1 INFO-VAX	Mon, 04 Jul 2005	Volume 2005 : Issue 370       Contents:0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET0 Re: A tale of two macro libraries LIB vs STARLET. Re: alpha instruction set - free to implement?. Re: alpha instruction set - free to implement?8 Re: ANNOUNCEMENT - OpenVMS Technical Journal - June 20058 Re: ANNOUNCEMENT - OpenVMS Technical Journal - June 2005 Re: CTRL-T - what's the T? GCC on OpenVMS Re: GCC on OpenVMS5 Re: Network card problems? Or something like that.... 5 Re: Network card problems? Or something like that.... 5 Re: Network card problems? Or something like that.... 5 Re: Network card problems? Or something like that.... 5 RE: Network card problems? Or something like that....  Re: OpenVMS and XML questions $ Re: The VMS Gateway FAQ is available$ Re: The VMS Gateway FAQ is available# Re: US military and confidentiality # Re: US military and confidentiality # Re: US military and confidentiality # RE: US military and confidentiality # Re: US military and confidentiality # Re: US military and confidentiality # Re: US military and confidentiality  Re: [Announce] FreeVMS 0.1.7 Re: [Announce] FreeVMS 0.1.7  F ----------------------------------------------------------------------  * Date: Mon, 4 Jul 2005 06:03:59 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> 9 Subject: Re: A tale of two macro libraries LIB vs STARLET ? Message-ID: <daajge$24k$1@nwrdmz02.dmz.ncs.ea.ibs-infra.bt.com>   
 Hi Robert,  . > Don't put 3rd party stuff in Starlet or lib.   Why not?  G Admittedly, I don't use the "I" option on provide_image and ship my own K shareable image library rather than use imagelib.olb but that's just 'cos I F like it like that. What would be wrong with sticking the following few  definitions in a system library?           .MACRO  $T3DEF,$GBL          $DEFINI T3,$GBL    $EQU    T3$M_INTERRUPT  1  $EQU    T3$M_MULTIPLE   2  $EQU    T3$M_MORE       2  $EQU    T3$M_DISCONNECT 4  $EQU    T3$M_CLOSE      8  $EQU    T3$M_NOW       16  $EQU    T3$M_OOB       32    $EQU    T3$V_INTERRUPT  0  $EQU    T3$V_MULTIPLE   1  $EQU    T3$V_MORE       1  $EQU    T3$V_DISCONNECT 2  $EQU    T3$V_CLOSE      3  $EQU    T3$V_NOW        4  $EQU    T3$V_OOB        5    $EQU    T3$K_DECNET     1  $EQU    T3$K_TCP_IP     2    $EQU    T3$K_SYSTEM     1  $EQU    T3$K_USER       2            $DEFEND T3,$GBL,DEF 
         .ENDM    Regards Richard Maher   H PS. I remember (before) $issdef that some bright-spark in VMS managed toL create two totally different Symbol Definition files ($persona service flagsI was one. can't remember the other) one in LIB and the other in Starlet. A  bit embarassing?  B "Robert Deininger" <rdeininger@mindspringdot.com> wrote in messageF news:rdeininger-0307052015030001@user-105n8e0.dialup.mindspring.com...J > In article <da9jq1$5d5$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>, "Richard- > Maher" <maher_rj@hotspamnotmail.com> wrote:  >  > >Hi, > > F > >If I'm shipping a product that has a macro symbol definitions file, should I > >(at VMSINSTAL time) > >  > >a) Stick it in STARLET.MLB  > >b) Stick it in LIB.MLB  > >c) Ship my own macro library  > >d) a n other  > 	 > C or D.  > . > Don't put 3rd party stuff in Starlet or lib.   ------------------------------  * Date: Mon, 4 Jul 2005 06:04:51 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> 9 Subject: Re: A tale of two macro libraries LIB vs STARLET ? Message-ID: <daaji2$egb$1@nwrdmz03.dmz.ncs.ea.ibs-infra.bt.com>   * > Don't even put it in the VMS directories   Yeah right.    Regards Richard Maher.  5 "David Froble" <davef@tsoft-inc.com> wrote in message * news:11chbjgeqovje27@corp.supernews.com... > Robert Deininger wrote: L > > In article <da9jq1$5d5$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>, "Richard/ > > Maher" <maher_rj@hotspamnotmail.com> wrote:  > >  > >  > >>Hi,  > >>G > >>If I'm shipping a product that has a macro symbol definitions file,  should I > >>(at VMSINSTAL time)  > >> > >>a) Stick it in STARLET.MLB > >>b) Stick it in LIB.MLB  > >>c) Ship my own macro library > >>d) a n other > >  > >  > > C or D.  > > 0 > > Don't put 3rd party stuff in Starlet or lib. > + > Don't even put it in the VMS directories.    ------------------------------  * Date: Mon, 4 Jul 2005 06:29:56 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> 9 Subject: Re: A tale of two macro libraries LIB vs STARLET ? Message-ID: <daal13$s4i$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>   	 Hi Larry,    > Only if both:  > : > A. You can guarantee the definitions will not change for >    a new version of VMS  > 7 > B. The names are all qualified by a VMS facility name 8 >    (with dollar sign) registered to you by the Product >    Registrar.  >   ! It looks like we've got a winner!   5 But I think I'll stick it in LIB anyway. Why? Well :-   D 1) Rightly or wrongly, I've always saw Starlet as the home of the SS, definitions and LIB as the symbol-def place.  J 2) There's already a $TT3DEF macro in Starlet (terminal stuff I'd imagine)+ so a $T3DEF there as well may be confusing?   + > Do not expect customers to code in Macro.   I Customers can do whatever they want! (At least that's the way it was with D the VMS that I knew and loved. Before the "Use C you ignorant scum!"I dictates were handed down.) However, I do not expect customers to code in L Macro. As discussed in previous posts, these symbols will just let customersK do binary ORs of flags at "compile" time. Here's your chance to tell us how J ADA could do it :-) But COBOL certainly can't. If you don't care about theK itsy bitsy CPU resources needed to do an OR at "runtime"? Then feel free to $ forget that the $T3DEF macro exists.  J Sure, all grass-hoppers were (and should continue to be) told that this isH "The VMS way"? or would you have us believe that mine is the first $*defL macro on VMS or that they are an anachronistic and nostalgic throw-back to a long forgoyyen time?   Regards Richard Maher     : "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message- news:IsZ5nkLMdUjd@eisner.encompasserve.org... J > In article <da9jq1$5d5$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>, "Richard, Maher" <maher_rj@hotspamnotmail.com> writes: > G > > If I'm shipping a product that has a macro symbol definitions file,  should I > > (at VMSINSTAL time)  > >  > > a) Stick it in STARLET.MLB >  > Only if both:  > : > A. You can guarantee the definitions will not change for >    a new version of VMS  > 7 > B. The names are all qualified by a VMS facility name 8 >    (with dollar sign) registered to you by the Product >    Registrar.  >  > > b) Stick it in LIB.MLB >  > Only if both:  > ; > A. You reasonably believe the definitions will change for  >    a new version of VMS  > 7 > B. The names are all qualified by a VMS facility name 8 >    (with dollar sign) registered to you by the Product >    Registrar.  >   > > c) Ship my own macro library > C > That always works, but the file name should by qualified by a VMS C > facility name (with dollar sign) registered to you by the Product  > Registrar. >  > > d) a n other > + > Do not expect customers to code in Macro.    ------------------------------  % Date: Mon, 04 Jul 2005 02:48:42 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 9 Subject: Re: A tale of two macro libraries LIB vs STARLET , Message-ID: <42C8DBB8.3846C698@teksavvy.com>   Richard Maher wrote: > , > > Don't even put it in the VMS directories > 
 > Yeah right.   M Yes. Customers want to know where a package resides, and need to know exactly H what needs to be moved if they decide to move the app to another system.  I Messing with system libraries and hiding some of your stuff in there make 6 application management that much harder for customers.  L And whenever the system manager upgrades VMS, you are at risk of losing yourM definitions if the upgrades provides a new  versions of the lib and starlets.   N Do you also provide C header files for starlet ? (there is a version for the C compiler as well).  L Keeping all your files in your own set of directories makes systems managersK that much easier, including shared images. Just define logicals to point to  the right files.   ------------------------------   Date: 4 Jul 2005 04:47:21 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) 9 Subject: Re: A tale of two macro libraries LIB vs STARLET 3 Message-ID: <mxJwnKFnXUmN@eisner.encompasserve.org>   u In article <daal13$s4i$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>, "Richard Maher" <maher_rj@hotspamnotmail.com> writes:    >> Only if both: >>; >> A. You can guarantee the definitions will not change for  >>    a new version of VMS >>8 >> B. The names are all qualified by a VMS facility name9 >>    (with dollar sign) registered to you by the Product  >>    Registrar. >> > # > It looks like we've got a winner!  > 7 > But I think I'll stick it in LIB anyway. Why? Well :-  > F > 1) Rightly or wrongly, I've always saw Starlet as the home of the SS. > definitions and LIB as the symbol-def place. > L > 2) There's already a $TT3DEF macro in Starlet (terminal stuff I'd imagine)- > so a $T3DEF there as well may be confusing?   @ No, I believe that your definition (presuming T3 is a registered= facility name) should be T3$whateverDEF, with the dollar sign  following the facility name.  , >> Do not expect customers to code in Macro. > K > Customers can do whatever they want! (At least that's the way it was with F > the VMS that I knew and loved. Before the "Use C you ignorant scum!"K > dictates were handed down.) However, I do not expect customers to code in N > Macro. As discussed in previous posts, these symbols will just let customersM > do binary ORs of flags at "compile" time. Here's your chance to tell us how L > ADA could do it :-) But COBOL certainly can't. If you don't care about theM > itsy bitsy CPU resources needed to do an OR at "runtime"? Then feel free to & > forget that the $T3DEF macro exists.  > If it is just bit flags, I would think using SDL to output the? definitions in multiple languages would be good, although there ? might not be an SDL back end for Cobol.  But writing one should B not be hard, especially if you only supported the pieces you need.= Shipping a Macro/Copybook/include/require/package file in all ; the languages would ensure the customer has what they need.    ------------------------------  % Date: Mon, 04 Jul 2005 09:51:10 -0400 4 From: Forrest Kenney <forrest_dot_kenney@hp_dot.com>9 Subject: Re: A tale of two macro libraries LIB vs STARLET * Message-ID: <42C93ECD.F3860439@hp_dot.com>  L     1) Do not put your stuff in either LIB or STARLET they belong to OpenVMS	 group and -         may be replaced or updated as needed.   O     2) Starlet is public definitions that are only supposed to be changed in an  upward compatible N          fashion.  LIB is mostly privileged semi-private definitions, that may change to suit theM          developers needs.  But typically if changed care is taken to make it  upward compatible.  M     3) The pseudo terminal driver stuff PTD$XXXX is a public interface and is 
 documented in ;         the I/O users section of the OpenVMS documentation.      Forrest Kenney OpenVMS    Richard Maher wrote:   > Hi Larry,  > D > > No, I believe that your definition (presuming T3 is a registeredA > > facility name) should be T3$whateverDEF, with the dollar sign   > > following the facility name. > K > Sounds good to me. Certainly makes more sense and is more consistent that J > way, but I couldn't find any fac$def stuff in my libs. (But then I don'tK > have a lot of layered products on this box at the moment. Pseudo-terminal : > driver seems to stand out.) Is this documented anywhere? > J > Also, I just re-read your earlier post and realized a big difference : - > 
 > Starlet.MLB > > > A. You can guarantee the definitions *WILL NOT* change for > >    a new version of VMS  > 	 > LIB.MLB ? > > A. You reasonably believe the definitions *WILL* change for  > >    a new version of VMS  > N > Why the difference? Are you saying that LIB gets replaced with a new version% > of VMS and STARLET is just updated?  >  > Thanks for the interest BTW. >  > Regards Richard Maher  > < > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message/ > news:mxJwnKFnXUmN@eisner.encompasserve.org... L > > In article <daal13$s4i$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>, "Richard. > Maher" <maher_rj@hotspamnotmail.com> writes: > >  > > >> Only if both: > > >>? > > >> A. You can guarantee the definitions will not change for  > > >>    a new version of VMS > > >>< > > >> B. The names are all qualified by a VMS facility name= > > >>    (with dollar sign) registered to you by the Product  > > >>    Registrar. > > >> > > > ' > > > It looks like we've got a winner!  > > > ; > > > But I think I'll stick it in LIB anyway. Why? Well :-  > > > J > > > 1) Rightly or wrongly, I've always saw Starlet as the home of the SS2 > > > definitions and LIB as the symbol-def place. > > > G > > > 2) There's already a $TT3DEF macro in Starlet (terminal stuff I'd 
 > imagine)1 > > > so a $T3DEF there as well may be confusing?  > > D > > No, I believe that your definition (presuming T3 is a registeredA > > facility name) should be T3$whateverDEF, with the dollar sign   > > following the facility name. > > 0 > > >> Do not expect customers to code in Macro. > > > J > > > Customers can do whatever they want! (At least that's the way it was > withJ > > > the VMS that I knew and loved. Before the "Use C you ignorant scum!"L > > > dictates were handed down.) However, I do not expect customers to code > inH > > > Macro. As discussed in previous posts, these symbols will just let > customers M > > > do binary ORs of flags at "compile" time. Here's your chance to tell us  > how L > > > ADA could do it :-) But COBOL certainly can't. If you don't care about > the N > > > itsy bitsy CPU resources needed to do an OR at "runtime"? Then feel free > to* > > > forget that the $T3DEF macro exists. > > B > > If it is just bit flags, I would think using SDL to output theC > > definitions in multiple languages would be good, although there C > > might not be an SDL back end for Cobol.  But writing one should F > > not be hard, especially if you only supported the pieces you need.A > > Shipping a Macro/Copybook/include/require/package file in all ? > > the languages would ensure the customer has what they need.    ------------------------------  % Date: Mon, 04 Jul 2005 10:29:13 -0400 4 From: Forrest Kenney <forrest_dot_kenney@hp_dot.com>9 Subject: Re: A tale of two macro libraries LIB vs STARLET * Message-ID: <42C947B9.1BDA8954@hp_dot.com>  I     I would treat imaglib like LIB and STARLET it belongs to the O.S. and L can get replaced.  If you stick a shareable in there you have to be preparedK on an upgrade to replace it.  In addition any visible symbols in there have J to be unique and not comflict with anything put in there.  So my advice isM don't put it in there.  Yes I know that means folks have to link against your 
 shareable.   Forrest        Richard Maher wrote:  
 > Hi Forrest,  > J > OK, I've got to make a decision on this (And it's already caused far tooL > much grief for the tiny bit of functionality it provides) so it looks like0 > I'll ship a macro library and be done with it. > M > But before I close the book (and to give others a chance to reply) I'd just I > like to ask you a question if I may. And that is "How do you feel about L > shareables in IMAGELIB.OLB?" I.E. Can anyone tell me why (philisophically)N > STARLET or LIB should be treated differently from IMAGELIB? in the "It's all > mine!" sense of things?  >  > Regards Richard Maher  > C > "Forrest Kenney" <forrest_dot_kenney@hp_dot.com> wrote in message & > news:42C93ECD.F3860439@hp_dot.com... > > H > >     1) Do not put your stuff in either LIB or STARLET they belong to	 > OpenVMS 
 > > group and 1 > >         may be replaced or updated as needed.  > > M > >     2) Starlet is public definitions that are only supposed to be changed  > in an  > > upward compatible N > >          fashion.  LIB is mostly privileged semi-private definitions, that > may  > > change to suit theN > >          developers needs.  But typically if changed care is taken to make > it > > upward compatible. > > N > >     3) The pseudo terminal driver stuff PTD$XXXX is a public interface and > is > > documented in ? > >         the I/O users section of the OpenVMS documentation.  > >  > >  > > Forrest Kenney > > OpenVMS  > >  > > Richard Maher wrote: > >  > > > Hi Larry,  > > > H > > > > No, I believe that your definition (presuming T3 is a registeredE > > > > facility name) should be T3$whateverDEF, with the dollar sign $ > > > > following the facility name. > > > J > > > Sounds good to me. Certainly makes more sense and is more consistent > thatN > > > way, but I couldn't find any fac$def stuff in my libs. (But then I don't? > > > have a lot of layered products on this box at the moment.  > Pseudo-terminal > > > > driver seems to stand out.) Is this documented anywhere? > > > N > > > Also, I just re-read your earlier post and realized a big difference : - > > >  > > > Starlet.MLB B > > > > A. You can guarantee the definitions *WILL NOT* change for > > > >    a new version of VMS  > > > 
 > > > LIB.MLB C > > > > A. You reasonably believe the definitions *WILL* change for  > > > >    a new version of VMS  > > > J > > > Why the difference? Are you saying that LIB gets replaced with a new	 > version ) > > > of VMS and STARLET is just updated?  > > > " > > > Thanks for the interest BTW. > > >  > > > Regards Richard Maher  > > > @ > > > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message3 > > > news:mxJwnKFnXUmN@eisner.encompasserve.org... G > > > > In article <daal13$s4i$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>, 
 > "Richard2 > > > Maher" <maher_rj@hotspamnotmail.com> writes: > > > >  > > > > >> Only if both:
 > > > > >>C > > > > >> A. You can guarantee the definitions will not change for " > > > > >>    a new version of VMS
 > > > > >>@ > > > > >> B. The names are all qualified by a VMS facility nameA > > > > >>    (with dollar sign) registered to you by the Product  > > > > >>    Registrar.
