Re: [discuss] Re: [Evolution] OO - GROUPWARE - Call for concluison- was: OpenOffice: Say it isn't so.



On Mon, 5 Feb 2001, Dan Kuykendall wrote:

== Recognizing The Potential:

Evo and OO both recognize that comprehensive groupware functions GREATLY
enhance the appeal of their respective products:  Evo sees shared
calendars, shared contacts, a user-administered back-end permissions
system, etc. all contribute to making Evo more valuable.

Can I mention that phpGroupWare (and other such solutions) would also
benefit greatly by a joint effort? We are already moving toward doing
this all by ourselves, but could benefit from a shared effort.


Of course anybody who awants to have something benefits if others go out
and start doing it for them 8-)

==  Facing The Reality

Evo and OO realize that building something that competes (at least on
some level) with Exchange and Notes is a significant undertaking.  Their
products are shaping up, but there is still a LOT of work to do.  Time
is of the essence.  Both groups conclude that their time is better spent
working on their core products; but gosh-darn-it, it sure would be nice
to have those back-end features.

Ahhh the ever present dilemma. I think that this can be solved. In fact
I am working on solving part of this with the phpGroupWare project (will
explain more later).


Everybody wants to see at least some light at the end of the tunnel before
running headlong into it.

== The Light Comes On

Evo and OO know what what open source is all about.  They
advocate/cheerlead/champion an open groupware initiative, put out a call
for developers, post notices on their websites, and sound the general
call to arms:  "We want to free countless corporate employees and system
administrators from the tyrannies of expensive, proprietary groupware
systems and the whims of their vendors!"  To get critical mass, they
might ask some other prominent open source groups - or - gasp - a
standards body - to get on board.

Why in the world isn't this happening?!

This is an OpenOffice list (at least where I read it) so it is hapenning,
at least to some point?

I spend countless hours trying to talk with other groupware projects and
constantly try and find out if they are building any kind of inter
operability between their and my project, but am shut down with little
to no real consideration. Im ready to play ball with other projects. Im
ready to make this dream reality. I just cannot do it alone.

== The Call Goes Out

Evo and OO are pretty well-known among the open source community;
developers hear about this "open groupware standards" thing and become
interested.  The founders of the 20 or so groupware projects that have
languished still-born on sourceforge are reinvigorated by this call to
action, thinking "Hey, I won't be ALONE this time, I'll have Ximian and
OO on my side, and all these other developers are interested too!"  They
call their project sidekicks in on it.  Eventually a persuasive
charismatic (or two) emerges, like a Jobs or de Icaza.  He spreads the
gospel at conventions:  "We're taking it to the next level, people!"

Why not start now? Im ready... what about leaders of other groupware
project? I will start a mailing list if needed, but I think that maybe
FSF would be more appropriate.
I am in the middle of creating the "Open Source GroupWare Foundation"
which is going to start by taking control of phpGroupWare, working as a
governing body to help push phpGW development in a focused direction.
Right now we are in the process of a complete cleanup of our code. As
soon as this process is done I will be starting a draft for an
XML-RPC/SOAP interface to the API and core apps (email, calendar, todo,
notes). This XML-RPC/SOAP interface will open up our interface and data

So where does this SOAP thingy leave LDAP? Well, at least Openoffice will
be covered when the UNO<->SOAP bridge becomes reality...

store to any number of programming languages. This is where the "Open
Source GroupWare Foundation" will help step in and maintain the
XML-RPC/SOAP interfaces and work with app developers who want to build
interfaces in GTK, VB, C++ or whatever. So the idea is to start with
what we have, but work toward opening it up to other languages.
At some point we will probably start development of the back end using C
which will be a full blown daemon. Of course it will have a built in
http server which will be responsible for transporting the xml-rpc/soap
data.


Why not work the other way around - starting with an existing
(modular) http server - say apatche - and then adding the things in? Or do
you really want to reinvent and write a http server yourselves?

I personally think that XML-RPC/SOAP is the best way to go for the back
end services required for all these GroupWare clients to use.
XML-RPC/SOAP is supported by most major programming languages, its
flexible and fairly well designed, its network centric and it is very
buzzword compliant. Just think of how kick ass it will be to offer a
free and open .NET groupware service before M$ does. I think it can be
done. I think with my start in phpGW it can be done. 


You might find out that lot's of people aren't interested at all to have
anything to do with MS .NET ...

Granted I don't think that PHP is the best long term solution, but I am
already working in this direction, and everyone else seems to be lost in
committee. The phpGW development is pushing ahead with or without anyone
else, but would of course rather have some heavy weights such as OO and
EVO working with us to make sure that the XML-RPC/SOAP structures we
build are properly re-usable by any groupware client. Building this
structure is the most important part. The backend code (PHP) can easily
be replaced by a better/faster solution if it becomes necessary.

== The Word Is Heard  (or: "Gentlemen: I give you - OOGS!")

<nitpick>What is the second O for?</nitpick>
I say we start drafting OGWS (Open GroupWare Standard) using
XML-RPC/SOAP now. I will build phpGW as the first implementation of the
backend (unless someone can beat me to it).



'Open Office Groupware Standard' as in an 'Office Groupware Standard'
(OGS) that is open? But that's just my guess.


If no one joins me, I will do it myself. I have already been working
toward this and will keep doing so. Those that join will simply make
sure I do it the best way possible.
 
As Evo and OO's interest in open groupware standards rises, they start
talking to each other more about interoperability.  Neither wants their
fate tied to the fate of the other, but they find areas where they can
cooperate and benefit without risking their own positions.

This is where a backend that is completely separated from the GUI's will
benefit everyone.
BTW, my XML-RPC/SOAP implementation will take into account vCard and the
iCalendar standards. I personally don't think a whole lot of iTIP since
I have yet to see a single iTIP daemon. Also as a stand alone daemon it
seems far too limited. You would end up talking to IMAP for email, iTIP
for calendar/todo, maybe some kind of vCard daemon for contacts... I
think it MUCH easier for everyone if the XML-RPC/SOAP services deal with
the data and leave the client side to giving the user a rich and useful
tool.

But you are probably biased for doing it with SOAP 8-)


Seek3r (aka Dan Kuykendall)


        Sander

One day a tortoise will learn to fly
        -- Terry Pratchett, 'Small Gods'





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]