opencms-dev Future features etc

[Robert Schrem]

Hi Alex,

>For he current alpha release (4.0aXX), we will focus on implementing all
the
>planned but still missing features. If all the features have been
>implemented, we will have the beta release (4.0bYY). In the beta, we will
>focus on improving performance and quality. If the system is stable and
fast
>enough, the production release (4.0rZZ) will be made. After that, we will
>have enough ideas for new features, and restart the process.

Sounds good. In the first place i didn't realize that it was just a ALPHA
Version.
I thought it would have been allready a final.

Do you have any Idea about a time schedule for a beta or final release ?

>
>> Another thing is the lack of documentation.
>
>The next release will contain example templates with Javadoc documentation.

Goood !

Maybe a RPMS is also a good idea, because setting up the hole System is
quiet tricky. I guess I wouldn't be able to do it, if I didn't had some
experience
with Apache/JServ/MySQL admin.

My two cents about your planned future features:

------ PostgreSQL support: Wait for a few months until InterBase 6 gets
released as
OpenSource. It's a full featured SQL Database for serval platforms and
is faster than PostgeSQL (see www.interbase.com) in current perfomance
benchmarks.
It has a solid SQL system, complete SQL implementation, Transactions, online
Backup etc...

------ LDAP over JINI:
Wouldn't it be great to store hierachical structured XML data in a LDAP
directory?
Data like user navigation trees or maybe even structured text documents?
Wouldn't that be quicker than using a table-based database system (like SQL)
because you wouldn't have to emulate a hieracical data structure in SQL?
Just let me recall the strengths of LDAP:

- FAST read access (at the cost of slow write access, for wich we don't
really care),
- standardisized way to replicate LDAP directorys over the net (with working
implementations),
- clever security features (tested and found to be good),
- regular expression search algorythms (that are usually usefull anytime),
- a free OpenSource implementation (www.openldap.org) etc. ppp.

To me it seems to be the perfect match for XML stoarge, or am I completly
wrong?
A starting point could be the LDAP Processor from the cocoon Project to get
XML datastructures out of a LDAP server sytsem via JINI (see
http://xml.apache.org).
I haven't tested this, but the docs look promising.

------ Full Text Search Engine: No need to search any further or
thinking about implementing it: take a look at www.htdig.org. A great
search engine that simulates a browsing person and follows all the Links -
even across HTML Frames. Just like AltaVista.

Aviable as OpenSource for free and for all relevant operating systems.
It has even a fault tolerant string matching algorythm implemented and can
handle differet languages (e.g. german language special characters that are
coded as ü in HTML etc.)

I think it would be hard and even unessesary to beat that just for the sake
to have it coded in JAVA.

that's it - fo the moment.

your robby