What would motivate you to contribute to KDE wikis such as Techbase or Userbase?
Within the Plasma team, developers put a fair amount of time and effort into writing tutorials, with some more taking shape on our Community wiki pages. We're about to start on content on Userbase, starting with documenting how Actitivies work from a user's point of view in the 4.5 release. Other teams within KDE are doing similarly.
While this is creating a very nice set of resources, the cost of making them is fairly high as we trade development (read: software quality improvement, feature) time for documentation time because it's the same people doing both sets of tasks. We regularly encourage people outside our core development group to contribute to this growing body of documentation, and occasionally they do. Usually .. they don't. Now, I'm used to the usual "talk to 10 people and one will actually do something" we all know and love from sales so I don't expect a very high success rate, but the success rate for documentation seems even lower than that.
Even when we seed a document with the essential information, people hesitate to add to it. When a new bit of information is discovered by someone using those tutorials, I have to urge them to add their findings to the page, and even then they often don't. I know we "all dislike writing documentation" (well, that overstates it a bit :), but I get the feeling we're doing something wrong here.
I could make guesses such as, "It isn't obvious that the wiki is something even me, as a casual third party developer, is encouraged to improve." Even then I'm not sure what would help fix that? A big "contribute, contribute!" header / footer on each page? The main landing page(s)?
So I'm looking for your feedback as to what would turn you into a contributing author. I look forward to your comments below. :)
28 comments:
I'm not sure how to make people contribute more, but I do think I know the reason why people don't.
I think it's as easy as:
It's one thing to have the know-how in how to accomplish X. It's quite another to be able to explain it in plain English so it's understandable to someone without that know-how. In my experience, most people don't have both the know-how and the pedagogical skills necessary.
And that goes not only for volunteers, but professional authors/writers as well. For example, since I started using Linux (and other unix-like systems) I've never seen a good explanation of how the file-security flags work. I've learnt that by trial-and-(t)error only. Which is a pity, since I probably would have had a deeper understanding of it if someone who _really_ knew the ins and out of it could have explained it to me better.
Hi Aaron.
I would certainly write end-user documentation, but I'm a spanish speaker. English is really hard to me. As an example, writing these two paragraphs, took me like 10 minutes, and Google Translate.
If Spanish documentation (Castillan) is in need of man-power, I can offer my services.
Best regards,
Norberto
maybe people are afraid they will be adding wrong things as they don't consider themselves experts..?
How do we know it will be reviewed? Are talk pages monitored?
should changes be communicated through talk or directly in the page..?
@zoolook: yes! Castillan documentation is in need of help! there is a KDE documentation / translation project:
https://mail.kde.org/mailman/listinfo/kde-l10n-es
and the general development list:
https://mail.kde.org/mailman/listinfo/kde-devel-es
@Jonas, mxttie:
so your feedback is, essentially (and please correct me if i'm misunderstanding you):
there is the (understandable) feeling that it takes someone who knows more than you do to document things.
there is the feeling that it takes a higher level of expertise, and a higher level of writing skills.
that's interesting, because neither is true in reality! even so-so writing skills are fine (we can and do edit pages) and every bit of information documented is a win.
so how can we make that more evident?
place notices on the wikis?
offer a mentorship program, so people who would like to write something can quickly have a "i'm looking for a guiding hand" message which would hook you up with a KDE wiki guide person?
something else?
personally, i think it would help if there would be some kind of review process. Just to know that technically the information one adds, has been validated.
another thing:
Some communities are sensible about editing pages directly and prefer you add changes in the talk page. Then again, it often feels like nobody looks there..
So I guess it must be made clear what steps one must take to contribute. In kde's case, I understand you can directly edit the page. It could be nice if you can post your revision for review and that it gets stamped as "approved". I'm not sure if it's feasible though :)
just my 2c
@Aaron,
"there is the (understandable) feeling that it takes someone who knows more than you do to document things.
there is the feeling that it takes a higher level of expertise, and a higher level of writing skills."
Not necessarily a higher level of expertise. After all, even someone who has no idea of how plasma (for example) works internally can probably think of ways to use it the developers didn't think of - at least not at first.
