Walter: I know this has been asked as lot, but
just what is the license on the D specification ?

That is, the files in dmd.zip available under:
dmd/html/d/* (total: 62 HTML files, 15 images)

As I understand it, at the moment it is just a
dump of the http://www.digitalmars.com/d/ site ?


If those are under a re-distributable license,
like dmd/src/dmd/* and dmd/src/phobos/* are now,
then the DMD sources/docs can be re-distributed -
after first stripping out the various X86 binaries...

If they're not, then the dmd/html directory also
needs to be stripped out of any re-distribution -
just like : dm/bin, dm/lib, dmd/bin, and dmd/lib
already have to be. (DMD, non-distributable license)


It would also make it more clear on how to treat
derivate documentation works, such as all comments
on the Wiki4D - which are released under the GNU FDL.
(see http://www.gnu.org/copyleft/fdl.html for license)

For instance, I have a local doc version which I
converted into XHTML - with a nice CSS stylesheet,
and one version that has been converted into a PDF,
but I am not really sure if I can re-distribute them ?
So now I just keep them to myself instead, sorry. :-(

(here are all markup errors found in the current
version: http://www.algonet.se/~afb/d/errors.txt)


It would be GREAT if the D docs could be open too!
(currently GDC ships *without* any D documentation)

The simplest would probably be to make this also
into a Dual License, just like the DMD front-end ?
GPL => FDL, not sure what is like the "Artistic";
maybe http://creativecommons.org/licenses/by/2.0/ ?


I think this too needs to be cleared up before D 1.0...
(as well as the few remaining license woes with Phobos)
--anders


PS.
Is Digital Mars™ a trademark of Walter Bright, or
does it belong to Digital Mars, Inc. or something ?

"Anders F Björklund" <afb@algonet.se> wrote in message news:d2n9vr$18ii$1@digitaldaemon.com...
> Walter: I know this has been asked as lot, but
> just what is the license on the D specification ?

It's just plain old copyrighted.

> That is, the files in dmd.zip available under:
> dmd/html/d/* (total: 62 HTML files, 15 images)
>
> As I understand it, at the moment it is just a
> dump of the http://www.digitalmars.com/d/ site ?

That's right.

> If those are under a re-distributable license,
> like dmd/src/dmd/* and dmd/src/phobos/* are now,
> then the DMD sources/docs can be re-distributed -
> after first stripping out the various X86 binaries...
>
> If they're not, then the dmd/html directory also
> needs to be stripped out of any re-distribution -
> just like : dm/bin, dm/lib, dmd/bin, and dmd/lib
> already have to be. (DMD, non-distributable license)

For the time being, the docs are not redistributable. The reason is I want to keep them from getting too fragmented.

> It would also make it more clear on how to treat
> derivate documentation works, such as all comments
> on the Wiki4D - which are released under the GNU FDL.
> (see http://www.gnu.org/copyleft/fdl.html for license)
>
> For instance, I have a local doc version which I
> converted into XHTML - with a nice CSS stylesheet,
> and one version that has been converted into a PDF,
> but I am not really sure if I can re-distribute them ?
> So now I just keep them to myself instead, sorry. :-(
>
> (here are all markup errors found in the current
> version: http://www.algonet.se/~afb/d/errors.txt)
>
>
> It would be GREAT if the D docs could be open too! (currently GDC ships *without* any D documentation)

The documentation is just a click away.

> The simplest would probably be to make this also
> into a Dual License, just like the DMD front-end ?
> GPL => FDL, not sure what is like the "Artistic";
> maybe http://creativecommons.org/licenses/by/2.0/ ?
>
>
> I think this too needs to be cleared up before D 1.0...
> (as well as the few remaining license woes with Phobos)

I thought the licensing issues with Phobos were cleared up?

> --anders
>
>
> PS.
> Is Digital MarsT a trademark of Walter Bright, or
> does it belong to Digital Mars, Inc. or something ?

Of Digital Mars.

Walter wrote:

> For the time being, the docs are not redistributable.
> The reason is I want to keep them from getting too fragmented.

That's a valid reason. And beyond D 1.0: when we get there ?
(it could be just me being paranoid and premature here, too)

Thanks for answering so quickly, I'll email the XHTML as a zip.

>>It would be GREAT if the D docs could be open too!
>>(currently GDC ships *without* any D documentation)
> 
> The documentation is just a click away.

