Grants talk:APG/Proposals/2015-2016 round1/Wikimedia Deutschland e.V./Proposal form

Questions from FDC members

feedback

@Julia Kirchner (WMDE):, can you please clarify the different between personal and non personal costs? are personal costs include only wage and employee cost? or also travels and others costs? Thank you --Itzike (talk) 07:03, 7 October 2015 (UTC)

Personnel cost includes salaries, employment taxes and benefits (incl. health insurance and employer contributions to retirement plans). All other cost is covered under non-personnel.--Nicola Zeuner (WMDE) (talk) 08:27, 7 October 2015 (UTC)

Operating reserves

Hello. You state that "In 2016 WMDE plans to set aside 114,940 € in operating reserves." - please could you clarify, is this amount your total operating reserves or the increase to your reserves this year? Please could you expand on your operating reserves policy/approach? Thanks. Mike Peel (talk) 19:45, 21 October 2015 (UTC)

The 114.940 € is the calculated surplus of the projected revenues/expenses for 2016. This amount is a financial cushion for the 2016 budget, rather than an operating reserve. The surplus will help to cover small budget gaps that result from currency fluctuations, unexpected expenses or less than projected income.

In 2015, we have a similar amount in the budget, and it will most likely be spent by the end of the year. There is no approach or policy regarding operating reserves at this time. Christof Pins (WMDE) (talk) 17:11, 4 November 2015 (UTC)

Administration costs

Hello. You give a value of €1,748,226 for "Administration (Board, ED, Operations, HR, IT)" - could you break this down further into those different sections, please? How does it compare to last year's expenditure? It sounds like it's quite an increase from the ~€970,000 in the current year. Thanks. Mike Peel (talk) 19:49, 21 October 2015 (UTC)

Hello Mike, many thanks for your questions about the budget details here and above. Unfortunately, our financial manager is out sick until the end of this week. Therefore we'll have to wait for him to return to the office before we can draft our answers. We hope for your understanding regarding this delay.--Nicola Zeuner (WMDE) (talk) 14:33, 27 October 2015 (UTC)

Thanks again Mike for this question. We are not sure where the €970K you cite comes from - it is not the correct number. Here are the budgeted breakdowns for 2016 (and 2015 for comparison):

IT: €297,176 (2015: €90,000)

Finance: €118,242 € (2015: €120,000)

Operations (incl. HR and works council): €832,742 (2015: € 715,000)

ED: €388,653 (2015: €465,000)

Board: €111,414 (2015: €115,000)

Total: €1,748,227 (2015: €1,505,000)

Please note that these figures do not yet reflect end-of-the-year re-allocations of admin cost to departments, and thus are slightly higher than the annual plan figures. In 2016, overall administration cost increases by around 16% (€243,227) compared to 2015 and by 4% (€56,878) compared to 2014. This is a result of the following developments:

• In 2015, administrative functions were underfunded. In 2016, we will need to implement infrastructure adjustments at the WMDE HQ to comply with federal workplace safety regulations.
• In 2016, to streamline purchasing processes, all IT supply and equipment purchases (such as laptops) are accounted for under the IT department, and no longer under the departments using the purchases.
• The office and admin teams are professionalizing their operations and improving processes, with the help of 1.6 additional FTEs.

Many thanks again for your questions. Christof Pins (WMDE) (talk) 17:11, 4 November 2015 (UTC)

@Christof Pins (WMDE): Thanks for the clarifications and extra info! The ~€970k number came from the "Board, ED, Operations, HR, IT" row under 'Administration' in Table 2, where you report €858,032 as 88.58% of the budget. I guess that excludes Finance and a large fraction of the Operations costs? Thanks. Mike Peel (talk) 18:32, 4 November 2015 (UTC)

@Mike Peel: Hello Mike, thank you very much for pointing this out. It was very helpful to reveal a potential misunderstanding in our presentation of the expenditures to date: The percentages stated in Table 2 do not reflect the part of the budget spent to date vs. the total amount budgeted. Instead, they represent the budget spent to date vs. the budget expected to be spent by 31 August 2015 (as planned). Therefore these figures cannot be used to make a projection for end of year expenditures.

