Scala Licence

Looks like they don't want to allow modified versions?


On Thu, Jul 15, 2010 at 8:53 AM, Sylvain HENRY <hsyl20@gmail.com> wrote:

Hi,

I posted a news about Scala 2.8 release on http://linuxfr.org/2010/07/15/27122.html (in French). In comments, someone wonders why the Scala licence is not BSD because the only difference seems to be the introduction:

BSD licence:
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

Becomes in Scala licence:
Permission to use, copy, modify, and distribute this software in source
or binary form for any purpose with or without fee is hereby granted,
provided that the following conditions are met:

Can someone explain why this rewriting was done?

Thanks,
Sylvain

>>>>> "Naftoli" == Naftoli Gugenheim writes:

Naftoli> Looks like they don't want to allow modified versions?

Huh? "Permission to use, copy, modify..."

On 16 July 2010 09:01, Seth Tisue wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

Here's a list of what rights can be granted under copyright law in the US: http://www.bitlaw.com/copyright/scope.html
Both licenses grant all the rights (if you assume the "use" means "perform" and "display")  The wording is different.  I do not read modification and distribution to be mutually exclusive.  Put another way, you've got the right to create a derivative work and you have the right to modify.  Reading the license to exclude the right to distribute derivative works would be like reading the license to exclude the use of derivative works.
Before I started my first software company, I wrote software licenses.
On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys <andrae.muys@cambia.org> wrote:

On 16 July 2010 09:01, Seth Tisue <seth@tisue.net> wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim <naftoligug@gmail.com> writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

--
Andrae Muys,
Senior Software Engineer, Cambia
Initiative for Open Innovation (IOI)
Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
+61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

I wouldn't read it as *excluding* distributing it in modified form, but it certainly doesn't mention permission to do so, no?To me "Permission to use, copy, modify, and distribute this software" is short for "Permission to use this software, and permission to copy this software, and permission to modify this software, and permission to distribute this software" [in source
or binary form for any purpose with or without fee is hereby granted...]So it's granting permission for four different things that are not inherently related, no?

On Thu, Jul 15, 2010 at 11:30 PM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

Here's a list of what rights can be granted under copyright law in the US: http://www.bitlaw.com/copyright/scope.html
Both licenses grant all the rights (if you assume the "use" means "perform" and "display")  The wording is different.  I do not read modification and distribution to be mutually exclusive.  Put another way, you've got the right to create a derivative work and you have the right to modify.  Reading the license to exclude the right to distribute derivative works would be like reading the license to exclude the use of derivative works.
Before I started my first software company, I wrote software licenses.
On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys <andrae.muys@cambia.org> wrote:

On 16 July 2010 09:01, Seth Tisue <seth@tisue.net> wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim <naftoligug@gmail.com> writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

--
Andrae Muys,
Senior Software Engineer, Cambia
Initiative for Open Innovation (IOI)
Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
+61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

(Tongue firmly in cheek this evening...)

See, any Scala guy work this chops would interpret the sentence to be a sequence of operations "use", "copy", "modify", and "distribute": for each one there is an implicit operating function passed as a parameter.

That is, "use" is an abstract operation on the software which requires a use-function in order to be instantiated as a real operation that can actually be performed. E.g., "use" the software by { performing some operation on the software }

In this framework, you would interpret "Permission to use, copy, modify, and distribute this software" -- to be a pipelined sequence of operation calls:

def useFunction(s: Software): Software // abstract def
def copyFunction(s: Software): Software // abstract def
def modificationFunction(s: Software): Software // abstract def
def distributeFunction(s: Software): Software // abstract def

distribute (modify (copy (use (the software) useFunction(_)) copyFunction(_) ) modificationFunction(_) ) distributionFunction(_)

With this interpretation you would have permission to invoke the following operation with any legally "reasonable" (useFunction, copyFunction, modificationFunction, distributeFunction) tuple:

First "use" the software with the useFunction;
and then "copy" the resulting software with the copyFunction;
and then "modify" the result of the copy operation with the modificationFunction;
and finally "distribute" the result of the modify operation with the distributeFunction;

For each step, an appropriate Null or Identity function can be applied, effctively skipping that step by making it a no-op.

This would mean you can modify and then distribute the modified copy.

On Thu, Jul 15, 2010 at 9:29 PM, Naftoli Gugenheim <naftoligug@gmail.com> wrote:

I wouldn't read it as *excluding* distributing it in modified form, but it certainly doesn't mention permission to do so, no? To me "Permission to use, copy, modify, and distribute this software" is short for "Permission to use this software, and permission to copy this software, and permission to modify this software, and permission to distribute this software" [in source
or binary form for any purpose with or without fee is hereby granted...]So it's granting permission for four different things that are not inherently related, no?

