Grants talk:PEG/User:Kipala/Swahili wikipedia longtime editors' meeting and workshop

GAC members who support this request

1. I personally find this a very interesting idea, also as a pilot project to expand community and outreach in the next 2-3 years in an area which has huge potential in terms of culture and human resources. My only concern (and rather a marginal one) is due to the fact that I have no idea on the cost of photocopies in Morogoro and of the volume of handouts you need to print :) We were actually discussing this with AWang: leaflet/handout volume is often overestimated. Can you give details on that, just to give us a complete overview? --Dry Martini (talk) 18:36, 30 January 2015 (UTC)
2. based on the Global South strategy, inability to use online connectivity tools (Skype, etc.) and, per my personal opinion, good chance of successful workshop at school, I would support this request rubin16 (talk) 14:18, 31 January 2015 (UTC)
3. Polimerek (talk) 10:15, 4 February 2015 (UTC)
4. Violetova (talk) 14:45, 5 February 2015 (UTC)
5. --Ilario (talk) 13:16, 6 February 2015 (UTC)
6. --109.131.99.226 --DerekvG (talk) 16:45, 6 February 2015 (UTC)

GAC comments

If I may make following suggestion to the authors : perhaps you could also discuss outreach issues, aims and strategies on how to get in touch with your reader base and get them to contribute, and creating their interest in contributing. --DerekvG (talk) 02:15, 28 January 2015 (UTC)

This is surely going to be a topic. That is why we hope to get the offline version. I do not expect big jumps forward because we have to be realistic about restraints any African language faces for written communication (monopoly of European languages in higher education), together with the restraints stemming from lack of infrastructure (around 20% of Tanzanians -our main "market"- have connection to electrical grid). Estimates of Internet access are 15% of Tanzanian population which looks high to me, and is practically restricted by poverty and connectivity. We discusssed among ourselves thru the years for whom we are doing it, and we stay on this work because we see things are changing: internet access is rising, Swahili use in population too (Swahili replacing the 100+ local languages).

So our priority is to continue creating quality content which will find its users. We have confidence that our work is worthwhile. Our readers are waiting to discover us. To get a foothold in a school environment is a great chance to get some more editors (5 more permanent editors would double our number!) and have feedback about what "regular people" see as useful and valuable.

We do have edits from casual contributors again and again, by general wikipedia standards of low quality (which is nothing to be astonished about). We have built the largest number of locality stubs for Tanzania among all wikipedias and this brings a permanent trickle of contributors adding some info about the place they are from - mostly unregistered editors with whom it is difficult to communicate. As we say in German "Mühsam nährt sich das Eichhörnchen" (roughly "the squirrel has to labour a lot for its food") but we go step by step. But still: by African language standards 2500+ page views per hour are encouraging for us. In some months the majority of views still come from the USA (Africans abroad, probably plus courious US - "African heritage fans" ) but the situation on the ground keeps changing. That is why we put "enhancing quality" as top priority. Kipala (talk) 09:18, 28 January 2015 (UTC)

Hi Kipala, I'm glad for your reaction, and I'm aware of the connectivity issues, young people who don't have access to a personal internet link (but rather a time-paid link (like internet cafe) are less inclined to use their time to contribute, they search material and use internet to communicate long distance. that is why I wanted to encourage you to think about other strategies to reach your public. I think you may want to stress that valid aim in your budget submission because it does have additional output value. I wish you every success in your endeavours. I'm close to the african community in belgium where a lot of people live form around teh great lakes, some of those people speak swahili, I will pledge my support also by promoting wikipedia in Swahili and encourage them to contribute to the swahili wikipedia from here in europe for the benefit of the readers in Africa--DerekvG (talk) 12:18, 30 January 2015 (UTC)

offline version

Hi, guys! Have you considered Kiwix? There are already ZIM files (database) available for your Swahili Wikipedia - [1] rubin16 (talk) 20:10, 28 January 2015 (UTC)

• Thanks for pointing that out! It works. Unfortunately without images, but still! Kipala (talk) 11:10, 29 January 2015 (UTC)

• Frankly speaking, it should have images - I downloaded it myself to test and I have images in the articles. Probably you should write to the developers rubin16 (talk) 20:18, 29 January 2015 (UTC)

