taskSavannah Administration - Tasks: task #14529, Submission of Real Time...

 
 

task #14529: Submission of Real Time Application Interface (for Linux)

Submitter:  Paolo Mantegazza <mante>
Submitted:  Tue 30 May 2017 12:40:02 PM UTC
   
 
Should Start On:  Tue 30 May 2017 12:00:00 AM UTC Should be Finished on:  Fri 09 Jun 2017 12:00:00 AM UTC
Category:  Project Approval Priority:  * 5 - Normal
Status:  Done Privacy:  Public
Assigned to:  ineiev Open/Closed:  Closed
* Mandatory Fields

Add a New Comment Rich Markup
   

Jump to the original submission

Thu 30 Nov 2017 08:07:32 AM UTC, comment #45: 

I believe storing the old CVS and starting a clean svn is acceptable.

Ineiev <ineiev>
Site Administrator
Wed 29 Nov 2017 04:56:19 PM UTC, comment #44: 


Everything is, in principle, possible. But right now we do not have a CVS server at all, see comment #14 for context.
We would need to set up a temporary local CVS server, delete lots of files, commit everything, double-check that we are not forgetting any file, etc...

What I'm proposing is to store the old CVS history for reference, leave those CVSs unused, and start from a clean state with a completely different cvs (or svn, if possible). In this way we would really be sure not to forget anything, still keeping the old project history somewhere.

Let me know how to proceed.

Marco Morandini <mmforrtai>
Wed 29 Nov 2017 04:22:59 PM UTC, comment #43: 

I think the way to go is to add a CVS commit fixing those notices and to import CVS history then. Is that possible?

Ineiev <ineiev>
Site Administrator
Wed 29 Nov 2017 11:46:31 AM UTC, comment #42: 

Thank you. We have discussed internally on how to proceed, since the changes required to fix copyright headers etc... were significant, and we were without CVS.

If you agree, we would like to

- import the saved CVS histories into Savannah, and leave them unused, only to save the code history. This CVS will not have the copyright fixed. Should we convert them to svn before uploading them?
- start a new svn, where we would like to upload, as revision 0, the approved tarball, and work from there.

Would this be ok?

Thank you again,

Marco

Marco Morandini <mmforrtai>
Mon 20 Nov 2017 04:38:35 PM UTC, comment #41: 

Thank you, approving (I wonder if you still could consider adding copyright and license notices to patches like base/arch/x86/patches/hal-linux-3.18.20-x86-6.patch).

Ineiev <ineiev>
Site Administrator
Fri 20 Oct 2017 08:11:37 AM UTC, comment #40: 

https://polimi365-my.sharepoint.com/personal/10112071_polimi_it/_layouts/15/guestaccess.aspx?docid=126d2376bcd4144ad87962ac1dba7bd36&authkey=Abg6RBNoR9F-cYDjM0jYPEE

We got rid of the SCILAB license by importing the files distributed by a scicoslab fork ( https://cermics.enpc.fr/~jpc/nsp-tiddly/mine.html ) that was relicensed under GPL.
The list of modified files is this (mostly cosmetic changes to the list of Copyright holders; "./" is the directory rtai-lab)

