Forum OpenACS Q&A: What is a WIKI
I must be dumb I guess, because everytime someone says this community needs a WIKI, I don't know what they mean. To me it is like saying "I need a program. The one I have doesn't work."
What does a WIKI do? How is it that you can't program this into an AOLserver module in a week of spare time?
So what is it?
- If it is the ability to edit someone elses contribution, use a few basic AOLserver/Tcl commands.
- Version control? The ACS has a 'manual' module that uses cvs, it takes an hour or so to setup.
- Access control? For a community of 3000 you could get by on ns_perm quite well.
- Let users build their own site? How about allowing users to post hyperlinks to their own website. Everytime someone even mentions "virtual" in a thread, I provide a link to my vhost modules. I don't think the physical location of information is that critical. The ability to link to it is. Maybe search engine accessibility is also important, so spider you site with my tool http://zmbh.com/discussion/submit-altavista.html and receive a list of links that submit these pages to AltaVista.
- Want the openacs.org name? Convince Ben to delegate a domain like members.openacs.org to mydomain.com (currently free) and then manage subdomains via the web. You could take requests by email or web form, and configure subdomains of members to use an ip address or frame redirection.
But without a list of what exact requirements are desired, no one can really argue or dispute that a WIKI is the only solution.
The examples above are not just a series of cheap plugs, but an illustration that everyone in this community has their own special pet interests. That diversity is good. I just think we should back up those interests with modules that work with AOLserver, even if the original ideas came from somewhere else.
However, I think it's more important to establish a certain standard framework for what the documentation should include, how it should be written, where it should be saved, etc. This would help the community collaborate simply by having a pretty straightforward starting point.
The problem with the documentation is not that there isn't a good tool, but there aren't any good docs! We do have the file manager, but that's not really the answer until something is written. There's the aD approach to documentation, but it's always hard for me to get a quick understanding of what a certain module is and how it works.
If something like this already exists, than I would be more than happy to be corrected. If not, then I think it's important that we think about setting one up before the OpenACS 4 release. I think there's some potential for a Wiki then, but I don't think a Wiki is paramount right now.
talli
Tell me more about the ACS manual module, I can't find any documentation for that in /doc nor in the usual place www/manual. Is it a 4.x thingy?
Anyway, here's why I like the idea of wikis:
1. It puts the users in control of the site without requiring much/any support from the webmaster.
2. Users can add their own information
3. Users can correct other's information
4. Users can add entire new sections
5. Users can reorganize other sections
6. There is a cvs / diff mechanism behind it so changes can be rolled back
7. The zope version appears to have a sophisticated permissions model, allowing for finer grain control of the above. the zope version also appears to have some standard wiki applications: for instance, lacking permission to edit a page, a user can just add comments to a page.
8. It allows rapid response to changes in our environment: when someone gets hot to trot about some initiative, lacking a sourceforge module, they can create an entire new section(s) within a wiki for discussing goals, requirements, tasks, documentation, etc.
9. 8 means that we can better capitalize on momentum. It allows grassroots projects to come together quickly with little intervention (or even approval).
10. It's a wonderful experiment in user collaboration.
If there's an ACS equivalent to wikis wonderful. If someone has time to put one together, terrific. I know at the moment that I have other fish to fry, hence my suggestions over time to just use one that works: the tcl wiki or the zope wiki. In addition to getting a solution that works in an afternoon, it would also provide us some competitive and user insights into other technologies that would better inform our own efforts to reimplement it in within the ACS or aolserver.
Well, I know I couldn't do it in a week of time. Or rather, I don't have that week of time. (And then there's the issue of getting folks to put it up.)
I like the idea of building more and better AOLserver modules and ACS modules, but my preference is to address a problem early, regardless of technology. I would rather use a tcl wiki today and toss it all out or migrate it all out when an ACS manual or ACS wiki comes along. My preference is for rapid satisficing solution today than wait for an optimal or maximal solution tomorrow.
Side note: do you really think that ns_perm would work well for a community of 3000? Maybe since that's in the user space. When I tried a few weeks back to stick a semi large number of rules into it to ensure that the public couldn't get into a site's other ACS modules, I was very unhappy with it's performance. (It was easier and probably better to just archive/delete those directories)
(I never had a birthday... His name is David. ... His is a tale of humanity. It's spielberific! AND IT's PINOCHIO too. Kill! Kill!!! )
The ACS Manuals module doc can be found at http://negeen.com/manuals/ There is also a gzipped file with what I think is the manuals package. This seems like it has pretty much what a WIKI has.