 > > > > >>	 > > > > > + > > > > > It looks like we've got a winner! 	 > > > > > ? > > > > > But I think I'll stick it in LIB anyway. Why? Well :- 	 > > > > > N > > > > > 1) Rightly or wrongly, I've always saw Starlet as the home of the SS6 > > > > > definitions and LIB as the symbol-def place.	 > > > > > K > > > > > 2) There's already a $TT3DEF macro in Starlet (terminal stuff I'd  > > > imagine)5 > > > > > so a $T3DEF there as well may be confusing?  > > > > H > > > > No, I believe that your definition (presuming T3 is a registeredE > > > > facility name) should be T3$whateverDEF, with the dollar sign $ > > > > following the facility name. > > > > 4 > > > > >> Do not expect customers to code in Macro.	 > > > > > N > > > > > Customers can do whatever they want! (At least that's the way it was
 > > > withN > > > > > the VMS that I knew and loved. Before the "Use C you ignorant scum!"K > > > > > dictates were handed down.) However, I do not expect customers to  > code > > > inL > > > > > Macro. As discussed in previous posts, these symbols will just let > > > customers N > > > > > do binary ORs of flags at "compile" time. Here's your chance to tell > us	 > > > how J > > > > > ADA could do it :-) But COBOL certainly can't. If you don't care > about 	 > > > the M > > > > > itsy bitsy CPU resources needed to do an OR at "runtime"? Then feel  > free > > > to. > > > > > forget that the $T3DEF macro exists. > > > > F > > > > If it is just bit flags, I would think using SDL to output theG > > > > definitions in multiple languages would be good, although there G > > > > might not be an SDL back end for Cobol.  But writing one should J > > > > not be hard, especially if you only supported the pieces you need.E > > > > Shipping a Macro/Copybook/include/require/package file in all C > > > > the languages would ensure the customer has what they need.  > >    ------------------------------  % Date: Mon, 04 Jul 2005 08:26:28 -0700 ( From: Jeff Cameron <roktsci@comcast.net>9 Subject: Re: A tale of two macro libraries LIB vs STARLET 0 Message-ID: <BEEEA334.1089C%roktsci@comcast.net>   On 7/3/05 11:03 PM, in articleB daajge$24k$1@nwrdmz02.dmz.ncs.ea.ibs-infra.bt.com, "Richard Maher"$ <maher_rj@hotspamnotmail.com> wrote:   > Hi Robert, > / >> Don't put 3rd party stuff in Starlet or lib.  > 
 > Why not?K Very simply, if HP comes out with a patch, update ore upgrade that replaces G the file, your  stuff goes away. Put it in your own library and make it  available to those who need it.    ------------------------------  * Date: Mon, 4 Jul 2005 16:06:19 +0000 (UTC) From: david20@alpha2.mdx.ac.uk9 Subject: Re: A tale of two macro libraries LIB vs STARLET ) Message-ID: <dabmpr$7tc$1@news.mdx.ac.uk>   [ In article <BEEEA334.1089C%roktsci@comcast.net>, Jeff Cameron <roktsci@comcast.net> writes:  >On 7/3/05 11:03 PM, in article C >daajge$24k$1@nwrdmz02.dmz.ncs.ea.ibs-infra.bt.com, "Richard Maher" % ><maher_rj@hotspamnotmail.com> wrote:  > 
 >> Hi Robert,  >>  0 >>> Don't put 3rd party stuff in Starlet or lib. >>   >> Why not? L >Very simply, if HP comes out with a patch, update ore upgrade that replacesH >the file, your  stuff goes away. Put it in your own library and make it  >available to those who need it. >   N Why is VMS engineering replacing these files rather than just merging updates  into each library ? N After all we can put third-party applications help information in the standardL help library and newer help information is just merged in when the system is upgraded to a new VMS version.  
 David Webb Security team leader CCSS Middlesex University   ------------------------------  * Date: Mon, 4 Jul 2005 16:23:17 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> 9 Subject: Re: A tale of two macro libraries LIB vs STARLET ? Message-ID: <dabnpk$se7$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>    Hi,   G > Why is VMS engineering replacing these files rather than just merging  updates  > into each library ?   D Ahhaah! Finally a dissenter. Good on ya David, I was giving up hope.  K Just what the hell *is* the "I" option on the Provide_Image Callback for if H it is not fully supported? So *is* IMAGELIB.OLB trashed everytime VMS isI upgraded? (Don't ask me; I don't know how to do it :-) Do you have to run L all these layered product "Library/Replace" scripts after post-install? DoesJ VMS engineering expect you re-install every layered product on your system after you upgrade VMS?   Regards Richard Maher.  + <david20@alpha2.mdx.ac.uk> wrote in message # news:dabmpr$7tc$1@news.mdx.ac.uk... ? > In article <BEEEA334.1089C%roktsci@comcast.net>, Jeff Cameron  <roktsci@comcast.net> writes: ! > >On 7/3/05 11:03 PM, in article E > >daajge$24k$1@nwrdmz02.dmz.ncs.ea.ibs-infra.bt.com, "Richard Maher" ' > ><maher_rj@hotspamnotmail.com> wrote:  > >  > >> Hi Robert,  > >>2 > >>> Don't put 3rd party stuff in Starlet or lib. > >>
 > >> Why not? E > >Very simply, if HP comes out with a patch, update ore upgrade that  replacesJ > >the file, your  stuff goes away. Put it in your own library and make it" > >available to those who need it. > >  > G > Why is VMS engineering replacing these files rather than just merging  updates  > into each library ? G > After all we can put third-party applications help information in the  standardK > help library and newer help information is just merged in when the system  is  > upgraded to a new VMS version. >  > David Webb > Security team leader > CCSS > Middlesex University >  >  >    ------------------------------  * Date: Mon, 4 Jul 2005 16:51:19 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> 9 Subject: Re: A tale of two macro libraries LIB vs STARLET ? Message-ID: <dabpe4$3j3$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>    Hi Forrest,   H >  I would treat imaglib like LIB and STARLET it belongs to the O.S. and > can get replaced.   L Just for curiosity's sake, when did this last happen? And, for example, whatH did you do with all the help calls from the SQL/Services API users whoseH apps just stopped linking? Rdb enginnering gets told and create a patch?  ; > If you stick a shareable in there you have to be prepared  > on an upgrade to replace it.  I How would you even begin to coordinate a "fix" to evry layered product on  the system?   / > In addition any visible symbols in there have ; > to be unique and not comflict with anything put in there.   = Yep. That's why we have the lovely Product Registry isn't it?    Product Id: TIER3 
 Facility: 333  Prefix: T3$  Registration# PRO908   Convention works!   G Anyway, as I've previously pointed out, I already ship my own shareable K image library so a macro library isn't a lot more skin off my beak, but I'd L just like to seperate any personal prejudices (and I have many) from what is the official VMS line on this.   Regards Richard Maher   A "Forrest Kenney" <forrest_dot_kenney@hp_dot.com> wrote in message $ news:42C947B9.1BDA8954@hp_dot.com...K >     I would treat imaglib like LIB and STARLET it belongs to the O.S. and E > can get replaced.  If you stick a shareable in there you have to be  preparedH > on an upgrade to replace it.  In addition any visible symbols in there haveL > to be unique and not comflict with anything put in there.  So my advice isJ > don't put it in there.  Yes I know that means folks have to link against your > shareable. > 	 > Forrest  >  >  >  > Richard Maher wrote: >  > > Hi Forrest,  > > L > > OK, I've got to make a decision on this (And it's already caused far tooI > > much grief for the tiny bit of functionality it provides) so it looks  like2 > > I'll ship a macro library and be done with it. > > J > > But before I close the book (and to give others a chance to reply) I'd justK > > like to ask you a question if I may. And that is "How do you feel about < > > shareables in IMAGELIB.OLB?" I.E. Can anyone tell me why (philisophically) L > > STARLET or LIB should be treated differently from IMAGELIB? in the "It's all  > > mine!" sense of things?  > >  > > Regards Richard Maher  > > E > > "Forrest Kenney" <forrest_dot_kenney@hp_dot.com> wrote in message ( > > news:42C93ECD.F3860439@hp_dot.com... > > > J > > >     1) Do not put your stuff in either LIB or STARLET they belong to > > OpenVMS  > > > group and 3 > > >         may be replaced or updated as needed.  > > > G > > >     2) Starlet is public definitions that are only supposed to be  changed 	 > > in an  > > > upward compatible K > > >          fashion.  LIB is mostly privileged semi-private definitions,  that > > may  > > > change to suit theK > > >          developers needs.  But typically if changed care is taken to  make > > it > > > upward compatible. > > > L > > >     3) The pseudo terminal driver stuff PTD$XXXX is a public interface and  > > is > > > documented in A > > >         the I/O users section of the OpenVMS documentation.  > > >  > > >  > > > Forrest Kenney
