1 INFO-VAX	Sun, 13 Mar 2005	Volume 2005 : Issue 143       Contents: Re: comparison: Rdb vs. Mimer  Re: comparison: Rdb vs. Mimer  Re: History of the VMS shark% Re: HP on fast track to finding a CEO % Re: HP on fast track to finding a CEO P Re: Transferring Executables to a VAX by a tortuous route - help please pleasepl  F ----------------------------------------------------------------------  % Date: Sat, 12 Mar 2005 14:02:35 -0500 # From: "John Smith" <a@nonymous.com> & Subject: Re: comparison: Rdb vs. Mimer, Message-ID: <-_qdnWpeqLZKoa7fRVn-iw@igs.net>   Joe Database wrote: E > I am interested in a comparison between Rdb and Mimer, primarily on E > Alpha but also on VAX.  (I suspect that the differences between Rdb F > and Mimer are greater than the differences between Alpha and VAX forF > the same product.)  I have some familiarity with Rdb, but none (yet)
 > with Mimer.  > F > In particular, I would be interested in hearing from people who haveG > switched from Rdb to Mimer (or vice versa), whether or not the switch > > was worth the trouble, and/or from folks familiar with both. > B > If you don't want to reply to the newsgroup, then email me afterD > removing the obvious capital-letter spam-blocker.  Any informationB > provided (even by Rdb engineers) will be treated with the utmost
 > discretion.  > 8 > Among others, I am interested in the following points: > D >    o  performance, especially for CPU-intensive queries on a large8 >       (several TB) database with many concurrent users > @ >    o  differences in SQL (in particular, what non-standard SQLE >       commands would I have to rewrite if moving from Rdb to Mimer)  > D >    o  using Mimer in a cluster (this is relatively straightforward >       with Rdb)  > G >    o  is the lock-avoidance strategy of Mimer all it is cracked up to  >       be?  > 8 >    o  pricing (commercial licenses, support, upgrades) > . >    o  ease of backup (for a few-TB database)    L For this kind of need, you also might want to contact Jim Starkey, principal> author of Interbase, now available in open source as Firebird.  J The original Interbase was first released on VMS. The VMS code is still inE Firebird but has not been built on VMS in some time, but Jim might be  interested in your situation.   C Interbase/Firebird is a very cool product. I last used v4.x on VMS.    --F OpenVMS    - The often imitated but never advertised operating system.   ------------------------------  % Date: Sat, 12 Mar 2005 13:24:54 -0600 ' From: "Tom M" <kryios@spam.comcast.net> & Subject: Re: comparison: Rdb vs. Mimer0 Message-ID: <Ic6dnVFjm7Jy367fRVn-hw@comcast.com>  J You might also want to look into Ingres.  The opensource version should beL available on VMS (Alpha only) within the next few months.  Documentation for) R3 can be found at www.opensource.ca.com.   . "John Smith" <a@nonymous.com> wrote in message& news:-_qdnWpeqLZKoa7fRVn-iw@igs.net... > Joe Database wrote: G > > I am interested in a comparison between Rdb and Mimer, primarily on G > > Alpha but also on VAX.  (I suspect that the differences between Rdb H > > and Mimer are greater than the differences between Alpha and VAX forH > > the same product.)  I have some familiarity with Rdb, but none (yet) > > with Mimer.  > > H > > In particular, I would be interested in hearing from people who haveI > > switched from Rdb to Mimer (or vice versa), whether or not the switch @ > > was worth the trouble, and/or from folks familiar with both. > > D > > If you don't want to reply to the newsgroup, then email me afterF > > removing the obvious capital-letter spam-blocker.  Any informationD > > provided (even by Rdb engineers) will be treated with the utmost > > discretion.  > > : > > Among others, I am interested in the following points: > > F > >    o  performance, especially for CPU-intensive queries on a large: > >       (several TB) database with many concurrent users > > B > >    o  differences in SQL (in particular, what non-standard SQLG > >       commands would I have to rewrite if moving from Rdb to Mimer)  > > F > >    o  using Mimer in a cluster (this is relatively straightforward > >       with Rdb)  > > I > >    o  is the lock-avoidance strategy of Mimer all it is cracked up to 
 > >       be?  > > : > >    o  pricing (commercial licenses, support, upgrades) > > 0 > >    o  ease of backup (for a few-TB database) >  > D > For this kind of need, you also might want to contact Jim Starkey,	 principal @ > author of Interbase, now available in open source as Firebird. > L > The original Interbase was first released on VMS. The VMS code is still inG > Firebird but has not been built on VMS in some time, but Jim might be  > interested in your situation.  > E > Interbase/Firebird is a very cool product. I last used v4.x on VMS.  >  > --H > OpenVMS    - The often imitated but never advertised operating system. >  >    ------------------------------  % Date: Sat, 12 Mar 2005 13:49:00 -0500 ) From: "Neil Rieck" <n.rieck@sympatico.ca> % Subject: Re: History of the VMS shark ; Message-ID: <JQGYd.35738$fW4.1048339@news20.bellglobal.com>   A "Kenneth Farmer" <kfarmer@NOSPAM.spyderbyte.com> wrote in message 7 news:D0%Xd.14093$_i3.244343@twister.southeast.rr.com...  >  [...snip...] > & > Can I get copies of that to post on: > - > http://www.openvms.org/pages.php?page=Logos  > , > ...or is this something you're doing solo? > J > It would be great to get a graphic artist to do a rendition of the GreatJ > White logo with a dead penguin in his mouth.  