Additionally, you are correct about the “Board, ED, Operations, HR, IT” line in Table 2: It does exclude several finance and other administrative costs, which are included in different line items above (e.g. “Lump Sum Project Department”).

For an overview of budget spent to date vs. the total amount budgeted, please refer to the table provided below with regard to Katy’s budget questions. There, all administrative costs are summarized, totalling to ~70% spent to date (30 September 2015), vs. the planned yearly budget (annual plan), which amounts to $1,505,000 for 2015. Cheers, Christof Pins (WMDE) (talk) 16:38, 5 November 2015 (UTC)

Thanks, the table below is very useful! I'm not sure I understand your clarification of Table 2, though - do you mean that the "Percentage spent do date" actually relates to the percentage of the Q1-3 budget that was actually spent in Q1-3 (which presumably would be end of September 2015, not end of August?), i.e. completely excluding Q4 expenditure? Thanks. Mike Peel (talk) 19:30, 6 November 2015 (UTC)

Yes, you understood it very well. All percentage figures in Tab 2 are calculated based on the budget spent until end of Q3 vs. the amount budgeted to be spent in Q1-Q3 only (and please excuse the August/ September confusion...). Cheers, Christof Pins (WMDE) (talk) 09:18, 9 November 2015 (UTC)

Questions from Risker

Thank you for your proposal, which I have read with significant interest.

Hello Risker, thanks a lot for your efforts and for giving us the opportunity to clarify and explain a bit more. Please find our answers below in-line with your questions.--Nicola Zeuner (WMDE) (talk) 14:33, 27 October 2015 (UTC)

• You have referred to the tools created by your developer employees for "the community". Could you please provide five examples of tools developed by WMDE staff that meet the following criteria:
  • Are available on the Tools Lab, and
  • Are in use on at least five other language projects. One of those projects should be at least 50% of the size of the comparable German-language project (e.g., at least one Wikipedia that has at least 800,000 articles, or a Wikisource that has at least half the number of content pages as German Wikisource).
  • If there are no tools that meet both of these criteria, please provide examples of tools developed by WMDE staff that are used on non-German projects.

The TCB Team – that is the product development team working on community requests – has been operative only since July 2015 and has barely started working on technical requests. The scale of replication across other ‘language projects’ requested here is not realistic after less than three months of this work. However, the participatory approach used assures that staff works only on requests that are generated by the community, prioritized by the community and relevant to the movement. The role of WMDE staff is not to pick the projects, but to facilitate the participatory process, and to follow up, together with community developers, on the prioritized requests. Regarding the use of tools among other ‘language projects’ - all features developed are based on our communities’ requests and needs, and are therefore likely to be utilized across language projects. The necessary documentation needed for project communities to implement tools and features of their choice is readily available online, and ad hoc support to other language communities is available through our team.

WMDE develops tools (also available on Tool Labs, if required for their purpose), as well as features and extensions to MediaWiki or Wikibase. These features then in turn enable volunteers and WMDE staff to develop tools. Following are a few examples:

Tools/products developed upon request by German community, available to other communities:

• Wiki Loves Monuments Upload Toolset, possibility for direct image upload for (German) WLM via monument lists in Wikipedia with a predefined category and object ID

This toolset was activated in September 2015 for the duration of the WLM contest. Therefore it is not accessible live post-contest

• Demo-site
• Concept and Feedback-loops with the community (in German)
• Documentation on Github and Phabricator

• Wiki Loves Downloads, a tool for generating lists of images from a certain Commons category to be used in a download manager

• Tool on Tool Labs and Github

Tools/products currently being developed upon request by German community, with international support and potential impact for other communities:

• Possibility to remove articles from watchlist after a custom timeframe

The concept is completed, next step is investigation, will probably be merged into MediaWiki core. (Phabricator)

• A meta data extractor for Commons (Github)

• Aside from MediaWiki improvements that are directly or indirectly developed to support Wikidata, please provide three examples of other significant extensions or MediaWiki core improvements that have been made by WMDE staff in their role as WMDE staff (as opposed to MediaWiki volunteer developers).