It is setup at https://pathfinderschool.com/manuals/ . Unfortunately you have to login, etc. to see it there.
boards where I feel comfortably within my zone of expertise. I
earn my living consulting organizations about how they can use
online tools to share knowledge. I've spent some time looking at
wikkis, and I'm not very impressed.
In there classic form, with no version control, wikki is an
interesting experiment with little practical value. Sure, it's easy to
post info, but it's even easier to write over that info or bury it in a
welter of poorly organized links. Self-organizing sites generally
don't work well in practice. The idea sounds great, but I've never
seen one work out to be useful.
Now, a system with more fine-grained permissions sounds
potentially more useful. But useful for what? As Tom points out,
I'm not sure I see what this could buy you that you couldn't
accomplish with file-storage, general comments, and related
links.
If somebody could give me a concrete use case, I might change
my mind, but I just don't see the value right now.
I posted a suggestion for how to approach documentation at the thread Jerry started here (https://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=00025b&topic_id=11&topic=OpenACS)
talli
Hmm...yeah, that doc really is from back in the bad old days of ACS documentation. I don't have the time to rewrite / expand it currently, but I would be happy to answer questions people have about the manuals module.
talli
For the life of me I can't see why someone who can't figure out how to use (not install) the Manuals module should be writing documentation anyway. But, at least on my instance of the module, linked to above, comments can be added at the bottom of the page.
Here is an Idea: download the current aolserver documentation. Run general comments on top of this and then users can add examples and comments on a per function basis. That only covers the AOLserver API, but it is a start. Or we could load the Manuals module with this info, and it looks like htmldoc would create the pdf files needed for a hardcopy. You can still run comments off these pages and collect the good examples into the docs from time to time.
But freeform editing isn't going to help anyone find anything. Tiny sub-sites popping up from time to time under the openacs banner also isn't going to help organize or enlighten. What might help is dumping the database contents into the filesystem into a search bot friendly url (https://openacs.org/bboard/11/00025d/message.html) and submitting these links to AltaVista.
If you doubt this works very well, consider saleonall.com. I submit one link every 6 seconds to AltaVista. Currently there are over 85,000 links from that site available on AV. Pick your favorite obscure manufacturer part number, type it in and more than likely several of the first page links will be from saleonall.com.
The bboard could be altered to allow the inclusion of a keyword or keywords for the particular message. When the messages are written out to disk, the page title should start with this word or words.
I'm not sure I understand your question. There is the documentation linked to above, and there is the code, which I assume matches that from some version of ACS 3.4. If those aren't sufficient to answer your questions, and I freely acknowledge that that documentation is pretty fast and hard, you can ask them here and I will answer them. But, that's pretty much all the documentation you'll find until someone writes more (which as I said, won't be me any time soon.)
We're not talking about a wiki that has no version control. I guess that most of the wikis I know of that I have used, the original perl wiki, the various tcl wikis, and the zope wiki all have version control.
I'm not sure what use cases you are looking for, but wikis are very successfully used by the Squeak project, http://minnow.cc.gatech.edu/Squeak/1, the Mailman project, http://www.zope.org/Members/bwarsaw/MailmanDesignNotes/FrontPage, the Tcl'ers project, http://mini.net/tcl/, the PHP Wiki project, http://phpwiki.sourceforge.net/, and by parsimony, a company that I am told offers over 25,000 different wikis, including of course, the all time favorite, the forum satanismus. You can find out more about wikis and who uses them at http://directory.google.com/Top/Computers/Software/Groupware/Wiki/
What do many of these projects have in common with OpenACS? They are communities of disparate and widely separated developers working collaboratively on some commonly used piece of software. I am not sure in what venues you have found wiki's disappointing, but there is, I gather a sizeable group of users that appear to love them and depend upon them.
One reason why implementing one wiki module might beat out the implementation of all of file-storage, general comments, and related links: with a wiki page, if you want to create a wholely new page, you just type in the link. The page, and it's versioning history, is created for you. Type in the link. Edit the page.
To quote from the php wiki project:
"The addictive quality of a Wiki is that making pages is as simple as making a link to them. If they don't yet exist, the page link will be followed by a hyperlinked question mark; follow that link and you can define the new page."
