Talk:Wikimedia Foundation Annual Plan/2017-2018/Draft

From Meta, a Wikimedia project coordination wiki

Overview[edit]

Place comments about the Annual Plan Overview here!

Financials[edit]

Other expenses[edit]

Hi, The other expenses section is planned to raise from 1,882 thousand USD to 4,864 thousand USD. Can you please share what will lead to a growth of approx 3 million USD? Thanks Matanya (talk) 17:59, 8 April 2017 (UTC)[reply]

Above comment moved from Talk:Wikimedia Foundation Annual Plan/2017-2018/Draft/Financials (now redirected here) as I belatedly merged the talk pages for all AP pages. Kbrown (WMF) (talk) 15:11, 10 April 2017 (UTC) [reply]
Hi Matanya, The increase of ~$2,982k in program Other Expenses is a result of a few things. Firstly, two additional expenses were included under this budget - full fiscal year funding for WikiData and the funding for the annual Wikimania event. For FY16-17 only partial year funding was required for Wikidata and for FY17-18 funding for the full year has been budgeted. That represents a $746k increase. Secondly, we included $533k for the event expenses related to Wikimania. In FY16-17, there was not a Wikimania event.
To conform our internal reporting to our external reporting, for FY 17-18, certain expenses have been classified to both program and non-program activities according to a percentage split. This classification is in compliance with US financial reporting practices and was done in consultation with our auditors. In previous years, we did this classification for our external IRS reporting, but not in our budget. Including it in our budget for the first time this year allows us to better describe the actual costs of specific activities. This did not result in an overall increase in spending by the Foundation. The primary change in classification related to very specific organizational expenses, the largest of which was our rent expense. This led to $989k being reclassified from non-program Other Expenses to program Other Expenses. --JBaldwin (WMF) (talk) 23:26, 11 April 2017 (UTC)[reply]
Thank you for your detailed answer! Matanya (talk) 05:55, 12 April 2017 (UTC)[reply]
@JBaldwin (WMF): This looks more like 9 million USD in "other expenses", including around 3 million USD solely under finance and administration. The plan also has around 5 million USD for "Outside Contract Services" - without any sort of description of what those services are. That is about the combined budget of *all* of the Wikimedia affiliates! Please can you provide a useful breakdown of these other expenses and outside contract services in the plan? Thanks. Mike Peel (talk) 23:58, 2 May 2017 (UTC)[reply]
Hi @Mike Peel:, Yes, we can provide more detail about what makes us these expenses. We will post a few additional tables shortly that will describe what is in each the Outside Contract Services and Other expenses. JBaldwin (WMF) (talk) 00:31, 6 May 2017 (UTC)[reply]
Hi @Mike Peel:, we have added additional tables to the bottom of the Annual Plan Financials Page to provide more detail into what the Outside Contract Services and Other Expense categories represent. We will also be adding an additional table to breakdown FTE allocation in greater detail tomorrow morning. JBaldwin (WMF) (talk) 22:22, 11 May 2017 (UTC)[reply]
@Mike Peel: Update: we posted the FTE allocation tables this morning on the Staff and Contractors page JBaldwin (WMF) (talk) 21:24, 12 May 2017 (UTC)[reply]

staff expenses: program vs non-program[edit]

How do you allocate expenses for staff if they involved in both types of activities? I see reduction in non-program staff expenses along with growth of program staff expenses: is it a real reduction of personnel costs or just a specific treatment of accounting policy for allocation? rubin16 (talk) 18:37, 10 April 2017 (UTC)[reply]

Hi rubin16 In short, For FY 17-16, we have incorporated a process that allows staff to budget estimated percentage of their planned actual workload associated with the outcomes they are committed to produce. This allows us to budget for a portion of staff time for different activities, some of which may be program and some non-program. In the past, we did not have the systems to allow us to do this. We have built the required infrastructure as part of our work to be able to budget for and report on our cross-departmental programs. In the past we could only budget systematically 100% of staff time to either program or non-program activities. As a result, we could not ask the staff to break out their time between program and non program work. The attribution of time allocation, on its own, did not increase our costs, but rather allow us to better understand the effective cost associated with our program and non-program work. For example, our Legal team’s staff time was previously allocated 100% to non-program activities. In FY17-18, we have allocated percentages of appropriate staff member’s time toward their program related work (ie. Trademarks, Public Policy, etc). -JBaldwin (WMF) (talk) 23:26, 11 April 2017 (UTC)[reply]
Thanks for both Rubin16's question and James' answer. However, in this case most of the difference is already there in FY16-17 forecast: comparing current period forecast to current period budget (both of which use the old accounting system, if I understand correctly), program staff expenses is in line while non-program staff expenses is 22% lower. Comparing next period plan to current forecast, there is a 20% increase for programs and 8% increase for non-programs. From these numbers, most of the shift (between two thirds and three quarters, I'd say) is relative to the current term, and is therefore not explained solely by a change in accounting. - Laurentius (talk) 09:39, 29 April 2017 (UTC)[reply]
Yes, there is another factor that you are correctly identifying that explains the larger growth in program staffing - both in the FY16-17 forecast and FY17-18 plan. In FY16-17 we added several positions mid-year that were not originally budgeted, all of which were program staff. In FY17-18 the majority of the new positions were program staff, which led to a larger growth in program staffing expenses.
The positions added in FY16-17 were for the Structured Data program and the Community Health Program. The cost of those additional positions in FY16-17 was somewhat offset by underspends in program staffing due to staff vacancies. We also had staff vacancies in non-program staff. Those additional program positions plus the staff vacancies explains why our projection shows some increase in program staff and a decrease in non-program staff. JBaldwin (WMF) (talk) 23:21, 2 May 2017 (UTC)[reply]

board designated expenses[edit]

There was no such line of expense before if I am not mistaken. How these costs will be approved? Will they be supervised? rubin16 (talk) 18:37, 10 April 2017 (UTC)[reply]

Yes, you are correct that this is a new item in our budget this year. Adding to the information we have described - Board designated funds are very common per our external auditors, and are most often used to earmark funds for very specific purpose, generally non-recurring activities or investments. The designated activities and associated funds can be proposed by an organization, and if approved by the governing board, must be used for the specific designations proposed and will not be used to fund any other initiative. The designation serves as a means to communicate both internally and externally a desired level of assurance of the Foundation’s intention to support/fund a specific project(s). The process requires that transactions incurred against such specific designations are tracked, reported and audited to ensure compliance to the specific purpose that the board designate. We will also be tracking and reporting on those activities in our annual plans and related reports. There are no other tax or accounting implications associated with these type of funds. --JBaldwin (WMF) (talk) 23:26, 11 April 2017 (UTC)[reply]

Tables[edit]

In this Annual Plan five tables are included about expenses. The FDC format asks for two. One about programs and explicitly excluding staff. And another which summarizes total expenses. Table #1 in the Annual Plan has three parts: program, non-program and board designated. The program part includes staffing, contrary to what the FDC asks for. What is the rationality for including staffing here? Will the FDC allow other affilitates to include staffing in the table of program expenses as well? The total of annual operating expenses is 76,800. That excludes the board designated expenses, isn't it? Why are the board designated expenses included in the total of expenses? Isn't the grand total 78,800? The collective revenue target is 81,800, of which 76,800 to cover expenses and 5,000 for the endowment. That is less than the grand total. Will the board designated expenses be covered from the reserves? Ad Huikeshoven (talk) 11:02, 17 April 2017 (UTC)[reply]

Hi @Ad Huikeshoven: Yes, we did present the financial tables in several different formats, including separating out all staffing related expenses by department. We want to help readers understand the expenses from several meaningful angles. We chose to also include the staffing expenses within program expenses and non-program expenses in Table #1 to be consistent with the way that we presented financial information in the current year annual plan and mid year report.
To address your other question, yes, the board designated expenses will be funded from the foundation’s surplus (reserves). We identified these activities because they are one-time, non recurring expenses. So, yes, the board designated expenses are not included in the $76.8M operating expenses.
JBaldwin (WMF) (talk) 22:30, 26 April 2017 (UTC)[reply]

Statement of Financial Position & Statement of Cash Flows[edit]

i see you are concentrating on expenses and revenue. are you going to report a Statement of Financial Position (balance sheet), and a Statement of Cash Flows? and example would be here [1] will you be including an endowment investment statement? Slowking4 (talk) 23:31, 19 April 2017 (UTC)[reply]

Hi Slowking4 - The WIkimedia Endowment Board passed an investment policy in January 2017. --Lgruwell-WMF (talk) 20:36, 27 April 2017 (UTC)[reply]
Hi @Slowking4:, yes, we will be reporting our Statement of Financial Position and Statement of Cash Flows on our audited financial statements which are published each year after our audit. Here is our 2015-16 Audited Financial Statements. JBaldwin (WMF) (talk) 20:53, 27 April 2017 (UTC)[reply]

Table 5 footnotes[edit]

From table 5 notes: «FY16-17 Forecast includes [...] $2M Movement Strategy Expenses in Executive Director». There is no "Executive Director" line: do you mean Governance? - Laurentius (talk) 09:45, 29 April 2017 (UTC)[reply]

Yes, thank you for catching that! We made the correction so that it now reads "Governance". JBaldwin (WMF) (talk) 23:21, 2 May 2017 (UTC)[reply]

Office move[edit]

There isn't much description of the "Office Move" budget of USD697k. Presumably this is all one-off costs of the refurbishment of the new offices, and the relocation of the existing furniture/equipment/etc., or does it also include recurring costs such as increased rent? Thanks. Mike Peel (talk) 00:06, 3 May 2017 (UTC)[reply]

@Mike Peel: The $697K is a preliminary estimate for the Office Move and established a board designation for these expenditures. The estimate was based on the information we had available at the time of preparing the Annual Plan. Firstly, it does not include rent expense - this expense has been budgeted under the rent general ledger account. Secondarily, we would like to clarify that with or without an office move, we would be incurring incremental, one-time expenditure spending because the other choice was to consolidate from 3 floors to 2 floors at 149 New Montgomery. The expenditures for the Office move fall under construction and infrastructure (e.g., cabling, network, hardware and systems) costs, furniture and relocation costs. We also plan to move a large amount of our existing furniture and technology hardware to the new office, which may reduce our cost estimate. We are estimating approximately $450K in time costs for internal technology, again for, cabling, network, systems and other hardware. We are estimating approximately $100K for facilities costs, including furniture purchases and services required for site development. Lastly, we are estimating $145K for other real estate support and relocation services to facilitate the move. JBaldwin (WMF) (talk) 20:56, 10 May 2017 (UTC)[reply]

