Jeff Vance wrote:

>
> Chuck wrote:
> ...
> > Then it is an exercise in futility.
> >
> > The fee will not be determined by what we consider
> > reasonable, but by what
> > these vendors will accept as a licensing fee.
>
> I understand the point about telling the vendor how much
> you'd be willing to pay, not knowing if they'd be willing
> to sell it for less -- that's not the way I'd buy something
> either.
>
> I appreciate Jon's effort to get this dialog going and it
> is important for CSY to test the waters that the pricing
> we are thinking of is at least reasonable.  So, hopefully
> without getting in trouble, I will tell you that we are
> currently considering a price of approximately $500 USD
> for one new MPE license. Most likely (not definite) this
> license would include all HP-owned subsystems, which, I
> *think* is all of them.  HP is not trying to make any profit
> on the sale of the new license. We do need to cover the
> royalties we pay to other vendors for the parts of MPE that
> we have licensed from them.  This includes the following:
>
>   - MKS for their POSIX shell and Utilities,
>   - Mentat for their MPS Streams code needed for pipes and some
>     other IPC functionality. MPS code is used by: 12H Raid Mgr,
>     XP Raid Mgr, POSIX pipes [pipe(), HPPIPE()], Predictive, CSTM,
>     and Patch/iX.
>   - MB Foster for their 32 bit ODBC driver.
>
> CSY also wants to try to cover our internal administration costs, but
> those should be pretty small, since we expect the volume to be fairly
> low.
>
> I hope this helps stimulate more constructive dialog.
>
>  Jeff Vance, "CSY"

Thanks to Jeff for shedding some light on CSY's thinking.

Philosophically, I still don't see how HP can justify keeping exclusive
control of MPE and charging for new licenses. However, practically speaking,
I can live with a $500 charge for new licenses. More importantly, $500 per
license should encourage the development of an emulator.

John