Realize the Potential of the Third-Party Ecosystem
Q 1 What is your Recommendation?
We think there is a large third-party user base potential that is not currently realized. Successful open source projects tend to receive resources (money, staff and volunteer time, etc.) from a diverse set of entities, which have different goals but work together to improve the tool they all need. MediaWiki on the other hand is a monoculture, almost entirely dependent on Wikimedia, even though there seems to be a much wider demand for the functionality it provides. If there is, in fact, the potential for such an ecosystem, the movement entity stewarding development should identify and resolve chokepoints which prevent it from being realized.
(MediaWiki is used in the wide sense here, including all technologies built around it, such as Wikibase or Parsoid.)
Q 2-1 What assumptions are you making about the future context that led you to make this Recommendation?
Becoming the essential infrastructure of free knowledge is a vast undertaking and will require all the resources the movement can muster, and more. Improving our platform ties down a significant fraction of our budget, and even so the rate of progress is much slower than we’d like. If there is an opportunity to resource platform improvements in from external sources, we should take it.
Q 3-1 What will change because of the Recommendation?
Directly, an analysis of user needs and market opportunities of the MediaWiki ecosystem will be performed to identify investments that make that ecosystem more self-sustaining, and those investments will be made. (Alternatively, the analysis will demonstrate that no such investments are viable, putting an end to a long and at times acrimonious debate). Indirectly, the MediaWiki ecosystem will be more thriving and more self-sustaining, freeing up movement resources to be put towards work more specific to Wikimedia projects.
Q 3-2 Who specifically will be influenced by this recommendation?
Immediately, the recommendation would instruct some group (presumably the WMF with the support of external expertise) to perform the research described in the previous sections, and follow up on its recommendations. Ultimately, if that has the expected results, third-party MediaWiki users will be influenced in that using MediaWiki will become easier for them and their numbers will grow, and the WMF and other movement organizations will see a reduced need of investment into platform and product features where the use cases are shared with third parties, due to increased involvement from them.
Q 4-1 Could this Recommendation have a negative impact/change?
The assumption that there is a wide potential for MediaWiki and related technologies might be wrong, in which case any effort invested into unlocking this potential would be wasted.
If Wikimedia and other entities with a significant stake in the platform end up with diverging technical visions they cannot reconcile, that might result in conflicts (such as a project fork) and bad publicity.
Compromises made to keep the Wikimedia tech platform appealing to the widest range of interested parties might hurt its efficiency for the Wikimedia use cases.
Q 4-2 What could be done to mitigate this risk?
The theory that the MediaWiki ecosystem has significant third-party potential should be confirmed by research or small-scale experiments with early involvement of third parties, before any major investment is made.
The risk of fallout between major stakeholders is expected to be both unlikely and low-impact, as long as the trademark and core development infrastructure is owned by Wikimedia.
The risk of making the architecture too generic or too flexible will have to be weighed whenever we make design decisions, and also as a cost during the business and market potential analysis.
Q 5 How does this Recommendation relate to the current structural reality?
If it’s successful, it would add a new region to the current stakeholder structure, although one that’s not part of the Wikimedia movement in the traditional sense.
Q 6-1 Does this Recommendation connect or depend on another of your Recommendations? If yes, how?
Not directly, although it complements Decentralization in some ways: it’s a different way of distributing the development burden in a more resilient and sustainable manner, and both will require similar improvements in communication and governance.
Q 6-2 Does this Recommendation connect or relate to your Scoping Questions? If yes, how?
It is a (provisional) answer to the scoping question “What is the relationship between funding the Wikimedia Movement and funding the Mediawiki ecosystem?” In a way it also answers “Should software/tech be considered as a possible revenue-generating avenue?” although instead of directly generating revenue for the Wikimedia movement, the third-party ecosystem is seen as saving resources by sharing some of the burden of evolving the platform.
Q 7 How is this Recommendation connected to other WGs?
It is relevant to Revenue Streams - even though the third-party ecosystem is not a source of revenue to the movement (to some extent it could be, e.g. by exploiting the MediaWiki trademark, but that is not expected to be significant), it would free up resources currently tied down by the needs of the platform to be used elsewhere, so in many ways its analogous to a new resource stream.
It would also set us up for collaboration with other actors sharing a technology platform with us, which in the scope of Partnerships. It is especially close to their “Technical Partnership For Developments under a shared vision and resource collaboration” recommendation, except that recommendation takes a traditional and narrow view on what counts as a partnership.
Q 8 Do you have anything to add that was not covered with previous questions, yet essential for understanding the recommendation?
While a proper business analysis was beyond the capabilities of the working group, here are some quick facts on which we based our assessment of the third-party ecosystem having high potential:
- MediaWiki is one of the two market leaders in wiki software; the other one, Confluence, which has a comparable user base (tens of thousands) and feature set, has over $100M annual revenue.
- Wordpress, which has 500x the downloads and 1000x the install base of MediaWiki, has been estimated to have an $1B plugin market.