I've given you some links, may I ask that you explore them once more and reexmine the question: can self-organizing sites work out to be useful?
I like Tom's suggestion that we turn general comments on the /doc and other pieces of documentation at OpenACS. Regrettably, when this was suggested back in April it died a quiet death and the community moved to using wp as documentation tool in early May. We quickly wrote four wps and no more have been written in a month. One problem with using wimpypoint, to the best of my knowledge there is no integration of wimpypoint and general comments yet, so there is no way for users to leave comments about any of the wimpies they read.
Of course, this message itself is just once more around the loop for most of us. Most of us here participated the first time in April, and then again in May, and then again again in May.
Sigh,
"I like Tom's suggestion that we turn general comments on the /doc and other pieces of documentation at OpenACS. Regrettably, when this was suggested back in April it died a quiet death and the community moved to using wp as documentation tool in early May. We quickly wrote four wps and no more have been written in a month. One problem with using wimpypoint, to the best of my knowledge there is no integration of wimpypoint and general comments yet, so there is no way for users to leave comments about any of the wimpies they read.
Of course, this message itself is just once more around the loop for most of us. Most of us here participated the first time in April, and then again in May, and then again again in May."
It occurs to me that we are talking a lot about changes lately, but nothing visible happens. My suggestions are:
1. We have to use polls a lot more. In order to make them visible for every member, the polls should appear on the first page as news. Additionally they should have a deadline and after that deadline they should be implemented as soon as possible.
2. Maybe we need a person that has the rights to implement changes that are wanted by the community. I have quite some examples in mind, where we came up with nice ideas to improve openacs.org, but they faded in the dark as they were never implemented. I understand that the founding members put a lot of work in getting 4.x done, but then again we need a webmaster that is not involved in the porting effort and therefore has enough time to improve openacs.org.
3. If the polls return the majority of the community members want a Wikki, a Wikki should be implemented ASAP period. Let's get some more democracy into openacs.
I believe that people slowly get frustrated with only suggesting improvements.
From a marketing perspective the appearance and features of openacs.org are at least as important as the 4.x porting...
Let's vote on a webmaster who is willing and has the ability to change the face and function of openacs.org...
In the OpenACS community, there are plenty of members who would be highly unaffected by these decisions, and therefore who would certainly vote on a whim and without much consideration for the long-term good of the community. For example, Jerry wants an insecure Wiki, but he doesn't want to be there if the box gets hacked and "someone" has to recover the damage. Sorry for picking on you, Jerry, but that strikes me as a rather irresponsible way of growing the community. And there are many members who would be *significantly* less affected than Jerry by many of these decisions.
The reason a small core of people has more power is that someone must have the strength (and authority) to respond and to VETO. Otherwise, the community, composed mostly of lurkers, will overwhelmingly say "yes" to everything. Do you think Linus takes a vote on all features to be included in the kernel? No, because it would lead to ridiculous feature bloat. And if we left all webmaster decisions up to the community at large, our web site would have every means of communication possible, which would lead to dozens of duplicate forums, hundreds of duplicate instructions, and overall confusion.
I'm spending about 20-30 hours per week right now talking with many members of this community, mostly in public, sometimes via private email, to figure out the *right* way to move forward. Some have proposed doing a new design, which I'm enthusiastically supporting. Someone else has submitted a new, more "business-like" logo, which I also support. I am in the process of creating a new, highly-advertised page that explains very clearly how to contribute to the site, in response to much of the criticism of the past few days. That should be ready by Tuesday. When Don comes back from his time off, I'm guessing he'll jump right in and help us figure out where to go. Guess what? We're not going to make everyone happy. Don and I had a long discussion about Wikis, and we came out with a resounding "no, this would hurt the community terribly, and we have no example to disprove this strong suspiscion." And so we made that decision. Frankly, I wish the discussion could somehow move off the darn Wiki issue and think about the higher-level problems. It's starting to get quite petty.
That said, I only have so much time and energy, and I'm starting to repeat myself without much support from other members. At the end of the day, my power (and Don's power) to guide the community (which sometimes requires overruling the "majority") can only work if most members are willing to trust us with that power.
I have set up a poll to see if the community wishes to elect a new webmaster. This poll will remain open for one week. If the vote is to elect a new webmaster, I will open up nominations on a discussion forum, and then attempt to run a somewhat fair election (probably closing off registration of new users for a while).
