(keitai-l) JINI on mobile terminals

From: <B.Beute_at_telecom.tno.nl>
Date: 05/14/03
Message-ID: <E5E837F7C17E5646B0BB305665E9C275090C2A@g07.oase.research.kpn.com>
Hi Giovanni,

| There is any plan at SUN for any Jini integration with mobile phones?

Jini was initially marketed as a technology for spontaneous networking =
between devices, but has recently become much more popular for =
networking between heterogeneous networks. The reason for that is that =
Java is much more common on big servers than it is on small devices. The =
latter is changing though, and slowly developers begin to realise the =
beautiful vision behind Jini and start applying that to the world of =
mobile terminals. One of the bigger hurdles at the moment is that Java =
RMI is the most practical protocol for communication between the client =
and the service backend, even though Jini is protocol agnostic (you =
might even use snail mail for it as far as Jini is concerned). Java RMI =
is unfortunately NOT (yet?) a standard part of CLDC/J2ME so many of the =
existing services are not yet consumable from J2ME devices. This is =
changing though, and Jini-projects like the 'surrogate project' =
(http://www.jini.org) are enabling non-jini devices to participate in a =
Jini network.

As far as security is concerned this discussion is timely: Jini 2.0 has =
just been released and provides a very thorough and elegant security =
framework (read an excellent introduction article about it here: =
http://www.javaworld.com/javaworld/jw-05-2003/jw-0509-jiniology_p.html). =
With the arriving of Java's notion of mobile code (according to its =
founders the original goal of the virtual machine, to ship code around) =
on mobile terminals, the need arises for a very good security framework. =
Java's sandbox and Jini's security model together make a strong =
proposition towards a secure, self-healing environment consisting of =
widely heterogeneous participants that require no human interaction to =
run. The icing on the cake is the ServiceUI project =
(http://www.artima.com/jini/serviceui/index.html) that was is now an =
official part of Jini and provides a mechanism to offer multiple =
interfaces with one service. Handy if you don't know the specifics of =
the device the service is consumed from.

Wrapping things up I can certainly recommend looking into the =
opportunities Jini offers for J2ME developers. Go check it out: =
http://www.jini.org

Kind regards
Berco Beute

____________________________________________________________

| Looks like a technology perfectly suited for heterogeneous
| network technologies connections, hardware devices
| and software running the applications.
|=20
| Cheers
|=20
| Giovanni
____________________________________________________________

| >> Ok MIDP 2.0 will still run things in a sandbox if they =3D
| are"untrusted".
| >> Public Key Infrastructure (PKI)  is utilized by MIDP 2.0=20
| to provide=3D20=3D
|=20
| >> secure
| >> connections and digital signatures for "trusted" MIDP=20
| application=3D20
| >> packages
| >> (also known as MIDlets).
| >
| > This is probably an abuse of the term "sandbox", all
| > byte-code is executed in a sandbox. It would probably be
| > clear to say that a MIDP 2.0 midlet that is trusted runs
| > with a less prohibitive security policy. Which is exactly
| > what you describe :-)
Received on Wed May 14 19:09:52 2003