Re: encrypted SPSS .sav files

[Bastián Díaz]

I think it's great. In this case, PSPP would complete interoperability with IBM SPSS native files.
Alternatively, if encryption is not just unsafe. Should we contact IBM to fix it?

What Michael said it makes sense, however, I believe that interoperability with SPSS not be broken, however as mentioned sometime in PSPP-users, it would be interesting to PSPP had a own file format without the artificial constraints imposed by IBM. (but that's additional time and effort on the part of the developers)

Regards

--
Bastián Díaz

---

De: Ben Pfaff <address@hidden>
Para: Michał Dubrawski <address@hidden>
CC: "address@hidden" <address@hidden>
Enviado: Miércoles, 30 de octubre, 2013 15:06:02
Asunto: Re: encrypted SPSS .sav files

I've had that suggestion a couple of times now.  It is a reasonable thought.
I will consider it.

On Wed, Oct 30, 2013 at 9:41 AM, Michał Dubrawski
<address@hidden> wrote:
> Dear Ben,
>
> That's great news :-) I'm really impressed. Reading your emails about what
> you have found out abut encryption and how was for me like reading very
> interesting detective story.
> This thought came to me: if SPSS encryption is so poor, maybe we should
> think about better encryption as additional option (working only on PSPP).
> It would be an option only for PSPP users, but users of previous versions of
> SPSS also cannot read encrypted files with encryption designed by SPSS. The
> difference is (if they have a password) they can always get PSPP, open it
> and save as non-encrypted file readable for any SPSS version (if they insist
> on using it ;-) ).
>
> warm regards,
> Michal
>
>>
>> Date: Tue, 29 Oct 2013 22:55:13 -0700
>> From: Ben Pfaff <address@hidden>
>> To: Basti?n D?az <address@hidden>
>> Cc: "address@hidden" <address@hidden>
>> Subject: Re: encrypted SPSS .sav files
>> Message-ID: <address@hidden>
>> Content-Type: text/plain; charset=us-ascii
>>
>>
>> I've now written code that reliably reads an encrypted .sav file, given
>> the password.  I'm concerned about poor crypto design, so I'm going to
>> ask people who are more expert than me to evaluate the design defects
>> before I publish the code.
>>
>> I've also written a program to decode an encoded password to obtain the
>> human readable password.  (This is much less interesting, but still
>> useful for interoperability.)
>>
>>
>>
>> ------------------------------
>>
>>
>> _______________________________________________
>> Pspp-users mailing list
>> address@hidden
>> https://lists.gnu.org/mailman/listinfo/pspp-users
>>
>>
>> End of Pspp-users Digest, Vol 89, Issue 28
>> ******************************************
>
>

_______________________________________________
Pspp-users mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/pspp-users