 > > > OpenVMS  > > >  > > > Richard Maher wrote: > > >  > > > > Hi Larry,  > > > > J > > > > > No, I believe that your definition (presuming T3 is a registeredG > > > > > facility name) should be T3$whateverDEF, with the dollar sign & > > > > > following the facility name. > > > > L > > > > Sounds good to me. Certainly makes more sense and is more consistent > > thatJ > > > > way, but I couldn't find any fac$def stuff in my libs. (But then I don't A > > > > have a lot of layered products on this box at the moment.  > > Pseudo-terminal @ > > > > driver seems to stand out.) Is this documented anywhere? > > > > L > > > > Also, I just re-read your earlier post and realized a big difference : -  > > > >  > > > > Starlet.MLB D > > > > > A. You can guarantee the definitions *WILL NOT* change for! > > > > >    a new version of VMS  > > > >  > > > > LIB.MLB E > > > > > A. You reasonably believe the definitions *WILL* change for ! > > > > >    a new version of VMS  > > > > L > > > > Why the difference? Are you saying that LIB gets replaced with a new > > version + > > > > of VMS and STARLET is just updated?  > > > > $ > > > > Thanks for the interest BTW. > > > >  > > > > Regards Richard Maher  > > > > B > > > > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message5 > > > > news:mxJwnKFnXUmN@eisner.encompasserve.org... I > > > > > In article <daal13$s4i$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>,  > > "Richard4 > > > > Maher" <maher_rj@hotspamnotmail.com> writes:	 > > > > >  > > > > > >> Only if both: > > > > > >>E > > > > > >> A. You can guarantee the definitions will not change for $ > > > > > >>    a new version of VMS > > > > > >>B > > > > > >> B. The names are all qualified by a VMS facility nameC > > > > > >>    (with dollar sign) registered to you by the Product  > > > > > >>    Registrar. > > > > > >> > > > > > > - > > > > > > It looks like we've got a winner!  > > > > > > A > > > > > > But I think I'll stick it in LIB anyway. Why? Well :-  > > > > > > I > > > > > > 1) Rightly or wrongly, I've always saw Starlet as the home of  the SS8 > > > > > > definitions and LIB as the symbol-def place. > > > > > > I > > > > > > 2) There's already a $TT3DEF macro in Starlet (terminal stuff  I'd  > > > > imagine)7 > > > > > > so a $T3DEF there as well may be confusing? 	 > > > > > J > > > > > No, I believe that your definition (presuming T3 is a registeredG > > > > > facility name) should be T3$whateverDEF, with the dollar sign & > > > > > following the facility name.	 > > > > > 6 > > > > > >> Do not expect customers to code in Macro. > > > > > > L > > > > > > Customers can do whatever they want! (At least that's the way it was  > > > > withI > > > > > > the VMS that I knew and loved. Before the "Use C you ignorant  scum!"J > > > > > > dictates were handed down.) However, I do not expect customers to > > code
 > > > > inJ > > > > > > Macro. As discussed in previous posts, these symbols will just let  > > > > customers K > > > > > > do binary ORs of flags at "compile" time. Here's your chance to  tell > > us > > > > how L > > > > > > ADA could do it :-) But COBOL certainly can't. If you don't care	 > > about  > > > > the J > > > > > > itsy bitsy CPU resources needed to do an OR at "runtime"? Then feel > > free
 > > > > to0 > > > > > > forget that the $T3DEF macro exists.	 > > > > > H > > > > > If it is just bit flags, I would think using SDL to output theI > > > > > definitions in multiple languages would be good, although there I > > > > > might not be an SDL back end for Cobol.  But writing one should L > > > > > not be hard, especially if you only supported the pieces you need.G > > > > > Shipping a Macro/Copybook/include/require/package file in all E > > > > > the languages would ensure the customer has what they need.  > > >  >    ------------------------------  % Date: Mon, 04 Jul 2005 13:36:04 -0400 4 From: Forrest Kenney <forrest_dot_kenney@hp_dot.com>9 Subject: Re: A tale of two macro libraries LIB vs STARLET * Message-ID: <42C97384.55027E2C@hp_dot.com>  2 1) Starlet and LIB have always belonged toOpenVMS.  L 2) The point about help is we told folks they can put stuff in there we take     pains to not break it   A 3)  IMAGELIB I am not 100% certain what the current or historical D     answer is.  The safe one it to not use.  The person who does the installationG     kits can give you a better answer about what steps are taken during 	 upgrades.      Forrest    Richard Maher wrote:  
 > Hi Forrest,  > J > >  I would treat imaglib like LIB and STARLET it belongs to the O.S. and > > can get replaced.  > N > Just for curiosity's sake, when did this last happen? And, for example, whatJ > did you do with all the help calls from the SQL/Services API users whoseJ > apps just stopped linking? Rdb enginnering gets told and create a patch? > = > > If you stick a shareable in there you have to be prepared   > > on an upgrade to replace it. > K > How would you even begin to coordinate a "fix" to evry layered product on 
 > the system?  > 1 > > In addition any visible symbols in there have = > > to be unique and not comflict with anything put in there.  > ? > Yep. That's why we have the lovely Product Registry isn't it?  >  > Product Id: TIER3  > Facility: 333 
 > Prefix: T3$  > Registration# PRO908 >  > Convention works!  > I > Anyway, as I've previously pointed out, I already ship my own shareable M > image library so a macro library isn't a lot more skin off my beak, but I'd N > just like to seperate any personal prejudices (and I have many) from what is  > the official VMS line on this. >  > Regards Richard Maher  > C > "Forrest Kenney" <forrest_dot_kenney@hp_dot.com> wrote in message & > news:42C947B9.1BDA8954@hp_dot.com...M > >     I would treat imaglib like LIB and STARLET it belongs to the O.S. and G > > can get replaced.  If you stick a shareable in there you have to be 
 > preparedJ > > on an upgrade to replace it.  In addition any visible symbols in there > haveN > > to be unique and not comflict with anything put in there.  So my advice isL > > don't put it in there.  Yes I know that means folks have to link against > your > > shareable. > >  > > Forrest  > >  > >  > >  > > Richard Maher wrote: > >  > > > Hi Forrest,  > > > N > > > OK, I've got to make a decision on this (And it's already caused far tooK > > > much grief for the tiny bit of functionality it provides) so it looks  > like4 > > > I'll ship a macro library and be done with it. > > > L > > > But before I close the book (and to give others a chance to reply) I'd > justM > > > like to ask you a question if I may. And that is "How do you feel about > > > > shareables in IMAGELIB.OLB?" I.E. Can anyone tell me why > (philisophically) N > > > STARLET or LIB should be treated differently from IMAGELIB? in the "It's > all  > > > mine!" sense of things?  > > >  > > > Regards Richard Maher  > > > G > > > "Forrest Kenney" <forrest_dot_kenney@hp_dot.com> wrote in message * > > > news:42C93ECD.F3860439@hp_dot.com... > > > > L > > > >     1) Do not put your stuff in either LIB or STARLET they belong to
 > > > OpenVMS  > > > > group and 5 > > > >         may be replaced or updated as needed.  > > > > I > > > >     2) Starlet is public definitions that are only supposed to be 	 > changed  > > > in an  > > > > upward compatible M > > > >          fashion.  LIB is mostly privileged semi-private definitions,  > that	 > > > may  > > > > change to suit theM > > > >          developers needs.  But typically if changed care is taken to  > make > > > it > > > > upward compatible. > > > > N > > > >     3) The pseudo terminal driver stuff PTD$XXXX is a public interface > and  > > > is > > > > documented in C > > > >         the I/O users section of the OpenVMS documentation.  > > > >  > > > >  > > > > Forrest Kenney > > > > OpenVMS  > > > >  > > > > Richard Maher wrote: > > > >  > > > > > Hi Larry, 	 > > > > > L > > > > > > No, I believe that your definition (presuming T3 is a registeredI > > > > > > facility name) should be T3$whateverDEF, with the dollar sign ( > > > > > > following the facility name.	 > > > > > N > > > > > Sounds good to me. Certainly makes more sense and is more consistent
 > > > thatL > > > > > way, but I couldn't find any fac$def stuff in my libs. (But then I > don't C > > > > > have a lot of layered products on this box at the moment.  > > > Pseudo-terminal B > > > > > driver seems to stand out.) Is this documented anywhere?	 > > > > > N > > > > > Also, I just re-read your earlier post and realized a big difference > : - 	 > > > > >  > > > > > Starlet.MLB F > > > > > > A. You can guarantee the definitions *WILL NOT* change for# > > > > > >    a new version of VMS 	 > > > > >  > > > > > LIB.MLB G > > > > > > A. You reasonably believe the definitions *WILL* change for # > > > > > >    a new version of VMS 	 > > > > > N > > > > > Why the difference? Are you saying that LIB gets replaced with a new
 > > > version - > > > > > of VMS and STARLET is just updated? 	 > > > > > & > > > > > Thanks for the interest BTW.	 > > > > >  > > > > > Regards Richard Maher 	 > > > > > D > > > > > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message7 > > > > > news:mxJwnKFnXUmN@eisner.encompasserve.org... K > > > > > > In article <daal13$s4i$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>,  > > > "Richard6 > > > > > Maher" <maher_rj@hotspamnotmail.com> writes: > > > > > >  > > > > > > >> Only if both: > > > > > > >>G > > > > > > >> A. You can guarantee the definitions will not change for & > > > > > > >>    a new version of VMS > > > > > > >>D > > > > > > >> B. The names are all qualified by a VMS facility nameE > > > > > > >>    (with dollar sign) registered to you by the Product  > > > > > > >>    Registrar. > > > > > > >>
 > > > > > > > / > > > > > > > It looks like we've got a winner! 
