Wiki, Take #whatever
hi
i like the idea even though i'm not part of the doc team .. i'm on the list for some reason but that fine ;) maybe we could get some help from gentoo-wiki people and/or 'merge'.. just my 2c (i don't know previous discussions or anything btw so if this was mentioned before just ignore me ;)) On Tue, 2008-11-11 at 01:16 -0800, Josh Saddler wrote: > Hey again. So there's some discussion (again) on starting up an official > Gentoo wiki. Official meaning it's hosted on our infrastructure; e.g. > wiki.gentoo.org. This time the discussion is coming from our fellow > developers and infra overlords. > > I know that the GDP discussed this any number of times on this list, and > some of us (most notably myself and neysx) have discussed it on the > forums, with occasional feedback from other developers. > > However, it's been quite awhile since the last time we (the GDP) talked > it over. Given our current issues of manpower and time (see > archives.gentoo.org for commit totals), perhaps a wiki could solve some > issues? > > The classic problems are: > 1) Who has access > 2) Who reports faulty articles > 3) Who fixes them > 4) Who verifies the article is correct > 5) ??? > 6) Profit > > So, I'd like to hear what you think. Should we have a wiki? Why or why > not? If so, for once, does anyone have some *sane* ways to admin and > moderate such a wiki? > > There's no question that having a properly-administered wiki can be a > powerful asset. Look at Ubuntu, Debian, Arch, the Xfce wiki, etc. What > about Gentoo? > > Halcyon has brought it up again on https://bugs.gentoo.org/75855, but > BEFORE I/we go over there and weigh in, let's hash it out here on our > list. :) |
Wiki, Take #whatever
Josh Saddler wrote:
> However, it's been quite awhile since the last time we (the GDP) talked > it over. Given our current issues of manpower and time (see > archives.gentoo.org for commit totals), perhaps a wiki could solve some > issues? > So, I'd like to hear what you think. Should we have a wiki? Why or why > not? If so, for once, does anyone have some *sane* ways to admin and > moderate such a wiki? I for one, am very much for a an officially Gentoo-hosted wiki. The unofficial wiki has been a very valuable resource, even with its shortcomings. I think we should bring it on board and offer the security of our infra resources. I am of the opinion that we should see the wiki more or less as we do the forums. It is a place where users can contribute to the Gentoo community. I would expect most of our users are internet-savvy enough to understand the nature of a wiki as user-generated and user-editable content, and therefore not being as reliable as say our official documentation. We could form a team of moderators (from both user and developer base) that would do some quality control, similar to what happens on Wikipedia. They could indicate articles of particular value and quality, as well as indicate if there are issues (outdated, incorrect, incomplete, etc) with specific articles. I don't see the "classic problems" as problems at all, as long as the nature of a wiki is taken into account. > 1) Who has access Everyone. To restrict this would be a mistake, in my opinion, and against the open and free nature of a wiki. > 2) Who reports faulty articles Wiki users (and mod team) > 3) Who fixes them Wiki users (and mod team) > 4) Who verifies the article is correct Wiki users (and mod team) We could add a disclaimer to the footer along the lines of: this wiki is open and free for everyone to edit, therefore Gentoo cannot guarantee the accuracy of its content. -- Ben de Groot Gentoo Linux developer (lxde, media, desktop-misc) Gentoo Linux Release Engineering PR liaison __________________________________________________ yngwin@gentoo.org http://ben.liveforge.org/ irc://chat.freenode.net/#gentoo-media irc://irc.oftc.net/#lxde __________________________________________________ |
Wiki, Take #whatever
On Tue, Nov 11, 2008 at 10:16 AM, Josh Saddler <nightmorph@gentoo.org> wrote:
> However, it's been quite awhile since the last time we (the GDP) talked > it over. Given our current issues of manpower and time (see > archives.gentoo.org for commit totals), perhaps a wiki could solve some > issues? > > The classic problems are: > 1) Who has access > 2) Who reports faulty articles > 3) Who fixes them > 4) Who verifies the article is correct > 5) ??? > 6) Profit I think we need to drop the incentive that the documentation on that wiki is validated by a developer. The moment you work with community-driven documentation, this is almost impossible to achieve. In my opinion, the moment we would start a wiki, we use it for what it is made for: community-driven documentation development. However, I would use the