The poll is
Sorry Ben, I didn't really start this thread to discuss the merits of WIKIs, only to point out the complete lack of meaning that the name has. A WIKI now appears to me to be complete anarchy. It seems to be a proxy for indecision and disorganiztion.
The basis of the intenet is the careful delegation of control and responsibility. Usually the people in control are given the opportunity to shoot themselves in the foot, but hopefully not anything (or anyone) higher up the chain of delegation. A WIKI appears to give ultimate control to those with the least responsibility.
The hyperlink is a prime example of a mechanism of delegation. The hyperlink has crosslinked the hierarchical structure of the internet making it much more useful.
My suggestion is that the hyperlink should be used instead of whatever a WIKI does to provide more than a WIKI ever could.
Now for those who wish to create files with abandon, try out zipi: http://negeen.com/zipi/. Zipi is going to be a collection of AOLserver scripts that can be written while directing half attention toward a late night talk show.
As far as using Wiki's on OpenACS.org, that's a judgement by the site administrators and I have enough faith and respect in Ben and Don to accept their decisions.
However, that doesn't stop anyone from implementing a wiki and seeing if there is some utility in such a tool. There are some communities that get a great deal of work accomplished through them, so they must have some use. Implemented with strong version control (if that's possible) might solve the most obvious problem.
Still, I don't think this is such a big issue. I have a feeling we're going to be yelling at each other on bigger issues soon. And Don hasn't even gotten back yet, and we're going need all of our collective breathes to deal with that dude.
talli
I think you're being a bit disingenuous, or maybe I just haven't made myself clear.
I'm not asking for a new webmaster, I've suggested, in this other thread, https://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=00025b, that we put together a steering committee to determine a bit of direction, a bit of policy, and a bit of organization.
I believe your new poll sets up a strawman to be defeated. We're not asking for a new webmaster. We all appreciate and are grateful for yours, Don's, Roberto's, and everyone's efforts. I would have preferred a poll that asks the question: would you like to see a steering committee formed? Why do I think a steering committee is needed? Because I cannot create a poll at this site /admin/poll, and I don't like strawmen arguments.
I never said I wanted an insecure Wiki. I said my preferences would be that given the alternatives, I would prefer to implement a wiki today, and migrate it/improve it later, and deal with the security risks today, with the tradeoff that we might get 0wn3d on occasion and as we learned.
By the way the first page of a google search for wiki and exploits reveals none known that lets the h4x0r 0wn a box, just the justified concern for letting arbitrary html into any wiki page and how that might effect users. So if there was any chance that a wiki might be implemented somewhere, I would suggest ensuring that there are filters in place, much as the ACS has filters too, to ensure that arbitrary and naughty html doesn't make it in.
I never said I didn't want to be there when the box gets hacked. I said that realistically the box is 3000 miles away from where I am, and that I won't be there when the box gets hacked. And I acknowleged upfront that that distance might make it too easy for me have the preference that I do.
I don't think that makes me irresponsible. I'd prefer to think it shows vision, confidence, and trust in the community.
I would like this community to be open to experimentation. Sometimes experiments go awry. Sometimes they don't. And sometimes we learn best through our failures. How will we know? How will we know if modules are scalable? This is probably one of the largest OpenACS sites, and my preference would be that it's okay for this site to occasionally be down, and even get restored, because I am confident that that's the best way to make it a showcase for open source, openACS philosophies and technologies. It would be one of the best showcases for an OpenForce, Furfly, Civilution, Digital-People, ybos, Museas, etc. to show.
I would like to see this be a learning community. Usually three steps forward and occasionally one step back. It would be a statement: we're open, we're honest, we share, we learn, and we're confident in our solutions. Experimentation. Boldness. Leadership. OpenACS.
I would like to see the OpenACS embrace web best practices that create well engineered solutions, to the greatest extent possible, regardless of underlying technology. I think we can learn something from our competition. And I think that by doing so, we can best our competition, coopt our competition, partner with our competitors, and leverage our competition. Okay hands now: which OpenACS companies out there are purposefully not exploring any of Java, Python, Zope, W2K, or C#? Who amongst us doesn't belong to a mailing list powered by Mailman?
If folks want W2K OpenACS, cool beans. If someone builds .NET hooks into OpenACS, okay. If someone suggests that we explore what a wiki might do, or that we turn general comments on /doc, I'd like to see that given a chance too.
