|
|
| Re: .sav format [message #52362 is a reply to message #52298] |
Tue, 30 January 2007 09:58   |
JD Smith
Messages: 850
Registered: December 1999
|
Senior Member |
|
|
On Tue, 30 Jan 2007 09:40:51 -0800, mgalloy@gmail.com wrote:
> On Jan 30, 6:21 am, "Haje Korth" <haje.ko...@nospam.jhuapl.edu> wrote:
>> Here is what I don't understand. In the first part of the notice that is
>> embedded in the sav files says that ANY reverse engineering is prohibited.
>> It does not distinguish between data and code elements. Thus it is my
>> understanding that nobody will ever be legally allowed to develop any
>> routine that messes with the sav file since it would automatically mean that
>> you would have to apply some sort of reverse engineering magic. The second
>> part of the notice says that I need a license for any non-RSI software that
>> reads/writes sav files. Doesn't that mean that anyone applying for such
>> license automatically incriminates himself/herself???
>
> I don't understand that either. Maybe apply for the license before
> doing the work? (I'm not a lawyer either.)
I think it was just a way to "grandfather in" Craig's work, which
occurred before they had such an explicit policy.
If you, like me, dislike closed and protected data formats (by the
DCMA, no less!), do let ITTVIS know. I mean, even Microsoft is being
forced to consider "open" XML formats for Word files, by government
agencies who refuse to be locked into data types over which they have
no control.
The ability to share data and use it in a variety of different ways is
fundamental to the progress of science. I use SAV files personally
since they are so convenient, but I will not distribute them, and I
always assume they have little long-term archival value. That, to me,
diminishes the value of IDL and its default file format. Luckily,
there are many other types of open, standard formats to choose from
(FITS, HDF, etc.). None offer the convenient mapping to IDL
variables/pointers/etc., but at least they can be read back in in 20
years.
JD
|
|
|
|
|
|
| Re: .sav format [message #52371 is a reply to message #52298] |
Tue, 30 January 2007 07:36 |
Paolo Grigis
Messages: 171
Registered: December 2003
|
Senior Member |
|
|
mgalloy@gmail.com wrote:
> Interesting. More poking around at Craig's site, reveals the following
> license for CMSVLIB:
>
> http://cow.physics.wisc.edu/~craigm/idl/down/LICENSE.RSI
>
> In particular, note point 5:
>
> ; 5. Allowed use of this software is limited to reading and writing
> ; IDL variable related portions of IDL Save files. It may not be
> ; used as a basis for reverse engineering, or otherwise
> ; accessing any other portions of an IDL save file, including but
> ; not limited to, those portions that encode executable IDL
> programs.
> ; Such use is in violation of the IDL EULA, and will be prosecuted
> ; to the fullest extent possible by Research Systems, Inc. It is
> ; permissible to read such sections of an IDL save file for the
> ; sole purpose of transferring it without examination or
> interpretation
> ; to another save file.
>
> So it appears you would need a similar license from ITTVIS for
> examining data .sav files (i.e. directly examining, there is an
> IDL_Savefile class in IDL 6.1 that can do a lot of this for data .sav
> files anyway) and it is strictly forbidden to examine code .sav files.
Do we also need a license for doing science? After all we are trying
to reverse-engineer the universe... let's hope the copyright expired
100 years after Galileo's death ;-)
Ciao,
Paolo
>
> Mike
> --
> www.michaelgalloy.com
>
|
|
|
|
| Re: .sav format [message #52377 is a reply to message #52298] |
Tue, 30 January 2007 05:21 |
Haje Korth
Messages: 651
Registered: May 1997
|
Senior Member |
|
|
Mike,
Here is what I don't understand. In the first part of the notice that is
embedded in the sav files says that ANY reverse engineering is prohibited.
It does not distinguish between data and code elements. Thus it is my
understanding that nobody will ever be legally allowed to develop any
routine that messes with the sav file since it would automatically mean that
you would have to apply some sort of reverse engineering magic. The second
part of the notice says that I need a license for any non-RSI software that
reads/writes sav files. Doesn't that mean that anyone applying for such
license automatically incriminates himself/herself???
I am a scientist not a lawyer (if I where a lawyer I probably would not use
IDL as my tool of the trade) and I don't really understand this legal mumbo
jumbo (nor will I ever intend to understand it). Therefore I stay away from
using sav files and use any of the other numerous file formats that are
cross plattform, cross program compatible.
I wonder why Christopher was set on sav files and needed the docs? I guess
he never explained, so I may be overlooking something.
Haje
<mgalloy@gmail.com> wrote in message
news:1170101297.922961.279850@q2g2000cwa.googlegroups.com...
