On Friday 16 April 2010 11:20:50 Rupert Horstkötter wrote: ...
I'm not on the same page as of the "de.o.o with features from wiki.o.o VS de.wiki.o.o" decision. We have to hold a strong timeline (June 1st 2010) for both the English and the German Wiki and after investigating the situation, de.wiki.o.o is required to be able to fully parallelize the process for our workforce we put upon this effort. Starting with a fresh instance and populate it with content (according to the new wiki.o.o structure) is the way to go here (just for the German Wiki, I do not propose this for every language wiki out there. Local maintainers should decide which road to go here).
Populating empty wiki with content is process that needs testing to make sure that we covered majority of topics that 11.3 users may need. When you get report that something is missing you need time to fix it. Old wikis are in the bad shape regarding ability to find appropriate articles and relevance of some articles, but they have some coverage for almost everything. I'm not worried about beginner finding nonsense articles: 1) There is few of them, (I don't count obsolete as nonsense) 2) Whatever is not indexed or categorized is lost, (which is not hard to prove) One of problems with en.o.o was removal of Portal pages from front page. While I had problem to keep them up to date, and at some point used opportunity to agree with idea to remove them, last time I did update they still covered very large percentage of the all wiki content; missing mostly articles that are not relevant to new users. To have that much coverage in a new created wiki you need a lot of man hours that can be better used to fix topics and create navigation structure that new users can actually use. New structure (schema, content layout) proposed by Henne is one way to navigate wiki, but not the only one. You can imagine that as creator of en.o.o Portal I'm for idea to have that, but I know that there are other ways and I want to keep other options open. I have problem to incorporate current Portal: design in my mind as a single navigation option in the new wiki concept: 1) It is so non opensource, and for that matter it is so non Wikipedia, and so far I see we want to follow it as an example, 2) Advantage is that user get potpourri if our offer in visually appealing form, which is what new users will like, 3) Disadvantage is that is complex which presents problem to create and maintain, so basing all navigation on it will create need for permanent maintainers, which means a lot more people engaged all the time 4) Current layout is overloaded with graphic elements that steal attention form content, we need something lighter with not so many content divider lines; in some parts of the world current Portal layout design can be appealing, as part of culture are visually busy motives with a lot of details, but even there simple is easier to use - basic human design (body) is essentially the same. Above is not strictly on topic with your requests, but it is explanation why I think that faster way to have something usable for June 1st is to copy MediaWiki software configuration (namespaces, extensions, CSS) from wiki.o.o to de.o.o and en.o.o, then to move content the other way around. I have problem to figure out how to use current Portal design, and Henne's explanation in another email opened more questions, but I have to see what to ask next, before I actually can ask. I'm not sure that when your wiki editor with a lot of experience is stuck with Portal implementation, that you will find many volunteers able to implement it fast. To buy the time, you need something that already works, and implement Portal ideas on top of it. That way you avoid danger to have half baked solution that lacks new user essentials. Fixing wiki problems on the run with user needing help "looking" at you is not recipe for good solutions. IMO, rushing transfer just to have something online on time can have another adverse effect, volunteers displeased with public perception of their effort leaving project.
According to henne, the master = wiki.o.o is almost ready for production use.
Keyword is almost and relates to bugs. There are no showstoppers, see: http://tinyurl.com/y2d72kz To me it seems that page needs update, as I'm sure that problem with Category:Portals is fixed. BTW, this is example what happens when page is not easy to find; it is not used even by those that created it in a first place. It was linked on a old Main page and now it can be found only in Special:All pages for openSUSE namespace. Even those that know it exists can't afford time to go trough labyrinth of links to reach it.
Thus we do not need a plain testing instance for another few months and can start copying what we have into a German transition instance = de.wiki.o.o in order to do the very same transition process as with the English master wiki. Actually the German Reviewing efforts even outperformed the English ones and we're in desperate need of de.wiki.o.o there.
After reading all above you think that you need de.wiki.o.o then you really need it. I'm sure that I need features of wiki.o.o in en.o.o to find out how to use Henne's Portal concept efficiently. It will help me to have content and features (extensions, namespaces, CSS) in one place without week(s) of work to move something the other way around, to find out that I need more, or that I have to remove something, or even worse that many articles need fixes. I'm pretty sure that I can have usable wiki by June 1st alone, not to mention if somebody jumps in and help. Then I have time to copy articles on a new wiki instance, specially because I'm sure what to copy.
... While you bring up some valid ideas here, this all isn't first glance scope of the problem with the German Wiki and the Multi-Language Concept. We'll decide on the fly how to deal with these once we move forward. I noted before and I respectfully ask again for it: "Please focus on show-stopping aspects". Time is pressing and we need to move forward.
I do focus on showstoppers, the only problem is that I can see more because I did wiki maintenance before and I can recognize problems a bit ahead of others that did not spent time trying to fix old (current) wiki. To make clear, I do want to clean up old wiki, and moving content to new instance is the least painful solution, but we need more time to learn how to use all nice features it has. Populating it now with articles adjusted for 5 year old wiki features will bring us no good in a long run. -- Regards Rajko, -- To unsubscribe, e-mail: opensuse-wiki+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-wiki+help@opensuse.org