Not that it'll happen to D and Digital Mars, but *other*
projects have had such online references "torn away"...

Especially without a free/open license, and redistribution.

>>I think this too needs to be cleared up before D 1.0...
>>(as well as the few remaining license woes with Phobos)
> 
> I thought the licensing issues with Phobos were cleared up?

Almost, just a few minor details left (see other thread)...

But again - thanks for taking the time to add "phoboslicense.txt"
And a simple way to "fix" recls is to use the standalone version ?

--anders

"Anders F Björklund" <afb@algonet.se> wrote in message news:d2ofls$27nh$1@digitaldaemon.com...
> Walter wrote:
> >>It would be GREAT if the D docs could be open too! (currently GDC ships *without* any D documentation)
> > The documentation is just a click away.
> Not that it'll happen to D and Digital Mars, but *other* projects have had such online references "torn away"... Especially without a free/open license, and redistribution.

I can't imagine why I'd want to do that, but I accept that it can be worrisome from another's perspective. Probably the worst thing I'd do is serve some modest ads on it.

> >>I think this too needs to be cleared up before D 1.0...
> >>(as well as the few remaining license woes with Phobos)
> >
> > I thought the licensing issues with Phobos were cleared up?
>
> Almost, just a few minor details left (see other thread)...

Thanks.

> But again - thanks for taking the time to add "phoboslicense.txt"

Yeah, that needed to be done.

> And a simple way to "fix" recls is to use the standalone version ?

For the moment, yes. I've been reluctant to upgrade it because the newer versions are so large. Matthew has promised a smaller version in the future.

"Walter" <newshound@digitalmars.com> wrote in message news:d2q3c0$igp$1@digitaldaemon.com...
>
> "Anders F Björklund" <afb@algonet.se> wrote in message news:d2ofls$27nh$1@digitaldaemon.com...
>> Walter wrote:
>> >>It would be GREAT if the D docs could be open too! (currently GDC ships *without* any D documentation)
>> > The documentation is just a click away.
>> Not that it'll happen to D and Digital Mars, but *other* projects have had such online references "torn away"... Especially without a free/open license, and redistribution.
>
> I can't imagine why I'd want to do that, but I accept that it can be worrisome from another's perspective. Probably the worst thing I'd do is serve some modest ads on it.
>
>> >>I think this too needs to be cleared up before D 1.0...
>> >>(as well as the few remaining license woes with Phobos)
>> >
>> > I thought the licensing issues with Phobos were cleared up?
>>
>> Almost, just a few minor details left (see other thread)...
>
> Thanks.
>
>> But again - thanks for taking the time to add "phoboslicense.txt"
>
> Yeah, that needed to be done.
>
>> And a simple way to "fix" recls is to use the standalone version ?
>
> For the moment, yes. I've been reluctant to upgrade it because the newer versions are so large. Matthew has promised a smaller version in the future.

Version 1.6 is significantly trimmed, and the effort in making it so was for no other reason than meeting the agreement to trim it for Phobos.

Although recls 2 will likely be significantly smaller, it's at least 3/4 months away. (Again, this was discussed with you when we agreed that 1.6 would the small-enough version for Phobos.)

Walter wrote:

> I can't imagine why I'd want to do that, but I accept that it can be
> worrisome from another's perspective. Probably the worst thing I'd do is
> serve some modest ads on it.

I'm not saying that you *want* to do it, just that it could happen.
And I don't anyone has anything against a few ads to cover hosting.

>>And a simple way to "fix" recls is to use the standalone version ?
> 
> For the moment, yes. I've been reluctant to upgrade it because the newer
> versions are so large. Matthew has promised a smaller version in the future.

You could just update std.recls, and leave the rest for download...
(and either leave the recls docs in, or take those out as well ?)

 48K    recls-1.6.1/mappings/D/std/recls.d
1.8M    recls-1.6.1.zip

It declares the private API (i.e. the "extern(C)") inline, but that
*could* be separated to a std.c.recls module and imported instead...

Of course any user would have to add the recls library themselves,
as it wouldn't be a part of Phobos anymore ? I think that's good.

1.2M    libphobos.a # without recls and zlib
1.4M    libphobos.a
1.7M    libphobos-contracts.a # without -release

--anders

Matthew wrote:

>>>And a simple way to "fix" recls is to use the standalone version ?
>>
>>For the moment, yes. I've been reluctant to upgrade it because the newer
>>versions are so large. Matthew has promised a smaller version in the future.
> 
> Version 1.6 is significantly trimmed, and the effort in making it so was
> for no other reason than meeting the agreement to trim it for Phobos.

