Is this a JSR170 port?

[Ruben]

Hi there,

Is SimPCore a JSR170 port?

I am actually building a PHP content repository myself at the moment
and was just browsing around a bit to see what others have done. It is
a complex task!

Regards, R

[Murphy]

hi ruben,

"Is SimPCore a JSR170 port?"

-- SimPCoRe is an attempt to mimic the JSR-170/283 API purely in PHP
... the backend can be whatever you want it to be ... shoot, it could
be JackRabbit :-) default example uses MySQL as the
backend/persistence manager :-)

murph

[Georg Wächter]

Hi,

is there somewhere a roadmap? The project looks very interesting, but
i'm wondering whether features like versioning and node types are
going to be implemented.

georg

[Murphy]

"I'm also familiar with the data first approach (due to this page
http://wiki.apache.org/jackrabbit/DavidsModel) and i really prefer
this method over the old RDBMS structure-first design."

-- amen to that ... also, while we are working out the kinks in the
PCR, we are also working on a PHP equivalent for Sling
(http://sling.apache.org/site/index.html) ... i think we will call it
Phling :-)

"Beside this node types are also useful in my opinion - not only for
documentation of the data but also when using the repository data with
more than one application."

-- blah ... all i know is whenever i am developing against the JCR, i
always use the node type "nt:unstructured" ... which is like not
having a node type at all ... but i get your point ...

"btw: your reply didnt show up on the public mailing list"

-- thanks ... doing so now :-)

****************************************************************************

"is there somewhere a roadmap?"

-- not really ... more like: what do we need/want next, and input from
members will drive this project ...

"i'm wondering whether features like versioning and node types are
going to be implemented."

-- versioning yes ... node types no ...

i was about to write a little diddy about how a CR allows for a "data
first" mentality, versus a "structure first" mentality ... "data
first" - basically it means, if you code it, it must be true ...
"structure first" - is what we have been doing for the last X number
of years (define the DB, define the tables, define the tables'
structure, define data types, etc.) ... to us, node types are a not
needed if a well defined development process is in place ... plus,
it's overhead that isn't really needed in all use cases ...

i will write more later, but that's basically our thinking for now ...