crypt and MD5 - still not wanted

Started by Karel Zakover 25 years ago8 messages
#1Karel Zak
zakkr@zf.jcu.cz

Befor one year ago, I ask if MD5 (free code from Debian) and DES crypt()
is possible include to PG.

Knows anyone if is possible include it to PG now?

For example GNU Debian distribute MD5 and crypt() in free software section
without some restriction.

BTW --- If I good keep track situation on this branch, USA some time ago
change some restriction for this matter. Or not?

Karel

#2Tom Lane
tgl@sss.pgh.pa.us
In reply to: Karel Zak (#1)
Re: crypt and MD5 - still not wanted

If I recall the prior discussion, MD5 is OK, crypt is still risky,
because MD5 is not an encryption algorithm so it doesn't fall under
the US export laws.

I believe Vince V. is working on improving the password challenge
code to use MD5, btw.

regards, tom lane

#3Vince Vielhaber
vev@michvhf.com
In reply to: Tom Lane (#2)
Re: crypt and MD5 - still not wanted

On Fri, 7 Jul 2000, Tom Lane wrote:

If I recall the prior discussion, MD5 is OK, crypt is still risky,
because MD5 is not an encryption algorithm so it doesn't fall under
the US export laws.

I believe Vince V. is working on improving the password challenge
code to use MD5, btw.

yep.

Vince.
--
==========================================================================
Vince Vielhaber -- KA8CSH email: vev@michvhf.com http://www.pop4.net
128K ISDN from $22.00/mo - 56K Dialup from $16.00/mo at Pop4 Networking
Online Campground Directory http://www.camping-usa.com
Online Giftshop Superstore http://www.cloudninegifts.com
==========================================================================

#4Karel Zak
zakkr@zf.jcu.cz
In reply to: Tom Lane (#2)
Re: crypt and MD5 - still not wanted

On Fri, 7 Jul 2000, Tom Lane wrote:

If I recall the prior discussion, MD5 is OK, crypt is still risky,
because MD5 is not an encryption algorithm so it doesn't fall under
the US export laws.

I believe Vince V. is working on improving the password challenge
code to use MD5, btw.

Not only passwords, but standard SQL functions a my drean is aggregate
function md5count() too. Cool --- that is MD5 OK.

Karel

#5Peter Eisentraut
peter_e@gmx.net
In reply to: Vince Vielhaber (#3)
Re: crypt and MD5 - still not wanted

Vince Vielhaber writes:

On Fri, 7 Jul 2000, Tom Lane wrote:

If I recall the prior discussion, MD5 is OK, crypt is still risky,
because MD5 is not an encryption algorithm so it doesn't fall under
the US export laws.

I believe Vince V. is working on improving the password challenge
code to use MD5, btw.

yep.

If you do that, maybe also look at the secondary password files. We
probably don't want those using a different encryption method.

(backward compatibility alarm goes off in the distance...)

--
Peter Eisentraut Sernanders v�g 10:115
peter_e@gmx.net 75262 Uppsala
http://yi.org/peter-e/ Sweden

#6Vince Vielhaber
vev@michvhf.com
In reply to: Peter Eisentraut (#5)
Re: crypt and MD5 - still not wanted

On Sat, 8 Jul 2000, Peter Eisentraut wrote:

Vince Vielhaber writes:

On Fri, 7 Jul 2000, Tom Lane wrote:

If I recall the prior discussion, MD5 is OK, crypt is still risky,
because MD5 is not an encryption algorithm so it doesn't fall under
the US export laws.

I believe Vince V. is working on improving the password challenge
code to use MD5, btw.

yep.

If you do that, maybe also look at the secondary password files. We
probably don't want those using a different encryption method.

(backward compatibility alarm goes off in the distance...)

Already thinking about that.

Vince.
--
==========================================================================
Vince Vielhaber -- KA8CSH email: vev@michvhf.com http://www.pop4.net
128K ISDN from $22.00/mo - 56K Dialup from $16.00/mo at Pop4 Networking
Online Campground Directory http://www.camping-usa.com
Online Giftshop Superstore http://www.cloudninegifts.com
==========================================================================

#7Tom Lane
tgl@sss.pgh.pa.us
In reply to: Karel Zak (#4)
Re: crypt and MD5 - still not wanted

Karel Zak <zakkr@zf.jcu.cz> writes:

Not only passwords, but standard SQL functions a my drean is aggregate
function md5count() too. Cool --- that is MD5 OK.

Er ... what? What would an "aggregate function md5count()" do?

Bear in mind that an aggregate function is useless if its result
depends on the order of its inputs ...

regards, tom lane

#8Karel Zak
zakkr@zf.jcu.cz
In reply to: Tom Lane (#7)
Re: crypt and MD5 - still not wanted

Karel Zak <zakkr@zf.jcu.cz> writes:

Not only passwords, but standard SQL functions a my drean is aggregate
function md5count() too. Cool --- that is MD5 OK.

Er ... what? What would an "aggregate function md5count()" do?

Count md5 sum from defined rows.

Bear in mind that an aggregate function is useless if its result
depends on the order of its inputs ...

Hmm, order is a problem in this idea (I not think of this) :-(

But, I mean that it is not total idiotism, make it is a differend
thing...

Well, I take back it.
Karel