harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Geir Magnusson Jr <g...@pobox.com>
Subject [OT} Re: So today Sun announced...
Date Fri, 19 May 2006 20:37:52 GMT
Some people complained about this, but I don't want some things 
unanswered.  So just ignore if you don't care.

Endre StĂžlsvik wrote:
> | 
> | I'm not sure how to answer this.  I believe you are a little confused about
> | how the JCP works and what we're doing here, and people are asking for.
> | 
> | First, we aren't advocating changes in the way Java SE specification is
> | defined.
> 
> So you still want the JCP? Run by Sun? Or by who?

My point is that we are implementing a specification, and our individual 
or collective opinions about the JCP are irrelevant here.

Join the jcp-open list (jcp-open-subscribe@apache.org) for JCP related 
discussion.

> 
> | 
> | Second, the Java SE specification - the 'standard' you refer to - is already
> | defined by multiple companies and individuals that work on the Java SE
> | specification Expert Group, of which Sun is the Spec Lead.
> 
> Hm.. I haven't quite come to see it that way: If you read some of the 
> blogs and whatnots on the Mustang pages: "Today I added this method, oh 
> yes!" and "Well, we whipped up this thing five minutes ago!". I don't 
> think a new method in java.lang.String is discussed back and forth in the 
> JCP before some Sun guy just makes it.
>   But I might just be mistaken.

I don't know.  If you think there's an issue, simply call them on it - 
or write to the JCP Executive Committee, or write to me (on jcp-open or 
privately) as the Apache EC rep.

> 
> | 
> | That's an orthogonal issue, independent of implementation of specifications.
> 
> I don't think so, and I guess that's where I feel something is lacking in 
> many comments I read around in the java communities: "oh my god, like, 
> open source java NOW, dummies!", while not thinking about the processes 
> behind the different scenes.

But I think that here, we have a very serious goal, and that's to 
implement the spec in OSS.  Sure, we'd love to see Sun OSS their 
implementation, but as we said before, that decision is up to them. 
It's their property.

We're putting our money where our mouth is, so to speak.

> 
> | 
> | In fact, it's clear to me that OSS in Java EE has had a very positive effect
> | on both evolving the spec for users (i.e. the Spring and Hibernate influence
> | on Java EE/EJB3) as well as making the technology available to end-users at a
> | faster pace.
> 
> I didn't know that there was that big a difference between EE and SE 
> regarding licensing / open source, so obviously I have missed out on some 
> points here.. Do you have any good links laying around?

I don't understand.  EE and SE are licensed the same.

> 
> | 
> | What's the "Swing issue"?
> 
> The "missing TCK trick":
>   http://danesecooper.blogs.com/divablog/2006/05/what_sun_doesnt.html
> 
> I read that in this thread, and have heard it mentioned before: The GUI 
> parts of "java standards" are a somewhat convoluted area.

I have no material information, but a UI is a hard thing to spec and a 
hard thing to test.

That said, there are no shared-source requirements in a JCP 2.5+ 
implementation or TCK.  We will implement according to the spec, and 
test with the TCK.  That's all we can do.

Our goal is to produce Swing/AWT so that users programs run as they 
expect.  It's going to be a lot of work.

[SNIP]

> But, open source the _implementation_, and keep the JCP as it is, with Sun 
> having a special position?

I'll happily share my personal opinion on jcp-open but this is 
irrelevant to Harmony. :)

We're just here to implement and certify an compatible implementation.

geir


---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org


Mime
View raw message