OK there were 2 versions, got it! Thx Kipala (talk) 15:03, 31 January 2015 (UTC)

quantity vs quality

First of all, I don't want to make you feel that I am saying that something is wrong, I just want to share some of my thoughts. Why do you want to restrict articles' creation? Why do you want to restrict new editors not only from creation at all but also want to make them work on some pre-selected topics but not topics that are interesting to them? From my personal point of view, every Wikipedia at the beginning of its big life starts from quantity - lots of new articles, lots of new editors that are posting texts about things that are interesting to them - pop-stars, sex, cars, etc. Having 1000 most important articles about physics, literature, etc. is really good but why don't you give people to get fun from editing? One of the first articles in Russian Wikipedia was devoted to Russia itself and firstly it contained the text translated as "Russia is the motherland of elephants". Somebody got fun and that text existed for quite a continuous time but it motivated to fix errors, to edit on your own. Now Russian Wikipedia has >1mln articles. So I am personally not supporting any restrictions on participation at the moment of initial development: some years later you'll get lots of editors and your quantity will turn to quality, but now... let them edit as a fun rubin16 (talk) 20:18, 28 January 2015 (UTC)

I am not sure if I get the point about "pre-selected topics". There is no limit whatsoever for topics any new editor wants to take up (not intended nor possible). I guess you react to "risk 1". I apologize that is not clearly worded. It refers only to the outcome of the intended discussion of our "longtime team" with teachers and the possibility that they tell us about topics they would like to see in order to use the content in their context. The "risk" is that we may not be able to react to that, because nobody of us might really have much interest in the topics they would like to see. Ideally we should be able to react to proposals from the school people we invite and try to get on board (maybe more biographies of people who fought for independence - fits into Swahili school curriculum as far as i remember). This would make a very "efficient" outcome for use of a grant in this proposal. (More ideal of course is every participant who starts to text him/herself - but probably on "their" topics and not on the "useful" ones.) Did this answer your concern? Should I reword the text? Kipala (talk) 21:07, 28 January 2015 (UTC)

Yes, my question came from the list of your risks, now I understand you better, thanks for comments rubin16 (talk) 20:00, 29 January 2015 (UTC)

budgeting

And some more comments... work of GAC wouldn't be so essential if we wouldn't be challenging budgets :) from the grants' proposal I didn't get full understanding of the upcomping meeting of Swahili contributors: do you have any agenda? What topics will you be discussing and what solutions are you looking for? Wouldn't be Webex or Skype more efficient in terms of personal communications, especially bearing in mind your schedules of business trips and so on? From my personal experience (of being Wikimedia Russia member), it's not efficient to meet just to see each other. We try to prepare agendas, to prepare questions and possible solutions and at the meeting we just discuss possible options and somehow "vote" on them: all other things could be done via emails, skype, hangouts, whatsapp, etc. When we came to the meetings without agendas and preparation (that happened before) we could speak for hours, stay till midnight but the result was equal to nil: everybody spoke about something, no positive impact on our activities was noted :) Nevertheless, I do think that meeting with teachers and students is a good idea, and you should concentrate on it. rubin16 (talk) 20:31, 28 January 2015 (UTC)

Thanks for sharing your experience. In RU you have more than 1000 major editors (25+ edits) and more than 3000 of 5+ edits, we are ~5 of 25+ and between 6-12 of 5+ edits. The bulk of development of sw-wikipedia over 7+ years depended on the cooperation of these 5 people (some very active non Sw-speakers used to contribute much but left in the last years). If 2 of the 5 left that would mean stagnation (and going back as there would no capacity be left for looking at new entries, advising new editors, deleting files - what we try to keep up but already now struggle with). Meeting once will be a boost to motivation - for which I have been fearing for years. There was a time when I was alone for months and there was a time when we were only 2 really active - that was on the brink. From the feedback amongst the team I see now it looks more optimistic than it was for some time because of the prospect of a meeting! The conference on Friday is the minor part but I see it as essential for the future perspective.

If wikimedia wants African languages - today after Afrikaans the only (moderate) success is Swahili and even if our editors structure may be unusual that is how it is and how we reached the point where we are!

Skype sounds nice - but between only sometimes reliable low-speed nets like Tanzania and Iran a conference? I gave that up years ago. Nairobi and Morocco should be able to do that as they should by now have the connection to the high speed cable - but all five of us? And a typical African problem: our Tanzanian on the team can only use his workplace computer - no idea if he has skype on that one or would be allowed to install. OK you motivate me to give it a try once more (my personal regular skype experience from Iran with family in high speed countries is not encouraging for a conference connection). The "conference part" on Friday evening is the minor part - and the agenda outline for the first evening you see in "activities 1". Of which preparation for the workshop is a topic.