Isn't it silly, isn't it bizarre, isn't it frustrating that postgresql.org has no db backed discussion groups? We think that those pg developers are sitting in the dark, using their mailing lists, and not turning to the light of the ACS. They're not interested in experimentation or outside technologies. How would you ever convince them otherwise?
Ben, let me apologize. I don't feel picked on, but I'm pretty sure you do, and for that, let me apologize once more.
I'm doing the best I can to understand how to proceed here:
- If you want some space on OpenACS.org to implement .NET hooks, SOAP, or any other project where you can manage the entire SDM segment, just email me. You will have that space within a few hours, and I will also personally support this effort because I, too, think it's a good thing.
- If you want to help PG have an OpenACS-backed site, I'll put you in touch with Marc Fournier. He, Don and I discussed this a while back, but it faded when everyone became too busy to make it happen. If you have the time, I'd love to have you take over that role and get the PG guys to use our stuff.
- If you want to really experiment with a Wiki, take risks, etc... do you think that OpenACS.org is the place to do this? I don't. However, if you want to set up a box and run a Wiki, I will point wiki.openacs.org to it, and I will make a link from the front page of OpenACS.org. How does that sound?
- My poll is no strawman. David said he wanted it. I won't make a decision on the steering committee until Don gets back, but I can make a decision on the webmaster situation given that I am the current webmaster.
Again, I am begging you to put forth serious reasons why a steering committee would improve the situation. "More organization" is not an end, it is a means. A means to what? What are we not accomplishing because we don't have enough organization?
I don't want a wiki. I want a solution. I don't understand why people keep asking me for configuration files for nsvhr/nsunix. I've put that information five gazillion times into the bboards here, at aD, at AOLserver, and in my own howto. But all I can do is put it into bboards, and my howto, and pray to Google. Why can't they find it? Why can't they find what other folks have said about it? As David Eison points out, I want any users to have an edit-this-page ability to create new content and help out. I want a way to take that content and reuse it (David, the jargon word for that is repurpose) and to promote it in different ways.
All those folks struggling right now to figure out new ways to document the ACS? I'd like them to have the ability to just go off and do it. On this site. openacs.org/project/new.tcl. With a CMS/CVS/Wiki store behind them.
I don't want to have to ask you for permission, nor do I want to have to wait for your help. I approve of policies. I approve of saying no. I approve of 24x7 development strategies that reach beyond one individual administrator. I like capturing energy. I don't like opaque policy decisions coming from folks that are admittedly overloaded.
I don't believe that all links must be forever links. I don't believe in six 9s reliability for a learning site.
loop {I don't like the loop collect into friction returning friction}
I do like discussions. But I also like closure. I don't like personal attacks on anyone. I don't like overloading anyone.
I do like having a committee that meets with intention, that listens, that can designate resources, that can offload resources, that can openly create policies and make policies open, that presents a face to complain to as opposed to people to complain about, that I can participate in, or just make comments to, and that can represent us to outside entities: press companies, etc. much as the Apache Software Foundation represents Apache.
Do I think that OpenACS is the place to experiment with a wiki? Yes I do. I certainly understand that you disagree, and that that makes it basically a noop. As I've said, I think it's okay to have a showcase/testbed OpenACS.org that goes down on occasion in the name of learning. Learn. Discuss. Act. break loop;
But I understand that you disagree.
Why doing it myself isn't a win. Dossy Shiobara has an AOLserver wiki. I've used it. It's useful. He keeps it up. Regrettably, and very unfortunately, having a random X buried in a random location, doesn't generate the network effect we need to learn about X or to turn X into a resource. A week or two ago, I put up a site-wide-search interface on my site that indexes the OpenACS. It's gotten about three dozen hits that are not my own. I've received absolutely no feedback about it, except for some scary reason, folks believe I am now the search engine expert. Yet gosh, all I've done is to implement htDig within the ACS, and that's something Aaron did a bazillion years ago I guess.
Right now there are three implementations of XML-RPC for tcl/ACS. There's Steve Ball's, there's Aaron's/Dave Bauer's, and there's Hafeez Bana's. Regrettably, not a one of them has written any documentation on how to use their package. They all apparently encourage others to use their package and document it. Today, I'm staring at all three packages trying to determine which is most apropos for my work.