And not necessarily the writing skills per se either. Rather the ability to be a good _teacher_ even if the writing is so-so. The grammar may be impeccable but if the text isn't cohesive enough or at the right level for the intended audience that won't matter much. On the other hand, a text that "gives away" (for a lack of better term) the author's own language can be edited to sound more idiomatic English as long as the content is good enough.
Still, the writing part can be remedied with "stern enough" proof-readers and some constructive criticism when needed.
Now that I've thought of it some more, I've begun to wonder if my original answer was entirely accurate. I still think those feelings are a big part of it, but I've begun to wonder: how many people are even aware of UserBase or the forums? Maybe they should be advertised more?
Now, I'm not an active contributor to UserBase (and TechBase is out of my league). Mostly because my time is very limited, but I think the mentor idea sounds promising. I wouldn't be able to do much good in explaining how a wiki works but there is one thing I would be able to do if there's a request for it:
Proof-read, give suggestions on how to improve readability, check grammar and so on. I teach English as a second language part-time and hopefully that can be put to some use.
This may be way off of what you are looking for, but ...
I have been playing around with OpenStreetMap.org these last few days, which in essence is a map wiki. They also have a wiki wiki, but I digress.
Maybe you can somehow get users who are already active in wiki environments to contribute the KDE wiki initiative by collaborating project??
So maybe if there was (is?) some sweet OSM integration somewhere, a tutorial could be written off of that, and maybe that would bring about the OSM community to contribute...
I use OSM as just an example. Don't know if this even makes sense, but it seemed like a good idea when I thought of it :P
Myabe this is crazy and most likely way off base, but I think it would be cool if there was a way to create a book on KDE4 and take the documentation that the community submits, track those who submit, by alias or real name if they'd like to offer it and publish it some where with everyones name listed. More people might want to document and edit if they're given a little credit. Not sure how possible it is and I know it's crazy, but I think it would put fun into documentation, which is normally not.
my issue is with commitment to the process -
To contribute a piece is (in my mind at least) a commitment to maintain and review it's relevance regularly.
I struggle to keep all my current threads in check without adding more -
I agree with mttie's call for someone else to handle or at least help, with the regular review and maintenance of content
I think its a matter of liking it or not. I personally like to write documentation for those things I didn't know before and I find out how to do them.
Also, now translating UserBase can be done through PO files, which is a big improvement, and I'm going to do it (I hope) soon.
The major poblem I guess it is that the people who likes or is good at writing documentation or are good with wikis also do many other things (other software translations, Wikipedia and similar projects, etc.).
An 'I would like more documentation' button on api.kde.org pages and overviews, so we can track demand for developer docs.
I am one of those new guys having lots of questions. I am also trying to document my findings. What motivates me? The fact that people are sacrificing their time explaining things to me on IRC.
So I guess, if you want people to contribute to the wiki: help them on irc, afterwards ask them to document it.
You could even mention this in a notice when you join the dev channel :)
To bring the threshold as low as possible you could create a dedicated IRC FAQ page where people can just post their question and the answer they got on it. Afterwards it could be integrated in some adequate other page.
People asking questions will automatically start to have the reflex to first look at the FAQ page which will releave devs from having to answer the same Q for the 10th time. And when it's still the case, they can provide a canned response.
Then why would people still not contribute? I think because often people start hacking when they actually have to go to bed. ;)
So they just want to achieve as much as possible in the shortest time span. It is not the best way to start hacking on KDE, but I guess it happens alot in this way :)
I think the question “who knows userbase and techbase” might hit it quite well.
Where do users look to find information?
I just tried it. I went to KDE.org and thought: OK, lets find the wiki.
I clicked on community.
So, where’s the wiki? right: not there. → “maybe they document stuff in the forums” → they aren’t there, either??
Why are community, dev platform and support split?
(apart from them having a hell of a lot of subtopics :) )
And while I’m at it: Why are the forums under support and not under community?
And why community? isn’t everything there community?
…
OK, enough ranting. Time to get constructive :)
There’s stuff „about KDE“ (news and project descriptions and such). Then there are „KDE tools“ (workspaces and applications). Also we have „using KDE for programming“ (dev platform → that fits quite well) and „using KDE“.