 > > > > > > > C > > > > > > > But I think I'll stick it in LIB anyway. Why? Well :- 
 > > > > > > > K > > > > > > > 1) Rightly or wrongly, I've always saw Starlet as the home of  > the SS: > > > > > > > definitions and LIB as the symbol-def place.
 > > > > > > > K > > > > > > > 2) There's already a $TT3DEF macro in Starlet (terminal stuff  > I'd  > > > > > imagine)9 > > > > > > > so a $T3DEF there as well may be confusing?  > > > > > > L > > > > > > No, I believe that your definition (presuming T3 is a registeredI > > > > > > facility name) should be T3$whateverDEF, with the dollar sign ( > > > > > > following the facility name. > > > > > > 8 > > > > > > >> Do not expect customers to code in Macro.
 > > > > > > > N > > > > > > > Customers can do whatever they want! (At least that's the way it > was  > > > > > withK > > > > > > > the VMS that I knew and loved. Before the "Use C you ignorant  > scum!"L > > > > > > > dictates were handed down.) However, I do not expect customers > to
 > > > code > > > > > inL > > > > > > > Macro. As discussed in previous posts, these symbols will just > let  > > > > > customers M > > > > > > > do binary ORs of flags at "compile" time. Here's your chance to  > tell > > > us
 > > > > > how N > > > > > > > ADA could do it :-) But COBOL certainly can't. If you don't care > > > about 
 > > > > > the L > > > > > > > itsy bitsy CPU resources needed to do an OR at "runtime"? Then > feel
 > > > free > > > > > to2 > > > > > > > forget that the $T3DEF macro exists. > > > > > > J > > > > > > If it is just bit flags, I would think using SDL to output theK > > > > > > definitions in multiple languages would be good, although there K > > > > > > might not be an SDL back end for Cobol.  But writing one should N > > > > > > not be hard, especially if you only supported the pieces you need.I > > > > > > Shipping a Macro/Copybook/include/require/package file in all G > > > > > > the languages would ensure the customer has what they need.  > > > >  > >    ------------------------------  % Date: Mon, 04 Jul 2005 20:28:25 +0800  From: prep@prep.synonet.com 7 Subject: Re: alpha instruction set - free to implement? - Message-ID: <873bqudac6.fsf@prep.synonet.com>   % "Tom Linden" <tom@kednos.com> writes:   @ > On 2 Jul 2005 12:23:05 -0700, mariuz <mapopa@gmail.com> wrote:  4 >> I have one question related to alpha architecture) >> Is alpha instruction set free to use ?   D > Why would you want to use it?  There are certainly better choices.   Such as?...    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Mon, 04 Jul 2005 09:13:55 -0700 # From: "Tom Linden" <tom@kednos.com> 7 Subject: Re: alpha instruction set - free to implement? ( Message-ID: <opstecphh3zgicya@hyrrokkin>  B On Mon, 04 Jul 2005 20:28:25 +0800, <prep@prep.synonet.com> wrote:  ' > "Tom Linden" <tom@kednos.com> writes:  > A >> On 2 Jul 2005 12:23:05 -0700, mariuz <mapopa@gmail.com> wrote:  > 5 >>> I have one question related to alpha architecture * >>> Is alpha instruction set free to use ? > E >> Why would you want to use it?  There are certainly better choices.  > 
 > Such as?...  > J Well, I like Mips, PPC, VAX.  I guess the OP asked because he was thinking of an Alpha emulator?    ------------------------------   Date: 4 Jul 2005 08:01:19 -0700 ! From: susan_skonetski@hotmail.com A Subject: Re: ANNOUNCEMENT - OpenVMS Technical Journal - June 2005 C Message-ID: <1120488312.926145.114720@g14g2000cwa.googlegroups.com>   E Thank you Richard, really the only thing I would like, is when I post F something like the technical journal which is a total volunteer effortD for it not to get trashed.  Folks from all over the world have spentF their time to help VMS and the message  gets lost, how?    But for anyF positive message to get covered by "BUT you did not do this" minimizesE the positives and makes it harder to get volunteers.  VMS Engineering G knows what our logo du jour has not done, probably better than you, but F we also know complaining about it here is not going to fix it.  And it> seems to be human nature that a negative message will get moreG discussion, more interest and I know that SUN and IBM troll here.  What E folks also have to know is that for every one negative message posted D here is the 100 email messages I get having to explain and its neverF ever the good ones.  Now I can spend my time helping or explaining you guess what I would rather do.   F Complain or do something.  Have you mentored a person under 25 on VMS,C don't give me an excuse on why not just a yes or no.  Are you still E fighting, no excuses, you love VMS or you don't.  Are you a hobbyiest G cool, do you show people what you can do?  Have you written an article, G submitted an OpenVMS Pearl?  Have you taken part in this forum? Are you D an Encompass memeber, yes they have free membership? Are you part ofG ITRC? Are you VMS Certified Are you a Partner? Are you a LUG member? Do G you go to meetings?  There are all kinds of questions, its real easy to E sit and type in a newsgroup and blame everyone else for not making it E easy for you.  Notice that I did not put things like when did you buy F anything, what is your service level contract.  Its not up to THEM its up to US   sue    ------------------------------  * Date: Mon, 4 Jul 2005 16:57:03 +0000 (UTC)3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> A Subject: Re: ANNOUNCEMENT - OpenVMS Technical Journal - June 2005 ? Message-ID: <dabpot$4g0$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>    Hi Sue,   J Do you want to take the time to point out that paragragh 2 was not pointed solely or directly at me?    I can do angry too.    Regards Richard Maher   . <susan_skonetski@hotmail.com> wrote in message= news:1120488312.926145.114720@g14g2000cwa.googlegroups.com... G > Thank you Richard, really the only thing I would like, is when I post H > something like the technical journal which is a total volunteer effortF > for it not to get trashed.  Folks from all over the world have spentH > their time to help VMS and the message  gets lost, how?    But for anyH > positive message to get covered by "BUT you did not do this" minimizesG > the positives and makes it harder to get volunteers.  VMS Engineering I > knows what our logo du jour has not done, probably better than you, but H > we also know complaining about it here is not going to fix it.  And it@ > seems to be human nature that a negative message will get moreI > discussion, more interest and I know that SUN and IBM troll here.  What G > folks also have to know is that for every one negative message posted F > here is the 100 email messages I get having to explain and its neverH > ever the good ones.  Now I can spend my time helping or explaining you > guess what I would rather do.  > H > Complain or do something.  Have you mentored a person under 25 on VMS,E > don't give me an excuse on why not just a yes or no.  Are you still G > fighting, no excuses, you love VMS or you don't.  Are you a hobbyiest I > cool, do you show people what you can do?  Have you written an article, I > submitted an OpenVMS Pearl?  Have you taken part in this forum? Are you F > an Encompass memeber, yes they have free membership? Are you part ofI > ITRC? Are you VMS Certified Are you a Partner? Are you a LUG member? Do I > you go to meetings?  There are all kinds of questions, its real easy to G > sit and type in a newsgroup and blame everyone else for not making it G > easy for you.  Notice that I did not put things like when did you buy H > anything, what is your service level contract.  Its not up to THEM its
 > up to US >  > sue  >    ------------------------------   Date: 4 Jul 2005 04:49:35 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) # Subject: Re: CTRL-T - what's the T? 3 Message-ID: <g3GAmIBU7aS1@eisner.encompasserve.org>   u In article <daahqq$kuh$1@nwrdmz01.dmz.ncs.ea.ibs-infra.bt.com>, "Richard Maher" <maher_rj@hotspamnotmail.com> writes:  > Hi,  >  >> "Why use a key for * >> status (Ctrl-T) that is directly beside& >> the key for 'Interrupt' (Ctrl-Y) ?" > 8 > God bless CONTINUE! What version did that come out in?  9 I thought it was there from the beginning.  The nature of = Control/Y is that it does not do image rundown.  The VAX DBMS = developers (actually KODA developers) always talked not about 0 handling Ctrl/Y, but about handling Ctrl/Y,STOP.   ------------------------------  % Date: Mon, 04 Jul 2005 11:09:54 +0200 % From: Harald Pollak <h.pollak@pke.at>  Subject: GCC on OpenVMS G Message-ID: <42c8fcb9$0$31228$91cee783@newsreader02.highway.telekom.at>   J I thought there was a port from GCC on OpenVMS ( i'm shure there was one )I but now I can't find on the net ( I think it was on vms.gnu.org - but the  site now unreachable) & Does anyone know what happens with it?" or where I can find it ( Gcc 3.x )  
