[sldev] [POLICY] Development by consensus (Re: Question regarding upcoming maintenance on 11/27-

Matthew Dowd matthew.dowd at hotmail.co.uk
Fri Nov 23 11:07:42 PST 2007

> There's also a lot of room between "open community development model"
> and "closed development model".  We're about as open as any market
> leading company out there.  Moreover, we've always said that opening up
> Second Life would be a gradual process.  Matthew's statement quoted
> above doesn't give us any credit for doing anything other than to
> "aspire" toward openness when we are far, far past that.  You all still
> wouldn't know a thing about this if we weren't more open company than most.

My original statement may have been blunt (it certainly had its intended aim of provoking a debate), but I do not believe it to be unfair (although on reflection it would have been more accurate to have said "*the* open community development model *which* it appears to aspire" rather than "*a* open community development model *as* it appears to aspire".

Whilst LL is more open than most, there are plenty of market leaders who having based their business models on an open source/community process from the outset are by necessity more open in their methodologies than LL.

LL shows many of the classical characteristics of a company who having released source code under an open source license, has realised that Open Source development is far more than that, and is still struggling with the thorny issue of how to integrate the open source development community with its own development process. The fact that there are two seperate JIRA's and and internal and external version control systems are good examples that LL has not achieved a completely open development model.

At the moment, I don't think it unfair to describe the current situation as having two fairly seperate groups - the internal developers and the external open source developers with some limited intermingling rather than a single development community involving both internal and external developers. Perhaps there is more of a mingling on IRC? but PJIRA, this list, the open source sections of the wiki etc. seem to be dominated by the external community with just a rew representative Lindens being visible (although we have been seeing more appearing on the e-mail list in recent months). All the public collaboration tools are mirrored by similar tools internally (we know there is a private jira, I think I'm pretty safe in guessing there are internal subversion, wikis and mailling lists!). I am not arguing that there will never be a need for internal/private discussion, issue tracking etc. just that the integration of the public and private discussion tools is still at a relatively early stage. A good example is the opacity of what happens after a PJIRA issue or patch is imported. Some issues are dealt with pretty efficiently. Others seem to stall for months with no visible progess or feedback!

The resultant situation is one where we do have two distinct communities rather than a single community, and whilst the information flow from the external community into the internal community, as many other have observed feedback, response and information flow the other way works less well. How much LL really listened to the isssues about the new authentication process, for example, is not clear. The impression given is that they have completely ignored the feedback and carried on regardless. The situation is not unsimilar to the chatterbox issue. The community gave a lot of feedback - much of it was (and remains) ignored. True - there is a judgement call how much was simply rejection of the new, however there was enough credible people arguing against the chatterbox approach for LL to take notice. There are now at least three third party viewers who have reverted to the older style, including the flagship third party viewer (Nicholaz), and a commercially licensed viewer (OnRez) whose launch via the CSI tie in was probably the largest publicity event for SL this half of 2007; this alone should be ring alarm bells at LL, as to how wrong they were not to listen to the external community on the issue.

Admittedly LL is a long way from a closed community development model - however it also has a long way to go to reach a true open community development model (e.g. having a single common version control system - albeit with some private areas for the non-redistributable stuff). It may be that where LL is, is perfectly appropriate to LL and it doesn't need to move any further towards full open community development. However, that is not reflected in the more aspirational of Philip Linden's statements about the future of LL and SL. I believe that where LL is now along the closed to open community development spectrum is not consistent with the vision of LL leading the development of an open, standards based 3D web. It is that mismatch between how open LL is currently compared to the openess required if LL is to reach the aspirations it expresses as regards LLs role in realising a true open 3D web, which is what my statement was really refering to.

I do think that LL needs some radically restructuring as we move to the Open Grid. The research lab model of LL may have been right to get SL to where it is now, but I'm not sure that it is right as we move into the future. There are now three activities that LL is trying to balance R&D/Innovation, Open Community Development of technologies and standards, and Virtual World hosting service provision (the latter representing the primary income source). These are very different activities requiring different skills, philosophies and methodologies. I wouldn't attempt to fault LL in the first of these, but whilst it has come some way in the other two areas, I do believe that it still has some work to do there. Philip, I think, admits as much about service provision in his latest blog, and Rob has admitted on this list that there are still things which need to be done better as regards community support. Whether he will take the risk of reworking LL (with its philosophy and skills) into the R&D/Innovation arm of a larger entity, and rebuild the service and open community arms from scratch with new skills/new philosophies, I'm not sure. True, doing so could be taking a high risk, but I also believe not doing so would be an equally if not higher risk!


The next generation of MSN Hotmail has arrived - Windows Live Hotmail

More information about the SLDev mailing list