./scicoslab/devices/machine.h
./scicoslab/devices/config.h
./scilab/devices/machine.h
./scicoslab/macros/RTAI/RTAICodeGen_.sci
./scilab/examples/test.cos -> add copy of ./scilab5/examples/README.license
./scilab/macros/RTAI/RTAICodeGen_.sci
./scilab5/scicos_src/urand.c
./scilab5/scicos_src/dynlib_randlib.h
./scilab5/scicos_src/sum.c
./scilab5/scicos_src/matz_cath.h
./scilab5/scicos_src/matz_catv.h
./scilab5/scicos_src/include/dynlib_scicos_blocks.h
./scilab5/include/scilab/realtime.h
./scilab5/include/scilab/HistoryManager.h
./scilab5/include/scilab/timer.h
./scilab5/include/scilab/getPartLine.h
./scilab5/include/scilab/api_error.h
./scilab5/include/scilab/dynlib_completion.h
./scilab5/include/scilab/api_scilab.h
./scilab5/include/scilab/MALLOC.h
./scilab5/include/scilab/ScilabStream.hxx
./scilab5/include/scilab/api_list.h
./scilab5/include/scilab/ScilabOptionsSetter.hxx
./scilab5/include/scilab/stackTypeVariable.h
./scilab5/include/scilab/sci_gateway.h
./scilab5/include/scilab/getmemory.h
./scilab5/include/scilab/sciprint.h
./scilab5/include/scilab/api_hypermat.h
./scilab5/include/scilab/freeArrayOfString.h
./scilab5/include/scilab/warningmode.h
./scilab5/include/scilab/call_scilab.h
./scilab5/include/scilab/api_int.h
./scilab5/include/scilab/dynlib_scicos.h
./scilab5/include/scilab/api_double.h
./scilab5/include/scilab/getScilabJavaVM.h
./scilab5/include/scilab/getScilabObject.h
./scilab5/include/scilab/blocks.h
./scilab5/include/scilab/api_optional.h
./scilab5/include/scilab/stackinfo.h
./scilab5/include/scilab/dynlib_history_manager.h
./scilab5/include/scilab/api_pointer.h
./scilab5/include/scilab/SCIHOME.h
./scilab5/include/scilab/libinter.h
./scilab5/include/scilab/getScilabJNIEnv.h
./scilab5/include/scilab/scicos-def.h
./scilab5/include/scilab/mode_exec.h
./scilab5/include/scilab/completeLine.h
./scilab5/include/scilab/scicos_math.h
./scilab5/include/scilab/gw_completion.h
./scilab5/include/scilab/addToClasspath.h
./scilab5/include/scilab/scilabDefaults.h
./scilab5/include/scilab/getos.h
./scilab5/include/scilab/api_boolean.h
./scilab5/include/scilab/charEncoding.h
./scilab5/include/scilab/ScilabAbstractEnvironmentWrapper.hxx
./scilab5/include/scilab/getCommonPart.h
./scilab5/include/scilab/sciquit.h
./scilab5/include/scilab/scisparse.h
./scilab5/include/scilab/api_string.h
./scilab5/include/scilab/InitializeHistoryManager.h
./scilab5/include/scilab/setgetlanguage.h
./scilab5/include/scilab/ScilabObjects.hxx
./scilab5/include/scilab/localization.h
./scilab5/include/scilab/ExternalObjects.h
./scilab5/include/scilab/dynamiclibrary.h
./scilab5/include/scilab/ScilabAbstractEnvironmentException.hxx
./scilab5/include/scilab/api_sparse.h
./scilab5/include/scilab/OptionsHelper.hxx
./scilab5/include/scilab/ScilabAbstractOptionsSetter.hxx
./scilab5/include/scilab/api_common.h
./scilab5/include/scilab/api_poly.h
./scilab5/include/scilab/Thread_Wrapper.h
./scilab5/include/scilab/sci_mem_alloc.h
./scilab5/include/scilab/gw_history_manager.h
./scilab5/include/scilab/ScilabAutoCleaner.hxx
./scilab5/include/scilab/ScilabGatewayOptions.hxx
./scilab5/include/scilab/setgetSCIpath.h
./scilab5/include/scilab/addToLibrarypath.h
./scilab5/include/scilab/ScilabGateway.hxx
./scilab5/include/scilab/completion.h
./scilab5/include/scilab/dynlib_fileio.h
./scilab5/include/scilab/api_boolean_sparse.h
./scilab5/include/scilab/getcommandlineargs.h
./scilab5/include/scilab/ScilabAbstractInvoker.hxx
./scilab5/include/scilab/ScilabAbstractMemoryAllocator.hxx
./scilab5/include/scilab/ScilabAbstractEnvironment.hxx
./scilab5/include/scilab/scilabmode.h
./scilab5/include/scilab/getFullFilename.h
./scilab5/include/scilab/doublecomplex.h
./scilab5/include/scilab/getversion.h
./scilab5/include/scilab/dynlib_external_objects_scilab.h
./scilab5/include/scilab/api_handle.h
./scilab5/include/scilab/Scierror.h
./scilab5/include/scilab/ScilabEnvironments.hxx
./scilab5/include/scilab/backtrace_print.h
./scilab5/include/scilab/do_error_number.h
./scilab5/include/scilab/dynamiclibrary_others.h
./scilab5/include/scilab/TerminateHistoryManager.h

Marco Morandini <mmforrtai>
Thu 05 Oct 2017 03:42:33 PM UTC, comment #39: 

Please check other files as well.

For example, rtai-lab/scicoslab/devices/machine.h has no
valid copyright notice, and it's released under
the SCILAB license, which makes that file proprietary,
https://www.gnu.org/licenses/license-list.html#Scilab

Ineiev <ineiev>
Site Administrator
Fri 15 Sep 2017 02:26:57 PM UTC, comment #38: 

New version:
https://polimi365-my.sharepoint.com/personal/10112071_polimi_it/_layouts/15/guestaccess.aspx?docid=160471041a2f54a69badc7e0693aff517&authkey=AZ94Zpbjkza2snN8AJRBKtA

> You can put there the date of the release of (the latest version of) these files.


