java -version output for JDK8

[Volker Simonis]

This is really just another crazy TM issue.

"Java" is a trademark and your not allowed to use it until you've licensed it.
In this case the bizarre thing is that the executable (i.e. the
launcher itself) is still called "java" even in the OpenJDK and
Sun/Oracle obviously could not change that. On the other hand they've
been so small-minded to prohibit the usage of the "java" word in the
version string.

That's not only annoying, it also breaks a lot of things. For example
you can not use a pre-installed ANT on many Linux distributions
together with a vanilla OpenJDK because ANT use various scripts from
the 'jpackage' project which do not understand the "openjdk" version
string either (see
http://hg.openjdk.java.net/ppc-aix-port/jdk7u/raw-file/tip/README-ppc.html#_TOC_ANCHOR_3_).

So I'm all in favour of fixing this upstream.

Any Oracle volunteers?

Regards,
Volker


On Fri, Jul 26, 2013 at 10:58 AM, Andrew Hughes <gnu.andrew at redhat.com> wrote:
> ----- Original Message -----
>> Hi Kumar,
>>
>> On 07/25/2013 07:05 PM, Kumar Srinivasan wrote:
>> > Changing any of these lines is very painful, because too many
>> > scripts and tests have made assumptions on these lines, and some of
>> > these are System properties which are baked at build time.
>>
>
> People have been building with this fixed since 2007, as have all your
> proprietary builds.
>
>> I am looking at it from the users' point of view. Oracle (proprietary)
>> JDKs claim to be "java version..." as do IBM (proprietary) JDKs. Isn't
>> openjdk the odd one out if it claims to be "openjdk version..."?
>>
>
> Yes.
>
>> I also think it will be much more painful for all our users to have to
>> fix their scripts and tests if any of those scripts or tests happen to
>> rely on the output of "java -version". Fixing our tests and scripts
>> would surely be less painful.
>>
>> Thanks,
>> Omair
>>
>
> No packages have ever shipped using "openjdk version", which is why I
> immediately noticed it as a regression.  So the discussion is moot really.
> It's still going to get changed back in local packages.  It would just be
> better to have it fixed upstream.
> --
> Andrew :)
>
> Free Java Software Engineer
> Red Hat, Inc. (http://www.redhat.com)
>
> PGP Key: 248BDC07 (https://keys.indymedia.org/)
> Fingerprint = EC5A 1F5E C0AD 1D15 8F1F  8F91 3B96 A578 248B DC07
>