following practices: - Specific documentation that is "dangerous" to execute should have a big red warning block, telling the users that this is not common practice, is dangerous to execute, might result in yielding support from developers, yada-yada. Examples of such topics could be bootstrapping, editing portage code, specific C(XX)FLAGS, ... - Translations of documentation are free to perform and should not be reigned by rules such as "must be based upon a revision of the English documentation". This does assume that the topic in the wiki is self-explanatory. - Wiki information pertaining to ~arch stuff should be in a different namespace or some other way of destinguishing them (if not, even a tag would help) that informs people that ~arch ebuilds are not tested enough and can contain bugs As manpower is important with wiki's (think of spam regulation), it would be nice if we could tie forum accounts to wiki accounts, and edits on the wiki are only allowed with accounts (no anonymous editing). The moment a spammer occurs, account deletion should result in some practice where all his/her edits are checked (I believe this also occurs on forums, but I'm not sure). Note that I'm not suggesting that forum admins should work on wiki's too - if they want to, that's great, but it's a different playground and I wouldn't want to push them into responsibilities they didn't ask for ;-) Wkr, Sven Vermeulen |
Wiki, Take #whatever
> So, I'd like to hear what you think. Should we have a wiki? Why or why
If you try to simplify things (suppose there's no spam, no obuse), wiki could be a great pool of potetial official documentation. As I see it, this is the greates value of wiki. "Just" pick good articles, refine them and move them to official docs. As a person who never administrated wiki, I have no clue how to secure it. -- FreeB(eer)S(ex)D(rugs) are the real daemons |
Wiki, Take #whatever
On Tue, 2008-11-11 at 11:41 +0100, Goran Mekić wrote:
> > So, I'd like to hear what you think. Should we have a wiki? Why or why > If you try to simplify things (suppose there's no spam, no obuse), > wiki could be a great pool of potetial official documentation. As I > see it, this is the greates value of wiki. "Just" pick good articles, > refine them and move them to official docs. As a person who never > administrated wiki, I have no clue how to secure it. How practical/possible is it to add some kind of 'verified by a developer / moderator' flag / tag which cannot be changed by normal users and only applies to a specific version of the page? |
Wiki, Take #whatever
Goran Mekić wrote:
>> So, I'd like to hear what you think. Should we have a wiki? Why or why > If you try to simplify things (suppose there's no spam, no obuse), > wiki could be a great pool of potetial official documentation. As I > see it, this is the greates value of wiki. "Just" pick good articles, > refine them and move them to official docs. . . . no. There's zero point in doing all the work twice (original article and then XMLifying it.) Especially since now you have twice the maintenance burden, and it IS harder to maintain XML docs than wiki articles. |
Wiki, Take #whatever
Seeing how many users are weeping for gentoo-wiki.com's content (its
database was lost) I'm sure that gentoo must have own wiki. В Втр, 11/11/2008 в 01:16 -0800, Josh Saddler пишет: > it's been quite awhile since the last time we (the GDP) talked > it over. Given our current issues of manpower and time (see > archives.gentoo.org for commit totals), perhaps a wiki could solve some > issues? I think wiki should never be associated with GDP project. It's a separate entity - it's primary goal is to provide a single place where our users could share solutions for their problems. Yes it may help GDP project - users and/or developers could write initial proposal there and when document will be considered good GDP will convert it to guidexml and publish it. But until document is on wiki GDP has no responsibility for content published there. > The classic problems are: > 1) Who has access Everybody. > 2) Who reports faulty articles > 3) Who fixes them > 4) Who verifies the article is correct It's wiki: if article faulty and user wishes to improve he/she fixes it. Mods are required to help to solve technical conflicts between users and to remove spam. Also they could help to organize content... -- Peter. |
Wiki, Take #whatever
Ben de Groot wrote:
> I for one, am very much for a an officially Gentoo-hosted wiki. The > unofficial wiki has been a very valuable resource, even with its > shortcomings. I think we should bring it on board and offer the security > of our infra resources. Nope. The gentoo-wiki.com owner has already stated on the forums that he doesn't see a need for it to be hosted on our infrastructure. More to the point, he told our infra guys this when we offered him a box (he got a better overpowered offer elsewhere). > I am of the opinion that we should see the wiki more or less as we do > the forums. It is a place where users can contribute to the Gentoo > community. I would expect most of our users are internet-savvy enough to > understand the nature of a wiki as user-generated and user-editable > content, and therefore not being as reliable as say our official > documentation. Unfortunately, they do *not* understand this. Just look around the forums. Users are greatly surprised when wiki or forums tutorials break their boxes, then get busy pointing fingers and wondering why no one's updated the article. Or they notice that no one really knows; there's not a "solution" as such for their issue. If users see a wiki on gentoo.org, it seems more like it counts as "official, verified" information. Maybe the smarter ones recognize that like the forums, it's limited and unofficial, but by and large we *cannot* depend on users understanding this. I think Ubuntu tackles the issue a bit differently -- there seems to be somewhat of a distinction between official-ish wiki/wiki articles and community articles. At least the basic stuff, like About, Installation, Desktops, etc. seems to be more or less Canonical-written/approved/official. Just try searching around to see where the differences start to creep in. Though their wiki sucks for searching and returning coherent results. > We could form a team of moderators (from both user and developer base) > that would do some quality control, similar to what happens on > Wikipedia. They could indicate articles of particular value and quality, > as well as indicate if there are issues (outdated, incorrect, > incomplete, etc) with specific articles. I suppose there would have to be flags/tags similar to wikipedia's "Out of date/needs review", with some sort of way of notifying admins/mods/devs/whoever about it. >(and mod team) Who is this mod team, really? I've seen some proposals for forum mods, but none of them have expressed any interest in it; they've enough work as it is. The ebuild devs aren't so interested in it; they're not interested in docs of any kind, and they've enough work as it is doing ebuilds. Basically, the developer pool is out. And really, I don't know that I trust the users, given what gentoo-wiki.com has turned into. We've seen how far most the users can go, and it's not enough. Infra has only said that they are willing to host one, and administer the server hardware itself. Spam and day-to-day article maintenance would not be performed by infra. > We could add a disclaimer to the footer along the lines of: this wiki is > open and free for everyone to edit, therefore Gentoo cannot guarantee > the accuracy of its content. > That's shooting ourself in the foot right there. Personally, I don't see the point of a resource that cannot be verified nor vetted for correctness. In my view, documentation simply must be accurate, otherwise we are doing ourselves and our users a disservice. |
Wiki, Take #whatever
Thomas Raschbacher (Gentoo) wrote:
> hi > > i like the idea even though i'm not part of the doc team .. i'm on the > list for some reason but that fine ;) > > maybe we could get some help from gentoo-wiki people and/or 'merge'.. > > just my 2c (i don't know previous discussions or anything btw so if this > was mentioned before just ignore me ;)) Yeah, it has. You may wanna search our archives or the Gentoo forums for posts by me on the subject. :) In short, a merge won't happen, as the gentoo-wiki owner has stated he's not interested in us hosting the site. More on that in another email in this thread. |
Wiki, Take #whatever
Thomas Raschbacher (Gentoo) wrote:
> On Tue, 2008-11-11 at 11:41 +0100, Goran Mekić wrote: >>> So, I'd like to hear what you think. Should we have a wiki? Why or why >> If you try to simplify things (suppose there's no spam, no obuse), >> wiki could be a great pool of potetial official documentation. As I >> see it, this is the greates value of wiki. "Just" pick good articles, >> refine them and move them to official docs. As a person who never >> administrated wiki, I have no clue how to secure it. > > How practical/possible is it to add some kind of 'verified by a > developer / moderator' flag / tag which cannot be changed by normal > users and only applies to a specific version of the page? No idea. That would depend on whatever wiki system is actually installed, I imagine. And I haven't heard from Infra what they'd be willing to install. |
| All times are GMT. The time now is 04:50 PM. |
VBulletin, Copyright ©2000 - 2013, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.