did ask other Skype developers. I'm convinced now it has to be said. So this post is more directed to my friends (please don't shoot the messenger) at Skype. It may not be encouraging to get a "negative report card." Still that traces more to a poor attendance record (lack of people) for some of these "streams." Overall the API and developer focus can't afford to be tactical. For example, they focus on what to expose (e.g. voice messaging) rather than on broader strategic issues like how do independent developers make money, create collective value for users. etc. Perhaps thinking more broadly will enable the "score" to change rapidly.
So what score would you give to Skype's Developer Programs? Am I being too harsh? Do you want more systematic metrics? Let me know!
If Skype really want to encourage usage of the API, they could really help us by providing language bindings and examples for all the major development languages. As it is COM is provided by two third parties and Delphi by another. They used to have links on the development area to all these but now they've all disappeared.
It's important here that Skype spport not just people who've produced applications but also people producing tools to help build applications.
I think part of the challenge lies in the how developers view Skype verus how Skype views it themselves. I personally saw it not only as a great VOIP app, but also as a platform for building collaborative apps. I think Skype was, for the most part, more worried about developing their IP-to-telephony infrastructure, which is their main way of deriving revenue. This make plenty of business sense, but with that done, I think they're ready to move on to the next phase.
Skype has said they will be adding app-to-app connectivity via the API. This will allow developers to build apps using Skype as an innexpesive but relatively secure transport layer, without worrying about firewalls and such. Unlike products like Groove, Skype is cross-platform which opens up a lot of opportunities. For example, salesforce automation applications can sync with a server or each other without having to have a central server or webservice platform (.NET, J2EE, etc.). Is Skype as secure or robust a platform as a full-fledged client-server or webservice platform? Probably not. But for SMB, it is plenty good enough
One unfortunate trend seems to be Skype adding functionality without regard to what has already been developed. One example is voice mail. A few parties went to great trouble to build voicemail apps, only to have Skype include it as an included feature (albeit not for free). Likewise, Skype has committed to an API/web-based mechanism for retrieving on-line status, which a couple of parties have already worked hard on developing, only to see the rug pulled out from under them. Skype could really do a better job of laying out what the roadmap is for planned (as well as possible) enhancements to their product, so developers don't bump into them further down the line.
In general, I would agree with Stuart's premises that Skype must do better on relationships with developers, especially with regard to legal matters and making sure that developers can create novel and useful applications. This is vital to maintain the independent developer enthusiasm that will help differentiate Skype from other, inevitable, "me-too" products yet to come. To their credit, Skype has made more of an effort over the last few weeks to do just that.
I think part of the challenge lies in the how developers view Skype verus how Skype views it themselves. I personally saw it not only as a great VOIP app, but also as a platform for building collaborative apps. I think Skype was, for the most part, more worried about developing their IP-to-telephony infrastructure, which is their main way of deriving revenue. This make plenty of business sense, but with that done, I think they're ready to move on to the next phase.
Skype has said they will be adding app-to-app connectivity via the API. This will allow developers to build apps using Skype as an innexpesive but relatively secure transport layer, without worrying about firewalls and such. Unlike products like Groove, Skype is cross-platform which opens up a lot of opportunities. For example, salesforce automation applications can sync with a server or each other without having to have a central server or webservice platform (.NET, J2EE, etc.). Is Skype as secure or robust a platform as a full-fledged client-server or webservice platform? Probably not. But for SMB, it is plenty good enough
One unfortunate trend seems to be Skype adding functionality without regard to what has already been developed. One example is voice mail. A few parties went to great trouble to build voicemail apps, only to have Skype include it as an included feature (albeit not for free). Likewise, Skype has committed to an API/web-based mechanism for retrieving on-line status, which a couple of parties have already worked hard on developing, only to see the rug pulled out from under them. Skype could really do a better job of laying out what the roadmap is for planned (as well as possible) enhancements to their product, so developers don't bump into them further down the line.
In general, I would agree with Stuart's premises that Skype must do better on relationships with developers, especially with regard to legal matters and making sure that developers can create novel and useful applications. This is vital to maintain the independent developer enthusiasm that will help differentiate Skype from other, inevitable, "me-too" products yet to come. To their credit, Skype has made more of an effort over the last few weeks to do just that.
If Skype really want to encourage usage of the API, they could really help us by providing language bindings and examples for all the major development languages. As it is COM is provided by two third parties and Delphi by another. They used to have links on the development area to all these but now they've all disappeared.
It's important here that Skype spport not just people who've produced applications but also people producing tools to help build applications.