⇒ I think a little reordering on the website would make the wiki more prominent.
But that’s only part of it.
Why isn’t there a link to a FAQ in the user wiki under „getting help“? That’s what I expected to see there.
I don’t want to ask questions. I want to find answers…
And why doesn’t my techbase login work in userbase? → unnecessary barrier for contributing.
I just wanted to add a small FAQ-site with links to the FAQ-pages of subprojects.
That would be something I’d add. It provides a very simple way to add information.
And please don’t make the FAQ too pretty (with rich markup or many nested pages). Every new contribution should be on equal ground, even if the contributor doesn’t know all the mediawiki markup. And the questions should be searcheable with strg-f (in browser).
…
sorry for the long rant. I thought I’d answer your question, and somehow it bit me…
Disclaimer: The KDE page is very impressive and extremely beautiful, and the same goes for the wiki (“you can do *that* with mediawiki?!”).
It’s just that obviously some problems exist, and some possible causes bit me while looking over it ;)
So even though my criticism is quite harsh at points, I admire what you built — very much so!
And a lot of my points might just be wrong…
PS: And at least under documentation there should be link to the wiki, that’s what a wiki is, after all.
Some random thought: make it absolutely clear it's not done by people who get profit/money for this. It's made by volunteers - and you can add your findings too.
Sounds silly, but even at my local scouting group parents believe we're getting paid for mentoring kids.
It creates the "someone else should do it" dynamic.
I believe the people talking about review are onto something. For a one-off contributor (who may eventually be enticed into repeat contribution), a 'somebody will review my contribution before it is seen by the rest of the world' security blanket could be reassuring. The idea of having mentors isn't bad, but doesn't fit at the very beginning of the road to repeat contribution - it adds a barrier to entry.
Just my $0.02
One thing I'd love, and might be (or maybe not, I'm not a coder) trivial to implement would be some kind of plasmoid which points the user to the wiki, streamlining the feedback process.
@Bernard: "To contribute a piece is (in my mind at least) a commitment to maintain and review it's relevance regularly."
interesting observation, particularly as it isn't true at all for the wikis. nobody would fault you for maintaining things that were added (by you or others), but future maintenance is not a requirement of adding content.
so it seems that the wikis have a certain amount of understandable "barrier to entry" to them that doesn't actually exist.
the next question is how to dispel those concepts effectively when someone lands on one of the wikis.
@arnebab: single sign on (SSO) is a big deal; they do support OpenID, but KDE is finally getting an in-house solution as well in the form of Gosa which is coming about as part of the git migration.
as to the rest of the issues you highlighted, how many of them have prevented you, personally, from contributing? (it's not clear which of the observations you made were problems you personally ran into and which are problems you identified as possibly barriers to others.)
For me, it's a combination of two things:
1) Whenever I check a page that I could possibly contribute to, I find it's already well filled out. I haven't looked to see if there's a section listing pages that need attention.
2) Pages such as http://techbase.kde.org/Getting_Started/Build/KDE4 are likely high profile. I've tried three times from this page to get a working build environment for KDE4, and I've failed three times (not counting past attempts during KDE3 times). I've never once had a working KDE build environment. So, if I can ever get one working, should I do any editing to the page on building KDE4? If there's a lot of unnecessary difficulty, should I simplify the page? Or will my changes be reverted back to the more complex version?
I guess it comes down to: 1) a feeling that a page is "complete enough" that I don't feel I can contribute to it (if there's room for contribution in the page, I don't realize it), and 2) a feeling that editing a potentially high profile page will not go well (which is probably an Asperger thing on my part, and may not affect most people).
Hopefully I made a little sense, because now I'm not sure if I did! I'd better submit this comment before I decide I'm being confusing and don't submit it.
Personally to me it just *feels* like most people don't look at them so what's the point? You always hear how they're deperately missing help but rarely hear about how much they helped. I've never heard anyone mention how incredibly useful anything they found on userbase was, or how they were stuck until they found a page there. It just *feels* like they're something that you'd contribute to if it didn't *feel" like it would be an effort that would help very few actual readers. Significant time required + little recognition + little percievable impact = low motivation.