Core improvements are made continuously in close collaboration with the WMF teams. Due to the integrated, collaborative nature of movement software development, it is not useful to look for improvements credited only to one team. Following are some examples of improvements based on requests from the German community that were developed with international support and have movement impact:

• CatGraph, in-memory graph database for Wikipedia category structure

The database is used by various tools, for example:

• The Render Article List Generator (uses CatGraph for searching the category tree), developed by WMDE
• CatCycle (finds cycles in the category structure, that is, categories containing one of their supercategories; it can also display root nodes and the shortest path between categories), developed by a volunteer
• Merlbot, developed by a volunteer

• Deepcat Gadget, intersection and subcategory search for Wikipedia, Wikimedia Commons and many other wikis.

International communication of the gadget will take place until the end of 2015.

• Documentation on Github and Phabricator

• Watch edits which add or remove pages from a category (merged into Mediawiki core)

Roll-out was reverted, issues were fixed, second roll-out on test-wiki planned for next week.

• Concept, Request for Comments, Documentation on Phabricator and Gerrit

• This question is directed to both WMDE and the WMF. In the FDC's 2014-15 funding recommendation for WMDE, we seemed pretty clear that we expected that the WMF and WMDE would find a funding solution for Wikidata that was outside of the FDC structure; indeed, last year was the first time that the FDC was asked to consider Wikidata in its funding decision for WMDE. Instead, this year we have received a proposal in which the FDC is asked to, essentially, fully fund Wikidata. What happened here? At what point was the negotiating ball dropped?

A joint answer will follow, until then please refer to the 'amount requested' section of the WMDE proposal.

Thanks for your question about how the funding for Wikidata is coming from the FDC's envelope, Risker. I understand your frustration in seeing this process move much more slowly than you would have liked. In early 2015, WMF leadership began exploring the possibility of funding Wikidata outside of the FDC / Annual Plan Grants Program. However, we’ve found that to be a significant change that has taken a lot of conversation to explore. And because significant transitions in WMF leadership, particularly in WMF engineering, were happening at the same time, WMF was delayed in moving this forward for the 2015/16 fiscal year. As you know, the intention was to include funding for this program in the WMF annual plan, but ultimately the annual plan timeline didn't allow for that. The only option for this fiscal then was to fund Wikidata in the way the movement has done for the last few years -- through the FDC. The change this year is of course that the FDC is being asked to consider only funding Wikidata and some of the software development costs.

Although it isn’t ideal, I do see some upsides to the current arrangement. WMDE continues to put forth its annual plan through the FDC process, opening it up for community review alongside other APG grant applicants. Community members from around the globe take part in reviewing these plans, and the FDC will soon deliberate on this proposal and make a public recommendation about this funding. I expect that will be a useful input to WMF as we consider supporting Wikidata via alternative channels for the next fiscal year.

We don't see the FDC as the best mechanism to continue supporting Wikidata, and appreciate the FDC’s guidance in this regard. My team is recommending to WMF leadership that the current approach be temporary. How else WMF might fund Wikidata is still under exploration as we go into annual planning for next year. WMF and WMDE leadership are meeting this week to discuss possibilities for scaling movement support to Wikidata. In that approach, funding for Wikidata would move out of the FDC's budget and into WMF engineering in WMF's next annual plan.

I appreciate that this is an unusual situation for the FDC: a restricted grant request for such a large technical project is not the kind of proposal you are used to seeing. This is yet another reason that we'd like to move this out of the FDC process.

