[Top][All Lists]

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

Re: [DotGNU]DotGNU.* namespaces -- Jabber & stuff

From: Jonathan P Springer
Subject: Re: [DotGNU]DotGNU.* namespaces -- Jabber & stuff
Date: Sun, 28 Apr 2002 08:45:08 -0400
User-agent: Mutt/1.3.28i


Rhys, S11001001 and I had an improptu discussion about this last night
on IRC.  We reached about the same conclusion.  I'm happy to hear that you've
mailed the "right contact" about that.  

S11001001 noted that it is also possible to instantiate WebRequest then 
set a delegate for the Jabber transport; however creating an easy-to-use
extension a la HttpWebRequest makes sense.

I'm continuing to look at integration of JabberBeans and Apache SOAP for
those operation on the Java side of the spectrum.  Apache SOAP in turn
provides support for Java classes, JavaBeans and EJBs to be served, so
there's a potential quick win/proof-of-concept here.  I've shifted my
thinking from immediate needs to what it would take to make a robust
Jabber-based SOAP server that could be run on an existing Java-based app
server (e.g. Tomcat).  I may propose some "jabber:dotgnu:*"
namespaces; I'll keep the group informed.

OK.  Here's the rub -- I really like the Apache software (Xerses, Xalan,
Xindice, SOAP and Tomcat) for the work I'm doing, but to quote the FSF:

We urge you not to use the Apache licenses for software you write.
However, there is no reason to avoid running programs that have been
released under this license, such as Apache. 

So, any "augmentations" can be released under GPL.  Heck, since Apache's
non-copyleft, we may just be able to further license its chunk under GPL (with
the Apache 1.1-required disclaimer, of course :-( ).  I'd like thoughts
from our license- and philosophy gurus on this question, as I know I
lean toward the less-zealotous on questions of licensing.


On Sun, Apr 28, 2002 at 03:12:48PM +0530, Gopal V wrote:
> Hi Everybody,
>      In relation to jabber support for DotGNU , it has been found that
> we really need some *real* support for Jabber in Portable.Net . So we're
> thinking about implementing a Jabber# library for it.
>     After giving much thought to how to implement Jabber I have a plan
> about how to about it. The System.Net.WebRequest should be extended to
> form a DotGNU.Net.JabberWebRequest like the System.Net.HttpWebRequest. 
> The DotGNU namespace should contain all the "innovations" we make. 
> It is also possible that someone from Jabber.Net already has such an
> implementation of the Jabber protocol , so I've mailed the Alpha Geek
> Joe HildeBrand <address@hidden> asking about the details on their 
> work. If that project may provide a Jabber.Client.dll (or someother
> stuff) , we may be able to abstract out the details in that .
> Until the System.Xml is finished (or usable) we will have to write code
> with our eyes closed. Hoping that the Jabber.Net will also include
> the client code stuff.
> Priorities for Network Development (High : two weeks)
> ----------------------------------
> * Socket.Select() Internal call
>    read pnet/engine/lib_socket.c (Hacker level 3)
> * pnetlib/System/Net/NetworkStream.cs
>    use the Platform.SocketMethods class and implement the Stream functions
>    using the recv , send in it. (Hacker level 2)
> * pnetlib/System/Net/Dns.cs & internalcalls
>    Really don't know the requirements or procedures 
>    (rhys: do you have uncommited code here ?)
> * Array declare woes in the compiler :( .
>   a quick fix in the form of an IL HelperClasses code 
>   ie a helpers assembly which has the ArrayHelper.NewByteArray(len)
>   see for example
> Jabber Specific (medium : next month)
> ---------------
> * pnetlib/System/Uri.cs 
>    implement the jabber addressing/ URI model here 
> * finish XML stuff
> * implement SHA1 for Jabber Auth (or was it Md5 ?)
> * Do the base64 stuff to make it send binary
> * needs XPath ? 
> Rpc Stuff (low : in the next 4 months)
> --------- 
> (after all that's what jabber is used for)
> * A [WebService] attribute to mark a static function as Rpc visible 
>   (Hacker level 1)
> * A reflection based Rpc server 
>   (Hacker level 2)
> * Implement MethodInfo.GetParameters()
> check out my Pnet demos at
> for an example of what Pnet can do (and some tricks off my sleeve) .
> This shows how we can work around most of cscc's limitations ....
> Gopal
> -- 
> The difference between insanity and genius is measured by success
> _______________________________________________
> Developers mailing list
> address@hidden

-Jonathan P Springer <address@hidden>
"A standard is an arbitrary solution to a recurring problem." - Joe Hazen

reply via email to

[Prev in Thread] Current Thread [Next in Thread]