Cross-Programs[edit]

The annual plan says that "we have or will be disclosing detailed budgets for each of these, showing the multi-team departments – Community Engagement, Advancement, Technology and Product – sub-department or team budget detail." When are we likely to get these? Bishdatta (talk) 22:29, 9 May 2017 (UTC)[reply]

Hi @Bishdatta: Sorry about the delay. Our tables with additional detail on the Cross-Departmental Programs were finished today and will be added to the annual plan tomorrow. We will be providing short descriptions of what is included in each of the major categories (Outside Contract Services, Other Expenses, etc) for each of the programs. These will be in table format and added to each programs' page of the Annual Plan. Please let us know if there are any specific follow up questions we can help with. JBaldwin (WMF) (talk) 22:50, 9 May 2017 (UTC)[reply]

New Readers[edit]

How will this project work with existing affiliates in the regions it is active in? Thanks. Mike Peel (talk) 22:49, 9 May 2017 (UTC)[reply]

Hi Mike Peel. Thanks for your question. We think it's really important to work with affiliates and volunteers wherever we are working, especially with country-specific projects. Starting with the research in the previous fiscal year, we've been in contact and hope to continue to work in partnership with local communities. One good example for how we're looking at this is in localized awareness campaigns. These are conducted with a local "marketing committee" of volunteers and based off of a country-wide survey of volunteers and engaged readers. See the awareness campaign in Nigeria for a good example of this approach. We're also working on how to systematically think about working with communities and consultations, and will be iterating how we think about this as we go. AGomez (WMF) (talk) 18:40, 10 May 2017 (UTC)[reply]

Wikimedia Brand and Identity[edit]

What our your initial thoughts and hypothesis for this program? The Wikimedia Foundation owns the Wikipedia brand, word mark and logo. The Wikipedia brand is very valuable. The general public probably won't recognize any other brand, wordmark or logo of the Wikimedia Foundation or Wikimedia movement, except Wikipedia. Is the plan to consolidate on a single brand, name, wordmark and logo, Wikipedia? Rename the Wikimedia Foundation to Wikipedia Foundation? Ad Huikeshoven (talk) 11:02, 17 April 2017 (UTC)[reply]

Hello @Ad Huikeshoven: first, my apologies for a slow response, it's nice to hear from you. I appreciate your question and the answer is important. The purpose of the brand and identity project is to explore exactly the issues you mention. Our first steps are to understand the positions of our brands and identities within the existing environment. This includes some research we've already done, some research and findings from the movement strategy process, and any other work we determine is needed for this project. The next step will likely be to determine what we need to reach our goals. There is no preconceived plan to consolidate. While we each have personal hypotheses, and opinions of what problems we have when it comes to Wikimedia brands, the most interesting purpose of this project is to surface the biggest opportunities we have and to develop a plan together. The team at the Foundation is interested in reviewing our brand architecture to best support the projects, affiliates, and individuals, and help us all achieve our shared vision of sharing all knowledge with every person. Heather Walls (WMF) (talk) 05:31, 4 May 2017 (UTC)[reply]

Sorry for the late question. Is there a reason why there is no mention of hiring an external contractor to do this work? Also, why have the three "key markets" not already been identified and included in this plan? In the "Trademarks" section, is this not an iteration of the current activities of the Legal Department? Thanks, Risker (talk) 04:52, 8 May 2017 (UTC)[reply]

@Risker: Sorry about the slow response. There is no particular reason for no mention of hiring an external contractor aside from accidental omission. I appreciate the patience of the FDC in helping us uncover where we haven't been clear. The three key markets are the markets already described in the New Readers plan. There are overlapping contexts but work that falls into different programs. The Trademarks section is both an iteration of the legal team's current work, and changes in that work due to new information and scope but not increased budget. You can find a longer, more contextual reply on the FDC response page. Same as my reply to Mike, we could arrange a call with our team if that would help. Thank you, Heather Walls (WMF) (talk) 03:44, 26 May 2017 (UTC)[reply]
I have been thinking that we also need to working on branding regarding how we describe our communities. In the past some branding has had negative connotations and this has likely negatively affects recruitment of editors and collaborators.
For example we should be describing those within our communities as:
  • Yes: Smart, intelligent, hardworking, curious, highly educated, professional, diligent, interesting, cool, altruistic, caring, etc
  • Not: Geeks, nerds, isolated, techies, kids, amateurs
This is IMO just as important as issues regarding branding the sites we work on. Doc James (talk · contribs · email) 02:24, 10 May 2017 (UTC)[reply]
@Doc James: that is very true! And it's something that we work on almost daily when we interact with journalists and media to clarify what people know and learn about the projects and communities. Your yes list is the way we talk about everyone, because it's true. When it comes to anything negative, we also want to be truthful, but it's about behavior and things we can work on together. I appreciate you bringing it up as an important point. Heather Walls (WMF) (talk) 03:44, 26 May 2017 (UTC)[reply]

Brand defense[edit]

The plan still lacks an indication of whether the money spend on brands/trademarks will be used in some way focused on community needs, for instance to address what the community has indicated as top priorities for legal action. --Nemo 09:55, 18 April 2017 (UTC)[reply]

Hi @Nemo bis: please also accept my apologies for a slow response. The community is a pretty broad audience — as you know, we have many communities with an array of needs. We hope that focus from the movement strategy process will uncover our shared priorities for both identity and brand protection. In my opinion, our brands are inextricably linked to the work and dedication of community members. We will be focused on outcomes developed from new and existing discussions, with and among community. The methods and strategies to achieve those outcomes can only be understood as next steps. Heather Walls (WMF) (talk) 05:31, 4 May 2017 (UTC)[reply]
Thanks for answering, but I would have expected an answer from the legal department. Are legal priorities for brand defense decided by the Communications department?
It feels unlikely to me that the general strategy could go into such details as which kind of legal defense to have for brands. Sure, we have many communities, so it would be rather easy to decide that each community defines the priority for the defense of "their own" brand. --Nemo 07:50, 2 June 2017 (UTC)[reply]

Partner brand usage[edit]

@Heather (WMF): How does this connect to the use of Wikimedia brands by affiliate organisations, and also the CC-BY-SA licensing of the Wikimedia project content? Thanks. Mike Peel (talk) 23:04, 9 May 2017 (UTC)[reply]

@Mike Peel: As far as I know this is not related to marks used by affiliate organizations, and does not affect the licensing of content. Does that answer your question? I may be misunderstanding it. Heather Walls (WMF) (talk) 05:49, 17 May 2017 (UTC)[reply]

Budget[edit]

@Heather (WMF): It seems that all of the budget for this activity is outsourced, although it's not clear to whom. A lot of this work (particularly for trademarks) has previously been done by the legal department - is that no longer the case? Or is there still legal staff time allocated to this work? Thanks. Mike Peel (talk) 23:18, 9 May 2017 (UTC)[reply]

@Mike Peel: I'm not sure what you mean by all of the budget being outsourced. By nature of program work, at least in Communications, the budget is for materials and outsourcing. There is work to be done by individuals on the team, but not enough to account for it separately. Trademark work is still handled by the legal department. heather walls (talk) 05:42, 17 May 2017 (UTC)[reply]
@Heatherawalls: Thanks for the reply. The "Wikimedia Brand and Identity" column in Table 2 shows $850k in the "Outside Contract Services" row, and nothing else. (Although in the budget table in the section, it's under "Legal Fees", which I presume is a typo). In the text, there is no indication that the work will be done internally or externally. This creates a strong impression that all of this work is being outsourced, not done in-house. Thanks. Mike Peel (talk) 06:44, 17 May 2017 (UTC)[reply]
@Mike Peel: Sorry again about the slow response, I wanted to make sure we had a complete one for the FDC's recommendation. The budget shown here does not include the traditional trademark work of the legal team. It is true that the budget is intended for expertise that is external to the Foundation. You can find a longer, more contextual reply on the FDC response page. With our small team here at the Foundation we will work with experts in their field and do the work of a client in those circumstances. That includes finding the right external consultants, explaining our history, integrating efforts with the communities and movement strategy, etc. Am I answering your question? We could arrange a call with our team if that would help.

We have deliberately budgeted this work as external services, as the Foundation does not have the full-time staff to support the scope of this work. We have also found through experience that external parties are often able to bring a wide range of comparative experiences, healthy perspective and remove, a user-centric process, and additional capacity and expertise. The three primary external costs covered under this budget are: Costs for contractors and agencies to redesign wikimediafoundation.org, a key fundraising property; Contractors, agencies, and/or consulting firms to facilitate several phases of a brand and identity project including: brand audit, architecture, story, recommendations and elements; Project management to facilitate the program

Thank you, Heather Walls (WMF) (talk) 03:19, 26 May 2017 (UTC)[reply]
When you say "redesign wikimediafoundation.org", will you use contractors with experience in using MediaWiki? Much of the recent work tended to focus on unviable piles of HTML which don't integrate well with our wiki community. Nemo 07:52, 2 June 2017 (UTC)[reply]

Community Health[edit]

Isn't there a positive and constructive program (rather than defensive) regarding community health as well. For example community building and or bonding of community leaders? Hans't the strategy process be designed in a way as a team building exercise for community leaders? Ad Huikeshoven (talk) 11:02, 17 April 2017 (UTC)[reply]

Hello, Ad Huikeshoven. While the "Community Health" program name is pretty broad, it isn't the only program related to improving community health in the roster as well as in the Foundation's regular core workflows, which are not singled out. :) For an example of time-based programs, there's "Gender diversity," "Community Capacity Development" (which last year included Conflict Management among its offerings), the bonding of community leaders that takes place through "Community leadership development & mentoring", the peer-to-peer mentorship of "Learning infrastructures: Wikimedia Resource Center" and the furtherance of "Training Modules". Community Health is a pretty pervasive theme. --Patrick Earley (WMF) (talk) 20:49, 18 April 2017 (UTC)[reply]

Targets[edit]

In segment 1 (Anti-harassment tools), targets for milestones are not set, and there is a "(source)" without a link that seems to be a mistake. Do you have targets, or when do you plan to publish them? For milestone 1, how do you measure whether a comment is a personal attack? - Laurentius (talk) 09:45, 30 April 2017 (UTC)[reply]