Thank you for your question, Risker. Siko (WMF) (talk) 20:33, 28 October 2015 (UTC)

Thank you Siko (WMF) for answering the question. WMDE agrees and we have nothing to add. Cheers, Christof Pins (WMDE) (talk) 09:41, 5 November 2015 (UTC)

• Given the significant increase in overhead and the higher-than-average cost of living in Berlin, is there a reason why you are not contemplating outsourcing more of the development and support work related to Wikidata?

WMDE is indeed not only contemplating, but actually practicing a mixed approach to sourcing labor. The team engages staff located in Berlin, staff located in other cities and countries, as well as contract labor locally and remotely. All of WMDE’s agile engineering and product planning processes allow for and constantly utilize remote work. Utilizing contract labor does not necessarily decrease overhead or overall costs, however.

Berlin in fact is the urban center in Europe with lower-than-average cost of living: It ranks 34th in cost of living, only followed by Southern and Eastern European cities. Yet Berlin is an extremely attractive place to live for developers and other professionals. Our efforts described in the proposal under Goal 2.b. are aimed at assuring a continuous pipeline and good integration of volunteer and staff developers.

I will look forward to your responses, and thank you in advance for them. Risker (talk) 03:42, 22 October 2015 (UTC)

Thank you all for your responses. I have one further question. How much of the €1,500,000 request is for Wikidata, and how much for the community engineering project? Staffing and other costs can be combined, but the proposal is not clear as to how much funding is intended for each of those two major projects. Risker (talk) 01:27, 29 October 2015 (UTC)

Hello Risker, please refer to Table 6b in the proposal for a breakdown of activities and their cost. While we implement all of the activities of the department in an integrated manner, and do not operate within the logic of ‘two projects’, the line item for agile development of community requests from the technical wishlist amounts to €272,676. Please also note that the total cost of Software Development for the movement is over €2M, with €538,644 covered by matching WMDE funds. Christof Pins (WMDE) (talk) 17:11, 4 November 2015 (UTC)

Questions from Wittylama

Hello Wittylama, thank you very much for pointing out these areas in need of clarification! Please find our answers below after each of your questions.--Nicola Zeuner (WMDE) (talk) 14:33, 27 October 2015 (UTC)

As a clarifier/extension to Risker's question regarding examples of existing work done by the community-focused software development team: Is that what the expandable box in the 2015 progress report "Targets: Software Development" is referring to?

No, the community-focused software development team (TCB) does not work on the 2015 APG targets for Wikidata and Commons. The targets you cite are being covered by the Wikidata product team.

That box cites 3 wikidata tools which are classified as on-track/successful, and 4 Wikimedia Commons tools which are classified as "changes in prioritization". Is that the same project/team that is now listed as "Agile development of community requests from the technical wishlist" in this year's proposal, and which is valued at €272,676?

No, again, this is not the same team. WMDE software development consists of two product teams: Wikidata and community-centered software development (TCB). The work of the latter is listed as "Agile development of community requests from the technical wishlist" in the budget of the current proposal.

Can you elaborate on where/how this team will receive and triage requests from the german-speaking community because I know that this the first problem that the WMF's new Community Tech team is facing.

For a detailed description of the community-centered approach please refer to the respective section in the progress report and the concept paper. In short: Community members submit requests through multiple channels, including the technical wishlist, a poll and f2f thematic or user-specific workshops. The community prioritizes the requests through voting. Then the requests are turned into structured work packages for the TCB Team. The team works on implementing the requests. Updates on the feasibility and status of each request are provided on the wikipage of the wish list as well as on phabricator. Other requests are addressed by community members themselves.

This approach was developed in 2015 together with the communities, and expanded components (the volunteer-initiated wishlist and the poll) that had been introduced in 2013/14. The TCB product team was established and started working in July of 2015. We are currently going through the second poll (German Wikipedia project page), and the 2nd iteration of the whole process is now in progress.

