[comp.sys.next] NeXT Software Distribution

sagar@psu-cs.UUCP (Arun Sagar) (03/06/89)

	Hi
	  I haven't used a NeXT yet, so this might be worth <$.02, but... :-)

One of the objections/hassles people seem to be obsessed with is the software
distribution for the machine.  Since the machine has an optical read/write
mechanism, can't it be configured to read CD-ROM discs? Since they cost
<$5.00 bucks(well, reasonable quantities) to mass produce, that should solve
the price problem.  Of course, with mega-bytes of space left over, the publisher/developer could put in 24-bit antialiased copies of their manuals on it.
Of course, small-time people like us wouldn't be able to distribute our programs
through this media because of prohibitive cd-mastering costs, but at least 
significant commercial software could be made available cheaply.

	Problems???????

mlee@awamore.cs.cornell.edu (mark lee) (10/02/89)

To: comp.sys.next
cc: 


  In an article in NeXTWeek (inside MacWeek 9/19/89), it was mentioned
  that NeXT users may be able to buy a software over the phone:

> 	  A demonstration version of the program (FrameMaker 2.0)  
>	with the save feature disabled will be included 
>	on Version 1.0 of the system software disk 
>	from NeXT.  Users can call Businessland or 
>	Computer Craft to buy the program and get 
>	documentation and a password to enable saving...
 
  My questions are:

  1. If there is a password to enable saving, how will FrameMaker
     keep NeXT users from telling other users the password?

  2. How does the new software distribution system proposed by
     Steve Jobs work?  Will all the programs and data be encrypted 
     so that a software key is necessary to decrypt them?

     If this is the case, how does NeXT plan on stopping people
     from giving other users the software key?

  Any information on these two questions would be greatly appreciated.

Mark Lee, 
mlee@awamore.cs.cornell.edu

jans@tekgvs.LABS.TEK.COM (Jan Steinman) (10/03/89)

<A demonstration version of the program (FrameMaker 2.0) with the save feature 
disabled will be included on Version 1.0 of the system software disk...  If 
there is a password to enable saving, how will FrameMaker keep NeXT users from 
telling other users the password?>

The password is a (non-trivial) function of the machine's ethernet address, 
which is supposedly unique for every machine in the universe.  Therefore, one 
must have a password for each machine such a program is to run on.

							   Jan Steinman - N7JDB
						  Electronic Systems Laboratory
					Box 500, MS 50-370, Beaverton, OR 97077
						(w)503/627-5881 (h)503/657-7703

scott@prism.gatech.EDU (Scott Holt) (10/03/89)

In article <32739@cornell.UUCP> mlee@cs.cornell.edu (mark lee) writes:
>  My questions are:
>
>  1. If there is a password to enable saving, how will FrameMaker
>     keep NeXT users from telling other users the password?
>
>  2. How does the new software distribution system proposed by
>     Steve Jobs work?  Will all the programs and data be encrypted 
>     so that a software key is necessary to decrypt them?
>
>     If this is the case, how does NeXT plan on stopping people
>     from giving other users the software key?
>
>  Any information on these two questions would be greatly appreciated.
>

	In both these cases, the key could be issued based on the hostid
  	of the system. The hostid is unique to each system. It could also
	be based on the ethernet address, which would again be unique to
	each system.
 
	Hostid based encryption schemes are commonly used throughout the
 	workstation software market. Frame uses a hostid based password
	in its floating license server for Suns to assure that the license
	server is run on the correct machine. 

	Of course, nothing can stop the determined hacker who likes making
	binary patches to code - but then I guess the marketing folks figure
	that the increased sales brought on from such schemes (one of the big
	reasons I picked frame for our suns was its demo) more than make up
	for the loses from a determined few.

>Mark Lee, 
>mlee@awamore.cs.cornell.edu


-- 
This is my signature. There are many like it, but this one is mine.
Scott Holt, Systems Analyst		Internet: scott@prism.gatech.edu
Georgia Tech 				BITNET:	  CCUSESH@GITNVE2
Office of Computing Services		404-894-6168

mnkonar@gorby.SRC.Honeywell.COM (Murat N. Konar) (10/03/89)

In article <2235@hydra.gatech.EDU> scott@prism.gatech.EDU () writes:
[discusion of software keys that are unique for each machine]
>	In both these cases, the key could be issued based on the hostid
>  	of the system. The hostid is unique to each system. It could also
>	be based on the ethernet address, which would again be unique to
>	each system.