Laurentius, great question, and great catch! The (source) links were artifacts from copying this segment from a word document to the wiki page. The milestones originally in that segment were aspirational and we were still discussing them as the time of this page’s publishing. We’ve updated them to be more realistic and obtainable:
  • Milestone 1: Increase the confidence of admins with their ability to make accurate decisions in conduct disputes, measured via focus groups or consultations.
  • Milestone 2: Release three features which empower Wikimedia administrators to better enforce community policies and reduce abusive activity.
For the now-outdated specific milestone of “Decrease percentage of personal attack comments on English-language Wikipedia” we were considering using Detox to measure whether a talk page edit contained a personal attack or not. We may explore this more in the future but for FY17-18 we will not be using Detox to measure our work’s success. — Trevor Bolliger, WMF Product Manager 🗨 17:34, 11 May 2017 (UTC)[reply]

Communication[edit]

I dunno where to put this, this part here seems to be quite OK.
Invest far more in the translation of your communication with your bosses, the communities. Up to now most messages are simply dumped somewhere in non-english project in english, and let those communities somehow deal with this, the WMF obviously doesn't care. There are two different aspects here: One more for processes: Give the translators enough time to do it, thus plan properly in advance before sending out a message, and never even think about sending a message (that ain't severe emergency) in less then 25 different languages. If time is not there, use professional translators instead of the normal volunteers. And that's why I put this here: There has to be some budget for this, and as communication with your superiors, the communities, should be extreme top priority, there should never be not enough money for this, considering the lush coffers of millions of dollars the WMF manages on behalf of the community. One good example of what should never ever happen again is mentioned here: The election notification about the elections of the highes body in the movement was only sent in english, with no chance for volunteer translators to translate it (far too late ready for translation) and no efforts whatsoever were made to make them readable for the whole movement by other means. It was a clear sign of negligence towards the communities by the WMF. Grüße vom Sänger ♫(Reden) 16:27, 4 May 2017 (UTC)[reply]

Hi Sänger. While translations aren’t the focus of this team, I agree that successful communication with a global group of people hinges on timely and high quality messaging in local languages. In my experience, the difficulty comes from the extremely high cost of translations - both the labour cost of volunteer work, and the very high monetary cost of commercial translating services. We have also had issues with professional translators not understanding the context of our communities and thus getting it wrong - community have voiced concerns about some paid translations in the past. Patrick Earley (WMF) (talk) 21:27, 11 May 2017 (UTC)[reply]
Feel free to move this to anywhere it fits, I don't know the precise internal structure of the WMF for such stuff. This time it was simply impossible to get the message out in a constructive way, so it was sent just in the arrogant read-English-you-buggers-way, because it was a) far too late for volunteer translations and b) no attempt was made for a professional translated version. That's simply no proper way of communication for an international project, where English is just one of many languages. I'm always tempted to write such stuff as this here just in German or Dutch, so that the English get the feeling about getting treated the way they treat others constantly.
Regarding the very high cost: How much of the multi-million-dollar-budget would such proper communication cost? Grüße vom Sänger ♫(Reden) 06:59, 12 May 2017 (UTC)[reply]
Sänger, how to caluculate those costs is a valid question. I think the first exercise would be to get a framework for deciding which parts of Wikimedia Foundation communications and documentation to prioritize. And, of course, which languages to prioritize - which is not an easy discussion. As you are aware, betwen announcements, documentation of projects and teams, resources, blog posts, etc., the volume of text is high, and the costs of professionally translating all of this into all langauges would be astronomical. I do think this is an important issue to explore. But, in this context (annual plan discussions around the specific program on anti-harassment), I'm afraid I can't offer more solutions. Patrick Earley (WMF) (talk) 21:18, 12 May 2017 (UTC)[reply]

Other topics?[edit]

Community health is about far more than harassment and the gender gap alone (although those are both big issues). Will the other issues (e.g., encouragement, enthusiasm, stress, workload) also be tackled by this project, or can the name for this project be revised? Thanks. Mike Peel (talk) 23:24, 9 May 2017 (UTC)[reply]

This is true, Mike. We are tackling two of the issues certainly not at the exclusion of some of the other problems that affect our communities. We see these two initiatives as part of community health, not the whole of it. This is a health initiative, not the health initiative. Our teams, due to size and resources, must of course focus on specific areas to see progress. But we in no way want to imply that these are the only problems that need attention, and we’d welcome ideas on how the Foundation should be tackling those other important areas you mention. (My reply to Ad above outlines some of the additional health-related work being done at present). Patrick Earley (WMF) (talk) 17:17, 11 May 2017 (UTC)[reply]

social interaction modeling research[edit]

Please can you provide a link to a summary of this research / the research plans? Thanks. Mike Peel (talk) 23:27, 9 May 2017 (UTC)[reply]

Hi Mike. Thanks for your question. I believe you’re referring to the objective “Based on social interaction modeling research, build a prototype to surface potential cases of wikihounding to admins” under Anti-harassment tools, correct? To answer your question: we don’t yet have a link to this summary as it does not yet exist. We know wikihounding (or stalking, to use a general term) is an important area of focus for the Anti-harassment tools segment but we’re still looking to hire the researcher responsible for this work.
We believe we’ll be able to evaluate known past cases of persistent unwanted contact to identify common patterns that could be used to identify potential new incidents. We plan to do this research on a large scale qualitative level (thus its inclusion in Segment 3: Research) but the exact scope depends on some initial planning that has yet to occur. — Trevor Bolliger, WMF Product Manager 🗨 21:10, 11 May 2017 (UTC)[reply]

Privacy, Security, and Data Management[edit]

Another late question, sorry! I see that ownership of the onboarding/offboarding (and presumably the changes in acces) are in the domain of Technology. It is good to see clearcut ownership of this responsibility. Will this include clear guidelines that specify exactly what work/volunteer assignments will lead to granting of access to (for example) checkuser-type data, suppressed revisions, or other non-public personal information of readers/editors/donors - or access to administrator/bureaucrat-like suites of permissions (including blocking, deletion, assignment of user permissions)? Will specific staff roles be assigned to the ongoing maintenance of specialized software that enables volunteers or others to access information that is covered under the current privacy policy (e.g., CheckUser extension)? Risker (talk) 05:34, 8 May 2017 (UTC)[reply]

Structured Data[edit]

Place comments about this section here!

Movement Strategy[edit]

Place comments about this section here!

Department Programs[edit]

Community Engagement[edit]

Hi CE department. I've got a series of questions across the range of programs. I've tried to keep them organise by the "program number" as listed in the CE section of the plan:

Hi Liam, Thanks for closely reviewing our annual plan for next year! I’m happy to answer your questions and will do so below. KLove (WMF) (talk) 04:44, 12 May 2017 (UTC)[reply]

Program 1: Resourcing communities

  • Can you explain what the milestones/targets mean?
When you say that "50,000 participants engaged through WMF grants” - what qualified as “engaged”? At the most stringent this could mean “receiving a grant personally” and at the most loose this could mean “receive access to new content which was created as part of activities which received a wmf grant”.
Does a person attending a conference [which received a WMF grant]' count as engaged? Does ‘a person retweeting a blogpost about that conference’ count as engaged? Does ‘the monthly pageview numbers of WP articles created as a result of a [WMF grant-funded editathon]’ count as the number of people engaged? Given your target is 50,000 'participants engaged' I would expect that it’s impossible to have 50,000 individual grantees in a year - but that it would be super-easy to have 50,000 views of a single Featured Article written as a result of a single small grant.
This metric will be an aggregation of the “Total participants” grant metric reported across all grants funded in the fiscal year. We have guided grantees to /not/ include people who are reached by mass communication (such as social media followers), but people attending a conference that was funded by a WMF grant would be included. I wouldn’t expect pageviews of articles to be included, as those who used pageviews as metric usually report that separately (and when they do they tend to be extremely high, in the hundreds of millions!).
  • When you say that there will be "70% of grants awarded to emerging communities” does this mean 70% of the value of all the grants given, or 70% of the grantees co-signing applications, or 70% of the grants awarded regardless of their value?
What we mean is 70% of the number of total grants awarded. As you probably know, we have been tracking this for a while now, and you can see our historical results by quarter (and by year--fiscal year that is!). We’re very excited to be officially using this information as a target going forward. We believe we have great potential to increase this number and contribute significantly to growth of the movement in new communities in global south/emerging communities.
  • When you say that milestone 3 is "Over 500K Wikimedia pages will be created or improved”, this appears to have the same difficulty as the Global Metric about ‘pages created/improved’ had, aggregating diverse activities from different wikiprojects under one 'catch-all' number. One bot could add a statement to that many thousands of Wikidata items, or batch upload that many files to Commons. This also devalues any grant which goes towards things other than new content creation. Are you sure that capturing this metric is more help than hinderance to your work?
You make a good point that this catch-all number obfuscates the work being done on quality, and on the many Wikimedia projects. Right now, the way we see this metric is: it's not great, but it's good enough until better tools are available to easily capture more detailed data. As we noted when we shifted from Global Metrics to grant metrics, there weren’t readily available, easy to use tools that would break down “pages created or improved” by Wikimedia project, and also separate content creation and improvement. While content creation, curation and improvement metrics were important and useful to many grantees, this disaggregation would have fallen on the grantee to pull manually. We decided against asking grantees to complete such a laborious task, and that meant we stuck with the catch-all number, until better tools can we created. And I’m happy to say that in partnership with Community Tech, we are now working on scoping such a tool. Broader community input and review will be forthcoming in the next month or two. Stay tuned!
  • The milestone 4 “leadership” has a series of statements about diversity - relating to which community they come from, gender, whether they become involved in other activities after their grant is concluded... These are all clever, nuanced targets - but there are no description of how/if these things would actually be measured/captured. Are these targets for this year, or aspirations for what would be nice to have in the future?
Glad to see you are asking us questions about this, Liam! Yes, seems this last target/milestones has pulled together many threads, and it certainly could have been more clear and SMART-er. Highlighting “leadership” as an outcome of grants is a new area for us. Over the last few years, we've been noting what seems to be a trend about leadership development as an outcome. We have seen and heard about this only anecdotally, and it is not something we have systematically tried to track or measure ever before. Instead of setting a quantitative target such as “X number of grantees take on leadership roles in the movement as a result of their work”, we will explore (qualitatively) a few simple questions: To what extent do we see grantees take on leadership roles in the movement after their grant? For those who do take on leadership roles, why is that? Is this leadership development related to their grant-related work, or something else? We expect to produce a few mini-case studies to understand this a bit better. The focus of this exercise will be to understand what’s happening at the interface of our “grants program” and the larger Wikimedia world. This work will help us begin to build our baseline understanding of what is happening with “grants” and “leadership”, which will hopefully lead us to measuring something useful and worthwhile in future years.
Finally, diversity (gender, emerging communities) and grantee satisfaction will remain metrics that we track internally and externally, but these will not necessarily be metrics we would use to measure “leadership”. Again, I see this got a bit confused in the narrative. Thanks for calling it out and asking for clarification. KLove (WMF) (talk) 05:16, 12 May 2017 (UTC)[reply]

