Mainframe emulator goes commercial

davejesc

Sep 30, 2009

Mainframe emulator goes commercial
Hercules, son of z/OS

http://www.theregister.co.uk/2009/09/25/turbohercules_goes_commercial/

9:59 am

Re: Mainframe emulator goes commercial

Michael Kerpan

Sep 30, 2009

So does this mean that Hercules is dead as a free project or merely
that boxes that have a pre-installed copy of Hercules are now being
sold. Depending on the answer this could either be great news or
downright tragic.

Mike

On 9/30/09, davejesc <davejesc@...> wrote:
> Mainframe emulator goes commercial
> Hercules, son of z/OS
>
> http://www.theregister.co.uk/2009/09/25/turbohercules_goes_commercial/
>
>
>

11:09 am

Re: Mainframe emulator goes commercial

Jay Maynard

Sep 30, 2009

On Wed, Sep 30, 2009 at 07:09:34AM -0400, Michael Kerpan wrote:
> So does this mean that Hercules is dead as a free project or merely
> that boxes that have a pre-installed copy of Hercules are now being
> sold. Depending on the answer this could either be great news or
> downright tragic.

The existence of TurboHercules changes nothing at all with the Hercules
project itself. It is, and will always remain, fully open source.
--
Jay Maynard, K5ZC http://www.conmicro.com
http://jmaynard.livejournal.com http://www.tronguy.net
http://www.hercules-390.org (Yes, that's me!)
Buy Hercules stuff at http://www.cafepress.com/hercules-390

11:14 am

RE: Mainframe emulator goes commercial

John P. Baker

Sep 30, 2009

It would be nice if IBM embraced Hercules as a disaster recovery platform.
However, I have serious doubts that IBM will see it that way.

IBM will almost certainly view it as direct competition with IBM's Business
Continuity Services organization (I think that I have the name right).

I do think that TurboHercules is pursuing the correct approach by seeking
IBM's blessing, rather than by simply violating IBM's licensing agreements,
which would certainly result in an immediate lawsuit and injunction.

We will just have to wait to see what response is received from IBM.

John P. Baker

From: hercules-390@yahoogroups.com [mailto:hercules-390@yahoogroups.com] On
Behalf Of davejesc
Sent: Wednesday, September 30, 2009 5:59 AM
To: hercules-390@yahoogroups.com
Subject: [hercules-390] Mainframe emulator goes commercial

Mainframe emulator goes commercial
Hercules, son of z/OS

http://www.theregister.co.uk/2009/09/25/turbohercules_goes_commercial/

[Non-text portions of this message have been removed]

1:55 pm

Re: Mainframe emulator goes commercial

Roger Bowler

Sep 30, 2009

On Wednesday, September 30, 2009 John P. Baker wrote:
> I do think that TurboHercules is pursuing the correct approach by
> seeking IBM's blessing, rather than by simply violating IBM's
> licensing agreements, which would certainly result in an immediate
> lawsuit and injunction.

Well I don't think it's necessarily an "either or" situation. You
could be in complete accordance with IBM licensing agreements and
still not have their blessing.

We know from the PSI case that IBM don't like competition. Having
alleged that PSI had violated licensing agreements and patents and
misappropriated trade secrets, IBM were unwilling to see the case
through to the end. Faced with the possibility that PSI might win, IBM
finished up by buying their way out of the case to get rid of PSI
http://www.tech-news.com/another/ap200807b.html

With Hercules the situation is somewhat different. As an open source
project, Hercules cannot be bought and turned into shelfware, in the
same way that buying out Red Hat and Novell would not destroy Linux.
Also, I think IBM would be reluctant to attack Hercules, because then
their claims to be "an industry leader helping governments move toward
greater openness and innovation" and "driving reforms for IP, Open
Source & Standards" might begin to look a bit hollow.
http://www-03.ibm.com/linux/ossstds/

--
Regards,
Roger Bowler
http://www.openmainframe.org
Hercules "I can't believe it's not a mainframe!"

2:51 pm

RE: Mainframe emulator goes commercial

John P. Baker

Oct 1, 2009

Roger,

People have tried to make as if the PSI case was solely about competition.

However, according to the various filings, PSI had in its possession IBM
confidential information that had been made available to Amdahl, and which
was reported to IBM as having been destroyed when Amdahl's assets were sold.

In addition, the PSI implementation apparently involved reading a page of
IBM object code, identifying executable instructions, translating those
instructions into native Itanium code (a direct violation of numerous IBM
licensing agreements), and then branching to the native Itanium code.

In my opinion, IBM would have prevailed at trial. However, IBM determined
that it was more cost effective to purchase PSI than to pay all of the court
costs and legal fees in a protracted fight. Financially, it would have been
a Pyrrhic victory.

Products which permit one machine's instruction set to run on another
machine's architecture fall into three classifications (emulator, simulator,
and translator). There has been a lot of argument over the years as to what
these classifications really mean. I prefer the definitions which were
prevalent when I first started programming some 41 years ago -

. Emulator

o Hardware-assisted execution of non-native instructions.

. Simulator

o Software-only execution of non-native instructions, principally through
the use of subroutines.

. Translator

o Software translation of non-native instructions into native
instructions, either on a one-time basis, or on a just-in-time basis.
However, when I started, just-in-time translation was cost-prohibitive,
except for laboratory experimentation.

PSI's product apparently fell into the "translator" category, and more
specifically into the "just-in-time" subcategory.

Hercules, on the other hand, is properly a "simulator" as opposed to an
"emulator", since as far as I know, there is no Hercules hardware-assist.

IBM licensing agreements specifically prohibit execution of IBM code by way
of a "translator". However, IBM uses the terms "emulator" and "simulator"
interchangeably, which does lead to some confusion.

John P. Baker

From: hercules-390@yahoogroups.com [mailto:hercules-390@yahoogroups.com] On
Behalf Of Roger Bowler
Sent: Wednesday, September 30, 2009 10:52 AM
To: hercules-390@yahoogroups.com
Subject: Re: [hercules-390] Mainframe emulator goes commercial

Well I don't think it's necessarily an "either or" situation. You
could be in complete accordance with IBM licensing agreements and
still not have their blessing.

We know from the PSI case that IBM don't like competition. Having
alleged that PSI had violated licensing agreements and patents and
misappropriated trade secrets, IBM were unwilling to see the case
through to the end. Faced with the possibility that PSI might win, IBM
finished up by buying their way out of the case to get rid of PSI
http://www.tech-news.com/another/ap200807b.html

With Hercules the situation is somewhat different. As an open source
project, Hercules cannot be bought and turned into shelfware, in the
same way that buying out Red Hat and Novell would not destroy Linux.
Also, I think IBM would be reluctant to attack Hercules, because then
their claims to be "an industry leader helping governments move toward
greater openness and innovation" and "driving reforms for IP, Open
Source & Standards" might begin to look a bit hollow.
http://www-03.ibm.com/linux/ossstds/

--
Regards,
Roger Bowler
http://www.openmainframe.org
Hercules "I can't believe it's not a mainframe!"

[Non-text portions of this message have been removed]

2:09 am