Done, this is the only change wrt the prev. version
beside two missing copyright notices added to two files called "rtai.gen" and to an include file (machine-I-dont-rememeber-the-exact-name.h)

Thank you,

Marco


Marco Morandini <mmforrtai>
Tue 12 Sep 2017 04:44:03 PM UTC, comment #37: 


> However, we have a problem: many files (mostly those in rtai-

lab/scilab5/scicos_src)

> do have a copyright notice, but without year. And, as these are not our files,
> but we have simply copied them, I really don't know what year I could put there.


You can put there the date of the release of (the latest version of) these files.

If you don't think this is an option, probably we'll have to ask licensing@fsf.org.

Ineiev <ineiev>
Site Administrator
Mon 04 Sep 2017 09:19:15 AM UTC, comment #36: 

New version:
https://polimi365-my.sharepoint.com/personal/10112071_polimi_it/_layouts/15/guestaccess.aspx?docid=1a88c832516544005b0f2aa13fe620f20&authkey=Ad1gcdjR4WGs7wV5c1q_hmc


> base/include/.rtai_shm.h.swp is still in the tarball.


deleted

> Is doc/doxygen/Doxyfile.in generated? It says Xapian
> is an "point.html open source" library; Savannah uses free
> software, not open source.


It works (and went) like this: a skeleton of the the configuration file Doxyfile.in is first auto-generated by Doxygen.
Then, Doxyfile.in is customized by hand. When a new Doxygen introduces new options, or deprecates old options, it gives to the user thee option to update the configuration file with sane default, without modifying, if possible, the user choices.
I made the error to follow Doxygen's suggestion, and updated the configuration file. Doxygen added the comment descibing Xapian as "open source", even if Xapian is GPL. I've changed that comment now. However, this is fragile, and I'm not completely comfortable with the change because Xapian, in his own home page, describes itself (wrongly, but this is not the point) as open source.
Anyway: fixed.

> rtai-lab/scilab5/RTAI/license.txt says:
>
> Copyright (C) 2009 Roberto Bucher
> 2010-2011 Holger Nahrstaedt


Fixed


> rtai-lab/scilab5/libs/scicoslibs.tgz lacks copyright and licensing notices.


This is more tricky, and I need your help in deciding what to do.
The tarball contained some precompiled libraries, with sources taken from scilab, version 5.
Now: if one gets a pre-built scilab, or builds it from scratch, he does not get those libraries build as we need them.
Thus, the initial choice to distribute the pre-built libraries.
However, I don't think distributing binaries is right.
Thus, we got rid of the pre-built libraries (the -tgz is gone), and are instead including the sources, taken from scilab 5, together with the Makefile required to build them. I went through the added files, and fixed the copyright notices (mostly removing the FSF address and adding the link to the GPL, but also adding the text of the licenses).
However, we have a problem: many files (mostly those in rtai-lab/scilab5/scicos_src) do have a copyright notice, but without year. And, as these are not our files, but we have simply copied them, I really don't know what year I could put there.
So the question: is this ok? If it's not ok (I know it is not a "valid copyright notice" as descibed in http://savannah.gnu.org/maintenance/ValidNotices/ ) then please don't waste your time on the tarball. And, if this is not ok, what would be better to do?

1) get rid of those sources, host somewhere a tarball with the sources, packaged as we need them, and modify the Makefile in suche a way that the tarball is automatically downloaded
2) do the same of 1), but with the pre-built libraries
3) give up out hope and go somewhere else (really hope this is not the case)
4) other ?

What do you suggest?

Thank you in advance,

Marco

Marco Morandini <mmforrtai>
Wed 23 Aug 2017 07:51:24 AM UTC, comment #35: 

Thank you!

base/include/.rtai_shm.h.swp is still in the tarball.

Is doc/doxygen/Doxyfile.in generated? It says Xapian
is an "point.html open source" library; Savannah uses free software, not open source.

rtai-lab/scilab5/RTAI/license.txt says:

    Copyright (C) 2009  Roberto Bucher
                  2010-2011  Holger  Nahrstaedt

rtai-lab/scilab5/libs/scicoslibs.tgz lacks copyright and licensing notices.

Ineiev <ineiev>
Site Administrator
Mon 14 Aug 2017 01:53:59 PM UTC, comment #34: 

Please check

https://polimi365-my.sharepoint.com/personal/10112071_polimi_it/_layouts/15/guestaccess.aspx?docid=1011e10aa04724a5d84a4225b82bb561e&authkey=AYJ8R4WmIya5m5dZL_GQTTQ

I hope I've fixed everything

Changes:

- deleted *.o.d
- deleted rtai-lab/matlab/ptinfo.tlc (it is distributed with matlab, we should not distribute it)
- deleted rtai-lab/scilab/examples/pid/pid.c and rtai-lab/scilab/examples/pid/pid_standalone.c (these should be generated by the user)
- added notices to files .sce, rtai-lab/scicoslab/macros/RTAI/rtai4_comedi_encoder.sci, rtai-lab/scicoslab/macros/RTAI/gif_icons/ , rtai-lab/scilab5/examples/test.cos
- added notice, and license file, for rtai-lab/scicoslab/devices/machine.h
- added notice to rtai-lab/matlab/examples/test.mdl and got rid of the additional LICENSE file I added there before
- changed notices from

  COPYRIGHT (C) 2002  Lorenzo Dozio (...@...)
  Paolo Mantegazza (...@...)