As a matter of fact - can you elaborate how how this WM-De team would work with that WMF team? Wittylama (talk) 20:35, 25 October 2015 (UTC)

WMDE has always strived to collaborate with WMF Engineering & Product Departments on community requests. Regular sync meetings are established and the collaboration will continue. However, with the new WMF Community Tech Team in place, in 2016 WMDE hopes that there will be additional and deeper opportunities for the two teams to collaborate on crucial ideas and community requests with movement impact.

questions from FDC staff

Thank you, colleagues at WMDE, for submitting this proposal for a restricted annual plan grant to the FDC. Here are some questions that we have in reviewing this proposal. Given that the request is limited to software development, we have limited our questions to these areas. If you have questions, let me know! KLove (WMF) (talk) 06:51, 29 October 2015 (UTC)

Hello KLove (WMF), thank you very much for your thorough questions and the opportunity to elaborate on these topics. Please find the answers to your questions in-line below. Cheers, Christof Pins (WMDE) (talk) 17:11, 4 November 2015 (UTC)

programs

• We see you've shared general baseline metrics for Wikidata. While you've not been able to set targets for the next year, you do expect an overall growth of 10% in all metrics. At what point do you expect to set targets? How will you know if you are succeeding without targets, or is a 10% growth a target for all metrics?

The 10% growth you are referring to is not a target, but the estimated scale of growth of most Wikidata Key Performance Indicators (KPIs), based on the observed growth rates in 2015, and it is a ballpark figure. Wikidata is – like all other international large organic Wikimedia projects – a project with a living, breathing community. As a result, KPIs are based on participation and performance of volunteers, just as in the large Wikipedias. Likewise, WMF does not set a specific target for the English-speaking Wikipedia to increase the number of authors by a certain percentage. In order to assess the overall and specific health of the project, and to know whether it is succeeding, we will constantly monitor the listed KPIs. They serve to let the team know if something is going wrong, and in what area, and allow the product team to make data-based decisions as part of the agile development process throughout the year.

• When do you expect to have baselines and set targets for all of the "important indicators / metrics" included in objectives under software development?

We provided three types of metrics in this proposal: Milestones, targets and indicators/metrics. Not all of the indicators/metrics are suitable to result in quantitative targets: Some metrics e.g. “percent of Wikidata statements with a non-Wikimedia reference as of a given month” serve as general Wikidata KPIs (as described above) and are part of a set of data quality markers for Wikidata. Based on groundwork done in 2015, we expect a further increase in this metric. But no single marker alone is sufficient to draw a complete picture about data quality in Wikidata, and additionally, the priority/importance of a single marker may change over time. As a result, a single quality metric will not be turned into a target, but will be monitored and published as part of a set of several quality metrics via the dashboard and in quarterly reviews.

Other metrics and indicators, e.g. the “number of articles created or translated after article placeholder was displayed in WM projects using article placeholder extension” are part of an early stage experimental setup and will help us to evaluate the impact of our work: We expect a positive effect on the creation/translation of articles by this extension, but it is not possible to estimate the actual scope of it at this point. After observing adoption and effect of this extension in some Wikis for a few months, we will be able to predict the size of this effect for further Wikipedias adopting it.

For the metrics related to working on technical requests from the editing communities (e.g. “percentage of products/release versions which are completed as estimated” or “usages and/or impact of implemented products on the Wikimedia projects”) we will set up a quarterly review and planning cycle that includes setting targets on a quarterly basis, integrated with the agile development processes, starting in the course of Q1 2016.

For a further elaboration about the rationale of success measurement in software engineering, please refer also to the expert input of WMF’s former Deputy Director Eric Moeller, which he provided for our 2014/2015 APG proposal.

• What will be the criteria for selecting the five projects on which you expect to support new article creation through the article placeholder extension?

