Forgot your password?
Cyhist Apr. 10 1998 b
Document Actions
========================================================================= Date: Fri, 10 Apr 1998 13:38:17 +0100
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> From: John Line <jml4@CUS.CAM.AC.UK>
Subject: Re: Archiving obsolete machines
In-Reply-To: <E0yNZn8-0000A3-00@taurus.cus.cam.ac.uk> from "Automatic digest
processor" at Apr 10, 98 00:56:47 am
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
Les Earnest <les@STEAM.STANFORD.EDU> wrote
>I believe that old hardware and software should be preserved separately. In order to be able to exercise the latter, write an emulator in some language that you expect to last for awhile. Inasmuch as hardware speeds keep increasing it will usually be possible to emulate old machines at a speed faster than they ever attained.
and
>Our plan is to make the archive available on a public FTP site but with personal text files encrypted. Owners of those files who can be located will be encouraged to make some or all of them public. There are a lot of public documentation files in the archive that describe research projects as well as application programs.
>
>If we can make an emulator available at the same site it will be possible to both read old documents and run ancient programs to see what they do or don't do. With moderate effort this archive can be migrated to still later hardware when it becomes available. After everyone involved is dead, the encryption can be removed entirely.
I'd certainly agree that is a realistic approach to keeping old software usable, but I would expect there to be problems in many cases, for example:
* software licence conditions may restrict use of critical software
(such as the operating system) to a specific system, or to the original purchaser (no right to transfer it to anyone else), or to execution on particular types of system (e.g. by implication, not emulators).
* Copyright may independently preclude making copies of the software
available publically, even if it is legal for the institution running an archive to possess and use the software - and where later versions of the same software are still in use and have commercial value, the vendors may refuse to give special permission.
* Software may depend on "undocumented" aspects of the hardware, needing
the hardware to be reverse-engineered to allow adequate emulation - e.g. where software authors (more perhaps for games rather than "serious" software) relied on knowledge of custom video hardware to get adequate performance. The shelves in bookstores with books about "undocumented" aspects of IBM PC hardware and software are a special case, and for many other systems there may well be no good sources of such information.
* For many systems, there may only be one or two published sources of
"definitive" information about the instruction set and related aspects of the system's hardware, and they may include mistakes or be sufficiently vague in places that while implementing an emulator is possible, it may not actually result in 100% accurate emulation of the original hardware, leading to "working" programs crashing or not behaving entirely as expected.
Of course, many of these difficulties can be overcome one way or another, at least if the emulators can be developed while there are enough people around with direct knowledge of the systems to help fill in any missing information. As long as legal issues don't prevent or delay that.
John Line
--
John Line - Cambridge University Computing Service, Computer Laboratory,
New Museums Site, Pembroke Street, Cambridge CB2 3QG, England. Internet: jml4@cus.cam.ac.uk Phone: +44 1223 334708 FAX: +44 1223 334679
______________________________________________________________________
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> From: John Line <jml4@CUS.CAM.AC.UK>
Subject: Re: Archiving obsolete machines
In-Reply-To: <E0yNZn8-0000A3-00@taurus.cus.cam.ac.uk> from "Automatic digest
processor" at Apr 10, 98 00:56:47 am
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
Les Earnest <les@STEAM.STANFORD.EDU> wrote
>I believe that old hardware and software should be preserved separately. In order to be able to exercise the latter, write an emulator in some language that you expect to last for awhile. Inasmuch as hardware speeds keep increasing it will usually be possible to emulate old machines at a speed faster than they ever attained.
and
>Our plan is to make the archive available on a public FTP site but with personal text files encrypted. Owners of those files who can be located will be encouraged to make some or all of them public. There are a lot of public documentation files in the archive that describe research projects as well as application programs.
>
>If we can make an emulator available at the same site it will be possible to both read old documents and run ancient programs to see what they do or don't do. With moderate effort this archive can be migrated to still later hardware when it becomes available. After everyone involved is dead, the encryption can be removed entirely.
I'd certainly agree that is a realistic approach to keeping old software usable, but I would expect there to be problems in many cases, for example:
* software licence conditions may restrict use of critical software
(such as the operating system) to a specific system, or to the original purchaser (no right to transfer it to anyone else), or to execution on particular types of system (e.g. by implication, not emulators).
* Copyright may independently preclude making copies of the software
available publically, even if it is legal for the institution running an archive to possess and use the software - and where later versions of the same software are still in use and have commercial value, the vendors may refuse to give special permission.
* Software may depend on "undocumented" aspects of the hardware, needing
the hardware to be reverse-engineered to allow adequate emulation - e.g. where software authors (more perhaps for games rather than "serious" software) relied on knowledge of custom video hardware to get adequate performance. The shelves in bookstores with books about "undocumented" aspects of IBM PC hardware and software are a special case, and for many other systems there may well be no good sources of such information.
* For many systems, there may only be one or two published sources of
"definitive" information about the instruction set and related aspects of the system's hardware, and they may include mistakes or be sufficiently vague in places that while implementing an emulator is possible, it may not actually result in 100% accurate emulation of the original hardware, leading to "working" programs crashing or not behaving entirely as expected.
Of course, many of these difficulties can be overcome one way or another, at least if the emulators can be developed while there are enough people around with direct knowledge of the systems to help fill in any missing information. As long as legal issues don't prevent or delay that.
John Line
--
John Line - Cambridge University Computing Service, Computer Laboratory,
New Museums Site, Pembroke Street, Cambridge CB2 3QG, England. Internet: jml4@cus.cam.ac.uk Phone: +44 1223 334708 FAX: +44 1223 334679
______________________________________________________________________
