Community effort to evaluate what 5.6 documentation applies to 5.7
Permalink
The core team has the challenging role of being the primary developers and also being responsible for documentation. With an active Github issues list, the available work hours have understandably been used for development.
In the meantime, there are many new users looking to do more with concrete5 (greater ability to offer community forum help, develop free blocks, make dashboards, contribute core code, etc), but limited by lack of documentation.
The idea is based on 5.7 bringing a lot of changes, but still sharing many aspects in common with 5.6.
Until more 5.7 specific documentation is created, would experienced community developers be open to picking a small section of 5.6 documentation and seeing if it applies to 5.7?
And to bring up an older, but important thread, related to community involvement in documentation and the idea of having it all found in one place.
1. Base documentation and API documentation by the core team and then expanded tutorials, screenshots, how-tos, infographics, screencasts, related answered forum questions, and examples by the community.
http://www.concrete5.org/community/forums/documentation_efforts/tow...
2. Include community based documentation in the current concrete5.org docs or a separate wiki?
http://www.concrete5.org/community/forums/documentation_efforts/tow...
In the meantime, there are many new users looking to do more with concrete5 (greater ability to offer community forum help, develop free blocks, make dashboards, contribute core code, etc), but limited by lack of documentation.
The idea is based on 5.7 bringing a lot of changes, but still sharing many aspects in common with 5.6.
Until more 5.7 specific documentation is created, would experienced community developers be open to picking a small section of 5.6 documentation and seeing if it applies to 5.7?
And to bring up an older, but important thread, related to community involvement in documentation and the idea of having it all found in one place.
1. Base documentation and API documentation by the core team and then expanded tutorials, screenshots, how-tos, infographics, screencasts, related answered forum questions, and examples by the community.
http://www.concrete5.org/community/forums/documentation_efforts/tow...
2. Include community based documentation in the current concrete5.org docs or a separate wiki?
http://www.concrete5.org/community/forums/documentation_efforts/tow...
@JohntheFish
When the idea of a wiki was being considered, was there any research on the wiki host?
Ideally, it would be best to find one that was free for open source projects.
Also, I have a few ideas related to the concrete5 IRC channel.
The IRC channel idlers are generally experienced, often very experienced, concrete5 developers. A lot of moderate to complex questions get answered there (questions not often seen in the forum). It seems like a lot of great information is getting lost by not sharing it.
At the minimum, finding out if the channel bot logs can be made accessible for manually browsing for answered questions.
Even better, possibly using a trigger system for questions and answers, a little like the forums. When someone asks a question they start it with "!question:". If a question is answered they type "!answer" and the chan bot makes a notation (user name and timestamp of !question: and timestamp of !answer). This process would make digging through logs more automated and less time intensive.
When the idea of a wiki was being considered, was there any research on the wiki host?
Ideally, it would be best to find one that was free for open source projects.
Also, I have a few ideas related to the concrete5 IRC channel.
The IRC channel idlers are generally experienced, often very experienced, concrete5 developers. A lot of moderate to complex questions get answered there (questions not often seen in the forum). It seems like a lot of great information is getting lost by not sharing it.
At the minimum, finding out if the channel bot logs can be made accessible for manually browsing for answered questions.
Even better, possibly using a trigger system for questions and answers, a little like the forums. When someone asks a question they start it with "!question:". If a question is answered they type "!answer" and the chan bot makes a notation (user name and timestamp of !question: and timestamp of !answer). This process would make digging through logs more automated and less time intensive.
That an IRC question gets answered, but the answer isn't easily (if at all - I don't know) preserved or searchable by others is one of the many reasons I prefer the forums.
Another is a poor signal to noise ratio on IRC. If IRC became too popular, it would soon become swamped.
I doubt if many would have the discipline to mark questions.
IRC was for a short while used as the chat channel during Totally Random live shows and there was some filtering from Totally Random questions on IRC into reddit. I haven't seen anything since TR was shelved.
The idea of a documentation wiki was turned down well before it got to any detailed thoughts on implementation.
Another is a poor signal to noise ratio on IRC. If IRC became too popular, it would soon become swamped.
I doubt if many would have the discipline to mark questions.
IRC was for a short while used as the chat channel during Totally Random live shows and there was some filtering from Totally Random questions on IRC into reddit. I haven't seen anything since TR was shelved.
The idea of a documentation wiki was turned down well before it got to any detailed thoughts on implementation.
Were there any specific reasons why a wiki was turned down?
I prefer not to have a wiki, instead including all information on concrete5.org in one well organized place.
I prefer not to have a wiki, instead including all information on concrete5.org in one well organized place.
The current emphasis on developing the core above documenting what is already there is short sighted (but typical of how top end developers work). If work on new core code was suspended for a week and that week spent on creating documentation, more developers would be able to contribute and that week would soon be recovered. The sooner such a decision is made, the bigger the effect it will have. It should have been done months ago.
One current business case for the core team is progressing Simple Sites as a revenue stream. That will be a closed environment maintained by the core team. So they don't need documentation for that.
PS. Themes are needed for simple sites, and theme development happens to be the part of 5.7 most documented.