By the way: Riccardo just told me that we should plan for 100 workshop participants according to the interest he sees. Sounds great! Kipala (talk) 22:28, 28 January 2015 (UTC)

Ok, I understand that it's not easy to use Internet as a perfect communication channel but, bearing in mind your comments about small meeting in Friday and full workshop the next day, I would join some comments below that in such case I would put more stress on the workshop that meeting and would set some targets and measures based on it: number of attendees, deliverables, agenda, etc. rubin16 (talk) 20:22, 29 January 2015 (UTC)

Rubin16 I support Kipala in his response to your e-communication suggestion. I have experience with working in and with Africa. Only a few countries on the african continent have reliable hi-speed public internet links, however they are also very clogged up with traffic, beside that only some Global corporations operate private networks and they have real hi-speed unclogged intranet traffic. Some of these corporations allow some of their staff to use these links for non business related traffic. I did voice/voip conference calls with Tangiers ( that means less then 30 mi from spain across the gibtaltar strait, and the ebst place to connect to morocco) and it was impossible to have an uninterupted meeting or a normal convernsation. Morrocco is (to the bbst of my knowledge) the best connected country on the african continent both internally and internationally (links to europe and across the atlantic to the US). You are right that a lot of the day to day work can be done electronically but i do agree with Kipala that at regular intervals they shoudl meet especially if they are so far apart in order to strengthen the collaborative effort, and straigthen out any misundertandings , becuase working throiug email and typed messages does genenrate over time misunderstandings en frictions that your need to get out of the way. --DerekvG (talk) 14:41, 31 January 2015 (UTC)

more evident impact

I would replicate here as done for the Hindi proposal. The request to have meetups doesn't make sense to me. I suggest to put in the deliveries a plan as result of this meetup to get more contributors and more content to be realized after the meetup. I don't see a specific benefit to have people knowing each other in presence when the majority of the time their activity is on line. Honestly the cost for a team building makes sense if this team building produces more tangible results like for instance the resolution of an internal conflict. --Ilario (talk) 14:44, 29 January 2015 (UTC)

I appreciate your critical view about the value of meetups in general. I ask you to consider

• keeping this small team (which more or less built the existing content) motivated is a tangible result. Motivation is at risk and endangers the development as it stands
• From our point of view I put the meeting in the first place in the interest of quality development. I ask you to consider the time effort we spend on the two parts of this weekend: A) shorter team meeting, and nearly half of it will have to be on coordinating for the Saturday workshop. B) the full day workshop with teachers and students aiming at gaining new editors.
• kindly consider the development since preparations for the workshop have started at Morogoro: number of interested participants has grown beyond expectation, we have to divide into 2 parallel workshops. Our Morogoro team member by now has come to contact also schools connected to Unawetanzania programme (we did not succed to get feedback from Unawe as such, but we found that 3 of their schools are in Morogoro and try to gwet teachers as participants and use that chance to point them to our relevant content and get feedback on usability in Primary schools (which are the only ones with Swahili as language of instruction, whereas for Secondary schools it is directly relevant only for the subject Kiswahili, as material for research, essays etc).
• kindly consider that this very welcome development is the result of the motivation push stemming from the prospect of doing something together Kipala (talk) 20:53, 2 February 2015 (UTC)

Workshop

First of all - I agree that the meeting of real-life old buddies who in fact created the Wikipedia in Swahili is worth supporting itself, and disagree that it might be replaced by skype meeting. Anyway - if supported it would be worth to have as much outcome as possible, which means more detailed schedule what are you about to discuss and what longer-term results you expect to have. For example the quite important question might be how to extend the base of devoted editors, and why there are only 5 of you...

1. Regarding the workshop part of the meeting: Do you have any preliminary information how many students and teachers from the school might be interested in the workshop? The website of the school claims that there are 3500 students - so it is quite a lot, but judging from pictures of class-rooms - there are very few computers there - so I guess school have to select somehow students to choose those who might benefit from workshop in the future.. Polimerek (talk) 11:48, 29 January 2015 (UTC)

The interest in the workshop has grown beyond expectation. So the school has to choose. There is a computer lab with 30 desks, if 2 can use 1 desk, it's up to 60. Today I hear of a request from some university people whom we should take in. Riccardo checks if they can set up a 2nd room. The offline KIWIX version does not help much for edit exercise... We have to discuss this now in the team how to go about - split, shifts?.

Why there are now only 5 of us - it is an African problem. I tried to answer that above under "GAC comments" and touched on it in this presentation from slide 17 onwards. It is pretty much the same in ALL African languages -except Afrikaans on the edit side (it is the ONLY African language medium of instruction in sec. schools and universities,,). The other AL-wikipedias with higher article count than us were mostly created by 1 editor: mg by a bot, yo by "Demmy" - both with heaps of entries without or with little content (cf slides 12-14 in the presentation). The following is an extract from today's https://meta.wikimedia.org/wiki/List_of_Wikipedias.

