Wikimedia Hackathon 2017/Lessons Learned

There are lessons learned from the Wikimedia Hackathon 2017 which took place in Vienna, Austria. This page is a combination of feedback from a survey sent to participants and ideas from the organizers.

This page will focus heavily on improvements for next time and things that went poorly. However remember that 98% of the participants that responded to the feedback survey stated that their overall experience with the hackathon was either "Positive" or "Very Positive."

Suggested changes for the next Wikimedia Hackathon

• Collect interests and information from newcomers to distribute to mentors in advance. Newcomers don't need to by the only ones approaching, mentors can approach newcomers in advance too!
• Extend MediaWiki installation session, hold more than one
• If there are pre-hackathons in the future publicize them more and put all the details about all the events in one place

The best!:

• This evolution of the mentoring program shows that we are heading in the right direction with newcomer support
• The people
• The atmosphere
• The venue

The worst!:

• Not enough sessions teaching skills, most sessions were lecture of discussion formats
• Room names were to hard to remember or say for non-german speakers
• Need more types of social events for people who don't like to drink / are more extroverted / like quiet activities

Feedback survey background

• rfarrand created the feedback survey using google forms based on feedback survey questions from past events.
• The Feedback Survey was sent out to participants on May 23rd, 2017
• The Feedback Survey closed on June 27, 2017
• The survey received 99 responses (41% response rate)

Considerations for next year

This section is based on the fill-in-the-blank/comment sections of the feedback form. Some of these comments will be contradictory. We are mostly trying to include common themes and issues that were felt by groups of people. This section is a bit more subjective than the #Data section below and suggestions may be paraphrased and combined with each other.

Mentoring Program and Newcomers

What worked well with the mentoring program?

• Dedicated / competent / friendly mentors who helped to form meaningful connections with newcomers
• Mentors who could not help easily redirected newcomers to other mentors / mentors worked well together to get the newcomers the best match
• Newcomers completed projects / work got done
• Good mentor to newcomer ratio
• Mentor expectations were clear

What should be improved about the mentoring program?

• Fewer meetings and check-ins, time at the hackathon is valuable
• More support for people arriving late (both mentors and newcomers). Somewhere for newcomers to go who are looking for a new project mid-event. Stress that newcomers can change projects / don't have to commit to one project for the entire event.
• More support for introverts / people who are not outgoing.
• Better ways to identify by mentors and newcomers
• Get newcomers on IRC (this is how Wikimedia technical communities communicate) / don't use telegram to coordinate because it is proprietary and some people don't want to download things onto their phones
• More / different advanced communications and expectations for mentors: Mentoring mailing list to get mentors familiar with each-other and each-others skill sets. Mentors should meet up the day in advance of the hackathon in person. Mentors should know more about the needs of the newcomers in advance of the hackathon.
• Keeping mentors / newcomers in a room separated from others did not encourage mixing with other contributors, find ways to engage other event participants to help newcomers even if they are not committing to being a mentor
• Don't spend so much time in the newcomers session having everyone talk to the group, writing what you are working on and then having 1:1 conversations is a better use of time
• Re-do poster session each morning / make newcomer & mentoring work visible to the rest of the event

Questions for Newcomers

How did you hear about and become interested in attending the Wikimedia Hackathon 2017.

• mailing list (wikitech-l)
• New WMF staff
• Wikimedia Austria
• pre-hackathon
• WMDE
• Wikipedia, meta, a wiki
• through my work
• Through Wikicite
• phabricator
• Wikimedia Ukraine
• Wikimania
• I don't know

How did you experience the amount of new information you got? (e.g. I wanted more, I felt overwhelmed, I followed along well)

• I followed along well / good / great (most)
• OK/ Fine
• Overwhelming
• Lots of info from talking to people
• Publish information earlier
• More info about registration deadline next time
• It made me curious

How did you experience the social atmosphere?