On Thu, Jul 15, 2010 at 11:30 PM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

Here's a list of what rights can be granted under copyright law in the US: http://www.bitlaw.com/copyright/scope.html
Both licenses grant all the rights (if you assume the "use" means "perform" and "display")  The wording is different.  I do not read modification and distribution to be mutually exclusive.  Put another way, you've got the right to create a derivative work and you have the right to modify.  Reading the license to exclude the right to distribute derivative works would be like reading the license to exclude the use of derivative works.
Before I started my first software company, I wrote software licenses.
On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys <andrae.muys@cambia.org> wrote:

On 16 July 2010 09:01, Seth Tisue <seth@tisue.net> wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim <naftoligug@gmail.com> writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

--
Andrae Muys,
Senior Software Engineer, Cambia
Initiative for Open Innovation (IOI)
Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
+61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

On Thu, Jul 15, 2010 at 9:29 PM, Naftoli Gugenheim <naftoligug@gmail.com> wrote:

I wouldn't read it as *excluding* distributing it in modified form, but it certainly doesn't mention permission to do so, no?To me "Permission to use, copy, modify, and distribute this software" is short for "Permission to use this software, and permission to copy this software, and permission to modify this software, and permission to distribute this software" [in source
or binary form for any purpose with or without fee is hereby granted...]So it's granting permission for four different things that are not inherently related, no?

When you're done with law school, let's have this discussion again.
English is a porous language and even the most carefully drafted language is open to interpretation.
Under your interpretation (you can do any of the four things listed, but none of them in combination), one could make a derivative work, but not use it or copy it.  Imagine how insane that would be... you would have to retype your modifications each time you wanted to make them and once you've re-typed them, you could not compile them (because compilation is making a copy, but altering the form).  I can't think of anyone making such an argument.  So, because it's reasonable to make modifications and then make copies of the modifications (compile) and perform (in the copyright sense) the modifications (run), why not distribute the modifications?  Further, EPFL has encouraged parties to engage in such behavior (support for PaulP's GitHub branch, support for the LLVM branch of Scala, the cross-pollination of Scala's XML libraries into Lift [we made a derivative work of some of Scala's XML libraries to enhance their functionality, EPFL then took the Lift enhancements and rolled those back into Scala], etc.)  It's clear to me that EPFL intends their license to allow distribution of derivative works by both words and actions.
Yeah, there is a very minor issue when folks change license language because the common open source licenses have been reviewed and are commonly agreed on.  However, very few of the open source licenses have ever been tested in court (to my knowledge, only the GPL has been tested at both the trial and appellate level).  So, none of us really know what any of the licenses mean because until a couple of circuit courts in the US (appellate) agree on an interpretation, it's just a bunch of guys on mailing lists wasting time.  

On Thu, Jul 15, 2010 at 11:30 PM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

Here's a list of what rights can be granted under copyright law in the US: http://www.bitlaw.com/copyright/scope.html
Both licenses grant all the rights (if you assume the "use" means "perform" and "display")  The wording is different.  I do not read modification and distribution to be mutually exclusive.  Put another way, you've got the right to create a derivative work and you have the right to modify.  Reading the license to exclude the right to distribute derivative works would be like reading the license to exclude the use of derivative works.
Before I started my first software company, I wrote software licenses.
On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys <andrae.muys@cambia.org> wrote:

On 16 July 2010 09:01, Seth Tisue <seth@tisue.net> wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim <naftoligug@gmail.com> writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

--
Andrae Muys,
Senior Software Engineer, Cambia
Initiative for Open Innovation (IOI)
Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
+61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

not to mention that their interpretation can be totally different in
other countries.