Program 4: Community leadership development & mentoring

  • Does "learning and collaboration events”, "special community convenings”, "peer leadership academy”, "Learning Day Events”, " pre-conference events”, "convenings to which the trainings are attached”, all refer to the same thing? That is: training workshops held in the days before other wikimedia conferences?
Thank you, User:Wittylama, happy to clarify. Leadership development and mentoring activities indeed have different routes of implementation; while some of these are synonymous, others are not. Specifically, there are three things we provide, two are continuations of our existing program and one will be a new effort related to the development of the Wikimedia Resource Center.
Importantly, “Learning Day Events” are preconferences and stand-alone workshop events by request that we work to document materials and best practices for so that key content and workshops may be replicated at any community convening.
We do this in two model ways:
  1. We organize and support two Learning Days pre-conferences; one at Wikimedia Conference, and one at Wikimania. For these pre-confereces we work iteratively to co-design programs and content working to cultivate content leadership among regional and thematic community mentors.
  2. We provide limited staff support for responsive workshop design and implementation at other regional or thematic conference events. As requested, and appropriate to the needs, we regularly partner with local Wikimedians preparing conference content for needs related to learning and evaluation. Here we may help from a distance, or may send a member of the team, to partner with regional or thematic community mentors in facilitating and presenting this content at their local or thematic gatherings.
In 2017/2018 we will continue to document and organize resources to better support community mentors in these ways by delivering key learning content at events in these two ways.
In addition, in 2017/2018 we will begin to explore better ways to increase the transparency of, and accessibility to, development opportunities throughout the movement by building peer academy space(s) to help in connecting community mentors and facilitators to one another, and to their shared learning, through further development of the Wikimedia Resource Center. Specifically, we are working to build space(s) for integrating and curating community created resources and in 2017/2018 we will encourage community leaders to populate that space with their shared guidance and resources for community engagement programs and initiatives. JAnstee (WMF) (talk) 17:10, 10 May 2017 (UTC)[reply]
  • Can you explain what Milestone 1 means? Does it mean that your target is to have a pre-conference ‘learning day’ at 80% of meetings/conferences which receive WMF grant funding?
Thanks again, for asking for clarification, User:Wittylama. No, this is not a target to increase the number of Learning Day events, but rather a target to explicitly monitor the ratio of presenters from avoiding WMF-heavy and encouraging community leadership. Here the target is to ensure that 80% or more of Learning Day events (e.g., our two annual pre-conferences as well as any responsive workshops at existing movement events) have community mentors leading as presenters or as peer collaborators and facilitators in session design and delivery. JAnstee (WMF) (talk) 17:10, 10 May 2017 (UTC)[reply]


Program 5: Learning Products. Community engagement Insights

  • For this survey, do you intend to allow non-WMF groups to submit questions and, pending review according to the design criteria you’re preparing, be included in the final product? This would allow affiliates to be involved, and not have to run their own, ‘competing’ surveys to gather probably-similar data.
Hi User:Wittylama - Yes - this is exactly what we would like to see happen at some point, but I'm not sure if it will start next year. The foundation benefits from having the knowledge and staff to run a very complex survey like this one, so we should also allow others to gain access to asking questions.We would like to design a proposal process not just from affiliates, but also from volunteers on the projects. This process, however will take some time. It will be likely that we need to form a committee to help us make decisions on which questions to include and which to not. But for now, we are working to make sure we have a system in place to collect, analyze, and report the results in a timely manner, and on an annual basis. Let me know if this answers your question. Thanks! --EGalvez (WMF) (talk) 20:42, 9 May 2017 (UTC)[reply]
  • Relatedly, Will you be able differentiate [some of] the questions provided to different slices of the community (especially by language and/or country)? This would facilitate the reduction in number competing surveys from different groups if they could all be supported in the one action.
Hi User:Wittylama - Yes - we will be doing this, either by request or we will publish it outright - depends on how much time it will take. Let me know if this answers your question. Thanks! --EGalvez (WMF) (talk) 20:42, 9 May 2017 (UTC)[reply]
Also - we will likely work to find a way to make the data open access OR allow affiliates to sign an NDA so they can access the data the data for themselves. Thanks! --EGalvez (WMF) (talk) 18:00, 10 May 2017 (UTC)[reply]


Program 6: learning infrastructures

  • If this portal on Meta is intended to be the ‘one place’ for all resources, will you also be investing time/effort in migrating the existing content across from the Outreach wiki?
Hi Liam. Thank you for bringing up this question. It has come up in conversations about the Wikimedia Resource Center (WRC) with other community members. Outreach Wikimedia is a project that has a community of contributors, and we would need to have this conversation with them. This question was already brought up in a consultation[1], with a negative response from the project’s community, which makes it especially important to have a dedicated conversation again. I believe Alex Stinson and Nichole Saad, who work to coordinate resources for GLAM and Education respectively, may have some thoughts to share here as well.
One final thought regarding the issue of having multiple platforms with resources. Outreach Wikimedia is not the only external portal. The WRC also links to resources on mediawiki.org, which focus mainly on volunteer developers. For both cases where we might be driving users outside of Meta Wikimedia, we are working on templates that can bring them back to the WRC. María (WMF) (talk) 23:30, 11 May 2017 (UTC)[reply]
Hi @Wittylama: I want to further elaborate on this a bit from the perspective of the community programs team: both the Education Team and myself in the GLAM-Wiki role, are in the process of examining and revising the communication and documentation tactics, in consultation and collaboration with community members and affiliates, for getting clear and accessible resources in front of our community leaders. One of the options on the table is definitely moving from Outreach to Meta: but we think that we need to do several things first: review the method for that kind of migration and make sure we have the resources to do it efficiently; do a community consultation to confirm interest/support in such a move and confirm that that opinion has changed since 2015; and find a window of time where that won't be disruptive. However, in my mind: the pageview statistics for outreach in comparison to Meta and the extant documentation on other wikis, paints a pretty clear picture: we aren't reaching the audiences that we want to with outreach --so its time to change tactics there. Astinson (WMF) (talk) 16:09, 12 May 2017 (UTC)[reply]

Program 7: Community Programs Infrastructure

  • Will the ‘library card’ signup platform indicate which resources are already provided for people in different jurisdictions by their own local/university/national libraries - to work with those library partners rather than compete with them?
    One of the goals of TWL has always been to lead editors and readers to full texts and information professionals at their local libraries. It has always been and remains a condition of the Library Card platform that editors not sign up if they already have free local/geographic/institutional access (https://wikipedialibrary.wmflabs.org/about/), but enforcing or even just informing this practice is quite difficult. There is no publicly or commercially available index of what every person in every region has access to. We worked for 2 years with OCLC on a proxy API that may have helped in this case, but it was dropped out of production. The task is highly non-trivial. We would need to take the user's ip address upon signup then search a massive (and, as of yet, not compiled) index which knew that a source they signed up for was already available for that ip or ip range, and then direct them to the appropriate library access point. This is technically doable but the data aggregation is outside of our means and even of our more technically sophisticated partners. Especially in phase 2 of 4 of the Library Card project, we're focusing on global access delivery and we want to give editors the most comprehensive and efficient research environment we can. That means, particularly with the advent of the 'bundle', sources at a click, when you need them, with barrier to entry only based on Wikipedia login credentials and qualifying account criteria. Competing with libraries is not our goal, but rather giving editors access who do not have sufficiently robust libraries nearby, or who do not have access to them (either because they are not permitted to use them, or more commonly because they simply haven't gone through procedures to procure a library card in person). The 25,000 potential editors we estimate will have access to the library card bundle do not represent a significant threat to library usage locally or globally. The best we can do at this point is remind editors that libraries are incredible institutions with not just unique collections and often varied subscriptions, but also reference experts to support individual contributions and allies that can support our community. Cheers, Jake Ocaasi (WMF) (talk) 21:31, 10 May 2017 (UTC)[reply]
  • OAuth library card bundle is very cool, with clear targets. But - aside from the specific technical developments (the library card ‘bundle’ and Structured Data on Commons) what is the difference between Program 7 & 8?
    @Wittylama: See Jake's response above, and the two here.
    The focus on technical tools is the difference. Infrastructure focuses explicitly on technical tools and software builds, while Capacity deals with "human capital"--leadership, skills, and relationships. For Community Programs, building significant tech inside the Community Engagement Department takes a new focus and skillset of technical project management. It's not necessarily what we have been trained to do, but we have learned to do it better on our own and increasingly with collaborations across Product and Engineering. Still, tech builds are distinct because of their workflows and the nature of the product being something based in code and usable by others online. Capacity on the other hand, yields a much "softer", but no less important, result. In the case of GLAM, Structured Data on Commons will empower hundreds or thousands to do their work more effectively, but merely building the tool is far from the end of the road. It takes significant community organizing and support to help GLAM leaders understand and utilize powerful new features -- and to ensure that software design anticipates community program applications for those features. Technology advocacy is just one component of the holistic knowledge-base of GLAM work and GLAM organizing leaderships. The latter is each part of the Community programs team's focus on capacity, because, while people use tools and need tools, people themselves are the most effective means we have of creating the interpersonal relationships that allow our programs to be effective. Astinson (WMF) (talk) 21:52, 10 May 2017 (UTC)[reply]


  • And, for that matter, much of the milestones/objectives for these two (7&8) seem to be the same as those in Programs 3, 4 and 6 - just with specific reference to the Education and GLAM communities. Does this indicate different teams duplicating effort, or the same activity being counted twice?
    Programs 7 and 8 differ from 3, 4, and 6 because of the scope of the audience and resources developed. The work that Asaf has been doing while coordinating community capacity development applies to individual communities, especially emerging communities (#3); Learning and Evaluation are building generalized program skillsets (#4); and the Resource Hub curates knowledge and guides for the entire Wikimedia community (#6). Our community programs (#7 and #8), focus on a different audience and level of specificity: zeroing in on particular domains of knowledge/capacity within the movement's activities, and working to develop the skills and resources that support those fields of knowledge to improve impact with very specific types of external partners. Our primary constituency and focus are around the community of practices working on Education programs, GLAM-Wiki partnerships and Research/Reference, respectively. We do support or cooperate with programs 3, 4, and 6 where their general guidance effects our specific focus on our 3 core areas or when our experience overlaps with their needs. However, we find that each of the CE teams reach different audiences, focused on acquiring different skills or tools, that target different needs. Astinson (WMF) (talk) 21:52, 10 May 2017 (UTC)[reply]