• Very good / or other very positive words (25+)
• Fine
• sometimes difficult to make contact with people, as if not interested +
• people were nice, but the event was not integrative to non-programmers

After attending the hackathon, would you feel prepared enough to keep working on MediaWiki projects?

• Yes (9)
• I get paid to do this
• Not my intention
• I was already prepared but now I have new tasks
• Not yet
• Somewhat more than before 

Did you feel accomplished at the hackathon? If so, why and If not, what might have helped?

• Yes (14)
• no (3)
• Yes and no / some
• I wanted to help newcomers but did not want to commit as a mentor

What did you find most confusing about the existing Wikimedia technical documentation?

• There is too much
• where to find things / time consuming
• Bad Landing pages/entry funnels/ overviews
• Too complicated
• not for beginners
• There is no single point of truth.
• no coherence
• I didn't look at that documentation
• Outdated information
• Missing OOUI tutorials

This answer is a copy paste of all answers without grouping them in order of most mentioned because they were all pretty different

• There is too much.
• where to find things (as usual)
• Tons of Wiki-pages, phabricator-tickets....
• Nothing
• not for beginners
• Landing pages/entry funnels. we put too many things in giant list form with too little differentiation.
• Still pretty scattered. Few, and "bad" examples.
• it is hard to find the bit of information you are looking for. sometimes it is on your wiki, sometimes on Meta, sometimes on MediaWiki.org etc.
• There is no single point of truth.
• no coherence
• How complicated it is to fetch image + thumbnail + coordinates via Commons API
• I didn't look at that documentation
• Lots of old documentation exists and doesn't not that it's wrong; the process to change things (e.g. merge code) doesn't automatically change the documentation on-wiki.
• if I look for an short answer - I get also a lot background infos on the actual topic
• That I have to dive into the source way too often.
• Almost everything :) In particular, Lab tools docs.
• That people seem to expect that it exists at all. No really. This is a bazaar space -- not a cathedral. User documentation is essential, but technical documentation is a high price. It's good that we cut paths through the forest, but I don't see why we need to build freeways of documentation for developers who are working to build new and different things anyway. I think we should focus on *literacy* for our newcomers. They should know how to *find out* about anything. We shouldn't be setting the bar at "wheres the docs". Instead it should be "do you know who to ask and how to find them?"
• OOjs UI is still hard to understand and write
• I did not touch any technical documentation during the hackathon at all
• That I did not quickly find a good general intro/overview into what there is available and where to start and where to go from that
• Missing OOUI tutorials
• There is no clear hierarchical structure for navigation/links to sub-topics or technical aspects of an extension or a module; at least, it's not easy to find. Some standardization of what must be included (templates with pre-defined fields or something) and of the overall content structure should be strictly enforced. It might be helpful to have for documentation something similar to OOjs or UI standards/guidlines?
• The layout is not very engaging. Tasks often felt like a very long list and main topics / focus areas were not very clear.

Which Wikimedia technical documentation pages were the most useful for you?

• What Wikimedia technical documentation did you want to find, but couldn't find?
• Start to finish tutorials on (things) is most missed. The sort of docs that a developer advocate would write.
• details on videos for commons
• Lab tools—logging, package installation, etc.
• OOUI tutorials
• The documentation for AbuseFilter is in place but it's not well structured; the info for FlaggedRevs documentation is spread out between three links that do not refer to each other in a clear way.

Pre Event Hackathons

Any comments about the pre-hackathons or suggestions for pre-hackathons during future years?

• Many people did not hear about the pre-hackathons until the event itself. They would have liked to get involved.
• Connect the work done at pre-hackathons more closely with the work to be done at the main hackathon

Programs and Collaboration

How could we improve the volunteering program?

