Hi all,
So, I really want to build a decoupled CMS and I want to invest my time
over the coming months to that purpose (providing I can actually find a
stable place to stay and work!).
As such, the first difficulty was how to do this rapidly in this
organization. As David has already mentioned he and Lukas are currently
further away from the organization and Wouter and I have been working on
the "new" things which I have introduced.
But all progress is painfully slow, even if this slowness often leads to
positive gains in code quality.
As such I have been considering the development of a new organization in
which I can rapidly develop the things that I "need" to create the
decoupled CMS.
The organization itself has the following features:
- One repository per library.
- Documentation contained within the library repository.
- One repository per "driver" (e.g. description-phpcr-odm,
content-type-doctrine-orm).
- Dedicated documentation site (readthedocs).
- Automated meta files via. Ansible (README, LICENSE, .gitignore, etc).
- Starts on PHP7 and Symfony 3.0.
Current components:
- resource-browser (no UI just model).
- description
- description-phpcr-odm
- content-type (I would transfer it from the Symfony CMF)
- meta (org management repository with ansible playbook)
The organization would be driven by the development of an actual CMS (in
a separate org), which may or may not be my experiment with SyCMS[1] but
the overall aim of the actual "new" organization would be to encourage
users to build their own CMS implementations, and the DX would be a
primary consideration.
So it is amazingly similar to the Symfony CMF but at the same time
running on a different track.
It is possible that it would not use any existing Symfony CMF
components other than Routing/Auto[2]. This means that if the new
components were developed in the current organization, it would mean
either branching efforts between this "new" stuff and the "current"
stuff, or deprecating all the "current" stuff[3].
Then there is the fact that I would be able to drive the organization
and take responsibility for it in order to produce the result that I
(we?) want.
As you can see I have already gone a long way down this decision path
and I actually cannot see a way to do what I want to do without this
"branching" of organizations. But maybe there is a way and I cannot see
it, hence this email.
It would not surprise me if if this all falls on its face. Perhaps again
more reason why it should be done in a separate organization as, as has
been mentioned numerous times, this organization does not have the
resources to maintain these things, more so I imagine when they are dead
ducks.
So I am interested in what you think,
Peace,
Dan
[1]
https://github.com/sycms (note this is *not* the new org).
[2] And it is conceivable that even RoutingAuto could be replaced by a
more generalized system.
[3] Here I am not indicating that the "current" stuff is bad, but that
it could be handled differently in the new organization.