-Stefan

2010/7/16 David Pollak :
>
>
> On Thu, Jul 15, 2010 at 9:29 PM, Naftoli Gugenheim
> wrote:
>>
>> I wouldn't read it as *excluding* distributing it in modified form, but it
>> certainly doesn't mention permission to do so, no?
>> To me "Permission to use, copy, modify, and distribute this software" is
>> short for "Permission to use this software, and permission to copy this
>> software, and permission to modify this software, and permission to
>> distribute this software" [in source
>> or binary form for any purpose with or without fee is hereby granted...]
>> So it's granting permission for four different things that are not
>> inherently related, no?
>
> When you're done with law school, let's have this discussion again.
> English is a porous language and even the most carefully drafted language is
> open to interpretation.
> Under your interpretation (you can do any of the four things listed, but
> none of them in combination), one could make a derivative work, but not use
> it or copy it.  Imagine how insane that would be... you would have to retype
> your modifications each time you wanted to make them and once you've
> re-typed them, you could not compile them (because compilation is making a
> copy, but altering the form).  I can't think of anyone making such an
> argument.  So, because it's reasonable to make modifications and then make
> copies of the modifications (compile) and perform (in the copyright sense)
> the modifications (run), why not distribute the modifications?  Further,
> EPFL has encouraged parties to engage in such behavior (support for PaulP's
> GitHub branch, support for the LLVM branch of Scala, the cross-pollination
> of Scala's XML libraries into Lift [we made a derivative work of some of
> Scala's XML libraries to enhance their functionality, EPFL then took the
> Lift enhancements and rolled those back into Scala], etc.)  It's clear to me
> that EPFL intends their license to allow distribution of derivative works by
> both words and actions.
> Yeah, there is a very minor issue when folks change license language because
> the common open source licenses have been reviewed and are commonly agreed
> on.  However, very few of the open source licenses have ever been tested in
> court (to my knowledge, only the GPL has been tested at both the trial
> and appellate level).  So, none of us really know what any of the licenses
> mean because until a couple of circuit courts in the US (appellate) agree on
> an interpretation, it's just a bunch of guys on mailing lists wasting time.
>
>>
>> On Thu, Jul 15, 2010 at 11:30 PM, David Pollak
>> wrote:
>>>
>>> Here's a list of what rights can be granted under copyright law in the
>>> US: http://www.bitlaw.com/copyright/scope.html
>>> Both licenses grant all the rights (if you assume the "use" means
>>> "perform" and "display")  The wording is different.  I do not read
>>> modification and distribution to be mutually exclusive.  Put another way,
>>> you've got the right to create a derivative work and you have the right to
>>> modify.  Reading the license to exclude the right to distribute derivative
>>> works would be like reading the license to exclude the use of derivative
>>> works.
>>> Before I started my first software company, I wrote software licenses.
>>> On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys
>>> wrote:
>>>>
>>>> On 16 July 2010 09:01, Seth Tisue wrote:
>>>> >>>>>> "Naftoli" == Naftoli Gugenheim writes:
>>>> >
>>>> >  Naftoli> Looks like they don't want to allow modified versions?
>>>> >
>>>> > Huh?  "Permission to use, copy, modify..."
>>>> >
>>>> > --
>>>> > Seth Tisue @ Northwestern University | http://tisue.net
>>>> > lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>>>> >
>>>>
>>>> The issue is the lack of permission to distribute modifications.
>>>> Normally the language wouldn't matter so much, but because it is
>>>> derived from a pre-existing, long standing license, it is definitely
>>>> possible that, absent an explicit statement of intent by the licensor,
>>>> that the redaction of permission to distribute modified versions may
>>>> be read by a court as explicitly forbidding it.
>>>>
>>>> Hence it is not unreasonable for a commenter to ask for a
>>>> clarification of intent.
>>>>
>>>> Andrae
>>>>
>>>> P.S. IANAL, but I have spent a lot of time advising lawyers on the
>>>> technical aspects of software copyright, and I'm pretty certain that
>>>> the reasoning above is sound; of course, I will check with my
>>>> copyright lawyer friends next time I talk to them.
>>>>
>>>> --
>>>> Andrae Muys,
>>>> Senior Software Engineer, Cambia
>>>> Initiative for Open Innovation (IOI)
>>>> Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
>>>> +61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)
>>>
>>>
>>>
>>> --
>>> Lift, the simply functional web framework http://liftweb.net
>>> Beginning Scala http://www.apress.com/book/view/1430219890
>>> Follow me: http://twitter.com/dpp
>>> Blog: http://goodstuff.im
>>> Surf the harmonics
>>
>
>
>
> --
> Lift, the simply functional web framework http://liftweb.net
> Beginning Scala http://www.apress.com/book/view/1430219890
> Follow me: http://twitter.com/dpp
> Blog: http://goodstuff.im
> Surf the harmonics
>

Regardless, the change was made and it would be nice to know why.  Programmers don't usually change the wording on legal documents just for fun.  There must be a motivation.

On Fri, Jul 16, 2010 at 9:04 AM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

On Thu, Jul 15, 2010 at 9:29 PM, Naftoli Gugenheim <naftoligug@gmail.com> wrote:

I wouldn't read it as *excluding* distributing it in modified form, but it certainly doesn't mention permission to do so, no?To me "Permission to use, copy, modify, and distribute this software" is short for "Permission to use this software, and permission to copy this software, and permission to modify this software, and permission to distribute this software" [in source
or binary form for any purpose with or without fee is hereby granted...]So it's granting permission for four different things that are not inherently related, no?

When you're done with law school, let's have this discussion again.
English is a porous language and even the most carefully drafted language is open to interpretation.
Under your interpretation (you can do any of the four things listed, but none of them in combination), one could make a derivative work, but not use it or copy it.  Imagine how insane that would be... you would have to retype your modifications each time you wanted to make them and once you've re-typed them, you could not compile them (because compilation is making a copy, but altering the form).  I can't think of anyone making such an argument.  So, because it's reasonable to make modifications and then make copies of the modifications (compile) and perform (in the copyright sense) the modifications (run), why not distribute the modifications?  Further, EPFL has encouraged parties to engage in such behavior (support for PaulP's GitHub branch, support for the LLVM branch of Scala, the cross-pollination of Scala's XML libraries into Lift [we made a derivative work of some of Scala's XML libraries to enhance their functionality, EPFL then took the Lift enhancements and rolled those back into Scala], etc.)  It's clear to me that EPFL intends their license to allow distribution of derivative works by both words and actions.
Yeah, there is a very minor issue when folks change license language because the common open source licenses have been reviewed and are commonly agreed on.  However, very few of the open source licenses have ever been tested in court (to my knowledge, only the GPL has been tested at both the trial and appellate level).  So, none of us really know what any of the licenses mean because until a couple of circuit courts in the US (appellate) agree on an interpretation, it's just a bunch of guys on mailing lists wasting time.  