I even managed to get it to compile, after a while of trying... :-)

Downloaded:
1.9M    http://synesis.com.au/downloads/recls/recls-1.6.1.zip
1.1M    http://synesis.com.au/downloads/stlsoft/stlsoft-1.8.3-beta4.zip

Issued:
unzip -d recls-1.6.1 -q recls-1.6.1.zip
unzip -d stlsoft-1.8.3 -q stlsoft-1.8.3-beta4.zip

make -f makefile.unix -s \
  -C recls-1.6.1/build/gcc33 \
  PROJ_BASE_DIR="$PWD/recls-1.6.1" \
  STLSOFT_INCLUDE="$PWD/stlsoft-1.8.3"

Built:
252K    recls-1.6.1/lib/recls.gcc33.a
344K    recls-1.6.1/lib/recls.gcc33.debug.a

With:
gcc (GCC) 3.3.2 20031022 (Red Hat Linux 3.3.2-1)
Copyright (C) 2003 Free Software Foundation, Inc.


But I could find any installation instructions, nor any packages ? :-(

I assume it's the Java way: copy anywhere, and mess with the ENV...
(So choosing /usr/include/stlsoft and /usr/include/recls could work)

Suppose it's just time for me to slap together a few RPM specfiles ?
(with a stlsoft-config and a recls-config script, to give the paths)

Funny how the recls library download is as big as the DMD compiler.

--anders

"Anders F Björklund" <afb@algonet.se> wrote in message news:d2quh2$1e6j$1@digitaldaemon.com...
> Built:
> 252K    recls-1.6.1/lib/recls.gcc33.a
> 344K    recls-1.6.1/lib/recls.gcc33.debug.a

The recls currently shipping with dmd is:

25K    librecls.a on linux
37K    recls.lib on Win32

Walter wrote:

>>Built:
>>252K    recls-1.6.1/lib/recls.gcc33.a
>>344K    recls-1.6.1/lib/recls.gcc33.debug.a
> 
> The recls currently shipping with dmd is:
> 25K    librecls.a on linux
> 37K    recls.lib on Win32

Note that it had all -g debug symbols left in,
before drawing any conclusions regarding recls.

I still think it should be an external library,
but not primarily because of the code size...


As I understood it, recls 1.6.1 was a stripped
down version - especially for D. And that the
main new additions would go in recls 2.0 instead ?
But Matthew Wilson would know all about that...

Also, recls 1.6.1 requires the latest beta of STLSoft
(i.e. 1.8.3 beta 3 or above). The stable lib won't do.

--anders

"Anders F Björklund" <afb@algonet.se> wrote in message news:d2rf87$218u$1@digitaldaemon.com...
> Walter wrote:
>
>>>Built:
>>>252K    recls-1.6.1/lib/recls.gcc33.a
>>>344K    recls-1.6.1/lib/recls.gcc33.debug.a
>>
>> The recls currently shipping with dmd is:
>> 25K    librecls.a on linux
>> 37K    recls.lib on Win32
>
> Note that it had all -g debug symbols left in,
> before drawing any conclusions regarding recls.

:-)

> I still think it should be an external library,
> but not primarily because of the code size...

[Do the following two paras make the point of why? If so, it's not clear. Can you rephrase/elucidate?]

> As I understood it, recls 1.6.1 was a stripped
> down version - especially for D.

Correct

> And that the
> main new additions would go in recls 2.0 instead ?

recls 2 will be huge change because it'll
    - do things more betterer, as a result of things learned in recls 1.x
    - it'll have plenty of new features, including more powerful behaviour and a wider variety of searchable things
(over and above just filesystem and FTP hosts)
    - it'll likely be largely written in C or 'tight' C++

But:
    - it'll be 2-3 months away. But because (AFAICT) that's still before 1.0 is released - at least I hope so! - I
really don't see a problem with going with updating Phobos with recls 1.x right now. To not do so seems to arbitrarily
single out recls for a fault (size) when the rest of Phobos is so embarassingly crappy (and I don't just mean the
features that I wrote).

> But Matthew Wilson would know all about that...
>
> Also, recls 1.6.1 requires the latest beta of STLSoft
> (i.e. 1.8.3 beta 3 or above). The stable lib won't do.

What's your point here?