  Modified 15.1.2003 Roberto Bucher ...@...
  Modified 15.8.2009 Henrik Slotholt ...@...

to

  COPYRIGHT (C) 2002  Lorenzo Dozio (...@...)
  COPYRIGHT (C) 2002  Paolo Mantegazza (...@...)

  COPYRIGHT (C) 2003  Roberto Bucher ...@...
  COPYRIGHT (C) 2009  Henrik Slotholt ...@...

in every file I could spot.


Marco Morandini <mmforrtai>
Fri 11 Aug 2017 12:57:00 PM UTC, comment #33: 

Thank you; this is the next pass.

These files don't seem relevant for a distribution:

./base/tasklets/.tasklets.o.d
./base/ipc/netrpc/.netrpc.o.d
./base/ipc/sem/.sem.o.d
./base/ipc/bits/.bits.o.d
./base/ipc/msg/.msg.o.d
./base/ipc/mbx/.mbx.o.d
./base/ipc/shm/.shm.o.d
./base/ipc/tbx/.tbx.o.d
./base/ipc/fifos/.fifos.o.d
./base/ipc/mq/.mq.o.d
./base/arch/x86/hal/.hal.o.d
./base/include/.rtai_shm.h.swp
./base/sched/.malloc.o.d
./base/sched/.sched.o.d
./base/sched/.api.o.d
./base/sched/.signal.o.d
./base/sched/.sys.o.d
./base/wd/.wd.o.d
./addons/drivers/serial/.serial.o.d

The status of files like
rtai-lab/scilab5/RTAI/builder.sce
rtai-lab/scilab5/RTAI/etc/RTAI.start
rtai-lab/scilab5/RTAI/src/c/builder_c.sce
rtai-lab/scilab5/RTAI/src/c/loader.sce
rtai-lab/scilab5/RTAI/loader.sce isn't clear; they are plain
text, but they say they are generated (from unspecified sources);
they have no license notice.

rtai-lab/matlab/ptinfo.tlc has no license notice.

rtai-lab/scilab5/examples/test.cos has no notices.

Is rtai-lab/scicoslab/devices/machine.h autogenerated?
It lacks valid notices.

rtai-lab/scicoslab/macros/RTAI/rtai4_comedi_encoder.sci lacks notices.

rtai-lab/scicoslab/macros/RTAI/gif_icons/* have no notices.

The format of rtai-lab/matlab/examples/test.mdl allows
comments; the notices should be written in the file itself.

In files like rtai-lab/Fl_Scope.cpp the copyright notice
isn't very clear:

COPYRIGHT (C) 2003  Lorenzo Dozio (...@...)
                    Paolo Mantegazza (...@...)
    Roberto Bucher (...@...)
    Peter Brier (...@...)

One can see that Lorenzo Dozio is a copyright holder, but what about the rest?
Did they contribute anything copyrightable, or are they mentioned
to give them credit for something else?

This is especially ambiguous because in some other files
one can read:

  COPYRIGHT (C) 2002  Lorenzo Dozio (...@...)
  Paolo Mantegazza (...@...)

  Modified 15.1.2003 Roberto Bucher ...@...
  Modified 15.8.2009 Henrik Slotholt ...@...

Did Roberto Bucher and Henrik Slotholt assign copyright to someone else?
Are their modifications just uncopyrightable?

Every copyright holder should have a separate copyright notice.

Ineiev <ineiev>
Site Administrator
Fri 11 Aug 2017 05:23:06 AM UTC, comment #32: 

Q&QWE/!"&$£!&"

Could you try with

https://polimi365-my.sharepoint.com/personal/10112071_polimi_it/_layouts/15/guestaccess.aspx?docid=1396b7264598744239e7e4c4515018133&authkey=AQzk9qf8u4ghmchNJxDMEUM

?

Apologize for using Microsoft drive, but I'm out of office and cannot upload into my usual web server space from here.

Marco Morandini <mmforrtai>
Thu 10 Aug 2017 06:07:57 PM UTC, comment #31: 

I'm awfully sorry, the file didn't attach.

Ineiev <ineiev>
Site Administrator
Mon 07 Aug 2017 09:13:01 PM UTC, comment #30: 

OK, see rtai-5.1-mora6.tgz (attached).

