At 12:48 PM 5/4/95 P, Ken Sletten b894 c331 x2525 wrote:
>>>>>    First the update:
>
>After post-IPROF discussion with HP, it is now my
>understanding that the "shareware" FSF GNU  C++
>compiler that HP talked about bringing to the 3000 at
>IPROF *will* be supported by HP;  i.e.:  The compiler
>itself will be downloadable from JAZZ or whatever,
>but users will have the option to purchase software
>support for that compiler directly from HP (I assume
>what this means is that HP will contract with the third
>party to act as the lab for C++).  This sounds like at
>least one big step in the right direction to me.
>
>Something else I wasn't aware of (maybe old news
>to many of you, but new to me):   Most other vendors
>who have C++ compilers on their platforms apparently
>use the shareware C++ product as a base for what
>they provide.  If that is the case, having a 3000 version
>of GNU  C++ should mean it will be easier to port C++
>programs from other vendor platforms, compared to
>porting C++ from HP-UX.  And a 3000 GNU C++ can
>be available much faster than a port from UX....  Since
>we need C++ on the 3000 ASAP for the client-server
>extension to our Transact system that we are working
>on, this is a major factor for our site....
>
 
I thought this was clear from HP's announcement. IMHO this gives us the best
possible solution for portability and evolutionary capability, giving us,
the MPE/iX developer community, unprecedented capability to drive the
enhancements and optimization of this critical development tool.
 
>
>======================================
>
>***   C++ on MPE/iX:   Requirements and Options   ***
>
>1)   If a compiler existed which handled only Unix-like
>constructs (at least initially) were supported, would
>you use it ?
Yes. In fact I think that a 'unix only' mode would be appropriate even for a
mature product, since it will often be used to compile code which needs to
be portable to unix.
 
>
>2)   If an MPE C++ compiler were enhanced after the
>initial port, what capabilities would it need:  indicate
>"don't care", "useful", or "must have" for each ?
>        Intrinsic support:
            useful (I can just declare them as externs, if I need to)
 
>        Access to IMAGE:
            It depends on the form.
            How would this differ from Intrinsic Support?
 
>        Access to KSAM:
            Again. It depends on the form.
            How would this differ from Intrinsic Support?
 
>        Long pointer support:
            MUST HAVE!!!
 
>        Ability to run in MPE Name Space:
            MUST HAVE!!!
 
>        Use default MPE naming conventions:
            MUST HAVE!!!
 
>        Other >________________________:
            A Well defined and documented means of building XL code.
            A fully XL resident runtime library, so that c++ code in the XL
could be called by a non c++ outer block (main) such as a
            CM SPL program, or a NM COBOL program.
 
>3)   Would you trust a third-party compiler that was
>supported by HP ?
Yes. and I would feel even better about it if source for the compiler were
accessible.
 
>
>4)   What would you be willing to pay HP for annual
>support of C++ on MPE/iX ?
Yes, provided the cost is not unreasonable.
 
>5)   If the cost of acquisition of the compiler were
>minimal or non-existent, would you be willing to pay
>higher support fees over a guaranteed term ?
Yes. Hoever I think that HP should take into consideration that supporting
this compiler will make it much much easier for 3rd party developers to port
products that will generate mucho business for HP. In other words, HP, Help
us help you.
 
>
>6)   Additional comments/remarks/concerns on this
>subject:
I have, throughout this thread, been very supportive of HP's decision to go
with the de facto standard in c++ compilers. IMHO this gives us the best
possible solution for portability and evolutionary capability, giving us,
the MPE/iX developer community, unprecedented capability to drive the
enhancements and optimization of this critical development tool. It also
paves the way for a true partnership between HP and those develooping
applications for MPE/iX.
 
 
 
 
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
:                                                                           :
: Scott Herman          [log in to unmask]   Yale-New Haven Hospital
:
: Dept of Lab Medicine                                       20 York Street :
: (203) 785-2449                                       New Haven, Ct. 06504 :
:                                                                           :
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::