 best regards   Harry    ------------------------------  % Date: Mon, 04 Jul 2005 14:47:20 +0200 0 From: Keith Cayemberg <keith.cayemberg@arcor.de> Subject: Re: GCC on OpenVMS B Message-ID: <42c92fd8$0$10802$9b4e6d93@newsread4.arcor-online.net>   Harald Pollak wrote:L > I thought there was a port from GCC on OpenVMS ( i'm shure there was one )K > but now I can't find on the net ( I think it was on vms.gnu.org - but the  > site now unreachable) ( > Does anyone know what happens with it?$ > or where I can find it ( Gcc 3.x ) >  > best regards   > Harry     I Please read the following COV Threads. To my knowledge the situation has   not changed since then.   U http://groups-beta.google.com/group/comp.os.vms/browse_thread/thread/4b5c627029f54f8d   U http://groups-beta.google.com/group/comp.os.vms/browse_thread/thread/e9167d05ab23bee8      OpenVMS FAQ 1 http://h71000.www7.hp.com/faq/vmsfaq_018.html#gcc     ! gcc 2.7.1 for VAX - planet mirror + http://public.planetmirror.com/pub/gcc-vms/    gcc 2.8.0 for Alpha - ctrl-c7 http://www.ctrl-c.liu.se/ftp/wku/vms/gcc-for-alpha/7_1/   - gcc 2.8.0 for OpenVMS/Alpha 6.2 and 7.1 - tmk 2 http://www.tmk.com/ftp/vms-freeware/gcc-for-alpha/  : gcc 2.8.1 for OpenVMS Alpha - used within GNAT-3.12p - NYU2 ftp://ftp.cs.nyu.edu/pub/gnat/private/old/openvms/  6 Nick Clifton - Re: Removal of VAX/VMS support from gcc; http://sources.redhat.com/ml/binutils/2003-07/msg00614.html   I VAX/VMS code has been obsoleted in the current GCC source tree - GCC 3.3   Release ' http://gcc.gnu.org/gcc-3.3/changes.html   / Updated header-files for GNU C 2.8.1 - Malmberg & ftp://ftp.qsl.net/pub/wb8tyw/gcc281_u/   Help GCC - SUNY NCSB3 http://nucwww.chem.sunysb.edu/helplib/@hvmsapps/GCC    Cheers!    Keith Cayemberg    ------------------------------  $ Date: Mon, 4 Jul 2005 21:28:43 +1000# From: "Gremlin" <not-here@all.mate> > Subject: Re: Network card problems? Or something like that..../ Message-ID: <42c91d6c$1@duster.adelaide.on.net>   	 Hi Martin   L Yes, very bad luck - it is a Netgear 24 port unmanaged switch and currently G supports autonegotiate for a HP Netserver, 5 Wintel boxes, 2 DS10Ls, a  J couple of Linux boxes and a Solaris box.  Only the AlphaServer is fussy - K the remainder are quite happy.  Setting it to any speed or duplex makes no  J difference, let alone auto-negotiate - and I have put two different cards  in, just in case.  Sad........   Thanks    @ "Martin Vorlaender" <martin@radiogaga.harz.de> wrote in message 5 news:42c8615a.524144494f47414741@radiogaga.harz.de... $ > Gremlin <not-here@all.mate> wrote:& >> set ewa0_mode without any arguments >> shows twisted-pair, >  > 10 Mbit half duplex  >  >> full duplex twisted pair, >  > 10 Mbit full duplex  >  >> aui, bnc, >> fast, >  > 100 Mbit half duplex > 	 >> fastfd  >  > 100 Mbit full duplex >  >> and> >> auto-negotiate.  When I try autonegotiate it won't allow it >  > Odd! > " >> and changes it to twisted pair. > * > ...which would mean 10 Mbit half duplex. > 6 >> This results in a VMS LANCP setting same as before, > % > So LANCP says it's set to 100 Mbit?  >  >> but no better connectivity! > I > Of course not. It's crucial to understand that auto-negotiate will only H > work if *both* sides are set to use it. Otherwise *both* sides must beH > set to the same fixed speed and duplex setting. If your switch doesn't> > allow you to configure that, and you can't set your ewa0 to  > auto-negotiate,  > it's very bad luck.  >  > cu, 	 >  Martin  > --  C > So long, and thanks        | Martin Vorlaender  |  OpenVMS rules! 6 > for all the books...       | work: mv@pdv-systeme.deI > In Memoriam Douglas Adams  |   http://www.pdv-systeme.de/users/martinv/ = >            1952-2001      | home: martin@radiogaga.harz.de     ------------------------------  * Date: Mon, 4 Jul 2005 11:55:29 +0000 (UTC)% From: Dan Foster <usenet@evilphb.org> > Subject: Re: Network card problems? Or something like that....6 Message-ID: <slrndci8tt.k0o.usenet@gaia.roc2.gblx.net>  R In article <42c91d6c$1@duster.adelaide.on.net>, Gremlin <not-here@all.mate> wrote: > N > Yes, very bad luck - it is a Netgear 24 port unmanaged switch and currently  > supports autonegotiate  G I had to order a managed switch, set the port to be fixed, and make the H PWS 600au match, to avoid all of the problems. (I have a Netgear 16 port> unmanaged switch, but had to buy a Cisco 2900 managed switch.)  G Not my preferred solution, but I couldn't find anything cheaper, except A for plugging in the PWS 600au directly to another machine using a  crossover UTP Ethernet cable.    -Dan   ------------------------------  $ Date: Mon, 4 Jul 2005 22:51:53 +1000# From: "Gremlin" <not-here@all.mate> > Subject: Re: Network card problems? Or something like that..../ Message-ID: <42c930ed$1@duster.adelaide.on.net>    Hi All  I After a lot of fiddling, I have found that, with this box only - and two  M different cards of the same model, the only way to get it to work is to tell  F SRM that it is FastFD but tell LANCP that it is /nofull/speed=10 - it H selects TP half duplex 10Mbit and works - no errors, no packet loss etc.  L Very confusing but at least operational - will be very slow with ftp though  :)  ( Thanks for all your thoughts and help...    6 "Richard Brodie" <R.Brodie@rl.ac.uk> wrote in message , news:dabauu$79f$1@blackmamba.itd.rl.ac.uk... > 1 > "Gremlin" <not-here@all.mate> wrote in message  + > news:42c91d6c$1@duster.adelaide.on.net...  > E >> Yes, very bad luck - it is a Netgear 24 port unmanaged switch and   >> currentlyI >> supports autonegotiate for a HP Netserver, 5 Wintel boxes, 2 DS10Ls, a L >> couple of Linux boxes and a Solaris box.  Only the AlphaServer is fussy -! >> the remainder are quite happy.  > M > If autonegotiate isn't working, then just setting it to Fast (half-duplex)   > is theE > easiest/cheapest thing to do. You're going to lose a little bit of  
 > performance B > compared with FastFD but be far better off than with a mismatch. > L > Having said that, it's highly unlikely that this is your main problem. If  > you K > lost 10% of your PING replies, maybe, but I assume you're not getting any H > back. Have a look at the LANCP counters, or with tcpdump from a remoteK > system, see if there are any packets going on the wire. If so, maybe you   > have! > a TCP/IP configuration problem.  >  >    ------------------------------  $ Date: Mon, 4 Jul 2005 13:44:14 +0100* From: "Richard Brodie" <R.Brodie@rl.ac.uk>> Subject: Re: Network card problems? Or something like that....2 Message-ID: <dabauu$79f$1@blackmamba.itd.rl.ac.uk>  X "Gremlin" <not-here@all.mate> wrote in message news:42c91d6c$1@duster.adelaide.on.net...  M > Yes, very bad luck - it is a Netgear 24 port unmanaged switch and currently H > supports autonegotiate for a HP Netserver, 5 Wintel boxes, 2 DS10Ls, aK > couple of Linux boxes and a Solaris box.  Only the AlphaServer is fussy -   > the remainder are quite happy.  Q If autonegotiate isn't working, then just setting it to Fast (half-duplex) is the N easiest/cheapest thing to do. You're going to lose a little bit of performance@ compared with FastFD but be far better off than with a mismatch.  M Having said that, it's highly unlikely that this is your main problem. If you I lost 10% of your PING replies, maybe, but I assume you're not getting any F back. Have a look at the LANCP counters, or with tcpdump from a remoteM system, see if there are any packets going on the wire. If so, maybe you have  a TCP/IP configuration problem.    ------------------------------  $ Date: Mon, 4 Jul 2005 08:59:09 -0400' From: "Main, Kerry" <kerry.main@hp.com> > Subject: RE: Network card problems? Or something like that....R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB650FEC@tayexc19.americas.cpqcorp.net>   > -----Original Message------ > From: Gremlin [mailto:not-here@all.mate]=20  > Sent: July 4, 2005 7:29 AM > To: Info-VAX@Mvb.Saic.Com @ > Subject: Re: Network card problems? Or something like that.... >=20 > Hi Martin  >=20B > Yes, very bad luck - it is a Netgear 24 port unmanaged switch=20 > and currently=20A > supports autonegotiate for a HP Netserver, 5 Wintel boxes, 2=20  > DS10Ls, a=207 > couple of Linux boxes and a Solaris box.  Only the=20  > AlphaServer is fussy -=20 ? > the remainder are quite happy.  Setting it to any speed or=20  > duplex makes no=20> > difference, let alone auto-negotiate - and I have put two=20 > different cards=20  > in, just in case.  Sad........ >=20 > Thanks >=20 >=20  H Just curious, but what version of the OS and patch levels are you using?    + Also, what type of NIC cards are you using?   H As I recall, there was some OpenVMS eco's to LAN drivers awhile ago that! fixed some auto-negotiate issues.   A Note - from what I have heard, there used to be a big gap in what G various network vendors viewed as the "standard" for how auto-negotiate B should have been handled. While most recent network devices do notC typically have problems, many older switches have had compatibility   issues with different NIC cards.   