Changes:

- got rid of doc/docbook (it was an unfinished work, and we are unable to identify the source of all files and/or their copyright holders); we'll work on it later, starting from scratch.
- got rid of doc/generated ; after discussing the matter internally we agreed that distributing the generated documentation is not such a great idea, and is very fragile: each and every time a maintainer regenerates the documentation there is the chance that new files are pulled in from e.g. doxygen, and we do not have a robust, automatic process in place to detect such additions and add the correct notices for those foreign files.
- few corrections here and there to configure.in, doc/doxygen/Doxyfile.in, and many GNUmakefile.am,
but without touching copyright notices.

Hope the first two changes simplify the matter.

Thank you,

Marco


Marco Morandini <mmforrtai>
Thu 03 Aug 2017 10:19:54 AM UTC, comment #29: 

Now, let us proceed with doc/docbook/; could you check files there? For example, I couldn't find a source for docbook-test/intro.xml, and that file has no notices.

Ineiev <ineiev>
Site Administrator
Mon 31 Jul 2017 08:58:36 PM UTC, comment #28: 

You are right onve more!

1)
Files

doc/generated/html/css/rtai.css
doc/docbook/css/rtai.css

were missing a copyright notice - added

2)
Files
doc/generated/html/api/doxygen.css

and

doc/generated/html/api/ftv2*.png

comes from doxygen when we generate the documentation.
I've added a corresponding
generated doc/generated/html/api/DOXYGEN_FILES.LICENSE
(I could add a copyright notice to doxygen.css, but it would be overwritten every time a developer re-generates the documentation)

All the other pngs in doc/generated/html/api/ are generated from RTAI sources by doxygen, likely calling dot.

New version: https://home.aero.polimi.it/morandini/Downloads/rtai-5.1-mora5.tgz

Marco

Marco Morandini <mmforrtai>
Mon 31 Jul 2017 02:04:39 PM UTC, comment #27: 

Thank you!

Files like doc/generated/html/api/{doxygen.css,*.png} still seem
to have no valid notices.

Are they all generated from RTAI source files?

Ineiev <ineiev>
Site Administrator
Thu 20 Jul 2017 03:59:49 PM UTC, comment #26: 

New version: https://home.aero.polimi.it/morandini/Downloads/rtai-5.1-mora4.tgz


> Yes, but probably you should add "specific licenses" to files like doc/generated/html/pictures/* (and possibly other files as well).


I did not realize that those file were being copied from docbook/ to generated/ while generating the documentation; added "specific licenses" also there.
Really hope to have not missed anything else and to stop wasting your time (btw: I really have to thank you for your patience).

> rtai-lab/matlab/examples/test.mdl still has no notices.


It is a text file, but the format is not well documented. For this reason we don't know how to add comment lines to the file itself. Thus, I've added a LICENSE file in rtai-lab/matlab/examples/

> rtai-lab/matlab/ptinfo.tlc has a copyright notice, but have no license notices.


I've checked with some developers who know that stuff better than me: that particular file is no more needed. I removed it.

> rtai-lab/matrixx/sa_utils.c refers to http://www.ni.com/legal/license, and that URL provides a proprietary software license. Could you check this?


You are right. After double-checking we are not sure that we can distribute and use it the way we used to. Luckily that portion of the code was likely completely bit rotten, as no one ever used it since at least ten years. We have thus decided to drop that part of code:
rm -rf rtai-lab/matrixx

Marco Morandini <mmforrtai>
Fri 14 Jul 2017 07:53:10 AM UTC, comment #25: 

Thank you!

> 3) added copyright notices for the images in doc/docbook/pictures
> 4) regenerated all the docs from sources
> 5) added a catch-all license file in doc (file "LICENSE.generated_files"). Hope it content is acceptable.


Yes, but probably you should add "specific licenses" to files like doc/generated/html/pictures/* (and possibly other files as well).

rtai-lab/matlab/examples/test.mdl still has no notices.

rtai-lab/matlab/ptinfo.tlc has a copyright notice,
but have no license notices.

rtai-lab/matrixx/sa_utils.c refers to
http://www.ni.com/legal/license, and that URL provides
a proprietary software license. Could you check this?

Ineiev <ineiev>
Site Administrator
Thu 13 Jul 2017 12:24:38 PM UTC, comment #24: 

New version:

https://home.aero.polimi.it/morandini/Downloads/rtai-5.1-mora3.tgz

Changes applied:

1) there must be a misunderstanding, rtai in under GPLv2+, with portions under LGPLv2+. We were erroneously distributing LGPLv3, now I've replaced it LGPLv2 (file "lgpl.txt")
2) removed the file "list": is should have not been there
3) added copyright notices for the images in doc/docbook/pictures
4) regenerated all the docs from sources
5) added a catch-all license file in doc (file "LICENSE.generated_files"). Hope it content is acceptable.
6) changed "open source" into "free software" (base/tasklets/README), and grepped for additional occurrences of "open source" (I found none)

Marco Morandini <mmforrtai>
Wed 05 Jul 2017 04:08:46 PM UTC, comment #23: 

Thank you!

