incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Brouwer <mark.brou...@cheiron.org>
Subject Re: Jini?
Date Tue, 08 Aug 2006 10:29:43 GMT
Sanjiva Weerawarana wrote:

> I know changing the name is a *really* tough thing for Jini. However, is
> Jini a *technology* or an *implementation*? If its the prior I'm afraid
> our current guidelines are not to do technology names.

I understand for those not very involved with the Jini Technology it is
hard to pinpoint what Jini exactly is and why some of us are willing to
go through great lengths to take this name with us, so let me try.

First Jini is a Technology, but with an extra handicap as the borders
where Jini begins and ends are not very well defined, even while in
1999/2000 there was already a document that described the Jini
Architecture and the Jini Technology Core Platform. It just lacks a
clear definition, when you ask 10 people to describe Jini chances are
high that you will get 10 different answers; some that will make you
happy or smile, and some that make you foam with rage ...

Many consider the implementation of Sun JTSK (Jini Technology Starter
Kit) as being 'Jini' but this is not correct (if you would ask me) and
while the proposal mainly centers around their code the proposal also
includes another Jini Community Approved Standard, namely ServiceUI (the
other trademark involved).

What is part of this proposal are most of the Jini Community Approved
Standards and the IMHO 'sad' thing is that the Jini Decision Process
that ratified these specifications as community approved standards
ceased to exist. Sun is no longer willing to provide the Executive to
run the process and there are not enough people in the community that
want to take it over (read don't want to spend time on it). The current
owner of the Standards also doesn't believe they can get accepted as JSR
in the JCP based on experience with some of the specifications that
ended up as Jini Standard while they should have become J2SE
specifications in the first place. The net result is that some really
important Jini Community aspects [1] (the Standards) we all circle
around are part of this proposal, and if we can stay clear of forks it
should be seen as the foundation on top of which the rest of the Jini
community will build their own stuff.

[1] as Jim mentioned the new http://jini.org/ is another aspect of the
community and is there for everything one could see as Jini related
(really open ended), http://jini.dev.java.net/ should be seen as the
yellow pages with regard to Jini related development projects and news
around that.

Renaming the Technology itself would be suicidal in my opinion as there
are dozens of people/companies that develop products/specifications on
top of 'Jini' so it would be harming them too. Of course it would be
possible to give the TLP a different name, but then again almost every
sentence would have a reference to the Jini Technology for which I guess
nobody could say where the 'Jini Technology' itself lives. Given the
fact the ASF project would be closest to defining the 'Jini Technology'
I think it is good to emphasize this by the name of the TLP.

IANAL and have no idea what the impact would be of handing over the Jini
trademark to the ASF, how the ASF will deal with other communities that
have Jini in their name, or other specifications that have Jini in their
name. I'm reluctant to say this given the fact that Sun has to protect
its trademarks, but I have the feeling that Jini has become rather
generic in the past years. People say I wrote a Jini service, I
developed a Jini Service Container, I do Jini, it is the Jini way, etc.
I also don't understand the implications of abandoning the trademark
itself, but I would like to see that Jini can be used in the future as
it is today, even when it makes your mouth foam.

To summarize as *I* see it at this very moment:

   - Jini is not a product.
   - Jini can't be used as a noun.
   - Jini in combination with another noun could serve as a specification
     name for which one can have multiple implementations, such as
     'Jini Helper and Utility Classes', 'Jini Platform',
     'Jini Service Container', etc.
   - the deliverables of an Apache Jini TLP should get their own distinct
     name without Jini in it, except when it represent a 'Jini ...
     Specification'.
   - Jini itself only represents the magic of doing distributed computing
     in a proper way and comes in a lamp, these are already hard to find
     these days so please don't make it even harder :-)

I would like to know whether people object against using Jini as name
for a TLP based on the above reasoning. I think for the discussion it
would be handy to tackle the appropriateness of the name first before we
deal with legal issues, as I guess that most people here have the IANAL
prefix, just like me.
-- 
Mark


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Mime
View raw message