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------   Date: 4 Jul 2005 03:12:05 -0700 + From: "Benji Mouse" <ben.burke-eds@eds.com> & Subject: Re: OpenVMS and XML questionsC Message-ID: <1120471925.896834.299760@o13g2000cwo.googlegroups.com>    Chris,  D I have had a pretty easy time with libxml2 - it really is an amazingA library and many, many helpful people online answering questions.   A And the kit I downloaded came with a dcl script to build on VMS - E worked fine (unless you have vaxes to consider;-( then it needed some  help)   C The problems you mention a couple of replies back sounds related to A /NAMES=AS_IS settings for decc and libxml - how did you build the  library?  C In terms of doco- what's there for Libxml2 is fine, once you get to E know it. In the mean time, check out the following article for a good  summary.   http://groups.google.com.au/group/comp.sources.d/browse_thread/thread/2c06ee888344f7e5/243cf63d60cbb2d0?q=xmlTextReaderPreservePattern&rnum=1&hl=en#243cf63d60cbb2d0   regards, bb   ------------------------------  % Date: Mon, 04 Jul 2005 17:23:01 +0200 - From: Didier Morandi <prenom.nom@freesurf.fr> - Subject: Re: The VMS Gateway FAQ is available 4 Message-ID: <42c9545a$0$32335$636a15ce@news.free.fr>   David J Dachtera wrote:    > Didier Morandi wrote:  > 1 >>there: http://www.vmsgateway.net/mig_corner.htm  >  > + > Blank white page in Netscape V4.77 on W2K   F and here: http://www.didiermorandi.com/vmsgateway/vms_gateway_faq.htm?   D.   ------------------------------  % Date: Mon, 04 Jul 2005 11:57:57 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> - Subject: Re: The VMS Gateway FAQ is available , Message-ID: <42C95C7A.CAF08855@teksavvy.com>   Didier Morandi wrote: 3 > >>there: http://www.vmsgateway.net/mig_corner.htm  > >  > > - > > Blank white page in Netscape V4.77 on W2K   R http://validator.w3.org/check?uri=http%3A%2F%2Fwww.vmsgateway.net%2Fmig_corner.htm  L Didier, you may wish to click on the above. It will tell you about errors onM your page. Once you add the proper DTD tag to tell it it is a fame definition M document, it will then parse your frameset and tell you you're missing ending  tags for <frame>  I Also, you have a frameset at mig_corner.htm which contains only one frame , which tries to load itself inside the frame.       > H > and here: http://www.didiermorandi.com/vmsgateway/vms_gateway_faq.htm?    g http://validator.w3.org/check?uri=http%3A%2F%2Fwww.didiermorandi.com%2Fvmsgateway%2Fvms_gateway_faq.htm   N The errors in this one are fairly straightforward. And most browsers can still render the contents.   ------------------------------  $ Date: Mon, 4 Jul 2005 07:16:33 -0400# From: "John Smith" <a@nonymous.com> , Subject: Re: US military and confidentiality4 Message-ID: <42c91ba9$1_1@spool9-west.superfeed.net>   Kerry,  J I ttruly appreciate each an every item you post, including the post below.  K One might say that progress is measured in steps, one ISV at a time. Yet HP H has known that VMS was to be launched on Itanic for how many *years* and, this is all we have to show for it ISV-wise?  K There should have been a flood of ISV's announce who have committed to, who K are actively, or who have completed their ports by now, not the 50-100 that  we have.  G The DSPP $2k program is just as much Intel/HP-UX/Linux as it is VMS, so K viewed a bit differently, out of each dollar in promoting this program only J about $0.25 goes towards VMS. How about a program that is 100 cents on theL dollar VMS-centric? I'd be happy with DSPP as it is right now + 75 cents per/ dollar towards VMS-dedicated advertising by HP.    Main, Kerry wrote: >> -----Original Message----- + >> From: John Smith [mailto:a@nonymous.com]  >> Sent: July 3, 2005 5:55 PM  >> To: Info-VAX@Mvb.Saic.Com/ >> Subject: Re: US military and confidentiality  >> >  > [Snip ...] >  >>D >> Q. When is HP going to advertise VMS's robustness in this regard? >>G >> A. There aren't enough COTS applications left on VMS for the average ? >> prospect to consider using VMS in the first place. Don't you  >> find it ironic ? >> that HP keeps saying it's the applications that sell VMS yet 
 >> they do no @ >> advertising to help convince ISV's that building apps for the >> VMS market is? >> a good use of the ISV's money? So the answer is why bother -  >> for both the  >> ISV and HP. >> >> >  > John,  > H > Re: HP working with ISV's and Customers on moving to OpenVMS Itanium -H > not sure if this was previously reported on the newsgroup, but just in; > case: (I believe Wayne posts occasionally here on c.o.v.)  > = > Just happened to be on the DSPP pages and noticed this url: J > http://h21007.www2.hp.com/dspp/bus/bus_BusDetailPage_IDX/1,1252,7721,00. > htmlF > "Winning with DSPP: First OpenVMS VAX to OpenVMS on Integrity Server% > Port Completed at a Workshop Event"  > B > "The HP & Intel(r) Developer Workshop, held April 26-28 in TampaB > Florida, was the site of the first OpenVMS VAX to OpenVMS I64 onB > Intel(r) Itanium(r) port to be completed at a Developer WorkshopF > event. Wayne Seifert, System Developer for Lieberman Technologies ofC > Evansville Indiana, completed the port within two days during the G > event. The event was available to companies in the HP DSPP program as  > a member-only benefit. > H > Mr. Seifert attended the workshop with the goal of porting and testingB > two of the company's applications as well as testing an array ofE > general utilities and subroutines.  "The workshop was superb. I was C > very pleased with how smooth the porting process was and with the C > level of support provided at the workshop".  After completing the F > ports, Mr. Seifert tested the HP Integrity server by running severalE > standard reports. Two reports, each of which had typically taken 45 G > minutes to an hour to complete, ran in less than one minute on the HP  > Integrity server.  > D > Lieberman Technologies provides consulting, design and developmentA > services to small and medium businesses.  In addition to custom C > software development, the company offers turnkey applications for ? > health center billing systems, scheduling packages and retail E > applications. Lieberman Technologies recently migrated all of their < > internal applications to Itanium, and is in the process ofE > transitioning all of their clients to HP Itanium Integrity servers. D > More information about Lieberman Technologies at http://LTnow.com" >  > [snip ..]  > G > Btw, they have nice web page for moving from VAX/Alpha to Integrity - # > click on link on their home page.  > 	 > Regards  >  > Kerry Main > Senior Consultant  > HP Services Canada > Voice: 613-592-4660  > Fax: 613-591-4477  > kerryDOTmainAThpDOTcom > (remove the DOT's and AT)  > 6 > OpenVMS - the secure, multi-site OS that just works.   --F OpenVMS - The never-advertised operating system with the dwindling ISV base.       O ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  % Date: Mon, 04 Jul 2005 15:40:13 +0200  From: S <soterroatyahoodotcom>, Subject: Re: US military and confidentiality& Message-ID: <42c93c38$1@news1.ethz.ch>   John Smith wrote: I > The DSPP $2k program is just as much Intel/HP-UX/Linux as it is VMS, so M > viewed a bit differently, out of each dollar in promoting this program only L > about $0.25 goes towards VMS. How about a program that is 100 cents on theN > dollar VMS-centric? I'd be happy with DSPP as it is right now + 75 cents per1 > dollar towards VMS-dedicated advertising by HP.   G A major surprise for me was to see the latest DSPP newsletter starting   with a VMS story. L On the other hand, VMS doesn't have its own section in the newsletter, so...   S    ------------------------------  % Date: Mon, 04 Jul 2005 10:12:20 -0400 2 From: "Stanley F. Quayle" <squayle@insight.rr.com>, Subject: Re: US military and confidentiality. Message-ID: <42C90B84.25384.8E194D8@localhost>  + On 3 Jul 2005 at 18:38, David Froble wrote: G > How are windoz systems attacked?  Usually via the network interface?  G > What if you're running the system without a network interface for the  > windoz OS? > C > Note, it's my very limited understanding that the emulator runs a H > seperate network interface, which windoz doesn't see.  Could be wrong.  B Most SIMH installations use pcap to set the port into promiscuous B mode.  Incoming packets are sorted by the host OS before going to F either the host stack(s) or the emulator.  This allows the host to be 	 attacked.   A In CHARON-VAX, a separate network interface is required.  A NDIS  D driver is used to send all incoming packets to the emulated VAX.    > This requires that no other protocols are running on the card.  F Some customers elect to connect the "Windows-side" interface to their + network for remote management.  Some don't.   E As for the frequent Windows patches, they're not required unless you   expose the host to a network.   