I wish, I just wish, there was some way I could get them all collaborating and discussing in a focused manner in their own section of this site. They are far smarter than I. They know alot more about how to put these packages together and how to turn XML-RPC into SOAP than I. And I would like to participate in that project. But I wish that project could have come about naturally on its own back in February when Dave and Hafeez apparently started simultaneous projects not knowing the other existed. I wish they could have mentioned it in one message, agreed in another, and then visited openacs.org/project/new.tcl, without having to ask you for approval or asking you for your time.
So I'm not sanguine about the outcome of my putting up a wiki at my site. Not enough traffic to test or convince or learn, and just not enough reasons for folks to try it out. And for gosh sakes, I'm at the tail end of a DSL Line in PG&E's block 14, and that's better than many developers here, who have no machines on the net, just a dial up line and an interest in participating. I gave one of those guys some space on my machine. He may not know it, but he took my machine down once or twice. But he won a(d) prize, and he's touring Boston.
Plus the time and resources issue. I'd like to see what happens with a wiki, and I know I'm not the only one around here to do so. But I am frying other fish, and I'd love the ability for any of us to suggest something and let others run with it in a supported fashion, on this site.
Jakarta. I'd like jakarta. I'd like the reputation of jakarta. I'd like the experimentation of jakarta. I'd like the functionality of jakarta. I'd like the network effects of jakarta.
Oh. Damn. Moments before clicking on confirm, you send me an email graciously asking me to try and resolve the issues. So what do I do with this?
Well, I don't have the answers. I really and truly believe that now is the time to cons a steering committee, to cons up a process. It's not a lose, it's really a win.
wikki. You know what? I find it harder to find things there than
here.
I've spent at least half a year trying to put together some kind of a
coalition or steering committee for an OpenACS-based solution
that shall remain nameless. You know what? The openacs.org
community is much more efficient at getting things done
informally.
Do we need a better system of organizing knowledge here?
Sure. What we need is basically a light-weight knowledge
management system.
You want a solution? Fine.
Here's an easy one:
Create an FAQ using the FAQ module. In the answer sections,
create hyperlinks to the HOWTOs, wimpypoint presentations,
documentation, bboard posts, and whatever else.
Problem solved. No wikki. No committee. "Easy peazy," as the
Naked Chef likes to say.
Let's all create a faq then /faq/admin. No wik[ki], no committee. Easy Peazy as they say.
your core argument is that people should trust you and Don with your judgement. I for my part can only say that I trust most members and especially you two as of your immense technical skills and the work that you have put into OpenACS which outrises mine by quite some factor.
I didn't mean to replace you as a webmaster, but suggested to additionally take somebody new into the boat, who increases time to market of changes to face and function of openacs.org. I must have expressed myself a little bit too fuzzy.
Out of my own real life community experience I can say that one of the most important factors to motivation are transparency and the ability of every member to change things.
Therefore I will not vote on your created poll, as it has not the options I would have voted for...
Maybe we could have a category inside bboard called "improvement of openacs.org" where propositions are discussed. After the true pros and cons are discussed up to a certain point or date the community can vote on it. The funny thing is that I never heard until your last thread that you and Don already decided against a Wiki and that security issues were one of the major reasons, if I understood you correctly. Note that I am only taking Wiki as an example!!!
After the community has voted or the "steering committee" has decided the result should be posted on an openacs.org "improvement roadmap". This roadmap would contain the proposed changes, whether or not they will be realized, until when if yes and who is the coordinator of that change.
Up to your last thread I didn't know that you and some other people were already discussing design changes to openacs.org. I mentioned design in the past, but never heard anybody responding. This I mean with transparency...
I like the idea of no taxation without representation. But keep in mind that citizens under 18 are not allowed to vote.
And now I hope that the fire that has been in some words doesn't burn the house but gives it a unique new color 😊
<dream>If the companies involved in the community decide to support a dedicated webmaster for openacs.org and showcase sites, I don't think Ben would object leaving the tedious work of maintaining a site (and subsites for each major project) to someone else. But does that mean we think he is not up to the job anymore. Well at least for me I can say, the main point is I'd like to see Ben write tons of papers like the Java one and evangelise about OpenACS than to work on setting up a bboard, check the error log of openacs.org and "fight" his way through all the requests for changes on the site.</dream>
That beeing said, let me come to my second point. I think most of the people asking for w*** and stuff like that want to have a way to store structured knowledge that is easy to find for others and easy to maintain. There are a lot of approaches to this, I've described the one I favour in german at http://www.sussdorff.org/km-article.html. aD implemented most of this at Siemens. I hope they will release this back to the community. Talking about it, I'll create one for Knowledge Management with OpenACS. Don't expect it to come up within the next three weeks though, as my final exams draw nearer.