On Thu, Jul 15, 2010 at 11:30 PM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

Here's a list of what rights can be granted under copyright law in the US: http://www.bitlaw.com/copyright/scope.html
Both licenses grant all the rights (if you assume the "use" means "perform" and "display")  The wording is different.  I do not read modification and distribution to be mutually exclusive.  Put another way, you've got the right to create a derivative work and you have the right to modify.  Reading the license to exclude the right to distribute derivative works would be like reading the license to exclude the use of derivative works.
Before I started my first software company, I wrote software licenses.
On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys <andrae.muys@cambia.org> wrote:

On 16 July 2010 09:01, Seth Tisue <seth@tisue.net> wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim <naftoligug@gmail.com> writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

--
Andrae Muys,
Senior Software Engineer, Cambia
Initiative for Open Innovation (IOI)
Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
+61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

--
http://erikengbrecht.blogspot.com/

Errr...wait...check svn before sending the message...
Just in case some other folks on the list are as dense as me...the Scala license has not changed beyond updating the copyright to 2010.
But it would still be interesting to know why the Scala license deviates from the BSD phrasing.
On Fri, Jul 16, 2010 at 9:17 AM, Erik Engbrecht <erik.engbrecht@gmail.com> wrote:

Regardless, the change was made and it would be nice to know why.  Programmers don't usually change the wording on legal documents just for fun.  There must be a motivation.

On Fri, Jul 16, 2010 at 9:04 AM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

On Thu, Jul 15, 2010 at 9:29 PM, Naftoli Gugenheim <naftoligug@gmail.com> wrote:

I wouldn't read it as *excluding* distributing it in modified form, but it certainly doesn't mention permission to do so, no?To me "Permission to use, copy, modify, and distribute this software" is short for "Permission to use this software, and permission to copy this software, and permission to modify this software, and permission to distribute this software" [in source
or binary form for any purpose with or without fee is hereby granted...]So it's granting permission for four different things that are not inherently related, no?

When you're done with law school, let's have this discussion again.
English is a porous language and even the most carefully drafted language is open to interpretation.
Under your interpretation (you can do any of the four things listed, but none of them in combination), one could make a derivative work, but not use it or copy it.  Imagine how insane that would be... you would have to retype your modifications each time you wanted to make them and once you've re-typed them, you could not compile them (because compilation is making a copy, but altering the form).  I can't think of anyone making such an argument.  So, because it's reasonable to make modifications and then make copies of the modifications (compile) and perform (in the copyright sense) the modifications (run), why not distribute the modifications?  Further, EPFL has encouraged parties to engage in such behavior (support for PaulP's GitHub branch, support for the LLVM branch of Scala, the cross-pollination of Scala's XML libraries into Lift [we made a derivative work of some of Scala's XML libraries to enhance their functionality, EPFL then took the Lift enhancements and rolled those back into Scala], etc.)  It's clear to me that EPFL intends their license to allow distribution of derivative works by both words and actions.
Yeah, there is a very minor issue when folks change license language because the common open source licenses have been reviewed and are commonly agreed on.  However, very few of the open source licenses have ever been tested in court (to my knowledge, only the GPL has been tested at both the trial and appellate level).  So, none of us really know what any of the licenses mean because until a couple of circuit courts in the US (appellate) agree on an interpretation, it's just a bunch of guys on mailing lists wasting time.  

On Thu, Jul 15, 2010 at 11:30 PM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

Here's a list of what rights can be granted under copyright law in the US: http://www.bitlaw.com/copyright/scope.html
Both licenses grant all the rights (if you assume the "use" means "perform" and "display")  The wording is different.  I do not read modification and distribution to be mutually exclusive.  Put another way, you've got the right to create a derivative work and you have the right to modify.  Reading the license to exclude the right to distribute derivative works would be like reading the license to exclude the use of derivative works.
Before I started my first software company, I wrote software licenses.
On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys <andrae.muys@cambia.org> wrote:

On 16 July 2010 09:01, Seth Tisue <seth@tisue.net> wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim <naftoligug@gmail.com> writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

--
Andrae Muys,
Senior Software Engineer, Cambia
Initiative for Open Innovation (IOI)
Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
+61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

--
http://erikengbrecht.blogspot.com/

--
http://erikengbrecht.blogspot.com/

I hear that -- thanks for the clarification!Still, I have a question... but it doesn't sound like anyone wants to hear it so I'll keep my mouth closed for now...