Program 11: Community Collaboration in Product Development

  • Beta Features is mentioned once in this plan - the goal of having "At least three wiki projects requesting beta features”. This is very positive as it shows that the Beta process is becoming increasingly embedded as a standard practice in new software/interfaces. However, there remains no documentation/process for how/when a feature ‘graduates from’ or is ‘dropped out of’ Beta. Is that part of your roadmap for providing clarity and consistent to the community on how/when things change?
For comparison, the Usability Initiative (which built the Vector Skin) had the excellent measure of a certain % opt-in retention rate needing to be hit, for each iteration before moving to the next phase. This allowed a sense of community ownership and an objective measure.
Thank you for your consideration, I hope I'm not too late in asking these questions,

Wittylama (talk) 22:40, 7 May 2017 (UTC)[reply]

Hi Wittylama, our use of the term "beta features" (note the lower case) is more generic than the specific implementation of mw:Beta Features and may include other ways to become early adopters of new features. I can see how this might be confusing, but we thought that this expression would be better understood in general than the more technical "early deployments". If needed, we can modify that text to reflect our meaning more accurately.
Regardless of the wording, your question about clarity and consistency about the process following early deployments still applies. While defining a single formula applying to all features in all cases might be too complex, documenting best practices in the context of the Technical Collaboration Guidance is feasible, and a good idea. Due to historical reasons, we have focused on recommendations about how communities can be involved in deployments as early or late adopters, how they can propose blockers, and so on. However, you are right that documenting how deployments are supposed to proceed from a product development perspective is just as important. I have added phab:T165143 to our backlog. Qgil-WMF (talk) 10:22, 12 May 2017 (UTC)[reply]


Resourcing communities[edit]

  • The department has discovered it can not serve from a central location grant programs in forty countries. Wikipedia is a starfish, not a spider. The success of Wikipedia is it's radical decentralization. Rather than extra staffing for this department, or closing one or two grant programs, isn't it far more viable to delegate and decentralize certain programs for small grants to affiliates, for example chapters? Ad Huikeshoven (talk) 11:11, 17 April 2017 (UTC)[reply]
    Hello Ad Huikeshoven, Nice to see you here. :) No, I wouldn't say we have learned what you suggest above. We have, for years, been making grants in 60+ countries. Our challenge is not the number of countries; in fact, we hope to increase the number of countries we make grants!
So what's changed since our last annual plan? We started a new grant program, Rapid grants. And Rapid grants has been highly successful in reducing barriers to entry and bringing in new participants to the movement. When we set the program up, we didn't give it a dedicated staff person. The reason for that is we didn't know whether it would be popular, and it certainly has been. What we have learned is that it takes a person to run a program. We've been doing the best we can this past year, passing it back and forth between people for a month at a time and sharing knowledge. We have learned a lot in this time!
At the end of this experience, we've learned the program needs a manager to succeed. So, our issue is in securing a new staff person, who would be needed whether this program was run at WMF or elsewhere, if run by another affiliate you suggest. I'm not sure I understand what would make it more viable to delegate the program elsewhere; it's still important work that needs a skilled person to run it and be dedicated to its success no matter where it's housed. KLove (WMF) (talk) 05:35, 19 April 2017 (UTC)[reply]
  • I understand that you believe the program, as it is now, is not sustainable, and you either need more resourcing or to reduce the activities. However, it's not clear to me what you plan to do, and if you have not decided yet, when you will decide and based on what. - Laurentius (talk) 10:09, 30 April 2017 (UTC)[reply]
    Hi Laurentius! Which part is unclear? Our plan is to secure a junior program officer so that we can give the Rapid grants the support it needs to be operational and successful. Rapid grants is a program that has been incredibly successful in bringing in new community members/organizers into the Wikimedia world. This is particularly true for emerging communities. You can read more about our annual plan if you'd like a longer version. If that doesn't answer what you're looking for, let me know and I'll be happy to answer! KLove (WMF) (talk) 18:19, 1 May 2017 (UTC)[reply]
    Thanks. What wasn't clear to me was whether the plan was actually to hire the junior program officer, given the stress on "otherwise we'll close some program", that seemed to leave also this option open. I understand the case is the first one. - Laurentius (talk) 06:57, 9 May 2017 (UTC)[reply]
Laurentius, right. Assuming all goes according to this plan, we will be able to resource our programs appropriately. Because of that, we won't have to choose between programs. I didn't see this note until now, since I didn't get a ping, but I agree--we're on the same page :) KLove (WMF) (talk) 05:23, 12 May 2017 (UTC)[reply]

Support and Safety[edit]

@MDennis (WMF):, is there a reason why SuSa doesn't appear in any of the programs outlined in this proposal? I'm well aware of the vital work that this group does, having worked closely with them on any number of occasions, but there isn't even a "program" that says "we will keep providing the @emergency service, supporting volunteers with advanced permissions (CU/OS/Steward), support the OTRS system, provide support to the Election Committee" etc. Some of these activities would present excellent opportunities for metrics (e.g., time between referral to @emergency and decision on action, OTRS has full complement of X administrators 85% of the time, provide one steward meetup per year, provide access to tools that support major (100+) sockpuppet investigations, etc). I'm not certain to what extent other members of the FDC are aware of the support provided by SuSa, and the absence of a more in-depth description leaves me in a position to try to defend (or even advocate for more staffing for) this important part of the WMF/community ecosystem, without any documentation in the actual plan. Can you point to some department metrics or plans that would help out here? Thanks. Risker (talk) 03:40, 10 May 2017 (UTC)[reply]

Hi, Risker. SuSa is here, but the way we wound up arranging them makes that less than obvious. :) SuSa plays a key role in the cross-department program Community Health and will also be supporting legal in Privacy, Security and Data Management. Additionally, they are part of Program 4: Community leadership development & mentoring and leaders of Program 10: Training modules. Unfortunately, I am only just now realizing that one of CE's programs was not copied over when this page was posted - this one deals with all of CE's support of community groups (not just the stewards, the Ombudsman Commission, CheckUsers, ArbComs, election committees, and the OTRS admins, handled by SuSa, but also L&E's support of AffCom, Community Resource's support of grantmaking bodies (like the FDC). I'm sorry to say I overlooked that it was missed. I checked for errors, but didn't think to check for a missing program. In terms of SuSa's other work, last year, we were a bit inconsistent with how we handled core vs. programs, so things like emergency@ got called out but other core work for SuSa and other teams did not. This year, we were encouraged to focus more on programs with specific time-based deliverables rather than ongoing "core" work. More about the work of Support & Safety is visible on the meta page, and Metrics are tracked and reported as part of our quarterly check-in. These are typically published at Community_Engagement#See_also and Wikimedia Foundation Annual Plan/Quarterly check-ins, but it looks like the former was missed this quarter. I'll poke the person who uploads those to make sure they're on both pages. (Metrics are now an appendix called "Scorecards" - SuSa is part of the second group.) --Maggie Dennis (WMF) (talk) 17:55, 10 May 2017 (UTC)[reply]
Thanks very much, Maggie. This is very helpful. Risker (talk) 18:02, 10 May 2017 (UTC)[reply]

Technology[edit]

Availability, performance, and maintenance[edit]

I see that the "data center expenses" are planned to stabilise at the current year's level, which was lower than expected. I suppose this is thanks to some hard work in reducing costs, which is commendable. But does the budget take into account

  • the potential need for additional caching centers to improve service for users in certain areas of the world, or
  • the potential costs/investments for a transition to more sustainable providers and/or an effort to make the infrastructure more energy efficient and/or the purchase of renewable energy (or renewable energy certificates from projects in the local grids or good quality carbon offsetting), given wmf:Resolution:Environmental Impact?

--Nemo 10:15, 18 April 2017 (UTC) P.s.: From skimming the complete text, I don't find any information on standard operations work alias "keep the lights on", which is quite funny given WMF's fundraising messaging.[reply]

Hi @Nemo bis:, our budgeted data center expenses do include the service expenditures that will be rendered by the new caching center in Asia, which is in active development right now. We are also incorporating the Resolution on Environmental Impact in our decision making for the new development, and sustainability is part of the procurement selection criteria.
Regarding “keeping the lights on”, we focused the bulk of our descriptions on our programmatic work and especially on our cross-departmental programs. However, we also did describe our routine operational work. For example, you can see our work maintaining the servers described by our Technical Operations team, and our other operational work by the Finance and Administration, Legal, and Advancement departments.
JBaldwin (WMF) (talk) 22:30, 26 April 2017 (UTC)[reply]


Hi @Nemo bis:, as James has confirmed, we are indeed budgeting for an edge caching center to better serve our community in Asia. While we do not have a plan to move our existing data centers next fiscal, and we have not budgeted effort and budget for that, we are continuously evaluating the available options especially in light of wmf:Resolution:Environmental Impact--VColeman (WMF) (talk) 23:41, 10 May 2017 (UTC)[reply]
Thanks, this is good. Nemo 07:56, 2 June 2017 (UTC)[reply]

Product[edit]

Audiences

Interestingly the product department show their understanding of Audiences. A kind of customer segments. What kind of model is Product using? Are they working on a Business Model Canvas? Are they working of Value Proposition Design and or Empathy Mapping for the different audiences / customer segments / personas? Has this picture of audiences been discussed with Movement roles project, affiliates, at Wikimedia Conference? Personally I would have expected the presentation of audiences to be part of the overall Annual Plan, and not part of a specific department. The segmentation is very detailed, Audiences A, B, C, D ... upto ... M, N, O. That is quite a number of audiences! Do you have an estimate of the number of individuals in the Audiences H, I, J and K? Is there a way to group these audiences to a hand full? Through which channel does the Wikimedia Foundation serve the needs of the audiences (please specify for each individual audience A to O). How does the Wikimedia Foundation maintain customer relations with each of these audiences? How does this segmentation relate to the current strategy process with four tracks A, B, C and D? Are those four tracks a grouping of all audiences A to O? Aren't some audiences excluded in the strategy process? Ad Huikeshoven (talk) 11:55, 17 April 2017 (UTC)[reply]

