0 INFO-VAX	Sun, 05 Feb 2006	Volume 2006 : Issue 72      Contents: A good day for VMS Re: A good day for VMS Re: A good day for VMS Re: A good day for VMS Re: A good day for VMS Re: C  History Re: C  History Re: DEC PWS433au memory   F ----------------------------------------------------------------------  % Date: Sun, 05 Feb 2006 08:34:28 +0000 ! From: "R.A.Omond" <Roy@Omond.net>  Subject: A good day for VMS 4 Message-ID: <ds4dal$kn1$1$8302bc10@news.demon.co.uk>  ; I am currently involved in setting up a "farm" of 12 DS15's @ at a very large customer in London (UK).  The same configuration; is likely to be duplicated at various other sites round the  globe.  5 This is new business, not a replacement for anything.   	 Roy Omond  Blue Bubble Ltd.   ------------------------------  % Date: Sun, 05 Feb 2006 04:00:16 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: A good day for VMS , Message-ID: <43E5BE9D.BE99CEB9@teksavvy.com>   "R.A.Omond" wrote: > = > I am currently involved in setting up a "farm" of 12 DS15's B > at a very large customer in London (UK).  The same configuration= > is likely to be duplicated at various other sites round the  > globe.  F Could you describe what type of application will be deployed on this ?! What systems is VMS replacing ?     > Is this one of the remaining "market niches" identified by VMSG management, or a customer that doesn't fit the mold that VMS management  has set for VMS customers ?    ------------------------------  $ Date: Sun, 5 Feb 2006 19:17:50 +02007 From: "Guy Peleg" <guy.peleg@remove_this_header@hp.com>  Subject: Re: A good day for VMS * Message-ID: <43e63342@usenet01.boi.hp.com>  , "R.A.Omond" <Roy@Omond.net> wrote in message. news:ds4dal$kn1$1$8302bc10@news.demon.co.uk...= > I am currently involved in setting up a "farm" of 12 DS15's B > at a very large customer in London (UK).  The same configuration= > is likely to be duplicated at various other sites round the  > globe. > 7 > This is new business, not a replacement for anything.  >  > Roy Omond  > Blue Bubble Ltd.   This is great news !   Thank you for sharing.  	 Guy Peleg  OpenVMS Engineering    ------------------------------  % Date: Sun, 05 Feb 2006 12:14:40 -0500 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: A good day for VMS / Message-ID: <r-adnR2n18HLr3veRVn-ig@libcom.com>    R.A.Omond wrote:= > I am currently involved in setting up a "farm" of 12 DS15's B > at a very large customer in London (UK).  The same configuration= > is likely to be duplicated at various other sites round the  > globe. > 7 > This is new business, not a replacement for anything.  >  > Roy Omond  > Blue Bubble Ltd.  , Better grab those 'likely' DS15s real quick.   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Sun, 05 Feb 2006 17:11:10 +0000 / From: "R.A.Omond" <Roy.Omond@BlueBubble.UK.Com>  Subject: Re: A good day for VMS 4 Message-ID: <ds5bjf$jjq$1$8302bc10@news.demon.co.uk>   JF Mezei wrote:  > "R.A.Omond" wrote: > = >>I am currently involved in setting up a "farm" of 12 DS15's B >>at a very large customer in London (UK).  The same configuration= >>is likely to be duplicated at various other sites round the  >>globe. >  > H > Could you describe what type of application will be deployed on this ?  % Sorry, I'm not at liberty to do this.   # > What systems is VMS replacing ?     @ As Larry K has already pointed out, it's not replacing anything.  @ > Is this one of the remaining "market niches" identified by VMSI > management, or a customer that doesn't fit the mold that VMS management  > has set for VMS customers ?   E This is a customer with a huge investment in VMS systems.  VMS drives G the world-wide backbone of this company.  Sure, I hear mutterings about C VMS on its way out (been going on for years), but the company still ! continues to buy new VMS systems.    ------------------------------  $ Date: Sun, 5 Feb 2006 16:01:14 +08003 From: "Richard Maher" <maher_rj@hotspamnotmail.com>  Subject: Re: C  History 1 Message-ID: <ds4bce$115$1@news-02.connect.com.au>    Gelnn, Glenn, Glenn,  G Your pitiful attempts to rationalize your Faustian bargain, betray your J growing discomfort with the price. Too late! The deal has been struck; the price will be paid :-)   Cheers Richard Maher  , PS. Oh yes. There were *many* other choices.  4 "Glenn Everhart" <Everhart@gce.com> wrote in message* news:11uakes22b8jd48@corp.supernews.com... > Richard Maher wrote: > > Hi Tom,  > > ) > > I liked it all except for this bit: - K > >> 1998 - Realizing that the applet thing is fading fast, Sun repositions I > >> Java again, this time as a server language. They steal the design of G > >> Microsoft Transaction Server and convince everyone to pretend they  > >> created the design. > [...] @ > The actual story of the coming of C was actually more amusing. > B > In the early 1970s, the Big Thing in computer science was systemD > programming languages. (Recall that at that time, if you wanted toE > be serious about computing, you had access to a pdp10 and had to be I > conversant with its assembler language and its instruction set, reputed K > back then to have been sufficiently well designed it obviously had Divine  > inspiration.)  > G > These languages obtained simplicity by having only one data type: the  machine J > word. People knew about bytes (a byte being any contiguous group of bitsC > within a word) but they weren't much used in systems programming.  > C > This seemed to work well until along came the pdp11, in which the  assumptions K > of such languages came apart badly. The pdp11 basically rubbed everyone's  noseI > in the fact that you could not on such a machine get away with claiming F > everything was a word. Some I/O devices could tell whether they were	 addressed I > as bytes or as words (unibus had separate DATO and DATOB operations for  wordE > and byte respectively) and could not be controlled unless both were 
 available.K > Skipping 2 bytes at a time to move to the next instruction or instruction  partH > didn't help either, and there were several other really tough problems that1 > turned these languages into real pigs on pdp11. L >    Thus the initial strength of C was that it reintroduced different sized dataK > objects (and not coincidentally happened to have operators for everything K > a pdp11 instruction could do...and pretty well operators for nothing that  a H > pdp11 instruction could not do). This made it a decent replacement for pdp11 J > assembly language <evil grin> and a suitable language for the many otherG > machines that started coming out around the same time which also used  sub-word< > addressing (and a much less general definition of 'byte'). > B > It is probably a good thing the pdp11 had a reasonably clean and
 orthogonalK > instruction set. If you think about some of the addressing idiosyncrasies  ofL > the pdp8 (e.g. autoincrementing addresses in page 0, return address storedH > in first location of subroutines...) it is difficult to imagine what a languageG > written to make these operations efficient might have looked like. Of  course, I > it is also fortunate that none of the, er, more unusual capabilities of  pdp11 ? > seem to have been strongly enshrined in C (or anything else).  > H > Consider divides on SP (changing SP and PC), ASHC on R6, programs likeK > "mov -(pc),-(pc)", the @pc addressing mode (in dos-11 generally commented  withH > ";CARE; USED AS LITERAL BY PREVIOUS INSTRUCTION"), or a few more. Also thank G > God the EIS instructions came out, as did the pdp11/45 floating point  unit, K > before someone was tempted to design a language to make the EAE efficient  to > use. > G > So rather than complain about C, just consider it is a cleaner way to  write J > macro-11 ;-) and be glad history did not present us with something truly weird L > so that every program would not have to have Page 0 autoincrment registers > or suchlike... > H > It is after all possible to write C with string descriptors. VMS users have > done that for years... >  > Glenn Everhart   ------------------------------  % Date: Sun, 05 Feb 2006 05:51:13 -0800 # From: "Tom Linden" <tom@kednos.com>  Subject: Re: C  History ( Message-ID: <ops4h53nh3zgicya@hyrrokkin>  G On Sat, 04 Feb 2006 20:11:09 -0500, Glenn Everhart <Everhart@gce.com>    wrote:   > Richard Maher wrote:
 >> Hi Tom,) >>  I liked it all except for this bit: - J >>> 1998 - Realizing that the applet thing is fading fast, Sun repositionsH >>> Java again, this time as a server language. They steal the design ofF >>> Microsoft Transaction Server and convince everyone to pretend they >>> created the design.  > [...] @ > The actual story of the coming of C was actually more amusing. > B > In the early 1970s, the Big Thing in computer science was systemD > programming languages. (Recall that at that time, if you wanted toE > be serious about computing, you had access to a pdp10 and had to be I > conversant with its assembler language and its instruction set, reputed K > back then to have been sufficiently well designed it obviously had Divine  > inspiration.)  > I > These languages obtained simplicity by having only one data type: the   	 > machine J > word. People knew about bytes (a byte being any contiguous group of bitsC > within a word) but they weren't much used in systems programming.   K I don't think that is an accurate reading of history.  Consider Algol and    PL/IJ they were and have been used as system programming languages.  Multics wasK written in PL/I Burroughs was Algol, VOS is written in PL/I and z/os (aka    MVS) is written in a PL/I dialect.  > E > This seemed to work well until along came the pdp11, in which the   