• Existing working groups before event from WMF / WMDE
• Don't use so many Phab sub-tasks
• Explain the volunteering roles more during registration including an expected time commitment  

Opening / Closing Sessions

• Cluster people who need browsers with links, people who don't need computers at all (encourage this) and people who need to connect their laptops, no live demos
• Opening is to long / boring (more info on wiki, less talked at people) / keep it interactive and energetic
• Showcase to long. Reduce applause time, concentration is hard for that long
• Showcase template, mentor people in advance for how the showcase should work, 1) What was the problem you were solving (from a user perspective)? 2) How did you solve it.
• Beamer switch / way for next presenter to connect before their turn
• Add a break to the showcase / closing
• Mentoring intro in the same room as the opening
• Restrict time per person during open mic
• How can participants ask questions to showcase presenters?
• More prep and intro for the structure of the showcase
• Showcase before closing

Sessions/Schedule

Are there any changes to the sessions that you would recommend for next year?

• Installing Mediawiki takes longer
• Allow people to extend sessions.
• Add a keynote “inspire with technological updates outside MW but well within the technological ecosystem (PHP, Javascript, Jenkins, ...)”
• I wanted to attend more of them
• More workshop sessions, learning sessons
• As usual: people had a hard time choosing between working and sessions
• Logistics

Comments on event logistics

There were no common themes here, each comment except for the top two waere only mentioned by one person. Even still, I am including them all for future consideration

• More vegan options, More food options in general
• More social events that are not about parties / social activities for quite people
• Room names were hard to remember and say
• All physical info should be found in one place instead of spread out
• The video projectors were resolution limited
• Allow for paper and plastic recycling
• Hotel registration was to slow
• Have a “welcome” party on the first evening
• Registration confirmation by email would have been appreciated
• People had to stand for the opening and closing
• The wifi was a problem 

Accomplishments

What, if any, coding activity/activities were you able to finish during the hackathon?

What other coding activity/activities did you begin during the hackathon?

What was your biggest accomplishment during the hackathon?

• What is the project that impressed you the most?
• Live editing on visual editor
• VisualDiff
• Mediawiki in Docker
• The audio recording tool for wiktionary.
• Hard to say. A lot of them were really great
• Gender Gap Dashboard
• Maps drawing tools
• locator-tool (https://phabricator.wikimedia.org/T165994)
• Android app 3D-model viewer
• Collaborative editing in VE
• I have to admit its again the Newsletter extension (sorry for this, but we were on this for like 2 years).
• TPT's SPARQL to SQL, many things in the showcase I can't recall due to the amount of cool things.
• The telegram translation bot
• MediaWiki in Docker
• Either the Telegram bot or the Wikibase documentation, both something I will use.
• Matching trending news articles to popular wiki articles
• KartoEditor
• Really hard to say, but if I really had to choose one I'll go with the ORES project.
• Collaborative Editing in VE, Visual Diff, New Android app for Commons, Commons games, blocking notifications from specific user.
• WikiFab.org
• Wikia development by User:TK-999. It's promising for a renewed cooperation upstream/downstream.
• Wikimedia Commons Android App
• Item quality prediction in Wikidata!
• Collaborative editing in VE probably but so hard to choose one.
• Flesch-Readability- https://phabricator.wikimedia.org/T155843
• openknowledgemap.org
• locator-tool geocoords in databases
• MW2SPARQL
• The Kartographer editing tool.
• There were several
• VE collaborative editing, it just keeps getting closer to being a reality that will let the Foundation use google docs less.
• harassment blacklist sounds like something really helpful in the communities
• so many!
• Many nice ones, and I'll just pick the first one that comes to mind: the user blacklisting project
• KartoEditor https://phabricator.wikimedia.org/T165984
• Real-time collaborative editing in VisualEditor
• ORES
• locatortool
• Collab thingie
• Visual diff
• simultaneous visual editing
• Make wiki pages translatable
• THERE WERE SO MANY WHAT
• GLAMs ... but unfortunatly I missed the session because I had another meeting
• Collaborative editing.
• Reference support in visual diffs (in one day!); OOJS UI widgets in Vue
• The newsletter extension.
• KartoEditor looked pretty cool
• The etherpad meets Wiki editing
• real time collaborative editing
• https://www.mediawiki.org/wiki/User:Khorn_(WMF)/Plantdata.io
• Sentiment analysis using ORES
• open knowledge maps
• Real-time VisualEditor

Anything else?