Hi Ad -- thanks for the questions. Product along with Comms are indeed using customer segments to try to understand our audiences better and make sure we allocate our resources effectively across them. The Reading team used a very basic segmentation model (based on geographies) to direct our energies towards the New Readers project last year and this audience map is the next step in a more systematic approach. Many of your questions are the next steps in this process.
We are not using the methodologies you mention because these types of models that are used in for-profit businesses generally break down in mission focused environments because we don’t have revenue to use as a prioritization heuristic. However, the ideas behind these methods are very valuable and to that end we are using a combination of design thinking combined with more non-profit appropriate theory of change/logframe models to identify focus areas, desired impact and user needs. The program model used in the annual plan is a direct result of this approach.
As far as the model itself -- yes -- there are a lot of audiences! Simply by listing them, we have had to deal with this fact and that the product org will have to make some choices, but acknowledging this reality is the first step in dealing with it constructively. We have similar questions about audience sizes. This diagram is an initial attempt to capture some audience sizes but we didn’t find it that useful and will probably seek to find other ways to measure current and potential audience sizes in the future.
This segmentation model was not quite ready to be integrated with the tracks from the strategy process but we have reviewed the model together and expect that the segments will be harmonized once the strategy process is concluded. I’ve pinged Suzie from the strategy team for more detail.
Finally, you are probably wondering what the impact of this segmentation project actually is given my responses. Here are some concrete examples found in this annual plan:
  • The Reading team uses the segments to target specific programs towards specific types of users and use cases.
  • The increased awareness around all of our audiences has helped product work better with Community Engagement, illustrated by the GLAM-focused components of structured data as well as Community Tech’s attention to program needs.
  • Editing has formed better understanding of our contributor communities, with features for the multiple types of knowledge stewards (B) rather than just article creators.
TNegrin (WMF) (talk) 01:21, 22 April 2017 (UTC)[reply]

Hi TNegrin (WMF), sorry for the late question, but could you (or someone?) help explain why "New Editors" is a a program listed within the Product department (despite having four teams listed in the 'teams' heading); yet "New Readers is considered a "cross-department program"? Wittylama (talk) 16:07, 9 May 2017 (UTC)[reply]

Hi Wittylama, agree that this is a bit confusing. This program will be done within the Editing team -- the other teams are listed as dependencies -- Tech for MediaWiki, Research for the generative research. Important partners for sure but not co-owners as in the cross-departmental programs. TNegrin (WMF) (talk) 14:13, 10 May 2017 (UTC)[reply]
  • TNegrin (WMF) I'm not sure if you're the best person for this question but if not perhaps you can direct me to the person who is. My impression is that Wikipedia has been falling in Google page rankings over the years. It is now not uncommon for me to get an entire 1st page of search results from Google from which Wikipedia isn't linked. It seems to me that investing in search engine optimization might be a good use of resources. I haven't looked into the New Readers initiative, but given the criticism from the FDC about the New Readers initiative, perhaps it would be good to shift funds from New Readers to SEO. What do you think? --Pine 03:37, 17 May 2017 (UTC)[reply]
Hi Pine, I'd invite you to consider this suggestion in the context of the audiences that we've mentioned earlier in the plan with the added dimension of geography. With this framing, it seems like you are saying that you are concerned about readers who use Google to find content on Wikipedia, predominately in the core geographies and are suggesting that we move resources from programs addressing new readers, primarily in geographies in Africa, Asia and South America, to those users. It's unlikely we'd move resources in this way, but instead we might consider focusing resources from programs already intended for these users towards working with Google making sure content is optimized for their search results. This is certainly reasonable but I'd be cautious when using your own search results as the only data point -- Google uses a variety of testing techniques and you never know what bucket you are in. Our external referral dashboard that the Discovery team built has been pretty consistent over the last 18 months. TNegrin (WMF) (talk) 18:27, 22 May 2017 (UTC)[reply]
Hi TNegrin (WMF), thanks for the link to the external referral dashboard; I may have seen it mentioned previously but I had forgotten about it, or this may be my first experience with it. Either way, that information is somewhat reassuring, although I would hope to see the trends going in an upward direction (as the number of people with access to the Internet continues to grow) instead of sideways. I would encourage you to consider what resources could be devoted to SEO to try to move the trend in a positive direction at least in sync with the growth rate of Internet users as a whole.
Regarding the New Readers initiative, I have briefly looked at it, and while I would support efforts to improve offline reading and reducing cost of access, it seems to me that some of the offline outreach work is costly and I am skeptical about its cost-effectiveness. The Wikimedia Foundation isn't an international development agency, and given my understanding of what happened in Brazil during Sue's tenure, it seems to me that elements of New Readers in their current form are high risk and high expense. I think that partnering with other international development efforts or local development efforts, such as with Doctors Without Borders or mobile network operators, makes sense; I'm a lot more cautious about the boots on the ground efforts.
What would be some metrics that you would use during the year, perhaps at quarterly reviews, to decide that one or more elements of the New Readers initiative is not meeting its objectives and to either downscale or eliminate that element? I want to support experimentation, but I also wouldn't want to keep investing resources into projects that aren't demonstrating meaningful and timely returns on investment. --Pine 19:29, 22 May 2017 (UTC)[reply]

Advancement[edit]

Offline content[edit]

In a few points you mention offline distribution of Wikipedia content. Do you plan do work with Kiwix for this, or it will be something different? I see that there are also plans to introduce offline functionalities in the Wikipedia app, but I imagine that it is a different use case. - Laurentius (talk) 15:18, 1 May 2017 (UTC)[reply]

Hi Laurentius: Thank you for your question, and apologies for a late response! Offline support is, in fact, a priority for both the New Readers cross-departmental program, as well as the programs for Reach of the Partnerships and Global Reach team. With this in mind, we have been looking into all of the different players in the offline arena (which we have been documenting), including Kiwix. Kiwix is definitely the one we're looking into and working the closest with so far. We have been in constant communication with the Kiwix team to find the best ways we can collaborate, and we are exploring the best ways to work together on this common goal.
As part of the New Readers offline focus, we are building and testing solutions both on mobile web and the Wikipedia Android app to support fully and partially offline readers. This is stated in the annual plan for New Readers. We are happy to continue this conversation, so feel free to reach out anytime! --JVargas (WMF) (talk) 18:55, 8 May 2017 (UTC)[reply]

Awareness in Mexico[edit]

Only 45% of Mexican have heard of Wikipedia, and only 45% of Mexican are online (data from Program 2): I would imagine that if you don't have access to Internet you are much less likely to know Wikipedia, and therefore, which this percentages, that most of Mexican internet users know Wikipedia. However, (from Program 6), more people learn about Wikipedia through word of mouth than by using internet and nearly half of internet users have never heard of Wikipedia; how is that possibile? It seems that having access to Wikipedia has almost no effect on whether you know about Wikipedia. I understand that you may have heard of Wikipedia without ever accessing it, and that you may use internet without ever going to Wikipedia, but I would still expect a significant correlation. - Laurentius (talk) 15:18, 1 May 2017 (UTC)[reply]

Hi Laurentius - Thanks for your comments, and apologies for a late response. First of all, let us clarify the sources of those percentages. The number in point 1 of Program 2 (45% of Mexicans have heard of Wikipedia) comes from our mobile phone survey, whereas the number in point 2 of Program 2 (55% of the Mexican population have never been online) comes from a different source, which only represents users who can access the Internet at home. Our phone survey showed that 80% of Mexicans use the internet, and the difference between these two statistics is likely due to the additional use of internet outside of the home, for instance school, work, libraries, internet cafes, among others.
When we looked at the phone survey data - the subset of people who said they use the internet - only 51% of them (internet users) had heard of Wikipedia. From that question on our phone survey, 41% of the people first learned about Wikipedia from using the internet, and the combination of friends + family(21%) and school (24%) is 45% - more than the 41% who discovered it from the internet.
Feel free to review the full results, flow chart, methodology, and documentation of the Mexico phone survey. If you have any more questions we will be more than happy to clarify them! --JVargas (WMF) (talk) 23:49, 11 May 2017 (UTC)[reply]

Legal[edit]

Place comments about this section here!

Communications[edit]

Place comments about this section here!

Non-Programs[edit]

Fundraising:

  • In the "Advancement: online fundraising" section, you refer to "relationship with existing donors through email optimization". Can you elaborate on what you mean by Relationship? I ask specifically because the overall GOAL for this section includes the line "...while educating readers about our movement" but all the Activities and Targets are focused specifically on improving the efficiency/effectiveness of the $ donations. Is this "relationship" through "email optimisation" in reference to the educational aspect of the team's goal? I note that in Program 2 (for major gifts) there is a line called "Create and release frequent updates to donors about the impact of their contributions" - which speaks to the kind of communication I'm talking about, but this is not included in the online fundraising section even though the content could be used for both, perhaps? Wittylama (talk) 13:19, 8 May 2017 (UTC)[reply]
Thanks for the question, Wittylama. There is definitely some overlap between these two efforts. We allow all of our donors to opt in to receive quarterly updates from us that highlight the foundation's work. It is primarily a roundup of the announcements in the blog from the previous three months. Any donor can sign up to receive this from us. This is a part of our effort to retain our donors for the long-term. We want to build a community of donors that will support us every year and really understand the organization in deeper ways than can be communicated in a banner. This will make the organization more resilient to changes in how our readers may consume Wikipedia content in the future. We also do more in-depth reports for major donors (mostly foundations) that are a part of the reporting requirements for each individual grant that we receive. --Lgruwell-WMF (talk) 16:53, 10 May 2017 (UTC)[reply]
We have not finalized the schedule of events for next year. This is referring to the cocktail parties that we hold for donors who give around $1000 USD or more. When we hold these events, we want to partner with the chapter if one exists or have local WIkipedians involved. In fact, we cannot do it without them. Local Wikipedians volunteer at all of our fundraising events and are critical for their success. Plus, our donors love meeting Wikipedians and ask a ton of questions about how Wikipedia works. --Lgruwell-WMF (talk) 16:53, 10 May 2017 (UTC)[reply]
Sorry for the jargon. :) We are trying to plan for a day when the majority of our readers might not be consuming Wikipedia content on WIkimedia sites or apps. That is what we mean by resilience and it is a big focus for us. We are trying to not only achieve the annual goal, but also make the organization resilient to technology changes that might affect the revenue model in the future. The endowment is a part of this strategy and so is deepening our relationship with our current donors so that they stick with us even if/when we can no longer reach them through a banner on a WMF site. We want to make our donor's connection to WMF stronger by deepening their understanding of our work, so they give to WMF every year regardless of where and how they consume the content. Also, we want to encourage a subset of them to make a planned gift to the endowment. That is the "stream" or "donor life-cycle" we are talking about.--Lgruwell-WMF (talk) 18:26, 10 May 2017 (UTC)[reply]

Staff & Contractors[edit]

In the second table (number of FTEs by department), there is a "FY16-17 Approved FTEs" column and a "FY17-18 Planned Total FTEs" column. This is phrased differently from the standard FDC model ("End of current funding period", "End of upcoming funding period", and explaination for the change; see for instance WMDE's section). Could you please confirm whether the meaning of the columns is the same, and, if not, explain what these ones mean; and add an explaination of the planned change?