On Fri, Jul 16, 2010 at 9:04 AM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

On Thu, Jul 15, 2010 at 9:29 PM, Naftoli Gugenheim <naftoligug@gmail.com> wrote:

I wouldn't read it as *excluding* distributing it in modified form, but it certainly doesn't mention permission to do so, no?To me "Permission to use, copy, modify, and distribute this software" is short for "Permission to use this software, and permission to copy this software, and permission to modify this software, and permission to distribute this software" [in source
or binary form for any purpose with or without fee is hereby granted...]So it's granting permission for four different things that are not inherently related, no?

When you're done with law school, let's have this discussion again.
English is a porous language and even the most carefully drafted language is open to interpretation.
Under your interpretation (you can do any of the four things listed, but none of them in combination), one could make a derivative work, but not use it or copy it.  Imagine how insane that would be... you would have to retype your modifications each time you wanted to make them and once you've re-typed them, you could not compile them (because compilation is making a copy, but altering the form).  I can't think of anyone making such an argument.  So, because it's reasonable to make modifications and then make copies of the modifications (compile) and perform (in the copyright sense) the modifications (run), why not distribute the modifications?  Further, EPFL has encouraged parties to engage in such behavior (support for PaulP's GitHub branch, support for the LLVM branch of Scala, the cross-pollination of Scala's XML libraries into Lift [we made a derivative work of some of Scala's XML libraries to enhance their functionality, EPFL then took the Lift enhancements and rolled those back into Scala], etc.)  It's clear to me that EPFL intends their license to allow distribution of derivative works by both words and actions.
Yeah, there is a very minor issue when folks change license language because the common open source licenses have been reviewed and are commonly agreed on.  However, very few of the open source licenses have ever been tested in court (to my knowledge, only the GPL has been tested at both the trial and appellate level).  So, none of us really know what any of the licenses mean because until a couple of circuit courts in the US (appellate) agree on an interpretation, it's just a bunch of guys on mailing lists wasting time.  

On Thu, Jul 15, 2010 at 11:30 PM, David Pollak <feeder.of.the.bears@gmail.com> wrote:

Here's a list of what rights can be granted under copyright law in the US: http://www.bitlaw.com/copyright/scope.html
Both licenses grant all the rights (if you assume the "use" means "perform" and "display")  The wording is different.  I do not read modification and distribution to be mutually exclusive.  Put another way, you've got the right to create a derivative work and you have the right to modify.  Reading the license to exclude the right to distribute derivative works would be like reading the license to exclude the use of derivative works.
Before I started my first software company, I wrote software licenses.
On Thu, Jul 15, 2010 at 8:12 PM, Andrae Muys <andrae.muys@cambia.org> wrote:

On 16 July 2010 09:01, Seth Tisue <seth@tisue.net> wrote:
>>>>>> "Naftoli" == Naftoli Gugenheim <naftoligug@gmail.com> writes:
>
>  Naftoli> Looks like they don't want to allow modified versions?
>
> Huh?  "Permission to use, copy, modify..."
>
> --
> Seth Tisue @ Northwestern University | http://tisue.net
> lead developer, NetLogo: http://ccl.northwestern.edu/netlogo/
>

The issue is the lack of permission to distribute modifications.
Normally the language wouldn't matter so much, but because it is
derived from a pre-existing, long standing license, it is definitely
possible that, absent an explicit statement of intent by the licensor,
that the redaction of permission to distribute modified versions may
be read by a court as explicitly forbidding it.

Hence it is not unreasonable for a commenter to ask for a
clarification of intent.

Andrae

P.S. IANAL, but I have spent a lot of time advising lawyers on the
technical aspects of software copyright, and I'm pretty certain that
the reasoning above is sound; of course, I will check with my
copyright lawyer friends next time I talk to them.

--
Andrae Muys,
Senior Software Engineer, Cambia
Initiative for Open Innovation (IOI)
Cambia@QUT, G301, 2 George Street, Brisbane Qld 4000, Australia
+61 414 517 882 (mobile)  +61 7 3138 4538 (work) +61 7 3138 4405 (fax)

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

On Fri, 2010-07-16 at 06:04 -0700, David Pollak wrote:
[ . . . ]

> When you're done with law school, let's have this discussion again.

It is interesting that when it comes to licences programmers think they
can be lawyers without first being trained in the law. And of course
(as noted below) there isn't just one law, there are many.

[ . . . ]

> copy it. Imagine how insane that would be... you would have to
> retype your modifications each time you wanted to make them and once
> you've re-typed them, you could not compile them (because compilation
> is making a copy,

Just because something is illogical doesn't mean that is not what the
law says. But that leads us back to the point above.

[ . . . ]

> Yeah, there is a very minor issue when folks change license language
> because the common open source licenses have been reviewed and are
> commonly agreed on. However, very few of the open source licenses
> have ever been tested in court (to my knowledge, only the GPL has been
> tested at both the trial and appellate level). So, none of us really
> know what any of the licenses mean because until a couple of circuit
> courts in the US (appellate) agree on an interpretation, it's just a
> bunch of guys on mailing lists wasting time.
[ . . . ]

The question should be: who provided the changed wording and if it
wasn't an appropriately trained lawyer, did an appropriately trained
lawyer review and agree that the changes were inline with the intention
of the original copyright owner?