> Interesting. More poking around at Craig's site, reveals the following
> license for CMSVLIB:
>
> http://cow.physics.wisc.edu/~craigm/idl/down/LICENSE.RSI
>
> In particular, note point 5:
>
> ; 5. Allowed use of this software is limited to reading and writing
> ; IDL variable related portions of IDL Save files. It may not be
> ; used as a basis for reverse engineering, or otherwise
> ; accessing any other portions of an IDL save file, including but
> ; not limited to, those portions that encode executable IDL
> programs.
> ; Such use is in violation of the IDL EULA, and will be prosecuted
> ; to the fullest extent possible by Research Systems, Inc. It is
> ; permissible to read such sections of an IDL save file for the
> ; sole purpose of transferring it without examination or
> interpretation
> ; to another save file.
>
> So it appears you would need a similar license from ITTVIS for
> examining data .sav files (i.e. directly examining, there is an
> IDL_Savefile class in IDL 6.1 that can do a lot of this for data .sav
> files anyway) and it is strictly forbidden to examine code .sav files.
>
> Mike
> --
> www.michaelgalloy.com
>
|
|
|
|
|
|
| Re: .sav format [message #52388 is a reply to message #52298] |
Mon, 29 January 2007 12:49 |
JD Smith
Messages: 850
Registered: December 1999
|
Senior Member |
|
|
On Mon, 29 Jan 2007 12:08:18 -0800, mgalloy@gmail.com wrote:
> Interesting. More poking around at Craig's site, reveals the following
> license for CMSVLIB:
>
> http://cow.physics.wisc.edu/~craigm/idl/down/LICENSE.RSI
>
> In particular, note point 5:
>
> ; 5. Allowed use of this software is limited to reading and writing ;
> IDL variable related portions of IDL Save files. It may not be ; used
> as a basis for reverse engineering, or otherwise ; accessing any other
> portions of an IDL save file, including but ; not limited to, those
> portions that encode executable IDL programs.
> ; Such use is in violation of the IDL EULA, and will be prosecuted ;
> to the fullest extent possible by Research Systems, Inc. It is ;
> permissible to read such sections of an IDL save file for the ; sole
> purpose of transferring it without examination or interpretation
> ; to another save file.
>
> So it appears you would need a similar license from ITTVIS for examining
> data .sav files (i.e. directly examining, there is an IDL_Savefile class
> in IDL 6.1 that can do a lot of this for data .sav files anyway) and it is
> strictly forbidden to examine code .sav files.
The license grants "the author of this software, and to all IDL users,
a license to use and redistribute this software in source or binary
form, subject to the following conditions", so they've already given
us the license (restrictive though it may be).
JD
|
|
|
|
| Re: .sav format [message #52490 is a reply to message #52298] |
Wed, 07 February 2007 05:10 |
Haje Korth
Messages: 651
Registered: May 1997
|
Senior Member |
|
|
JD,
That depends on what functionality your code relies on. The VM has some
restrictions such as the EXECUTE function not working. If your code is not
affected by the restrictions and the VM splash screen does not represent an
eye sore to you, the VM has indeed replaced embedded or runtime license. I
personally do not know anyone working with embedded licensing.
Cheers,
Haje
"JD Smith" <jdsmith@as.arizona.edu> wrote in message
news:pan.2007.02.06.23.45.39.837420@as.arizona.edu...
> On Mon, 05 Feb 2007 18:02:36 -0500, Haje Korth wrote:
>
>> What I was referring to is Embedded licensing, which is produced if you
>> choose "Licensed IDL sav file" in Project options. You need the
>> development
>> kit license for this (I don't have one so for me the option is greyed
>> out).
>> For example, demo.sav has an embedded IDL license. It will run as long as
>> you want (no demo mode timer) without presence of any IDL license.
>>
>> Embedded licensing has the advantage that you can provide your customer
>> with
>> a product without them having to buy a runtime license. The runtime
>> license
>> is embedded in the sav file itself together with your code. Hope this
>> makes
>> more sense.
>
> Aha, thanks. The IDLVM has pretty well killed that business model, yes?
>
> JD
|
|
|
|
|
|
| Re: .sav format [message #52497 is a reply to message #52298] |
Tue, 06 February 2007 15:44 |
JD Smith
Messages: 850
Registered: December 1999
|
Senior Member |
|
|
On Tue, 06 Feb 2007 15:00:47 -0500, Haje Korth wrote:
> "Guillermo" <gcastill@ucalgary.ca> wrote in message
> news:1170788649.657637.173590@l53g2000cwa.googlegroups.com.. .