The focus will be on smaller Wikipedias, because they are most likely to see the largest benefit and impact from using the placeholder. We will invite a number of these smaller projects and select those with the most interest in growing content, and the highest willingness to adopt the extension.

• Does the organization intend to solicit further content donations to Wikidata, or do you instead plan to prioritize responding to offers that come to you?
• Can you offer some further context as to why you've set your target to two new data partnerships of high quality/relevance for 2016?

Answer to both questions above: Organizations and institutions offer data donations and collaboration on a regular basis. There is a high number of and a large variety in these offers, and the processing takes up a lot of staff time. WMDE’s challenge lies not in solicitation, but in strategic selection and efficient response. So the crucial factor here is not the quantity of partnerships, but their quality, as well as the relevance of the data. In order to identify which donations will add relevant items to Wikidata without compromising quality, the team needs a strategy that guides the prioritization and selection.

Consequently, as outlined in our annual plan and the proposal, WMDE will create and test a strategy and a structure for data partnerships. The objective is to be ready for the next data donations in 2016. The number of “two new partnerships” is a minimum amount placeholder. Small data donations or well prepared data donations can be easy to process, and larger, diverse ones, such as Freebase, could take up several FTEs. So the objective is rather to smartly select and efficiently process a yet unknown number of high quality data donations.

• Under Goal 2b "Assure sustainability of Berlin-based software development hub", Objective 1 focuses on engaging junior developers. WMDE includes one important indicator as mentees continued involvement in Wikimedia projects. Can you tell us what kind of participation you hope to see and how you will work to support it?

There are many potential benefits to working with mentees. In the mid-to long-term we anticipate that mentees join Wikimedia as staff, continue to contribute to WM projects as volunteers and/or act as multipliers and ambassadors for WM projects in their future work environments. Some of these outcomes have already occurred in recent years, and we now wish to formalize and strengthen this work to sustain the impact, and to assure the pipeline of young developers who will work for the movement. Some of the activities (either in place or under consideration) to maintain relationships are: Regular events and f2f get-togethers, travel support for movement events, a structured exchange among alumni (via IRC and google groups), as well as staff personally staying in contact with their former mentees (for example through including them in the notorious WMDE staff karaoke evenings....).

budget & finances

• We need the 2015 full year budget projections for Table 2 to understand your projections for 2015 (current grant year).

Pls. refer to the budget table below:

Updated Revenues and Expenditures Table for 2015

Financials for the current funding period
Revenues or expenses	Planned (budgeted)		Actual, until 30 September 2015		Projected
	EUR	US dollars	EUR	US dollars	EUR	US dollars
Revenues (from all sources)
Donations	2,150,000 €	$2,411,849	2,127,089 €	$2,386,147	2,127,089 €	$2,386,147
APG Grant	840,000 €	$942,304	840,000 €	$942,304	840,000 €	$942,304
Membership dues	1,100,000 €	$1,233,969	1,039,955 €	$1,166,611	1,121,801 €	$1,258,425
Carry-over from 2014	30,000 €	$33,654	30,000 €	$33,654	30,000 €	$33,654
Fundraising reimbursement WMFG	150,000 €	$168,269	150,000 €	$168,269	150,000 €	$168,269
Federal Contract OER1	0 €	$0	113,616 €	$127,453	451,040 €	$505,972
Unrestricted Grants2	500,000 €	$560,895	69,360 €	$77,807	105,000 €	$117,788
Other Revenues	100,000 €	$112,179	63,218 €	$70,917	63,218 €	$70,917
Wikimedia Conference	100,000 €	$112,179	108,882 €	$122,143	108,882 €	$122,143
Total revenues	4,970,000 €	$5,575,296	4,278,504 €	$4,799,583	4,997,030 €	$5,605,618
Expenses
Software Development (APG)	1,250,000 €	$1,402,238	872,595 €	$978,868	1,200,511 €	$1,346,721
Non-APG Programs	1,310,000 €	$1,469,545	690,764 €	$774,892	1,517,182 €	$1,701,960
Adminstration & Program Support
Board, ED, Operations, HR, IT3	1,505,000 €	$1,688,294	1,042,428 €	$1,169,385	1,414,502 €	$1,586,774
Communication	380,000 €	$426,280	235,600 €	$264,294	334,150 €	$374,846
Partnerships and Development	220,000 €	$246,794	153,606 €	$172,314	208,606 €	$234,012
Event Management	185,000 €	$207,531	167,610 €	$188,023	197,165 €	$221,178
Operating Reserves4	120,000 €	$134,615	0 €	$0	120,000 €	$134,615
Total expenses	4,970,000 €	$5,575,296	3,162,603 €	$3,547,776	4,992,116 €	$5,600,106