There is also the issue of jurisdiction: just because some US court
makes a ruling, doesn't mean that applies in any other jurisdiction.
And there is the issue of statute vs. case law -- for those
jurisdictions where this is an issue.

Although this is seemingly well off topic, it is nonetheless important
stuff.

Call me a contrarian, but I for one do not think that being a lawyer automatically makes one qualified to comment on some particular aspect of the law, nor do I think that not being one makes one's opinion invalid. For one thing, "the law" is a pretty broad topic, and just because someone is an expert in, say, bankruptcy or tax law, doesn't mean that the person knows anything about software licenses. Not to mention that there are moron lawyers along with the morons of every other profession.

As far as non-lawyers go, there are plenty of them who have extensive legal experience who likely know more about their particular area of expertise than a "real" lawyer who specializes in something else. I'd likely trust David Pollak's opinion about a software license issue over that of the "real" lawyer who wrote my will.

In this as in all intellectual domains, a person's certifications or educational qualifications are secondary to their ability to explain their point of view and make a persuasive argument. People who have attended law school and passed the bar exam can probably speak to some aspect of the law better than someone who doesn't have that background, but if they have to fall back on "well I'm a lawyer, so that means I'm right!" then that is pretty much an admission of failure. Certainly lawyers don't get to use that argument in court!

W

On Jul 17, 2010, at 2:14 AM, Russel Winder wrote:

> On Fri, 2010-07-16 at 06:04 -0700, David Pollak wrote:
> [ . . . ]
>
>> When you're done with law school, let's have this discussion again.
>
> It is interesting that when it comes to licences programmers think they
> can be lawyers without first being trained in the law. And of course
> (as noted below) there isn't just one law, there are many.
>
> [ . . . ]
>
>> copy it. Imagine how insane that would be... you would have to
>> retype your modifications each time you wanted to make them and once
>> you've re-typed them, you could not compile them (because compilation
>> is making a copy,
>
> Just because something is illogical doesn't mean that is not what the
> law says. But that leads us back to the point above.
>
> [ . . . ]
>
>> Yeah, there is a very minor issue when folks change license language
>> because the common open source licenses have been reviewed and are
>> commonly agreed on. However, very few of the open source licenses
>> have ever been tested in court (to my knowledge, only the GPL has been
>> tested at both the trial and appellate level). So, none of us really
>> know what any of the licenses mean because until a couple of circuit
>> courts in the US (appellate) agree on an interpretation, it's just a
>> bunch of guys on mailing lists wasting time.
> [ . . . ]
>
> The question should be: who provided the changed wording and if it
> wasn't an appropriately trained lawyer, did an appropriately trained
> lawyer review and agree that the changes were inline with the intention
> of the original copyright owner?
>
> There is also the issue of jurisdiction: just because some US court
> makes a ruling, doesn't mean that applies in any other jurisdiction.
> And there is the issue of statute vs. case law -- for those
> jurisdictions where this is an issue.
>
> Although this is seemingly well off topic, it is nonetheless important
> stuff.
>

On 16 July 2010 14:04, David Pollak <feeder.of.the.bears@gmail.com> wrote:

Yeah, there is a very minor issue when folks change license language because the common open source licenses have been reviewed and are commonly agreed on.  However, very few of the open source licenses have ever been tested in court (to my knowledge, only the GPL has been tested at both the trial and appellate level).  So, none of us really know what any of the licenses mean because until a couple of circuit courts in the US (appellate) agree on an interpretation, it's just a bunch of guys on mailing lists wasting time.

I agree that because software licenses have rarely been tested in the courts it's true that none of us (even the lawyers) actually know what they mean in practise however I disagree that changing the language is a minor issue.
For my sins, I work for a large multi-national and this would be a significant barrier to adoption for us.  Why?  Well, a number of the main FOSS licenses have been cleared by our legal team.  In the midst of a large bureaucracy this took months to achieve and is not something anybody will be rushing to do again soon.  FreeBSD was one of the licenses approved - however due to this modification the Scala license would have to be cleared separately.  This means the project wishing to use it would bear a cost and the waiting time would quite possibly rule it's use out.
Of course, big dumb risk averse corporations aren't everything and some companies are smarter than others at this sort of thing, this issue probably won't affect many projects.  But it would still seem to me to be a significant enough barrier to make trying to stick to a widely recognised license worthwhile, particularly when the modifications appear to the untrained eye to be cosmetic.  
So I think it's a valid, useful question - what were the reasons for the decision to use a modified FreeBSD license over the FreeBSD one?  Leading to:  Can that decision be re-evaluated in the interests of easier adoption?  

Hi,

Got some strange behaviour with named parameters:

val x = C (x = "...")

