பயனர்:Asaf (WMF)/Community Capacity Development/On-wiki technical skills survey results

This page presents results from a survey about on-wiki technical skills conducted in the Tamil Wikipedia community, Nov 4th through 11th 2015, in the context of the Wikimedia Foundation's Community Capacity Development pilot project.

47 responses were received, of which 85% were male. 50% have been contributing to Wikimedia projects for longer than three years.

Most respondents have either limited or no skill in common programming languages. Those who do have more than basic skills tend to be familiar with Wiki template programming (33%) and JavaScript (25%).

Among the choices presented, the most common on-wiki technical activities are using Wikidata for interwiki links to articles in other languages and using/inserting templates.

Among the choices presented, the most common off-wiki technical activities are Using tools from Tool Labs and “Programming other technical tools”, both done by about 13% of respondents at least once a month. Overall, 4% indicate that they engage in some off-wiki technical activity at least once a week.

Respondents mostly have basic familiarity with Wikidata (46%). Some don't know Wikidata at all (18%). Some contribute directly to Wikidata items (filling in labels and property values) or Wikidata ontology (26%).

Lua, a scripting language available for use on-wiki since 2013, is regularly or occasionally used by 21% of respondents. The rest are not familiar with Lua programming or they do not find it useful (78%).

Few respondents are involved in the global technical Wikimedia community (15%-20%). However, a large majority is receiving technical news via the village pump (74%). A few are subscribed to some technical mailing lists (16%).

Regarding preferred training topics, individual preferences tracked quite closely with individual assessments of what would be most useful to the community at large.

The training topics most interesting to respondents are Bot programming for beginners (72%), An overview and tutorials on many existing tools (68%) and Wikidata ontology -- items, properties, values, units (64%).

The training method preferred by a large (82%) majority is Workshops with frequent hands-on exercises. Other preferred methods are Self-study using written material (74%) and Self-study using training videos (73%).

Regarding times of availability for training, most indicated their preference Weekday evenings and Weekend full days.

These are some general additional comments provided by respondents (most appearing only once):

• Frustrated with the display of errors when templates are copied and pasted from the English Wikipedia.
• More software localization wanted, including the introduction of spelling/grammar checkers, OCR and dictation technologies for Tamil wiki projects.
• Requested that training sessions be customized to the Tamil community instead of offering generic training.
• Suggested assembling a dedicated set of community volunteer programmers to regularly interact with WMF technical experts.

• The responses gathered suggest that most of the respondents who do not regularly perform technical activities lack the requisite skills to do so.
• Respondents expressed diverse preferences for both personal and community training activities. However, cross-tabulation of these differences suggests that the most preferred training topics are: An overview and tutorials on many existing tools, Bot programming for beginners, Statistics and data mining using Quarry and Wikidata Query (WDQ), and Wikidata ontology -- items, properties, values, units.
• Knowledge of programming languages is relatively rare, and the existing programmers have skill in disparate languages. This could be a challenge for an effective bot programming tutorial in the short time available.