 > assumptions B > of such languages came apart badly. The pdp11 basically rubbed   > everyone's nose I > in the fact that you could not on such a machine get away with claiming H > everything was a word. Some I/O devices could tell whether they were   > addressed K > as bytes or as words (unibus had separate DATO and DATOB operations for    > wordG > and byte respectively) and could not be controlled unless both were    > available.A > Skipping 2 bytes at a time to move to the next instruction or    > instruction partJ > didn't help either, and there were several other really tough problems   > that1 > turned these languages into real pigs on pdp11. H >    Thus the initial strength of C was that it reintroduced different   > sized dataK > objects (and not coincidentally happened to have operators for everything H > a pdp11 instruction could do...and pretty well operators for nothing   > that aJ > pdp11 instruction could not do). This made it a decent replacement for   > pdp11 J > assembly language <evil grin> and a suitable language for the many otherI > machines that started coming out around the same time which also used   
 > sub-word< > addressing (and a much less general definition of 'byte'). > D > It is probably a good thing the pdp11 had a reasonably clean and   > orthogonal> > instruction set. If you think about some of the addressing   > idiosyncrasies of G > the pdp8 (e.g. autoincrementing addresses in page 0, return address    > storedJ > in first location of subroutines...) it is difficult to imagine what a  
 > languageI > written to make these operations efficient might have looked like. Of   	 > course, K > it is also fortunate that none of the, er, more unusual capabilities of    > pdp11 ? > seem to have been strongly enshrined in C (or anything else).  > H > Consider divides on SP (changing SP and PC), ASHC on R6, programs likeC > "mov -(pc),-(pc)", the @pc addressing mode (in dos-11 generally    > commented withJ > ";CARE; USED AS LITERAL BY PREVIOUS INSTRUCTION"), or a few more. Also   > thank I > God the EIS instructions came out, as did the pdp11/45 floating point    > unit, C > before someone was tempted to design a language to make the EAE    > efficient to > use. > I > So rather than complain about C, just consider it is a cleaner way to    > write L > macro-11 ;-) and be glad history did not present us with something truly   > weird D > so that every program would not have to have Page 0 autoincrment   > registers  > or suchlike... > J > It is after all possible to write C with string descriptors. VMS users   > have > done that for years..   I It is also possible to use high-level languages for better efficiency and  reliability.   >  > Glenn Everhart   ------------------------------  $ Date: Sun, 5 Feb 2006 12:07:00 -0500- From: William Webb <william.w.webb@gmail.com>   Subject: Re: DEC PWS433au memoryI Message-ID: <8660a3a10602050907v4e627998p9ec27d96563681ff@mail.gmail.com>   & On 2/3/06, Eccles <nospam@home> wrote: > Keith A. Lewis wrote: L > > Eccles <nospam@home> writes in article <luidndRm6-kzeX7eRVnyjQ@brightvi=. ew.com> dated Fri, 03 Feb 2006 23:18:11 +0000: > > K > >>I picked up a DEC PWS 433au from ebay but it doesn't have any memory in K > >>it.  Can I use any 168-pin PC100 ECC SDRAM (so long as they're matched) 1 > >>or does it need to be proprietry DEC memory ? K > >>Also, given that I want to install VMS purely for hobbyist use, nothing E > >>serious or heavy going - how much would you reckon I would need ?  > >  > > J > > Yes, generic memory works, in matched pairs.  I think you can even use > > non-ECC. > > L > > Define "hobbyist use".  I run an Alphastation 500 for Apache and audio = withL > > only 192 MB, but I use 500 MB in my 500au which I use for DECwindows an= d  > > Mozilla. > > 4 > > --Keith Lewis              klewis {at} mitre.orgB > > The above may not (yet) represent the opinions of my employer. > J > I tried non-ecc ram in it and all it did was beep the 1-3-3 code so I'll: > have to try ecc ram instead.  Thank goodness for ebay !! > G > All I'll probably end up doing with it is possibly running Apache and I > DCL coding and doing things to it that I wouldn't dream of doing to our I > Alpha systems at work, like trying some other software/freeware just to E > see what it's like. I can't do this at work because the systems are H > validated, if I make a small change then it needs to be documented and1 > checked so a box of my very own would be great.  > 	 > Thanks, 	 > Martin.  >   $ Have you downloaded the docs for it?   WWWebb --C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------   End of INFO-VAX 2006.072 ************************                                                                                                                                                                                                                                                                                                                                                                                                  9<l_!`)Zcy╵U3xsd\0ޡֈdx~H0^tWWuA2uT2KZᕢ+_CZ>eC
~l؁EG*6ןrIQO?St	)O]֧*J9z~UKG>x֎]='z5{;s ϿۛrZoh_a]"+vr-O-y~ןj!V|޾H?IVD\&?j7x7'/e
;9^GZSO>\?Ǭ_a/7,˶3BB<&>cy{"95?MC~K>
?{^EbLymQx3{k>޲tbx[j͞=f9~BB3VPxx-@^Grx%^9
ո@zXf;[t;OUo$\?>enveh7{}$^![XH
vvg?A;`_Q|~'x0F4+ޫyE2<p3zfqA/!M;\,_^dk44d{޽|lG'83@ca2f'}mz?cvM-7Y% ak<X#Sz.2,eT-||~
>|P|Y|Dj3Þ~A^?3hm\>	"? (ֳxX#z>jK
<CB蟢2{>?kx=F#KiS
9֯<O9y:z
-|_u"wEY"q"+S)5SW+?9.}L9SՕr20⣟T