(where C.apply is a method with an argument that is named
"x") gives the following message:

error: recursive value x needs type

while:

val x : T = C (x = "...")

compiles OK. To me it is obvious that the "x" in "x = ..." is
a parameter name. However, the parser doesn't seem to see
that. So either I or the parser am/is missing some clue here.

Who's right?

Job

i assume the compiler cannot decide wether you want to assign the value
"..." to a parameter named x or assign the value "..." to the member x
and use x itself as a parameter. once you tell the compiler that val x
is T and not "...", it understands what you want.

Job Honig schrieb:
> Hi,
>
> Got some strange behaviour with named parameters:
>
> val x = C (x = "...")
>
> (where C.apply is a method with an argument that is named
> "x") gives the following message:
>
> error: recursive value x needs type
>
> while:
>
> val x : T = C (x = "...")
>
> compiles OK. To me it is obvious that the "x" in "x = ..." is
> a parameter name. However, the parser doesn't seem to see
> that. So either I or the parser am/is missing some clue here.
>
> Who's right?
>
> Job
>
>
>
>

> i assume the compiler cannot decide wether you want to assign the value
> "..." to a parameter named x or assign the value "..." to the member x
> and use x itself as a parameter. once you tell the compiler that val x
> is T and not "...", it understands what you want.

Yes, that's what I thought. But is this implied by the language spec or is
it a compiler "problem"?

Job

On Sat, Jul 17, 2010 at 6:25 AM, Willis Blackburn <wboyce@panix.com> wrote:

Call me a contrarian, but I for one do not think that being a lawyer automatically makes one qualified to comment on some particular aspect of the law, nor do I think that not being one makes one's opinion invalid.  For one thing, "the law" is a pretty broad topic, and just because someone is an expert in, say, bankruptcy or tax law, doesn't mean that the person knows anything about software licenses.  Not to mention that there are moron lawyers along with the morons of every other profession.

As far as non-lawyers go, there are plenty of them who have extensive legal experience who likely know more about their particular area of expertise than a "real" lawyer who specializes in something else.  I'd likely trust David Pollak's opinion about a software license issue over that of the "real" lawyer who wrote my will.

Just to be clear, I graduated BU Law '91.  I was admitted to the Rhode Island bar in 1992 and the US First Circuit bar in 1994.  The highest profile work I did was representing Eugene Kashpureff in the civil side of the AlterNIC case (http://en.wikipedia.org/wiki/AlterNIC).  Or maybe it was when I filed a motion in the Lotus v. Borland case (http://www.nextcomputers.org/NeXTfiles/Articles/NeXTWORLD/NeXTWORLD_Extra/93.11.Nov.NWE/93.11.Nov.NWExtra27.html ) claiming computer programs were speech under the first amendment and thus a programming language could not be protected by copyright law (the implementation could, but the language itself could not.)  I lost my case, but the 9th circuit later held that computer programs were protected speech.

I no longer practice, but have a reasonable grasp of both the tenets of the law (I'm no longer current on the cases) and technology.
 

In this as in all intellectual domains, a person's certifications or educational qualifications are secondary to their ability to explain their point of view and make a persuasive argument.  People who have attended law school and passed the bar exam can probably speak to some aspect of the law better than someone who doesn't have that background, but if they have to fall back on "well I'm a lawyer, so that means I'm right!" then that is pretty much an admission of failure.  Certainly lawyers don't get to use that argument in court!

W



On Jul 17, 2010, at 2:14 AM, Russel Winder wrote:

> On Fri, 2010-07-16 at 06:04 -0700, David Pollak wrote:
> [ . . . ]
>
>> When you're done with law school, let's have this discussion again.
>
> It is interesting that when it comes to licences programmers think they
> can be lawyers without first being trained in the law.  And of course
> (as noted below) there isn't just one law, there are many.
>
> [ . . . ]
>
>> copy it.  Imagine how insane that would be... you would have to
>> retype your modifications each time you wanted to make them and once
>> you've re-typed them, you could not compile them (because compilation
>> is making a copy,
>
> Just because something is illogical doesn't mean that is not what the
> law says.  But that leads us back to the point above.
>
> [ . . . ]
>
>> Yeah, there is a very minor issue when folks change license language
>> because the common open source licenses have been reviewed and are
>> commonly agreed on.  However, very few of the open source licenses
>> have ever been tested in court (to my knowledge, only the GPL has been
>> tested at both the trial and appellate level).  So, none of us really
>> know what any of the licenses mean because until a couple of circuit
>> courts in the US (appellate) agree on an interpretation, it's just a
>> bunch of guys on mailing lists wasting time.
> [ . . . ]
>
> The question should be:  who provided the changed wording and if it
> wasn't an appropriately trained lawyer, did an appropriately trained
> lawyer review and agree that the changes were inline with the intention
> of the original copyright owner?
>
> There is also the issue of jurisdiction:  just because some US court
> makes a ruling, doesn't mean that applies in any other jurisdiction.
> And there is the issue of statute vs. case law -- for those
> jurisdictions where this is an issue.
>
> Although this is seemingly well off topic, it is nonetheless important
> stuff.
>
> --
> Russel.
> =============================================================================
> Dr Russel Winder      t: +44 20 7585 2200   voip: 3Arussel [dot] winder [at] ekiga [dot] net" rel="nofollow">sip:russel.winder@ekiga.net
> 41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel@russel.org.uk
> London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder

--
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Blog: http://goodstuff.im
Surf the harmonics

On Sat, Jul 17, 2010 at 9:25 AM, Willis Blackburn <wboyce@panix.com> wrote:

Certainly lawyers don't get to use that argument in court!

