i recently needed a java-to-lisp translator.
as i couldn't find one, i extracted the parser generator from cl-xml, found a
tractable bnf, and implemented a rudimentary translator.
in the event that it might be generally useful, i've posted it under setf.de.
(see http://www.setf.de/code.html)
...
james anderson <··············@setf.de> writes:
> i recently needed a java-to-lisp translator.
>
> as i couldn't find one, i extracted the parser generator from cl-xml, found a
> tractable bnf, and implemented a rudimentary translator.
>
> in the event that it might be generally useful, i've posted it under setf.de.
> (see http://www.setf.de/code.html)
I'm not a lawyer, but...
I would recommend that you be careful with what you call this and what
kinds of disclaimers you associate with it. Sun is quite protective
of the use of "Java" as a trademark and what you can say implements
Java semantics.
I don't see any obvious problem with the program itself, in the sense
that the code you are translating will _presumably_ be owned by the
person doing the translation, so they have the right to translate it
and can take responsibility for the consequencs of having done so.
However, if they had expectations that Java was reliable and that your
translation would preserve that degree of reliability, that's where
you could run into trouble, if you appeared to be placing the output
under the same "happy safety umbrella" that the Java name is intended
to evoke for the original source code.
(If anyone knows of any specific legal cases in this area, I'd be curious.
But I'd bet most of this is duked out "out of court" by mere intimidation,
which is what the legal system seems to prefer people do, to spare the
courts expense...)
Kent M Pitman wrote:
>
> james anderson <··············@setf.de> writes:
>
> > i recently needed a java-to-lisp translator.
> >
> > as i couldn't find one, i extracted the parser generator from cl-xml, found a
> > tractable bnf, and implemented a rudimentary translator.
> >
> > in the event that it might be generally useful, i've posted it under setf.de.
> > (see http://www.setf.de/code.html)
>
> I'm not a lawyer, but...
>
> I would recommend that you be careful with what you call this and what
> kinds of disclaimers you associate with it. Sun is quite protective
> of the use of "Java" as a trademark and what you can say implements
> Java semantics.
the translation is _really_ literal. and i do not provide a library. lacking
function bindings for the target symbols in java.lang, etc, does the
translated program have any sematics? i wonder if that undermines their intent
even more severely?
>
> I don't see any obvious problem with the program itself, in the sense
> that the code you are translating will _presumably_ be owned by the
> person doing the translation, so they have the right to translate it
> and can take responsibility for the consequences of having done so.
the grammar was in the public domain, but i didn't consider the issue of
making available a tool which could be used to engineer a dmca violation.
> However, if they had expectations that Java was reliable and that your
> translation would preserve that degree of reliability, that's where
> you could run into trouble, if you appeared to be placing the output
> under the same "happy safety umbrella" that the Java name is intended
> to evoke for the original source code.
see above wrt. "no library".
...
james anderson wrote:
>
> Kent M Pitman wrote:
>
>>james anderson <··············@setf.de> writes:
>>
>>
>>>i recently needed a java-to-lisp translator.
>>>
>>>as i couldn't find one, i extracted the parser generator from cl-xml, found a
>>>tractable bnf, and implemented a rudimentary translator.
>>>
>>>in the event that it might be generally useful, i've posted it under setf.de.
>>>(see http://www.setf.de/code.html)
>>
>>I'm not a lawyer, but...
>>
>>I would recommend that you be careful with what you call this and what
>>kinds of disclaimers you associate with it. Sun is quite protective
>>of the use of "Java" as a trademark and what you can say implements
>>Java semantics.
>
>
> the translation is _really_ literal. and i do not provide a library. lacking
> function bindings for the target symbols in java.lang, etc, does the
> translated program have any sematics? i wonder if that undermines their intent
> even more severely?
AFAIK, Sun doesn't care what you actually provide under the name Java. I
also have the impression that they are very restrictive. That's why all
the tools for Java only use the letter "J" to indicate the relationship,
and not the full name.
One prominent example is the JavaClass library that had to be renamed
because of pressure from Sun, and is now called BCEL. Microsoft's J++ is
another example. (I guess that's the main reason for their strictness,
the troubles between them and Microsoft. Even Sun employees have to be
careful how they call what. Guy Steele made a remark to that extent in
his "Growing a Language" paper.)
So better be safe than sorry.
Pascal
--
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."
Kent M Pitman <······@nhplace.com> writes:
> I'm not a lawyer, but...
That's ok, nor am I ...
> However, if [users] had expectations that Java was reliable and that your
> translation would preserve that degree of reliability, that's where
> you could run into trouble, if you appeared to be placing the output
> under the same "happy safety umbrella" that the Java name is intended
> to evoke for the original source code.
I wonder whether a reasonable person could hold that expectation.
It's been a while since I clicked on a Java licence agreement, but I'm
pretty sure it (as is typical for most of the rest of the software
industry) it comes with big scary warnings about disclaimed warranties
and not-for-use-in-power-stations and the rest of that schtick.
If such a case were to be brought, and to succeed, would it have
knockon implications for the typical software industry EULA?
-dan
--
http://web.metacircles.com/ - Open Source software development and support
Kent M Pitman <······@nhplace.com> writes:
> I'm not a lawyer, but...
>
> I would recommend that you be careful with what you call this and what
> kinds of disclaimers you associate with it. Sun is quite protective
> of the use of "Java" as a trademark and what you can say implements
> Java semantics.
Indonesia has somewhere between 13,557 and 17,508 islands. Surely
they cannot *all* be taken.
Kent M Pitman wrote:
> (If anyone knows of any specific legal cases in this area, I'd be curious.
> But I'd bet most of this is duked out "out of court" by mere intimidation,
> which is what the legal system seems to prefer people do, to spare the
> courts expense...)
The Sun vs Microsoft Java lawsuit is obviously relevant here even
though it was settled out of court. As a result of the settlement,
the court entered a permanent injunction that bars Microsoft from
claiming that any of their products are Java compatible.
The text of the settlement is online at
http://java.sun.com/lawsuit/settlement.html
Here is the story, as spun by Sun and by Microsoft:
http://java.sun.com/lawsuit/
http://www.microsoft.com/presspass/press/2001/jan01/01-23SunPR.asp
Will
William D Clinger wrote:
>
> Kent M Pitman wrote:
> > (If anyone knows of any specific legal cases in this area, I'd be curious.
> > But I'd bet most of this is duked out "out of court" by mere intimidation,
> > which is what the legal system seems to prefer people do, to spare the
> > courts expense...)
>
> The Sun vs Microsoft Java lawsuit is obviously relevant here even
> though it was settled out of court. As a result of the settlement,
> the court entered a permanent injunction that bars Microsoft from
> claiming that any of their products are Java compatible.
Yep. They've repackaged and fiddled with the syntax and implemented
a new VM, and now they're calling their version C#. No more claims
of Java compatibility. Interestesting reading, though.
Bear
james anderson wrote:
> i recently needed a java-to-lisp translator.
>
> as i couldn't find one, i extracted the parser generator from cl-xml, found a
> tractable bnf, and implemented a rudimentary translator.
>
> in the event that it might be generally useful, i've posted it under setf.de.
> (see http://www.setf.de/code.html)
There's an XML dialect called JavaML which might be a reasonable
intermediate step. See http://www.cs.washington.edu/homes/gjb/JavaML/
(I don't have any experience with that tool, though.)
Pascal
--
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."