 --Stan Quayle  Quayle Consulting Inc.  
 ----------- Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 3 8572 North Spring Ct., Pickerington, OH  43147  USA 0 stan-at-stanq-dot-com       http://www.stanq.com) "OpenVMS, when downtime is not an option"    ------------------------------  $ Date: Mon, 4 Jul 2005 10:26:04 -0400' From: "Main, Kerry" <kerry.main@hp.com> , Subject: RE: US military and confidentialityR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB650FEF@tayexc19.americas.cpqcorp.net>   > -----Original Message-----A > From: S [mailto:soterroatyahoodotcom@ccerelrim02.cce.hp.com]=20  > Sent: July 4, 2005 9:40 AM > To: Info-VAX@Mvb.Saic.Com . > Subject: Re: US military and confidentiality >=20 > John Smith wrote: @ > > The DSPP $2k program is just as much Intel/HP-UX/Linux as=20 > it is VMS, so @ > > viewed a bit differently, out of each dollar in promoting=20 > this program only @ > > about $0.25 goes towards VMS. How about a program that is=20 > 100 cents on the@ > > dollar VMS-centric? I'd be happy with DSPP as it is right=20 > now + 75 cents per3 > > dollar towards VMS-dedicated advertising by HP.  >=20B > A major surprise for me was to see the latest DSPP newsletter=20
 > starting=20  > with a VMS story. ? > On the other hand, VMS doesn't have its own section in the=20  > newsletter, so...  >=20 > S  >=20  1 Actually, it does further down in the newsletter.    :-)   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Mon, 04 Jul 2005 09:05:37 -0700 # From: "Tom Linden" <tom@kednos.com> , Subject: Re: US military and confidentiality( Message-ID: <opstecbnvzzgicya@hyrrokkin>  7 On Mon, 04 Jul 2005 10:12:20 -0400, Stanley F. Quayle    <squayle@insight.rr.com> wrote:   - > On 3 Jul 2005 at 18:38, David Froble wrote: G >> How are windoz systems attacked?  Usually via the network interface? H >> What if you're running the system without a network interface for the
 >> windoz OS?  >>D >> Note, it's my very limited understanding that the emulator runs aI >> seperate network interface, which windoz doesn't see.  Could be wrong.  > C > Most SIMH installations use pcap to set the port into promiscuous C > mode.  Incoming packets are sorted by the host OS before going to G > either the host stack(s) or the emulator.  This allows the host to be  > attacked.    That should clearly be avoided.  > B > In CHARON-VAX, a separate network interface is required.  A NDISB > driver is used to send all incoming packets to the emulated VAX.@ > This requires that no other protocols are running on the card. > G > Some customers elect to connect the "Windows-side" interface to their - > network for remote management.  Some don't.  > F > As for the frequent Windows patches, they're not required unless you > expose the host to a network.  >  > --Stan Quayle  > Quayle Consulting Inc. >  > ----------/ > Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 5 > 8572 North Spring Ct., Pickerington, OH  43147  USA 2 > stan-at-stanq-dot-com       http://www.stanq.com+ > "OpenVMS, when downtime is not an option"  >  >    ------------------------------   Date: 04 Jul 2005 16:51:47 GMT$ From: "Doc." <doc@openvms-rocks.com>, Subject: Re: US military and confidentiality7 Message-ID: <Xns9689C00605F9Adcovmsrox@212.100.160.126>   C %NEWS-I-NEWMSG, Tom Linden wrote in news:opstecbnvzzgicya@hyrrokkin    <snip>  D >> Most SIMH installations use pcap to set the port into promiscuousD >> mode.  Incoming packets are sorted by the host OS before going toH >> either the host stack(s) or the emulator.  This allows the host to be >> attacked. > ! > That should clearly be avoided.   H I suppose it depends on what you're doing.  If you're a Hobbyist who is I just getting started with VMS, you're going to want to do that as you'll  H also want to use the underlying OS.  At Deathrow we've seen quite a few J people get curious from non-privved access, move on to SIM-H to play with J elevated privileges, then buy second hand hardware to experience the real  thing.  D Obviously, as Charon is a solution for use in a commercial setting, 2 you've a little more budget to do things properly.     Doc. --  G OpenVMS:     Eight out of ten hackers prefer *other* operating systems. G http://www.openvms-rocks.com    Deathrow Public-Access OpenVMS Cluster.    ------------------------------  % Date: Mon, 04 Jul 2005 10:07:26 -0700 # From: "Tom Linden" <tom@kednos.com> , Subject: Re: US military and confidentiality( Message-ID: <opstee6opgzgicya@hyrrokkin>  @ On 04 Jul 2005 16:51:47 GMT, Doc. <doc@openvms-rocks.com> wrote:  E > %NEWS-I-NEWMSG, Tom Linden wrote in news:opstecbnvzzgicya@hyrrokkin  >  > <snip> > E >>> Most SIMH installations use pcap to set the port into promiscuous E >>> mode.  Incoming packets are sorted by the host OS before going to I >>> either the host stack(s) or the emulator.  This allows the host to be 
 >>> attacked.  >>" >> That should clearly be avoided. > I > I suppose it depends on what you're doing.  If you're a Hobbyist who is J > just getting started with VMS, you're going to want to do that as you'llI > also want to use the underlying OS.  At Deathrow we've seen quite a few K > people get curious from non-privved access, move on to SIM-H to play with K > elevated privileges, then buy second hand hardware to experience the real  > thing.  H Hobbyists are one thing, but of course, they don't directly generate anyI revenue.  I haven't used SIMH, but i got the impression from the way Stan H worded his response that you didn't have to run the nic in promiscuous   mode.   K The emulator I am using under Linux provides me with an environment no less  secure than a real VAX.    > E > Obviously, as Charon is a solution for use in a commercial setting, 4 > you've a little more budget to do things properly. >  >  > Doc.   ------------------------------    Date: 04 Jul 2005 11:33:28 +0200( From: Andreas Davour <ante@update.uu.se>% Subject: Re: [Announce] FreeVMS 0.1.7 4 Message-ID: <cs9d5py6hlj.fsf@Psilocybe.Update.UU.SE>  % "Tom Linden" <tom@kednos.com> writes:   L > >>> So, who actually uses FreeVMS, or would be interested in doing so, andK > >>> for what?  Other than hacker types, of course, just doing it for fun. H > >>   Fancy an operating system feeling like VMS, but on an x86?  Where > >> if  somethingD > >> doesn't work like you like, you can change it yourself (even if > >> most  users4 > >> probably wouldn't make use of this feature ;-))
 > >>  Thierry  > J > How about OpenSolaris?  Anecdotally it seems few if any would modify it.L > http://www.informationweek.com/story/showArticle.jhtml?articleID=164902561  @ Well, it has only been out in the open a few weeks. Give it some5 time. Even the unix habit of splitting up takes time.    /andreas   --  A A: Because it fouls the order in which people normally read text. ' Q: Why is top-posting such a bad thing?  A: Top-posting. ; Q: What is the most annoying thing on usenet and in e-mail?    ------------------------------  % Date: Mon, 04 Jul 2005 19:51:59 +0800  From: prep@prep.synonet.com % Subject: Re: [Announce] FreeVMS 0.1.7 - Message-ID: <87br5idc0w.fsf@prep.synonet.com>   ) Glenn Everhart <Everhart@gce.com> writes:    > Uses for FreeVMS?   E >    If it should become popular, it could provide a way to get a LOT @ > of applications running from that system service set. That theF > system services have been designed to be more secure than most otherB > OSs will help, and will be a more effective tool for selling theA > commercial built-like-a-battleship VMS code than most any other ! > measure I see being undertaken.   D And as part of the Free VMS work, CLI$, RMS$, descriptors, and other@ stuff could be implemented on the various U boxen so as to raise the quality bar.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   End of INFO-VAX 2005.370 ************************