Moreover, comparing the "FY16-17 Approved" to 2016-17 annual plan there are some significant changes, both in the total and in specific departments. Could you please explain the reason behind those changes?

I also noticed that this section is not shown in the 1-page view. I thought it was a problem with a noinclude tag and I tried to fix it, but apparently unsuccessfully (unless some caching is preventing me to see the change...). - Laurentius (talk) 10:40, 30 April 2017 (UTC)[reply]

Hi Laurentius. The Finance team will be coming along to answer the bulk of your question, but I wanted to thank you for pointing out the transclusion error on this section! You were correct that it was an issue with an unclosed noinclude tag, and your fix indeed fixed it. The page just needed to be re-marked for translation so your fix propagated to the translation subpages that people are actually shown when they view the transcluded content. I've done the marking, and the tables should be displaying properly on the 1-page view now. Thanks again! Kbrown (WMF) (talk) 14:54, 1 May 2017 (UTC)[reply]
@Laurentius: We missed this question when you first asked it and it somehow fell off our radar. I apologize about that. Yes, they mean the same thing as in the FDC’s tables. The meaning of "FY16-17 Approved FTEs" is equivalent to “End of current funding period.” Likewise, “FY17-18 Planned Total FTEs” is equivalent to “End of upcoming fiscal period."
The increase in FTEs from the end of FY16-17 to FY17-18 will fill operational gaps in several departments. Each department's requirements for staffing are slightly different. Finance and Administration, for example, has a need to expand our accounting team by one person to handle the increasing sophistication of our workflows and growing volume and complexity of our accounting transactions.
The changes in our original FY16-17 Annual Plan, as compared to where we will end the FY16-17 year, are primarily the result of two things: the separation of our Product and Technology departments (with the relocation of Team Practices team into the Product department) and the addition of two major programs added mid-year: Structured Data on Commons and Community Health. Together, these two programs added 12 new FTEs that were not included in the FY16-17 Annual Plan. The other additional FTEs above our original plan came from a few other minor structural changes. JBaldwin (WMF) (talk) 00:22, 11 May 2017 (UTC)[reply]

Verification[edit]

Does "This question is not applicable to the Foundation." mean that the WMF is not planning on spending any funding on "political or legislative activities" (and so forth), or that it is? If so, how much funding is being spent on that work? Thanks. Mike Peel (talk) 00:08, 10 May 2017 (UTC)[reply]

Appendix[edit]

Place comments about the Annual Plan Appendix here!

General comments[edit]

Meta-question[edit]

Shouldn't there be a note on the home page that encourages review of this (and maybe other orgs') plan? Elitre (WMF) (talk) 15:01, 20 April 2017 (UTC)[reply]

Hey, Elitre (WMF). Not sure how to answer this except "maybe." :) Alerting people to use of movement funds is a good thing. Adding notice to the main page when annual plans are being discussed might be a great idea. That sounds like an idea you might want to either implement directly, in the wiki way, or take up as a discussion with whomever works on the main page, in the wiki way. --Maggie Dennis (WMF) (talk) 13:46, 26 April 2017 (UTC)[reply]
Hey Mrs.Dennis, looks like Delphine (WMF) already took care of that - I think the current notice is fine. Thanks. Elitre (WMF) (talk) 13:54, 26 April 2017 (UTC)[reply]

Apply knowledge that we produced in another Wikipedia[edit]

Why don't Wikipedia apply stuff, tools, software that will become Wikipedia in all languages more easier, will become Wikimedia safiest, bring people that don't trust in Wikipedia using experiences, stuff, tools from German Wikipedia, English Wikipedia, Spanish Wikipedia, Japanese Wikipedia and bring more good stuff from some European Wikipedia/Wikimedia to improve software, become some editions easier, the fonts used in Wikimedia, become faster Wikimedia to all users?--Ultragp7 (talk) 01:14, 24 April 2017 (UTC)--Ultragp7 (talk) 01:24, 24 April 2017 (UTC)[reply]

Hi Ultragp7 -- thank you for asking! There is actually a lot of shared software across the projects. All of the projects use the Mediawiki platform stack of course, and there is a high degree of shared use in our datacenters, hardware, etc. Features like lazy loading of images are enabled on the all wikis.
Moving up the stack, things become more complicated. Communities are able to in some cases choose the features that are available to their projects. This may create the impression that software isn't available everywhere. In addition, due to some technical constraints, sharing code becomes more difficult -- the discussion about global gadgets illustrates some of the concerns and also the popularity of the request. TNegrin (WMF) (talk) 20:35, 27 April 2017 (UTC)[reply]

SMART metrics[edit]

In general, the metrics here are not SMART - in particular, while they are often specific, responsible, and time-related, they are not measurable (and as a result, it's unclear if the intentions are achievable). Taking as an example the very first outcome here (not picking on this particular point - this is true throughout this proposal), "More people recognize and understand the value of Wikipedia content, so they seek out Wikipedia (either in search results, by coming to our projects, or following a link) when they are looking to learn." - how will you actually measure that to know that it will be achieved? Both in terms of the actual measurement (website statistics), and also the target that you are aiming for (what is currently achieved, and what are you aiming for by "more" - this is where "achievable" comes in, as well as a budgeted amount to know if this is a worthwhile thing to do). This is where the WMF-defined grant metrics are useful - in this case, you are looking at the "total participants" metric in this case.

Please consider adding measurable metrics as soon as possible. I would encourage you to use the grant metrics that the WMF has defined wherever possible, and clearly define a set of grantee-defined metrics in the same way that the other organisations have done. Thanks. Mike Peel (talk) 23:49, 2 May 2017 (UTC)[reply]

@Mike Peel: We will also respond to this question next week. I apologize about the delay. JBaldwin (WMF) (talk) 00:57, 6 May 2017 (UTC)[reply]
@JBaldwin (WMF): Please respond in the next couple of days - before the FDC discusses this! I view this (along with the lack of budget information) as a huge, fundamental problem with this annual plan, and this either needs to be fixed or there needs to be a clear path to fixing it. Thanks. Mike Peel (talk) 23:06, 9 May 2017 (UTC)[reply]
Hi, @Mike Peel:. Thank you for the feedback. Having one short set of defined metrics is not appropriate for the Foundation, given our scope of audiences, programs, and objectives. As you may have seen in our statement in the annual plan (under Programs), “the Foundation's programs do not directly produce impact in the metrics used for other organizations, namely the metrics of participants, newly registered, and content pages. When it does, it is through its support of individuals, groups, and organizations that develop the activities allowing the tracking of these metrics.
The evolution in grants metrics in the past few years makes it difficult to consolidate results across the whole movement into those three metrics.” Because the Foundation works across many different areas (technology and product, community engagement, etc.), we currently have program-defined metrics as our “grantee-defined metrics.” These are monitored on a quarterly basis and reported via a mid-year report to show programmatic progress. All of these are publicly posted.
In some cases, we are engaging work that is strategic, and by nature speculative. In these instances, the objectives are necessarily less measurable, as the specifics will evolve in response to proper scoping during the development of the program. This scoping will generally occur as part of Q1 efforts, following the approval of the annual plan. These more refined (SMART) objectives and milestones will be what is monitored and reported against during the upcoming year.
We expect that once the new movement strategy direction is set, the Foundation will be able to generate stronger predictive metrics.JVillagomez (WMF) (talk) 23:41, 11 May 2017 (UTC)[reply]

Links to more detailed information?[edit]

Most of the segments don't include links - and where they do, they are to very general pages about a given topic. Please can links to specific information about the teams and work involved in the various segments be added? Thanks. Mike Peel (talk) 00:18, 3 May 2017 (UTC)[reply]

Hi, @Mike Peel:. Thanks for pointing this out. We will ask the teams to add in their team meta pages as well as more detail on these projects, as it is available. Some of these are new initiatives, so there is less supporting details available. JBaldwin (WMF) (talk) 00:31, 6 May 2017 (UTC)[reply]
Thank you, JBaldwin (WMF). What we are looking for are pages like this, many of which were available in last year's plan. If Technology/Product pages are on MediaWikiWiki, links to those pages will be fine. Risker (talk) 02:25, 10 May 2017 (UTC)[reply]
@Risker: Thank you for the example. We have updated the plan with more links with the support detail and a few more links are getting posted right now. JBaldwin (WMF) (talk) 22:22, 11 May 2017 (UTC)[reply]

This feels more like a list of projects rather than a functional work plan[edit]

In some areas, it's appropriate that the annual plan will read more or less like a series of projects; however, in several areas there seems to be a lack of information on what the teams are doing when they *aren't* working on a project. An annual plan includes the already-occurring activities (with appropriate metrics) as well as any project-like or focused work. The lack of this information is genuinely problematic in assessing the value-for-money of the annual plan; to be honest, if all the WMF is going to do in the next fiscal year is the programs listed here, it would be difficult to justify even half the proposed budget. Adding it all together, it looks more like you need 125 FTE rather than 300; there's just not enough work in this plan to justify this large a cadre of employees. I know this is coming late in the game, but I labour under the belief that the kind of information we're looking for exists somewhere or other (although quite likely on a non-public wiki/server used only by WMF staff). Having worked with many departments within the WMF, and with staff from every department at some time or other, I know there's a lot more involved in the work being done than shows up in this document. It makes it difficult, though, to support that position when we don't have the documentation to back it up. Risker (talk) 04:01, 10 May 2017 (UTC)[reply]

Hi @Risker: First, thank you so much for the comment and appreciation for all the work that the Foundation does. The recognition and good faith are genuinely appreciated by Foundation staff. You are right, this plan showcases our most important projects, and is not a detailed timesheet of all activities of all staff members. We do indeed have much more detail (both internally and on meta). We have asked the teams to provide more links, which is happening now. In addition to adding more links, we are preparing more detail about the deployment of the FTEs. This should help readers better understand the staff resource across programs.
In general, due to the scope of work of the organization, every year we confront this question: how do we balance the need to provide transparency and accountability into our work with the significant investment of staff time required to produce a detailed list of individual staff activities? We also generally believe that the most significant, strategic, high-impact, or new activities are those that should receive the greatest attention in annual plan reporting, rather than detailed explanations of day-to-day work that is already largely familiar to the community and Board. This is a perpetual balancing exercise, as the annual planning document is already extensive, and is just one part of our regular transparent planning and reporting processes (alongside monthly metrics, quarterly goals and check-ins (formerly known as quarterly reviews), and the mid-year report).
Our intent with the annual plan is to give the FDC and Board enough information to understand the scope of new work and major projects. To that end, we welcome feedback on what is most helpful information about those types (new, significant, cross-departmental, strategic) of work, so that we can improve reviewer understanding or planning clarity. And in the spirit of appreciation: Thank you to you and the rest of the FDC members for your time, diligence, and efforts this week in Warsaw. Yours, Katherine (WMF) (talk) 19:00, 11 May 2017 (UTC)[reply]

Product and tech budget[edit]

Budget shift[edit]

Comparing the numbers in WMF's submission from previous year and the present one, you can see that over 1.1 Million USD has been moved between last year's budget and this report (table #5) from Talent and Culture to Product and Technology (the latter move from $27,927,324 (plan) to $15,078 + $13,983 = $29,061 (plan), $15,686 + $12,659 = $28,345 (projection) ).