>> On Feb 6, 8:42 am, David Fanning <d...@dfanning.com> wrote:
>>> I think the more obvious reason is that ENVI consists
>>> of IDL save files. :-)
>>
>> Hold on! Excuse my ignorance, but do this thread and the thing of the
>> 'reverse engineering' mean that anyone having the right tools (eg. the
>> company proprietary of the language) can retrieve the source code from
>> a code.sav file?? Sounds scary...
> Sure the idl code can be recovered, just like you can disassemble any
> executable file. You won't recover the variables names but you can see the
> logic of the code. The question is more whether its worth the effort. I
> would be curious in seeing how it's done but asking myself whether I would
> waste my time on it, the answer is absolutely NO. What's the sense? Even the
> ENVI routines are in the end just implementations of publicly available
> algorithms.
I wouldn't be so sure. Long ago I used to run IDLSPEC, an IDL
benchmark site. To ensure people didn't "game the system" I added a
little checksum to the data, and distributed the routine in a .sav to
prevent peeking. I got an email one day with this section of the code
as I had written it, line by line (well, different capitalization and
block elements, but anyway). It's in there. Variable names, function
calls, etc.
Does anyone aside from ITTVIS use .SAV as some form of proprietary
binary distribution channel? It's sad that all this protectiveness of
their "compiled binary" format may limit the utility of .SAV as a binary
data format.
JD
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Re: .sav format [message #52530 is a reply to message #52298] |
Mon, 05 February 2007 10:27 |
Haje Korth
Messages: 651
Registered: May 1997
|
Senior Member |
|
|
JD,
one reason for guarding the format is obvious. The sav files allow an IDL
license to be embedded. If one were to figure out how this is done, one
could probably extract that license and inject it into any other sav files.
I don't use IDL run time licensing so I don't see much use in this. However
such action would clearly illegal and would warrant guarding the format. If
the variables were stored in a separate file format than code+license,
reverse engineering the variable save files wouldn't be such a delicate
issue. My guess is that storing variable data was not the primary reason for
creating the sav file format.
Cheers,
Haje
"JD Smith" <jdsmith@as.arizona.edu> wrote in message
news:pan.2007.02.05.17.58.05.570514@as.arizona.edu...
> On Sun, 04 Feb 2007 02:04:09 -0500, Craig Markwardt wrote:
>
>>
>> "mgalloy@gmail.com" <mgalloy@gmail.com> writes:
>>> On Jan 29, 1:49 pm, JD Smith <jdsm...@as.arizona.edu> wrote:
>>>> The license grants "the author of this software, and to all IDL users,
>>>> a license to use and redistribute this software in source or binary
>>>> form, subject to the following conditions", so they've already given
>>>> us the license (restrictive though it may be).
>>>
>>> To "use and redistribute this software" (i.e. CMSVLIB), but the
>>> original poster wanted documentation of the .sav format (presumably so
>>> they could write their own software).
>>
>> The documentation of the format is unencumbered by any license other
>> than Creative Commons.
>
> The "license" does seem to indicate that *anyone* needs permission to
> create a tool to read .SAV files, whether or not they have agreed to an
> EULA:
>
> "Non-RSI supplied software that reads or writes files in the IDL
> Save/Restore format must have a license from Research Systems
> explicitly granting the right to do so. In this case, the license
> will be included with the software for your inspection. Please
> report software that does not have such a license to
> Research Systems, Inc."
>
> They haven't yet wielded the DMCA hammer, but I don't doubt that they
> would if they felt it necessary. There is a reverse engineering clause
> in the DMCA, but I'm not sure to what degree it applies:
>
> http://cyber.law.harvard.edu/openlaw/DVD/1201.html#f
>
> I for once don't see the point in so jealously guarding a format: it's a
> perfect way to ensure it never sees wide use.
>
> JD
>
|
|
|
|
| Re: .sav format [message #52531 is a reply to message #52298] |
Mon, 05 February 2007 09:58 |
JD Smith
Messages: 850
Registered: December 1999
|
Senior Member |
|
|
On Sun, 04 Feb 2007 02:04:09 -0500, Craig Markwardt wrote:
>
> "mgalloy@gmail.com" <mgalloy@gmail.com> writes:
>> On Jan 29, 1:49 pm, JD Smith <jdsm...@as.arizona.edu> wrote:
>>> The license grants "the author of this software, and to all IDL users,
>>> a license to use and redistribute this software in source or binary
>>> form, subject to the following conditions", so they've already given
>>> us the license (restrictive though it may be).