If the hostid is not hardcoded into the machine, I guess it would be
ok.  What really sucks is something like what Mentor Graphics does
which is setup authorization codes that are dependent on the hard-
wired nodeID of the Apollo workstation.  When that workstation 
breaks down, your software is unusable on any other node!

Disclaimer: I'm no UNIX dude (I don't even play one on TV) so
forgive me if my naivete is showing.

____________________________________________________________________
Have a day. :^|
Murat N. Konar        Honeywell Systems & Research Center, Camden, MN
mnkonar@SRC.honeywell.com (internet) {umn-cs,ems,bthpyd}!srcsip!mnkonar(UUCP)

scott@prism.gatech.EDU (Scott Holt) (10/03/89)

In article <33442@srcsip.UUCP> mnkonar@gorby.UUCP (Murat N. Konar) writes:
>
>If the hostid is not hardcoded into the machine, I guess it would be
>ok.  What really sucks is something like what Mentor Graphics does
>which is setup authorization codes that are dependent on the hard-
>wired nodeID of the Apollo workstation.  When that workstation 
>breaks down, your software is unusable on any other node!
>

Well, it is hard coded - on the Suns (I may be wrong, someone correct me
if I am), the value is in ROM...I would imagine that its the same for
the NeXT. 

On their Sun Floating License server, Frame allows you to define a backup 
server so that if the primary server ever goes down, you can start up the 
license process on the back up. When the license server starts, it first
tries to find out if other license servers are running - if one with the
same serial number is, it won't start...thus, both the primary and backup
cannot be active at the same time. Alas, no "hot" backup, but at least
some sort of safety net. 

Another problem with hostid based copy protection is that when you swap the
CPU board, your hostid can change. On Suns this is avoided by replacing
the ROM from the dead board in the replacement - that way, even though its
essentailly a different system, the hostid remains the same. I don't think,
however, this saves you when you upgrade to a new system or do something
else which makes the old ROMS unusable.

All this generate one very important question to ask when dealing with vendors
(like Frame) who use hostid/ethernet based copy protection: "what does it
take to transfer the license." This is important becuase you don't want to
get stuck with software that becomes obsolete when your hardware does.

Of course, the whole floating license server concept is designed to prevent
problems like the one you have with Mentor Graphics. You purchase licenses
which are not keyed to a particular machine - the license server doles them
out on a first come first served basis and makes sure that no more than x 
copies of the product are in use at one time. Only the license server software
is keyed to a particular machine (which hopefully is a more reliable one).
I really (hint hint hint Frame) wish Frame had implemented this for their
NeXT product - especially since many NeXTs are going to be incorporated in
academic and corporate networks...the fact that they didn't implement it 
indicates a certain level of PC think - that is, treating the NeXT as a PC
rather than the networked workstation that it is.


-- 
This is my signature. There are many like it, but this one is mine.
Scott Holt, Systems Analyst		Internet: scott@prism.gatech.edu
Georgia Tech 				BITNET:	  CCUSESH@GITNVE2
Office of Computing Services		404-894-6168

fischer@iesd.auc.dk (Lars P. Fischer) (10/09/89)

In article <2246@hydra.gatech.EDU> scott@prism.gatech.EDU (Scott Holt) writes:
>Another problem with hostid based copy protection is that when you swap the
>CPU board, your hostid can change. On Suns this is avoided by replacing
>the ROM from the dead board in the replacement - that way, even though its
>essentailly a different system, the hostid remains the same. I don't think,
>however, this saves you when you upgrade to a new system or do something
>else which makes the old ROMS unusable.

Like disaster striking while swapping ROM's (hey, Murphy, knew you'd
drop by today :-). Of course, you can get a replacement ROM, but that
takes *time* (I know, 'cause I've tried it)

>All this generate one very important question to ask when dealing with vendors
>(like Frame) who use hostid/ethernet based copy protection: "what does it
>take to transfer the license." This is important becuase you don't want to
>get stuck with software that becomes obsolete when your hardware does.

And the problem then is: how to design a copy protection scheme that
allows transfers, e.g. allows passwords to be called back?

/Lars
--
Copyright 1989 Lars Fischer; you can redistribute only if your recipients can.
Lars Fischer,  fischer@iesd.auc.dk, {...}!mcvax!iesd!fischer
Department of Computer Science, University of Aalborg, DENMARK.

"That makes 100 errors; please try again" --TeX