What did you like most about the hackathon?

• The people / meeting people / in person / working together (25)
• Relaxed / motivated atmosphere (10)
• Hotel and hacking space together in the same location / Venue (8)
• Mentoring program / focus on newcomers (7)
• Getting stuff done (4)
• The organizers/organization/support (4)
• Showcase (3)
• Table football (2)
• wifi (2)
• Telegram
• Learning
• The balance of attending sessions / getting work done
• Kids at the venue
• badge stickers
• Dance party
• Info desk
• Close to home

What did you like least about the hackathon?

• no time to explore the city (3)
• Too short / Not enough time to do everything (3)
• Arena party (3)
• Food (3)
• Long showcase (2)
• There should be more varied social events (3)
• Fire alarm
• Saturday night rain
• To many people / to many things going on
• Introduction
• Pizza
• scattered rooms made it hard to find people
• I could not attend more sessions
• I didn't get to talk to everyone I wanted
• Telegram
• Accommodation should have been available for Thursday and Sunday nights
• Some wi-fi issues

Data

How would you describe your experience as a developer?

I am not a developer	17	17.2%
Novice	7	7.1%
Beginner	6	6.1%
Intermediate	27	27.3%
Advanced	42	42.4%

Including this event, how many Wikimedia hackathons have you attended? This can be any Wikimedia Hackathon of any size.

1	26	26.3%
2	18	18.2%
3	12	12.1%
4	7	7.1%
5	7	7.1%
6 or more	29	29.3%

How would you describe your overall experience at the Hackathon?

Very Positive	72	72.7%
Positive	25	25.3%
Neither Negative nor Positive	2	2%
Negative	0	0%
Very Negative	0	0%

Mentoring Program and Newcomers

Did you participate in the mentoring and newcomers program

Yes, as a newcomer	7	12.3%
Yes, a mentor	9	15.8%
I helped in another way	7	12.3%
No	33	57.9%
Other	1	1.8%

If you participated please rate your overall experience with the mentoring program

Very Positive	9	40.9%
Postive	7	31.8%
Neither Negative nor Positive	5	22.7%
Negative	1	4.5%
Very Negative	0	0%

How useful did you find the sessions for newcomers? (a question for newcomers)

Completely Useful	2	6.1%
Mostly Useful	5	15.2%
Moderately Useful	2	6.1%
Somewhat Useful	0	0%
Not at all Useful	0	0%
Did not attend	24	72.7%

Pre Event Hackathons

Did you attend one of the pre-hackathons?

Yes, Greece	1	1.8%
Yes, Romania	0	0%
Yes, Hungary	2	3.5%
Yes: Czech Republic	0	0%
Yes, One or more of the pre-hackathons in Vienna	3	5.3%
No	51	89.5%

If you attended, has the pre-hackathon fulfilled your expectations?

Yes	5	17.9%
Not sure	1	3.6%
No	0	0%
I did not attend	22	78.6%
Other	0	0%

If you attended, did the pre-hackathon lead to improved results of your Wikimedia Hacakthon 2017 experience?

Yes	4	13.3%
Not sure	1	3.3%
No	1	3.3%
I did not attend	24	80%
Other	0	0%

Programs and Collaboration

How useful did you find the skill share board?

Completely Useful	5	6.6%
Mostly Useful	8	10.5%
Moderately Useful	10	13.2%
Somewhat Useful	15	19.7%
Not at all Useful	1	1.3%
I did not use it	37	48.7%

How was it to find people to work with?