Hi Aaron,
Many thanks first for all your hard work on KDE :)
This may already exist, but if someone was willing to go and find pages that need doing, stubs, incorrect stuff and so on and basically create a "to do" list for the wiki, that would definately help. (at least for me).
for a lot of people, being able to tick something off as "done" or close the bug etc gives an immense sense of satisfaction and so this kind of todo list would be great.
I havent looked at the wiki recently, but perhaps also having a "flag" for users to say that this needs work would be great, if it doesnt exist already.
if these things do exist, perhaps then advertising that they are there and encouraging people to do this.
I guess the point for me is I hear requests from ubuntu, kde and about a 1000 different projects for my time, but if they are not specific enough ie. We need help with the wiki" instead of "here is a list of pages that need updating" then its just becomes another meh, next thing. If I see in the list things I might know about, then I can spend some time looking to see if its something I can help with.
So I guess my answer is get a bit more specific, help people to better see what needs to be done.
After looking at the wiki again, perhaps it could be done this way.
We put 3 items on each page, Flag page for inaccuracy, mark page as obsolete, request page have additional information added (or different wording, but to that effect).
When these buttons are pressed, a box appears where people can give more info - this in itself is contributing to the wiki.
All these flagged pages get put on a list, which contributors can continually work through. This list gets advertised when we ask for more contributors to the wiki.
Just ideas...
@Aaron: The main issue for me was the single sign on. The second is that the userbase wiki just isn’t visible. It isn’t part of my workflow.
When I need help I look in the pages of my distribution (Gentoo); the wiki page I visited the most during the last few months is the Schedules-Page… (und I always used Google for that after not having found it immediately on the wiki a few times).
@Jussi01: How about also adding a flattr button to every page?
Or at least some way to say“this is cool! Thanks a lot!”
I think what makes a huge difference is how "alive" and "open" a wiki appears to the reader (= potential contributor).
Despite of the standard mediawiki toolboxes etc., the KDE wiki pages "feel" like static pages to me.
In fact I'd hide all those standard toolboxes (maybe collapse them like in the new wikipedia.org theme), to make space for displaying more newbie-friendly stuff instead that makes the reader feel like the site he's visiting is at the center of a live and active online community, rather than some external "information dump" that's outside of the sphere of normal KDE community activity.
An example of a site that achieves this very well is wikiHow (Sample page: http://www.wikihow.com/Install-KDE-4-on-Ubuntu).
Just to list some of the things they have that make it *very* welcoming for outsiders to contribute:
- big "Write an Article" button
- big "Edit" button for every page section
- "Things to Do" box (in friendly every-day words, not mediawiki-jargon!)
- "Recent Changes" box
- "Featured Articles" box
- "Meet a Community Member" box
- "Like" button
- simple vote-based feedback mechanism for each page
- number of page views shown at bottom of every page
- "Watch this article" link
- possibility to request articles to be written, and to easily see what articles others have requested
- very useful introduction message that new users get after creating an acount
- a "Sandbox" page for getting familiar with wiki editing
- "Explore Categories" box at bottom of every page
I believe that all these things, however small each might seem on it's own, help to make a potential contributer feel like:
- that this is a real, active community site that many people visit frequently, both for reading and editing (and to some dergree, discussing)
- that there are many places he/she can start contributing
- that it won't be difficult to contribute
- that his/her contribution will be very much appreciated, and that it will not be a waste of time
At least that's how *I* feel..
(Maybe I'm too sentimental?)
@uetsah: those are all good points and ideas imho ...
(and no, i don't think you are too sentimental :)
I totally agree with some posts before. As pure user, I've never noticed that everybody who wants to edit the wiki can do this. Therfore a edit button in the upper corner seems to be necessary. It should be visibly to all users. If anybody wants to edit an article, he/she has to click on edit and gets to the login-page.
The second issue concerns the layout. The fixed width restricts the readability. In my oppinion the wiki of the german ubuntu community is a good exemple for readability.
Look at: http://wiki.ubuntuusers.de/Plasma
But stating “we don’t do that for money” creates the “making money is bad” dynamic, which is bad in itself.
Post a Comment