All files should have valid copyright and license notices;
if any copyrightable files lack them, they are (and the package
as whole) proprietary (technically, at the very least).

If file format (like many image formats) makes it impractical
to put such notices in the file itself, a README file in
the same directory should provide copyright notices
and explain licensing terms for such files (ideally,
with a list of the licensed files).

This applies to doc/ and to list; please check other files as well.

Then, you say your package is GPLv3+, but your tarball doesn't
include the text of the GPLv3 (COPYING is the GPLv2, and
you also include the LGPLv3, which requires including the GPLv3
in the distribution; so you require that your users do something
you don't do yourselves).

Next point: Savannah is a hosting site for free software rather than open source. Please label your package as "free software" (base/tasklets/README).

Ineiev <ineiev>
Site Administrator
Tue 04 Jul 2017 08:42:12 AM UTC, comment #22: 

My fault, I was too sloppy in checking the result.

What about

https://home.aero.polimi.it/morandini/Downloads/rtai-5.1-mora2.tgz

?

Thank you again,

Marco

Marco Morandini <mmforrtai>
Wed 21 Jun 2017 12:23:12 PM UTC, comment #21: 

Thank you.

Files like addons/comedi/kcomedi.c have a wrong license notice,
it ends like


 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, write to the Free Software
 *
 */


All non-trivial files should have copyright and license notices; in your
rtai-5.1-mora.tgz, ChangeLog and some README files have none.

I feel I failed to explain this so far; do you think there is
something hard to understand about it?

Ineiev <ineiev>
Site Administrator
Wed 21 Jun 2017 09:22:44 AM UTC, comment #20: 

I tried to follow Paolo Mantegazza's advice, and uploaded a modified .tgz at

https://home.aero.polimi.it/morandini/Downloads/rtai-5.1-mora.tgz

Could you please have a look ? We are not sure to have got everything right.

Note that the files ending with .patch are still carrying a copyright notice with wrong FSF address in the diff. This is because these files come straight from a different (GPLed) project, and we would rather avoid messing with them.

Thank you for your time,

Marco

Marco Morandini <mmforrtai>
Wed 07 Jun 2017 01:07:48 PM UTC, comment #19: 
Ineiev <ineiev>
Site Administrator
Wed 07 Jun 2017 07:27:33 AM UTC, comment #18: 

After contacting those I could, for the scripts-makefiles, we have agreed to use the following copyright scheme:
"# Copyright (C) 2004 The RTAI project
# This [file] is free software; the RTAI project
# gives unlimited permission to copy and/or distribute it,
# with or without modifications, as long as this notice is preserved.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY, to the extent permitted by law; without
# even the implied warranty of MERCHANTABILITY or FITNESS FOR A
# PARTICULAR PURPOSE."
We are also aware of the suggestions in:
https://www.gnu.org/licenses/license-list.en.html#informal,
but do not think they invalidate what we would like to use.
Does what above meet your request?

Paolo Mantegazza <mante>
Mon 05 Jun 2017 01:28:41 PM UTC, comment #17: 

Savannah has no issue with waiting until the package is fixed.

Ineiev <ineiev>
Site Administrator
Mon 05 Jun 2017 09:42:32 AM UTC, comment #16: 

I confirm what Marco said: right now I've no intention to withdraw. The satisfaction of you requests will take time though. As it has been said, I've to contact the originator(s) of the makefiles and know what they want me to do. It is not said I will succeed, as the all stuff goes back to more then 10 years. The only request I can fulfill at once it the addition of the (L)GPL licence(s) (which are already in the CVS we want to recover) and the change of the outdated links.

Paolo Mantegazza <mante>
Sun 04 Jun 2017 04:54:21 PM UTC, comment #15: 


> 5)  I don't think that Paolo, or anyone else, can commit to change the already distributed tarball. Nor to change the old, pre-existing CVS history, where the first commit of some files was without copyright notices.


I don't think it's necessary to rewrite history in such ways.  It would be sufficient if you got the most recent working copy, make those changes and produce a new tarball from it.  When you import VCS history, those changes will result in a single commit, so it shouldn't be very complicated.

If this is still too hard for you, you'll have to speak to gnu-advisory or rms about an exception; Savannah admins are not in the position to make such decisions.

Ineiev <ineiev>
Site Administrator
Sat 03 Jun 2017 12:47:32 PM UTC, comment #14: 

I made an account here to comment on this, because I fear that there is some misunderstanding, possibly on both sides. What I'm writing do represent my point of view.

