search

openstax / napkin-notes

:notebook: virtual back-of-a-napkin notes (persistent non-structured discussions)
3 stars 8 forks source link

:art: fix typos #50

Closed philschatz closed 8 years ago

philschatz commented 8 years ago

@pumazi

Fixed some typos while reading; I think the 4 phases make sense. How much of the code could be reused in each phase between the DMS-solution and the final implementation? For example

Also, I wonder if the service could be designed to work with GitHub Webhooks to provide validation (and preview PDF generation) on push 😄 . It would require:

/cc @rich-hart

mmulich commented 8 years ago

Yes, as we discussed in previous meetings it’s likely that the source material will be pulled from github. Making it compatible with the file structure is still a requirement.

mmulich commented 8 years ago

This tool does not involve cnx-authoring at all. So, yes, cnx-authoring be completely bypassed. The document mentions cnx-authoring because the http service (casually named DMS Press) moves from the publishing component layer to the authoring component layer through the iterations.

I probably need to diagram the component layer parts of this, as it's not clear what this means to anyone. Essentially there are 3-4 components in the cnx-suite. archive and publishing can be seen as the same component, but we'll count them as 1 and 2. The implementations for these components are cnx-archive and cnx-publishing. These makeup the core of our system.

The 3rd component is authoring. It's more likely to see more than one implementation at this component layer. Our implementation of this at this time is cnx-authoring. Basically, the authoring component allows for editing and submission to a publishing component. The 4th component is a little fuzzy at this time, but it comprises worker processes that are pre & post publication processes. We don't have any of these at this time, because publishing hasn't been factored to separate them from itself.

With all that said, the DMS press service at first lives as a publishing component, because it persists content directly to the archive database. Later it becomes an authoring component after it has dropped support for directly persisting the content in the archive. Instead it uses the publishing component to achieve that goal. In the middle it is a hybrid component of both authoring and publishing.