Table notes:

¹ A restricted grant awarded in February.

² The planned amount was not achieved.

³ These amounts differ slightly from the annual plan, because they do not yet consider re-allocations of admin cost to project/departmental budgets done at the end of the year.

⁴ The financial cushion will be spent to cover budget shortfall.

• Can you clarify the total costs for Wikidata for WMDE in 2015 and requested for 2016, including staff costs and administration costs as distinct from other programs included in software development?

The cost of Wikidata is not separable from that of other ‘programs’. Wikidata cannot be done without the community work. Wikidata itself, as a volunteer-driven project, is per definition community engineering. The community-centered software development approach and product team grew out of our successful experiences with the Wikidata community, and serves to support Wikidata, as well as related improvements to MediaWiki. Specifically:

• Even though we have two product teams, the work of the two teams is continuously integrated through joint management, floating personnel, and linked agile processes.
• Improvements to MediaWiki based on community requests are also often related to Wikidata or work towards a better integration of Wikipedia and Wikidata.
• The FOSS outreach activities serve to build the human resource foundation for Wikidata and other development activities.
• For a breakdown of the activities and related cost estimates for 2016, please refer to table 6b in the proposal. Direct development work on Wikidata core and features is largely covered by the first eight activities.
• Please note that more than 25% of the overall cost of €2M is covered by WMDE matching funds.

• Can you give us a high-level sense of your fundraising plan to support Wikidata in the longer term?

The fund development strategy for Wikidata will most likely include a mix of WMF, WMDE and carefully selected external funds. Details are still subject to negotiation with WMF as part of the Wikidata Business Case, and it would be premature to provide more detail at this time. Please also refer to Siko’s statement above.

• You note that administration costs are 448,370 euros out of the 2 million euros. Is this a standard 22% administration charge for all programs at WMDE? If not, how was this number calculated?

WMDE’s actual administrative overhead rate for 2016 is 28%, which amounts to €448,370 calculated on the basis of the €1,590,274 actual program cost. The rate is calculated annually by dividing the administrative cost by the total annual budget. This rate is not particularly high when compared to the average rate of nonprofit organizations the size of WMDE. For more information on evaluating a nonprofit based on its financial ratios, see here and here.

• Can you tell us more about the "Floating Capacity" line item, set at 119,778 €? The note reads "Management, coordination and developer capacity that is flexibly allocated to activities as needed throughout the year." Has WMDE had a floating capacity budget line item before? How is it allocated, and what kinds of expenses might be included there?

What we call floating capacity reflects the budgetary/personnel side of agile planning processes. Floating capacity allows us to allocate personnel resources flexibly to the product teams as workloads shift, tasks arise and sprints start and end. The FTEs covered here have functions and tasks across teams, products and projects. In addition to .5 FTE developer, this line item includes some of the departmental administrative and management capacity, including FTE portions of the head of department and his assistant, as well as non-personnel funds for travel, administration and hardware. In the past, this cost was allocated to projects proportionally, and did not show up as a line item. Showing this as a separate item more realistically reflects how personnel and other resources are allocated in an agile manner across integrated projects.

staffing

• WMDE is requesting funding to increase its FTE by 4.5 people with APG funds for software development work. Can you share more about these positions and the increase in 4.5 FTE?