1) RTAI is an old project, whose CVS was hosted on GNA. It is led by Paolo Mantegazza.
Unfortunately, due to sloppiness on our side, we did not plan to move everything before GNA's shutdown. Thus, we lost some CVS history, because the last GNA backup predates some of our commits.
2) We do strongly believe in free software.
3) RTAI has an own website ( www.rtai.org ) and mailing list, and the tarballs are distributed from there.
4) We understand the requirement to add Copyright notice to the CVS files. I read http://savannah.gnu.org/maintenance/ValidNotices/ and understand the motivations behind it. We can commit ourselves to do that, in the CVS, for all the files for which we can trace back the contributors. As a matter of fact, RTAI does not ask nor copyright assignment, especially for files that were not considered (By RTAI's developers) as "code", such as Makfiles. Thus, I'm not 100% sure we can correctly list all the copyright holders, especially because for some years RTAI was developed without a CVS.
5) I don't think that Paolo, or anyone else, can commit to change the already distributed tarball. Nor to change the old, pre-existing CVS history, where the first commit of some files was without copyright notices.
6) Right now we would rather avoid messing with the backup CVS copy we have. We think it would be much safer to have a new CVS home, import there the CVS history, and then add the missing Copyright notices and the License file.
7) Of course, as you say, RTAI could go elsewhere. However, we (I) would prefer to stay here, or on a service like this. Mainly because we dislike the motivations behind alternatives such as github, sourceforge, and the likes. Not to talk about the confusion between free and open source software they do not help to clarify. As Paolo wrote in the initial project submission, "we have decided to pass it over to SAVANNAH, because they have so much in common, especially in relation to what free software is ..."

Marco Morandini <mmforrtai>
Fri 02 Jun 2017 08:24:52 AM UTC, comment #13: 
  • Yes, this is a common requirement.
  • In carbonkernel Makefile.am is likely to be uncopyrightable (it's effectively less than 10 lines long); for longer files, please file a bug report in their tracker.  Likewise for other projects.
  • One can get an impression that you aren't really willing to follow our policies.  In such cases it may be better for all if people withdraw their registration.
Ineiev <ineiev>
Site Administrator
Thu 01 Jun 2017 10:02:34 PM UTC, comment #12: 

Pavel (Kharitonov) wrote: "Now, all files, including documentation, makefiles and configure scripts, should have valid license and copyright notices --- not just .c and .py files."
Is that savannah's rule for everybody? My questions comes out just because I'm curious. So I searched savannah for "real time". A first instance, I was acquainted with, has been "carbonkernel", whose first member is Loic Dachary, a name you should be acquainted with, I suppose.
Well, just by way of example, if you browse that project sources repository and check Makefile.am, Makefile.in, configure.in and configure, at the very beginning, you'll see that they are the same as in RTAI. In particular Makefile.am and configure.in have no copyright-licence mentioned while Makefile.in and configure have one (likely as generated by autotools).
Since in "carbonkernel" Makefile.am and configure.in have no copyright-licence mentioned, it seems your theorem above either fails.
If you surf a few other projects you'll see that "carbonkernel" is not the only instance.

Paolo Mantegazza <mante>
Thu 01 Jun 2017 08:18:54 PM UTC, comment #11: 

Does that mean that I've to revise the whole tarball I sent you, to put into it the copyrights everywhere, plus the GPL licences, and resend it to you, for a fresh restart?

Paolo Mantegazza <mante>
Thu 01 Jun 2017 05:42:40 PM UTC, comment #10: 

If you are not a VCS expert, Savannah admins will help you; however, I'm afraid having necessary texts in place is a prerequisite for registering on Savannah.

Ineiev <ineiev>
Site Administrator
Thu 01 Jun 2017 12:28:25 PM UTC, comment #9: 

Pavel (Kharitonov) said: " ... you'll get a write access to its newly created repositories, so you'll be free to import any history as you wish."
I'm not a real expert of this stuff, so I'll trust you.
In the same way, could you please trust me for a while, allowing me to add a recent (L)GPL licence, the missing copyrights and fix the existing ones once the seminal CVS has been set up and I can begin my alignment procedure?

Paolo Mantegazza <mante>
Thu 01 Jun 2017 12:06:29 PM UTC, comment #8: 

Thank you, this tarball is just what I wanted.

When your package is registered on Savannah, you'll get a write access to its newly created repositories, so you'll be free to import any history as you wish.

Now, all files, including documentation, makefiles and configure scripts, should have valid license and copyright notices --- not just .c and .py files.

I also couldn't find the text of the GPL in your tarball.

At last, the postal address of the FSF used in your notices is outdated, I recommend use the URL like GPL HowTo suggests.