№	Language	Language (local)	Wiki	Articles	Users	Active Users (1 edit/month)	Depth
59	Malagasy	Malagasy	mg	76,315	8,173	25	11
80	Afrikaans	Afrikaans	af	34,282	64,984	168	34
88	Yoruba	Yorùbá	yo	31,092	11,722	26	5
93	Swahili	Kiswahili	sw	28,024	19,840	52	33
109	Amharic	አማርኛ	am	16,252	17,818	38	20
168	Somali	Soomaali	so	3,743	10,912	66	67

From http://stats.wikimedia.org/EN/Sitemap.htm you can see numbers of editors at the moment (in "summary": active 5+, 25+,100+ monthly). Of course we take the challenge up with the workshop. Africa ticks different - and we go against the stream. Stubborn & gladly. Kipala (talk) 22:59, 29 January 2015 (UTC)

Community comments

Comment to Dry Martini/ copy cost

(do not want to type up there): no real idea, I will ask. When I drafted it we thought of a 20 teachers workshop, by now the interest has come to be so strong (Saturday is the day off!) that we just discussed setting up a store as second room and using 50-60 computers (pray for connectivity), for the students 2pax/desk so a total of 70-80 people. Still copy estimate may be on the high side - then it will be visible in the final statement and returned.Kipala (talk) 20:59, 30 January 2015 (UTC)

WMF comments

Thank you for this grant proposal and for your (and the GAC's) engagement in the discussion. We appreciate all the work your small group of active editors have done over the years to grow the Swahili Wikipedia. We do believe there is value in in-person communication and coordination for both motivation sake and to discuss key issues on the Swahili Wikipedia. We are happy to support this type of meeting, but do have a few remaining questions/comments:

1. As mentioned above, it would be great to add to your measures of success some specific outputs from the discussion and workshop. For example, # of participants (split by teachers and students), # of people that learn basic editing skills, # of people that continue to edit 2 months after the workshop, list of the 50 (?) most important articles relevant to the teachers/students you engage with, 12-month development plan for Swahili Wikipedia, establishment of wiki club. This are just examples to give you an idea!
2. Since you have budgeted for 15 t-shirts, we suggest you give them out to the top participants in the workshop.
3. How will you manage the increased interest in the workshop? You will be training both teachers and students, correct? If you need to prioritize, it might be best to focus on the teachers who can then share their learnings with students later. We would caution you against taking on too many participants. We have found 7-10 people per experienced editor to be a good ratio. Here are some good resources on planning editing workshops: Edit-a-thon Program Resources

Looking forward to your responses. Cheers, Alex Wang (WMF) (talk) 02:18, 4 February 2015 (UTC)

The workshop can be considered as a success

• with 15-20 teachers participating (on their day off!)
• with 20 and more students participating
• all of them learning basic editing skills, registering as users, editing or starting 1 entry
• having initiated a students sw-wikipedia club
• having 3 additional users who continue as active editors 2 months after the workshop.
• after receiving a list of 50 topics useful for schools as recommended by teachers

The idea about t-shirts was to leave them in Morogoro as "prize" for active editors in the sw-wikipedia club... we could use some during the workshop for "very active editors"

The increased interest we take up by preparing an additional room for splitting the workshop. Our internal debate agreed that it would not be good to mix students and teachers for "heshima" i.e. respecting status in the school environment.

Kipala (talk) 10:32, 6 February 2015 (UTC)

Hi Kipala. Thanks for your reasonable responses to our questions. In terms of tracking active editors after the event, it would be great if you used Wikimetrics to track the new users registered at the workshop and in the wiki club in the future. Please read through the tool's training materials and do let us know if you have questions on how to use it. Alex Wang (WMF) (talk) 19:07, 6 February 2015 (UTC)

Grant Approved

We are happy to approve this grant for a meet-up of the active Swahili Wikimedians and an editing workshop. While we understand some GAC member's concerns about the value of in-person meet-ups for essentially online activity, we do see value in supporting an in-person meeting for this community. These 5 editors have built the Swahili Wikipedia over the last 10 years and an in-person meeting provides an opportunity for community building, problem-solving, and strategic planning. With 80 million speakers (5-15 million native speakers) of Swahili, the Swahili Wikipedia has the potential to offer high quality content to a large population. If we can support both the active editors and bring on new editors, we would like to do this. Even if the workshops result in just a couple of active editors, that would be a significant boost to the editor pool! And the establishment of a wiki club under the guidance of a long-term editor is a great strategy. Looking forward to seeing how the meetings go! Alex Wang (WMF) (talk) 19:24, 6 February 2015 (UTC)