P.S.: Oh yeah, if someone wants to translate my paper or help with translating, just do it. I wont have the time to do it till middle of October.
"Out of my own real life community experience I can say that one of the most important factors to motivation are transparency and the ability of every member to change things."
Ability to change things is one thing, but ability to change everything to everybody is insane, utopic and unrealistic.IMHO what is need is more action. Do you have the talent to jazz up a better website for openacs.org? What's stopping you from e-mail'ing Ben and letting him know? You want to help with documentation? What's stopping you from e-mail'ing me and writing it?
Malte, the members of this community always had the power. They have the power now. Thing is some people are more committed than others, and some people don't want to commit as much. Want to be a leader? Show me your actions first. Show me what you're doing or have done for the community and then we can talk about a steering committee.
More bureaucracy will only lead to more "paper" work and taking the fun out of working on OpenACS, which is the whole point, IMHO.
I absolutely agree with you that we need more action.
The problem is that the infrastructure for talking (bboard) is great, so people enjoy talking on openacs.
The infrastructure for improving the site is not closely as efficient as the infrastructure for talking, as we have recently seen from some people's postings.
I would like to give another example for improvements that never found their way to openacs.org:
1. We had a discussion on SDM. Pascal created an improved SDM, but it never got implemented.
Like Malte I am tumbling towards my final exams at University here in Germany. So I unluckily cannot contribute to a better face and function at the moment. But as soon as they are over and I get back from a deserved vacation, I will stop talking and take over some responsibility for the appearance and functionality of this site, in case the openacs godfathers let me 😉 (Mid of August...)
I just created a new thread for issues and priorities of openacs.org improvements:
https://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=00026j&topic_id=11&topic=OpenACS
The purpose would not be for people to ask questions, but would instead be for people to contribute more info to the docs in a rapid manner and make it easier for the writers/ maintainers to know what is missing or needs clarification.
After a comment is read, and the docs changed (if needed), then the
comment should be deleted.
The SDM is just not easy enough when it comes to making suggestions
from the floor for something like the documentation; it is great for code
changes, bugs, and feature requests.
Give people good tools, and give several people (if they are willing to step up) the job of reviewing the comments and improving the docs. I think it could do wonders.
Sam
I said:
> ...software requirements... PostgreSQL + ??
Vince Vielhaber [vev@michvhf.com] said:
It's set up on apache w/ PHP 4.0.x and PostgreSQL.
I wrote a script that read each html doc file and encoded it so
it would go into the database (escaped apostrophes, etc) and
stored that along with a language tag, page number, reference
number, filename and a couple of other things. The php script
that does the work (index.php) does a global replace upon display
(not on storing it in the database) that puts:
http://www.postgresql.org/idocs/index.php?
in front of all urls so anything displayed or clicked on will go
thru index.php. if it's a non-local url a redirect is made to
the
raw url. The search simply does an ILIKE search, after removing
any malicious characters (there are morons everywhere that just
live to break things, one even had the gonads to send me email
about it!) so I only allow a-z A-Z 0-9 and spaces, all else is
converted to a space. The user comments are converted before
being
stored to their printable forms (a PHP function does this) and
that's about it. The source isn't available just yet but I want
it
to be in the not too distant future - but don't wait on it :)
Vince.
He also said:
>There aren't really any requirements to the interactive docs beyond
that
>they're not supposed to be used as a support forum. Before I
delete any
>of them I look at them long and hard, quite often something that appears
>to be a support question (and may very well be) will be an excellent
way
>of saying the docs are omitting something fairly obvious. On
the other
>hand the "I don't know how to do this" type of questions are deleted
>without giving it a second thought. Source code and config examples,
>corrections, corrections to the source code or config examples, additions
>of explanations, and other stuff like that is mainly what it's about.
>If you take a look at the interactive docs that the PHP folks use
>(www.php.net) you'll get an idea since that's what I modeled it after
>somewhat.
Best Regards,
Carl Garland