Actually, the budget states that APG funds would pay for an increase in 2 FTEs in the WMDE Software Development Department: 1 FTE Project Manager Community Communication Wikidata, and 1 FTE Engineering Manager who will be responsible for the development and management of the engineering teams.

The remaining 2.5 FTE are NOT part of APG. They are paid out of WMDE’s other revenue sources and result from the transfer of 2.5 FTEs from WMFG - the Wikimedia Fundraising Corporation- to WMDE’s Software Development. These developers are providing support to the fundraising campaign.

Many thanks, WMDE, for your responses. Cheers, KLove (WMF) (talk) 06:51, 29 October 2015 (UTC)

Thanks again, KLove (WMF), for your questions. Please let us know if anything remains unclear. Cheers, Christof Pins (WMDE) (talk) 17:11, 4 November 2015 (UTC)

Subject-matter expert feedback regarding software development programs

Wikimedia Deutschland has requested subject-matter expert input on this proposal, specifically related to the chapter's software engineering work. This assessment is based on my leadership role with product/engineering initiatives at the Wikimedia Foundation, a team summary review comprised of project managers and engineers from WMF, and the working relationship between the Wikimedia Foundation and Wikimedia Deutschland software development on the Wikidata project.

What is your overall reaction to the plan as proposed?

The overall request is respective to the to challenges the Wikidata team is facing as the project begins to scale. Wikidata is strategically aligned with both WMF goals, mission and long term focus. Wikidata is a solid team with strong software development expertise and would benefit from having the relevant data expertise, and engineering management positions filled so they can mature the project.

Do you believe Wikimedia Deutschland is qualified to undertake this?

The Wikimedia Deutschland software development team is qualified to continue development.

What are the major strengths of the plan, from your perspective?

The Wikimedia Deutschland software teams use common product and engineering development standards and execute well in both areas. Additionally they do a good job with dependency discussions and collaboration across foundations and with the community. So the technical assessment is that they have a good level of competency.

What are the major weaknesses of the plan, from your perspective?

There are a few costs associated with "Floating Capacity", international and FOSS requests that are not clear enough in the capacity to directly support specific development initiatives tied to the Foundation. The costs seem high for the long term impact to the projects.

If the plan were executed, what if any suggestions would you have for the Wikimedia Deutschland, that you feel would increase their likelihood of success?

Focusing on the completion of the ability to track performance of Wikidata and the movement of the core engagement metrics as measures of success against the primary development goals. The focus on quality and intake are good and will require solid understanding to know the impact of the work.

If the plan were executed, do you think it would make a small, medium or large contribution to the goals of the Wikimedia movement, and to its ability to achieve global impact?

Large. The spread of structured knowledge to a diverse audience will help internal features as well as community driven project development. Strengthening items from the community wishlist will also help improve additional software used by other development teams.

Does the plan propose to break new ground in ways that could create important learnings for the global Wikimedia movement? In other words, if the plan is executed, will the Wikimedia movement have learned anything new and important that could be applied outside this plan, that would help it achieve its global goals? Would that learning be a small, medium or large contribution?

Large. The contribution to editing and new types of content creation are positive. The usage of Wikidata is also being explored as a tool to increase relevancy in search and improve the conveyance of additional relevant results from across projects. There are benefits to growing smaller projects and potential inclusion of more sources of knowledge by having a focus on improving quality.

Do you believe the budget (for the area that you are particularly evaluating) is feasible?

Yes, in general the other costs line up with comparable software development projects, however I do have concerns around some of the areas related to "floating capacity", which is not well defined and should be provided for in other areas.

— The preceding unsigned comment was added by WMoran (WMF) (talk) 19:23, 10 November 2015 (UTC)