>>
>> To "use and redistribute this software" (i.e. CMSVLIB), but the
>> original poster wanted documentation of the .sav format (presumably so
>> they could write their own software).
>
> The documentation of the format is unencumbered by any license other
> than Creative Commons.
The "license" does seem to indicate that *anyone* needs permission to
create a tool to read .SAV files, whether or not they have agreed to an
EULA:
"Non-RSI supplied software that reads or writes files in the IDL
Save/Restore format must have a license from Research Systems
explicitly granting the right to do so. In this case, the license
will be included with the software for your inspection. Please
report software that does not have such a license to
Research Systems, Inc."
They haven't yet wielded the DMCA hammer, but I don't doubt that they
would if they felt it necessary. There is a reverse engineering clause
in the DMCA, but I'm not sure to what degree it applies:
http://cyber.law.harvard.edu/openlaw/DVD/1201.html#f
I for once don't see the point in so jealously guarding a format: it's a
perfect way to ensure it never sees wide use.
JD
|
|
|
|
| Re: .sav format [message #52539 is a reply to message #52386] |
Sat, 03 February 2007 23:04 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
"mgalloy@gmail.com" <mgalloy@gmail.com> writes:
> On Jan 29, 1:49 pm, JD Smith <jdsm...@as.arizona.edu> wrote:
>> The license grants "the author of this software, and to all IDL users,
>> a license to use and redistribute this software in source or binary
>> form, subject to the following conditions", so they've already given
>> us the license (restrictive though it may be).
>
> To "use and redistribute this software" (i.e. CMSVLIB), but the
> original poster wanted documentation of the .sav format (presumably so
> they could write their own software).
The documentation of the format is unencumbered by any license other
than Creative Commons.
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
| Re: .sav format [message #52578 is a reply to message #52503] |
Sun, 11 February 2007 13:53 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
"Haje Korth" <haje.korth@nospam.jhuapl.edu> writes:
> Sure the idl code can be recovered, just like you can disassemble any
> executable file. You won't recover the variables names but you can see the
> logic of the code. The question is more whether its worth the effort. I
> would be curious in seeing how it's done but asking myself whether I would
> waste my time on it, the answer is absolutely NO. What's the sense? Even the
> ENVI routines are in the end just implementations of publicly available
> algorithms.
It can be done. I did it. I had a legit reason at the time, which
was to recover some routines whose source I had lost... but then it
became a challenge in its own right. The format was not particularly
difficult to decipher: it was almost a direct translation of IDL
structure into a byte-code format, with special records that described
variables, commons, etc. However, since then, I am led to believe
that RSI has encrypted save-files with code (although I haven't
checked).
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
| Re: .sav format [message #52579 is a reply to message #52531] |
Sun, 11 February 2007 13:47 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
JD Smith <jdsmith@as.arizona.edu> writes:
> On Sun, 04 Feb 2007 02:04:09 -0500, Craig Markwardt wrote:
>
>>
>> "mgalloy@gmail.com" <mgalloy@gmail.com> writes:
>>> On Jan 29, 1:49 pm, JD Smith <jdsm...@as.arizona.edu> wrote:
>>>> The license grants "the author of this software, and to all IDL users,
>>>> a license to use and redistribute this software in source or binary
>>>> form, subject to the following conditions", so they've already given
>>>> us the license (restrictive though it may be).
>>>
>>> To "use and redistribute this software" (i.e. CMSVLIB), but the
>>> original poster wanted documentation of the .sav format (presumably so
>>> they could write their own software).
>>
>> The documentation of the format is unencumbered by any license other
>> than Creative Commons.
>
> The "license" does seem to indicate that *anyone* needs permission to
> create a tool to read .SAV files, whether or not they have agreed to an
> EULA:
>
> "Non-RSI supplied software that reads or writes files in the IDL
> Save/Restore format must have a license from Research Systems
> explicitly granting the right to do so. In this case, the license
> will be included with the software for your inspection. Please
> report software that does not have such a license to
> Research Systems, Inc."
J.D., RSI can claim whatever it wants, but legally they have no
control over users who do not hold an IDL license.
> They haven't yet wielded the DMCA hammer, but I don't doubt that they
> would if they felt it necessary. There is a reverse engineering clause
> in the DMCA, but I'm not sure to what degree it applies:
Since IDL save files (with data only) are not encrypted or scrambled
in any way, they are not covered by the anti-circumvention provisions
of the DMCA. Furthermore, since the data is (presumably) *your own*,
you have your own authority to grant permission to extract the
information from the file.
> I for once don't see the point in so jealously guarding a format: it's a
> perfect way to ensure it never sees wide use.
A good point!
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