What is the story behind it? Is it a transfer of some activities and their budget to a different department, or some shift in actual programmes and expenses?

Hi @Aegis Maelstrom:, Yes, we did move one of our teams, the Team Practices Group, from out of the Talent and Culture department into the Product department. We have been considering how best to structure this team within WMF. In FY17-18, we are planning another, smaller, shift in the Team Practices Group’s structure with a small part of that team potentially moving back under the Talent and Culture department. JBaldwin (WMF) (talk) 18:30, 12 May 2017 (UTC)[reply]

Largest grow[edit]

Product and Tech grows from $29.06 Million (new plan 16/17) to $33.34 Million (17/18). These two departments are heavily project-based (so e.g. comparing two years is difficult) but unfortunately not only the budgets of particular projects are not provided but also it is unclear which projects were enabled or strengthened thanks to this budgetary increase. How are the activities prioritized?

@Aegis Maelstrom: The largest increases in the Product and Technology departments’ budgets are coming from the addition costs of the Structured Data on Commons program, the Community Health program, and the cost of the Wikidata program. Together, they represent $3.1M in expenses captured within the Product and Technology departments in FY17-18. (Note: There are additional costs of these programs captured in other departments). We did provide separate budgets for the Structured Data and Community Health programs in our Cross Departmental Program financial table #2. You can also see the total cost of the Wikidata expense in financial table #7.
There are other budget increases in the Product and Technology related to the New Readers program, the Privacy and Security Program and various costs associated with filling operational gaps in our existing programs and projects. JBaldwin (WMF) (talk) 18:30, 12 May 2017 (UTC)[reply]

Geographic expansion[edit]

Bay Area is not precisely the least expensive place, and AFAIK there is a long tradition of off-site tech collaborators. Did you consider having more activities outside from SF, like Wikidata team?

Thanks for your insight! aegis maelstrom δ 23:13, 11 May 2017 (UTC) [bumping @JBaldwin (WMF) here :) ][reply]

Yes, @Aegis Maelstrom:, there is a tradition of conducting technical collaboration outside of San Francisco, including hackathons, Wikimania, and the Wikimedia Affiliate Conference. We have a Technical Operations team and a Technical Collaboration team that are predominantly based outside of San Francisco. JBaldwin (WMF) (talk) 21:24, 12 May 2017 (UTC)[reply]
Thank you very much @JBaldwin (WMF):, these are pretty helpful! Regarding the employment, I noticed that the tech collab is quite dispersed :). As we already have a Wikidata development office in Berlin, and we are going through a strategic process now, I've been wondering if the involved WMF staff considered the feasibility and possible opportunities/disadvantages of a further dispersion or establishing a similar competence centre. Best Regards, aegis maelstrom δ 08:50, 13 May 2017 (UTC)[reply]
Or just close the office altogether. --Nemo 09:36, 4 July 2017 (UTC)[reply]

Red link[edit]

At the bottom of Wikimedia_Foundation_Annual_Plan/2017-2018/Draft/Programs/Product#Program_6:_Community_Wishlist. --Elitre (WMF) (talk) 08:38, 1 June 2017 (UTC)[reply]

Thank you for spotting that and letting us know, Elitre (WMF)! I've fixed it. Kbrown (WMF) (talk) 11:57, 1 June 2017 (UTC)[reply]

Some corrections regarding gender data[edit]

Regarding this part in Community Engagement Program 2:

Based on the incomplete data we have, we know that only 13-23% of our contributors on the Wikimedia projects are female, and we know nearly nothing about the participation rates of other genders. ... The 2010 UNU-MERIT survey was opt-in and found that less than 13% of Wikipedia contributors were female. The 2013 Benjamin Mako Hill survey, “The Wikipedia Gender Gap Revisited: Characterizing Survey Response Bias with Propensity Score Estimate”, estimated the proportion of female U.S. adult editors was 22.7%, and the total proportion of female editors around the world to be 16.1%.

First, kudos for quoting the more accurate numbers from the Hill/Shaw paper instead of the still often-used "one in ten" etc.! However, some things are still a bit mixed up here, and even though I guess these factual inaccuracies don't materially affect the rationale for the planned activities, we should correct them before they go on record with the finalized plan:

  • There was no "2013 Benjamin Mako Hill survey". The linked article by Benjamin Mako Hill and Aaron Shaw did not present a new survey, but rather reexamined the UNU-MERIT data with a novel correction method, adjusting the aforementioned 13% to 16.1%. (see also the summary here)
  • Also, that UNU-MERIT Wikipedia survey was conducted in 2008, not in 2010.
  • "we know nearly nothing about the participation rates of other genders" - that's not quite true:
    • The 2008 UNU-MERIT survey included "other" as response option alongside "male" and "female", with the (uncorrected) ratio among editors, 0.63%, being somewhat higher than that among readers in general, 0.49%. (In other words, in contrast to women, nonbinary genders were over- instead of underrepresented among Wikipedia editors.)
    • In the April 2011 Editor Survey, the question was modified (IIRC in response to some criticism) to include both transsexual and transgender as response options, and according to the results "Fewer than 1% of editors self-identified as transsexual or transgender."
    • The November 2011 Editor Survey reported 0% transsexual and 1% transgender editors (rounded, obviously)
    • In the 2012 Editor Survey, 0.1% of editors identified as transsexual and 0.2% as transgender. (NB: these results from the three 2011/12 surveys have not been subject to the same propensity score correction as the 2008 survey, because the corresponding reader data needed for that method is not available for these three surveys.)
  • Lastly, regarding "was opt-in": all of these surveys are opt-in (voluntary), and they probably always will be. In contrast to e.g. Facebook or a government census, Wikipedia editors are simply never required to state their gender. What the 2013 Hill/Shaw paper did was to correct the opt-in data using data from a source (Pew) that was still based on voluntary responses, but with a much higher response rate/less response bias (phone surveys).

So altogether, here is one suggestion on how this part of the text could be reworded to keep it accurate:

Based on the incomplete data we have, we know that only about 13-23% of our contributors on the Wikimedia projects are female. A 2013 study by Benjamin Mako Hill and Aaron Shaw, “The Wikipedia Gender Gap Revisited: Characterizing Survey Response Bias with Propensity Score Estimate”, estimated the proportion of female U.S. adult editors was 22.7%, and the total proportion of female editors around the world to be 16.1%, based on data from a 2008 UNU-MERIT survey. [leaving out the non-binary claims, but I would be happy to help with a one-sentence summary there too]

PS: If I may, here is an overview I published in 2015 (in collaboration with the then CE team) about the available results regarding female contributor rates, also discussing methodology questions: https://blog.wikimedia.org/2015/04/30/how-many-women-edit-wikipedia/

Regards, Tbayer (WMF) (talk) 21:33, 4 June 2017 (UTC)[reply]

Staff numbers reshuffle[edit]

I suppose [2] is related to mw:Wikimedia Engineering/June 2017 changes, right? --Nemo 20:58, 18 June 2017 (UTC)[reply]

Minassian Media, Inc.[edit]

Insofar as Minassian Media, Inc. is an S-Corporation based in New York run by Craig Minassian who is the Chief Communications Officer for the Clinton Foundation and who—according to the 2015 C.F. 990 form (p. 43 of 117)— was paid $200K for a 50 hour work week. I think "the Community" should learn more about the arrangement with this public relations firm, because according to the Wikimedia Foundation's 2015 990 form this firm was paid close to half a million dollars (see page 58/58 or 60/60) for public relations work on behalf of the foundation. Kbrown (WMF), JSutherland (WMF), Heather (WMF), Quiddity (WMF) could you provide any further information on this? SashiRolls (talk) 22:39, 18 June 2017 (UTC)[reply]

@SashiRolls: Thank you for your interest in our work. Minassian Media, Inc. has been a consultant to the Wikimedia Foundation since 2010. They are a communications firm that works primarily with global organizations which have social good or non-profit missions. The Wikimedia Foundation uses a number of communications agencies around the world to help support our efforts to engage in the worldwide attention the Wikimedia movement receives. It is very typical for a brand of Wikimedia’s size to have consistent agency support like this in order to bring high-level expertise and staffing to our public relations work. As an example of that work, last quarter the Communications department gave substantive support for more than 25 interviews, replied to 125 press requests, and handled over a 100 other press interactions.
At any given time, the firm has up to 4 staff working with us on our ongoing interactions with international press, coordination for public speaking and events, responding to emerging situations in the media, and supplementing our communications capacity as needed. As always, our goal is to support the values and communities that make Wikimedia possible, and we’ve found that the talented team at Minassian Media is an excellent partner.
I would like to take this opportunity to remind community members around the world that we welcome collaboration in our communications and outreach efforts. Our daily work with Wikimedia volunteers and entities is core to our work and ability to serve the movement. Please feel free to email us (communications(_AT_)wikimedia.org) with any ideas or specific interests. Thank you! --Heather Walls (WMF) (talk) 21:14, 29 June 2017 (UTC)[reply]
@Heather (WMF): -- Thank you for your response. I'm afraid you haven't responded to the point of my question, though. Could you indicate whether this work done during the 2016 US elections was done by the current Chief Communications Officer of the Clinton Foundation (Craig Minassian) or by Jove Oliver (who previously held that job title) or by someone else entirely? The problem is that this seems like a potential conflict of interest, especially given the strong, documented bias on English Wikipedia towards Clinton during the 2016 US elections. Do you continue to employ current Clinton Foundation staff? SashiRolls (talk) 23:04, 4 July 2017 (UTC)[reply]
It would probably be better to respond without fear over on this page where some good points have been made by someone who is less immediately frustrated at the moment about the "harassment" labels that get thrown around in communications reports and on en.wp. There's also a question about the culture of advertising value equivalencies there. SashiRolls (talk) 21:54, 6 July 2017 (UTC)[reply]
  1. Proposal to merge Outreach wiki into Meta Wiki on Outreach Village Pump (February 5, 2015)