Judge Linda Reade seems to disagree...

On Sat, Jul 17, 2010 at 2:14 AM, Russel Winder <russel@russel.org.uk> wrote:

It is interesting that when it comes to licences programmers think they
can be lawyers without first being trained in the law.  And of course
(as noted below) there isn't just one law, there are many.

Whoa! Did I presume to understand American law? I asked a question, hoping to get a clarification. I'm sorry if non-lawyers are not capable of understanding explanations pertaining to law. I for one enjoy getting a better understanding of things for the intellectual pleasure of it. If you don't then the question wasn't directed to you.
Since the discussion hasn't ended, I'll take the liberty of asking the question I suppressed before, with the disclaimer that if anyone thinks it's a waste of his time to answer me, then please don't answer. And if reading it is wasting your time, then don't read further, and I apologize for having made you read these two paragraphs.

I wouldn't read it as *excluding* distributing it in modified form, but it certainly doesn't mention permission to do so, no?

To me "Permission to use, copy, modify, and distribute this software" is short for "Permission to use this software, and permission to copy this software, and permission to modify this software, and permission to distribute this software" [in source

or binary form for any purpose with or without fee is hereby granted...]

So it's granting permission for four different things that are not inherently related, no?

 

Under your interpretation (you can do any of the four things listed, but none of them in combination), one could make a derivative work, but not use it or copy it.  Imagine how insane that would be... you would have to retype your modifications each time you wanted to make them and once you've re-typed them, you could not compile them (because compilation is making a copy, but altering the form).  I can't think of anyone making such an argument.  

Exactly! 

So, because it's reasonable to make modifications and then make copies of the modifications (compile) and perform (in the copyright sense) the modifications (run), why not distribute the modifications?

Because as quoted above, one not is reasonable and to prohibit it, and one is reasonable to prohibit. So again, no combination of the four actions was mentioned explicitly. Therefore they have changed the phraseology to remove explicit permission to (among other things) distribute modifications, and that's something that one shouldn't automatically assume is permitted. However that argument doesn't extend to using or making personal copies of the derivative, because it's clear that they don't mean to prohibit it.
 

 Further, EPFL has encouraged parties to engage in such behavior (support for PaulP's GitHub branch, support for the LLVM branch of Scala, the cross-pollination of Scala's XML libraries into Lift [we made a derivative work of some of Scala's XML libraries to enhance their functionality, EPFL then took the Lift enhancements and rolled those back into Scala], etc.)  It's clear to me that EPFL intends their license to allow distribution of derivative works by both words and actions.

That's definitely true. Perhaps they just wanted the wording to be more neutral on derivative distribution, so people "in the wild" would give pause before automatically assuming it was okay? Or perhaps they mean to prohibit it by default except in certain cases where it's clearly to their benefit.

On Sat, Jul 17, 2010 at 09:25:15AM -0400, Willis Blackburn wrote:
> Call me a contrarian, but I for one do not think that being a lawyer
> automatically makes one qualified to comment on some particular aspect
> of the law, nor do I think that not being one makes one's opinion
> invalid. For one thing, "the law" is a pretty broad topic, and just
> because someone is an expert in, say, bankruptcy or tax law, doesn't
> mean that the person knows anything about software licenses.

While true in theory, twenty years of watching programmers speculate
about legal issues has led me to one inescapable conclusion: programmers
are grievously unsuited to speculate about legal issues. I'm with you
that non-professionals should not automatically defer to professionals,
it's just that something in the programmer DNA seems to demand they
follow what seems like a logical progression in what seems like a
logical way. And in most domains it works pretty well. But in this one
there is a meta-logic that takes a long time to acquire, and without it
you are miscalibrated from the start.

For the record I would be very happy if scala used an unmodified BSD
license for all the good reasons already stated. And another good
reason would be to give threads like this one no reason to exist,
because nothing is sadder than a bunch of people who could be writing
code authoring legal opinions instead.

It's a bug, I filed a ticket (http://lampsvn.epfl.ch/trac/scala/ticket/3685).Thanks for reporting!

On Sat, Jul 17, 2010 at 18:12, Job Honig <joho@xs4all.nl> wrote:

> i assume the compiler cannot decide wether you want to assign the value
> "..." to a parameter named x or assign the value "..." to the member x
> and use x itself as a parameter. once you tell the compiler that val x
> is T and not "...", it understands what you want.

Yes, that's what I thought.  But is this implied by the language spec or is
it a compiler "problem"?

Job