Could you fix that?

Ineiev <ineiev>
Site Administrator
Thu 01 Jun 2017 11:15:33 AM UTC, comment #7: 

If I got you correctly, here it is:
https://www.rtai.org/userfiles/downloads/RTAI/rtai-5.0.1.tar.bz2.
Take into account that what above is just the "product" we distribute to our end users. Nevertheless, please, since we want to recover the whole project, including its all history, we would like to have the image of our CVS I sent you reinstated, as it was at GNA at the end of the last February.

Paolo Mantegazza <mante>
Thu 01 Jun 2017 10:41:44 AM UTC, comment #6: 

This looks like a repository. Could you provide a link to a distribution tarball, without any VCS files?

Ineiev <ineiev>
Site Administrator
Thu 01 Jun 2017 09:20:06 AM UTC, comment #5: 
Paolo Mantegazza <mante>
Thu 01 Jun 2017 08:58:32 AM UTC, comment #4: 

Please post a link to your tarball.

Ineiev <ineiev>
Site Administrator
Thu 01 Jun 2017 07:56:10 AM UTC, comment #3: 

As a further remark you shoudl take into account that my bz2 compressed tarball is about 57 MB (far in excess of likely accepatable 16?)

Paolo Mantegazza <mante>
Thu 01 Jun 2017 07:52:23 AM UTC, comment #2: 

Sorry if I was not clear enough. I took for granted that, since It is a Linux extension, using kernel patches, Linux is required. Clearly along to the GNU making suite (automake, gcc, etc. etc.)
As for the name, I'm sorry but it has to remain so. I do not want to confuse anybody by changing a name that has been well established since the very beginnig of 2000. Take into account the it was so at www.gna.org already (since 2004) and there is also a home site at: www.rtai.org.
In relation to my tarball I must admit that I thought to have uploaded it within my hosting request, but you are now telling me I failed. Is it possible that my failure is due to the use of a bz2 tarball? I assumed that the request of a ".tar.gz" was just by way of suggestion. Please let me know if it the gz extension is compulsory and what I should do to successfully upload it again.

Paolo Mantegazza <mante>
Wed 31 May 2017 03:35:55 PM UTC, comment #1: 

I'm confused a little bit: it says that it's a Linux extension, but no software is needed; does it really run on bare hardware without any other software?

Then, I can't download your tarball.

At last, we discourage too short abbreviated names for new packages; could you suggest a longer one?

Ineiev <ineiev>
Site Administrator
Tue 30 May 2017 12:40:02 PM UTC, original submission:  

A new project has been registered at Savannah
This project account will remain inactive until a site admin approves or discards the registration.


Registration Administration


While this item will be useful to track the registration process, approving or discarding the registration must be done using the specific Group Administration page, accessible only to site administrators, effectively logged as site administrators (superuser):



Registration Details


  • Name: Real Time Application Interface (for Linux)
  • System Name:  rtai
  • Type: non-GNU software and documentation
  • License: GNU General Public License v2 or later (None needed)





Description:

Since around year 2000, RTAI has been a hard real time extension for Linux, interoperating a distributed kernel and user space supports. It is based on a dual kernel approach, nowadays exploiting its own modified "ipipe" kernel patches. Within such a framework it provides a large set of interprocess communication and synchronization APIs, including sem-mutexes, interprocess messages, (typed)mailboxes, msg queues, shared memory and a limited posix; all of them usable in a distributed way (through netrpc). It has an own application (RTAILab), which runs controllers whose code is generated through Matlab-Symulink-RealTimeWorkshop and ScicosLab.


Other Software Required:

None needed


Other Comments:

The attached file should be the complete original image, including its all history, of the rtai cvs, as it had been hosted at gna.org, since the late 2004. Following the loss of GNA support we have decided to pass it over to SAVANNAH, because they have so much in common, especially in relation to what free software is, hopefully including how I could mange the project. Another important point is the availability of CVS, which should make the transition faster. Nonetheless, once the transition has been verified to be stabilized and well working, it is likely that we will pass to SVN. Anticipated thanks for your support, hoping we will be accepted and everything will go well.


Tarball URL:

/home/mante/Downloads/TMP/cvs/rtai.tar.bz2


Paolo Mantegazza <mante>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by mmforrtai (Posted a comment)
  • -email is unavailable- added by ineiev (Posted a comment)
  • -email is unavailable- added by mante (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only logged-in users can vote.

     

    Follow 4 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2017-11-20 ineiev StatusIn Progress Done
        Open/ClosedOpen Closed
    2017-05-31 ineiev StatusNone In Progress
        Assigned toNone ineiev

    Back to the top

    Powered by Savane 3.13-f8d8.
    Corresponding source code