Document publishing process

[kelson42]

To secure the quality of the publishing process, things have to be done following a clear and rigourous process. So far this is not documented properly, We should document it to:

• As a reminder/howto for content responsible
• As an input for the CMS developers

I would put the documentation in the wiki (in this repo).

[kelson42]

@RavanJAltaie @Popolechien Please work on top priority on this. Content quality and dev people suffer uselessly beceause this document is missing. @rgaudin and myself will be happy to work as reviewers.

[kelson42]

@RavanJAltaie @Popolechien Any feedback?

I see as well that the recipe which fail seem totally ignored (I keep discovering recipes not working for months and which are not reported), nobody seem to take care of them in a regular and frequent manner. But they have to be taken care to identify problem and maybe fix them either with a new configuration or maybe a software fix.

[RavanJAltaie]

Hi Emmanuel, Actually this has been done long time ago

• The documentation on how to open an issue for a new zim file request has been published here on Github after reviewing it with @Popolechien (https://github.com/openzim/zim-requests#readme)
• Kiwix content manual on how to create a recipe for mwoffliner & zimit scrappers and how to report an issue are created as well and reviewed by @Popolechien You can have a look here ([https://docs.google.com/document/d/1XT8RTddCNkIIBH_5_cpEtsqVZ7yyYkMmaJGPbbHji7Q/edit#heading=h.qerpdy92de9u]) Any modification and feedback are welcome.

[kelson42]

@Popolechien @RavanJAltaie Thanks, it's helpful, but they are mostly howtos, not really process (and for sure not policy). Any reason why the second part is not public? We follow open developement principles, things should be made public and available to anybody AFAP.

[rgaudin]

Thanks for sharing the document but:

• it's directive (do this) but doesn't explain the reasons behind the choices so reader cannot make an informed one.
• Many recommendations are questionable and lack precision.
• Covers only a small portion of zimit and mwoffliner.

There is close to no knowledge in this document. Most of it is self-explanatory and I doubt anyone goes back to it after the initial walk around the Zimfarm.

[RavanJAltaie]

@rgaudin @kelson42 Thanks for your feedback, I'll discuss this subject deeply with @Popolechien to update the process

[rgaudin]

As I said somewhere else, I think we need to set goals and expectations collectively so we can help you with documenting the technical aspects of it. But without a process to apply, there's nowhere to start. Identifying what task you do, why and in reaction to what event would be my methodology.