Very Easy	17	23.9%
Somewhat Easy	32	45.1%
Neither Easy Nor Difficult	15	21.1%
Somewhat Difficult	7	9.9%
Very Difficult	0	0%

Opening / Closing Sessions

Hackathon Opening Session [How useful were each of the following sessions? ]

Not at all	1	1%
Somewhat	5	5.1%
Moderately	11	11.1%
Mostly	33	33.3%
Completely	36	36.4%
Did not attend	13	13.1%

Hackathon Closing/Showcase Session [How useful were each of the following sessions? ]

Not at all	0	0%
Somewhat	5	5.1%
Moderately	8	8.1%
Mostly	26	26.3%
Completely	49	49.5%
Did not attend	11	11.1%

Sessions/Schedule

How satisfied were you with the selection of breakout sessions?

not satisfied: 1	0	0%
2	2	2.3%
3	24	27.6%
4	36	41.4%
very satisfied: 5	25	28.7%

Logistics

The food at JUFA [How would you rate:]

Excellent	38	38.4%
Good	42	42.4%
Fair	13	13.1%
Poor	6	6.1%
Very bad	0	0%
N/A or didn't attend	0	0%

Wifi quality at JUFA [How would you rate:]

Excellent	44	44.4%
Good	41	41.4%
Fair	13	13.1%
Poor	1	1%
Very bad	0	0%
N/A or didn't attend	0	0%

Breakout room equipment  [How would you rate:]

Excellent	30	31.6%
Good	40	42.1%
Fair	7	7.4%
Poor	0	0%
Very bad	0	0%
N/A or didn't attend	18	18.9%

This event overall [Ratings of the Hackathon Logistics]

Very Satisfied	77	77.8%
Satisfied	22	22.2%
Neither dissatisfied or satisfied	0	0%
Dissatisfied	0	0%
Very Dissatisfied	0	0%
No Answer	0	0%

Overall organization of the hackathon [Ratings of the Hackathon Logistics]

Very Satisfied	80	80.8%
Satisfied	19	19.2%
Neither dissatisfied or satisfied	0	0%
Dissatisfied	0	0%
Very Dissatisfied	0	0%
No Answer	0	0%

Organized social activities [Ratings of the Hackathon Logistics]

Very Satisfied	52	53.1%
Satisfied	27	27.6%
Neither dissatisfied or satisfied	9	9.2%
Dissatisfied	1	1%
Very Dissatisfied	1	1%
No Answer	8	8.2%

Availability of space for informal discussions  [Ratings of the Hackathon Logistics]

Very Satisfied	77	78.6%
Satisfied	18	18.4%
Neither dissatisfied or satisfied	2	2%
Dissatisfied	0	0%
Very Dissatisfied	0	0%
No Answer	1	1%

Accomplishments

What different tasks did you accomplish during your hackathon time?

Improved a Community Wishlist task	21	21.9%
Wrote code for a new tool or project	36	37.5%
Added features to an existing tool	44	45.8%
Fixed bugs	49	51%
Created templates	4	4.2%
Answered research questions	22	22.9%
Worked on  research / analytics	21	21.9%
Agreed on a solution to an old problem	44	45.8%
Discussed future plans in a formal meeting	45	46.9%
Other	24	25%

Anything else?

Attending this event was worth my time [Please tell us how much you agree or disagree with the following statements.]

Strongly Agree	79	80.6%
Agree	15	15.3%
Neither agree or disagree	1	1%
Disagree	1	1%
Strongly Disagree	2	2%

The opportunity to meet fellow developers and Wikimedians was valuable to me [Please tell us how much you agree or disagree with the following statements.]

Strongly Agree	83	83.8%
Agree	16	16.2%
Neither agree or disagree	0	0%
Disagree	0	0%
Strongly Disagree	0	0%

I would like to attend this event again next year. [Please tell us how much you agree or disagree with the following statements.]

Strongly Agree	74	75.5%
Agree	16	16.3%
Neither agree or disagree	6	6.1%
Disagree	2	2%
Strongly Disagree	0	0%

See also

• Wikimedia Hackathon 2016/Lessons Learned
• Wikimedia Hackathon 2015/Lessons Learned