Someone had mentioned that > somewhere. >  > Ken  > # OK so grabbed some logos from here: + http://www.openvms.org/pages.php?page=Logos G then merged them using "Animation Shop 3" (part of Paint Shop Pro) from  www.jasc.com into this: 8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html   Artist's unspoken message:D HP now owns OpenVMS but |d|i|g|i|t|a|l| will always be at the heart.  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html   ------------------------------   Date: 12 Mar 2005 20:33:07 GMT( From: bill@cs.uofs.edu (Bill Gunshannon). Subject: Re: HP on fast track to finding a CEO, Message-ID: <39h203F5umojqU1@individual.net>  = In article <42333034$0$78287$157c6196@dreader1.cybercity.dk>, . 	Karsten Nyblad <nospam@nospam.nospam> writes: > prep@prep.synonet.com wrote:, >> "Main, Kerry" <kerry.main@hp.com> writes: >>   >>  @ >>>For things on Linux that require kernel changes, IBM can onlyA >>>request features. Linus (and associates) are the only ones who 2 >>>decide what and *when* kernel changes get done. >>   >>   >> Horse stuff...  >>  D >> They have the code, there is NOTHING to stop them from writing itC >> themselves! Zip, nadda, nil. But they DO have to give the source  >> to their customers. >>  7 >> What other parts of Open Souce don't you understand?  >>  I > Well, Kerry is spinning the truth to make VMS & HP look good.  I think  G > Kerry is right, but IBM's options look much better if the wording is   > changed a little.  > G > Of course IBM can make the modifications that minimizes that work of  I > moving from AIX to Linux, but the customers will want to run Redhat or  C > SuSE.  Thus IBM will have a strong interest in getting the Linux  $ > community to accept their changes.  E If they put their minds to it, do you think it would be all that long G before IBM supplanted RedHat and/or SuSE as the ruler of the commercial C Linux world?  Even Linus has nothing to protect his position as the 
 God of Linux.    > K > Alternatively IBM could make a runtime environment, that made Linux look  J > like AIX to programs.  That could be runtime libraries, a virtual Power D > machine, or any of the other tricks that have been used by, e.g., J > Digital to make VAX code run on Alpha, various companies and OSS people : > to make Windows code run on none Windows platforms, etc. > I > Please note that there are environments for making Windows code run on  H > Linux, and that people have been writing code to run on multiple *nix I > platforms for years.  It seems feasible to make Linux look like AIX to  / > programs without changing that much of Linux.   H Assuming the availbility of AIX Libraries (and I assume IBM has that :-)E there would be very little involved in making AIX software run on any K of the Open Source systems.  Linux code runs on BSD using this very method. K All it is going to take is for IBM to finally get serious about this sector  of the market.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Sun, 13 Mar 2005 06:37:49 +0100 + From: Karsten Nyblad <nospam@nospam.nospam> . Subject: Re: HP on fast track to finding a CEO= Message-ID: <4233d1f5$0$80886$157c6196@dreader2.cybercity.dk>    Bill Gunshannon wrote:? > In article <42333034$0$78287$157c6196@dreader1.cybercity.dk>, 0 > 	Karsten Nyblad <nospam@nospam.nospam> writes:G >>Of course IBM can make the modifications that minimizes that work of  I >>moving from AIX to Linux, but the customers will want to run Redhat or  C >>SuSE.  Thus IBM will have a strong interest in getting the Linux  $ >>community to accept their changes. >  > G > If they put their minds to it, do you think it would be all that long I > before IBM supplanted RedHat and/or SuSE as the ruler of the commercial E > Linux world?  Even Linus has nothing to protect his position as the  > God of Linux.   D Right now you need support for Oracle databases if you want to be a F Linux vendor.  Oracle would be dead dumb if they helped their largest C competitor on DB software getting a monopoly on the Linux platform.   F IBM has been spending much money on Linux, but it is difficult to see E that they have got that much revenue from that.  Linux fans may like  @ what IBM do, but the fans do not let that decide their purchase G decisions.  Sun have written important OSS like OpenOffice and J2EE is  B the only competitor to .NET, but Sun is in trouble.  IBM might be F writing great software for Linux, but I doubt they can wind that many  customers that way.   F And now my own point of view.  I am a geek that hates monopolies, one G that likes the idea of software being free as in speech, not only free   as in beer.   F The VMS zealots always points out that closed source software has the H advantage that if you have a problem, you can go to the software vendor A and make the vendor fix it.  That might be true if you are a big  G customer of the vendor.  Otherwise you can send in bug reports, and if  D you are lucky, you get a fix 6-12 months later.  If you are a small I customer, than you are in a constant risk of running into a problem that  H is a major issue to you, but not to other customers, and the vendor not H willing to fix the problem.  Face it:  The support model of open source I software has problems, but so has the support model of closed source.  I  I can't see that the support model of closed source is any better to small  ' customer than the support model of OSS.   E Then you have the advantage that you can call the vendor if you have  D some dumb questions on how to use their software.  Well I have only F tried it with VMS, and the Danish support could only answer questions G were you could read the answer in the manuals plus questions on how to  I get the system parameters right.  Not that much help to me.  I would say  E that reading the manuals plus c.o.v would put most people i a better  ( situation than using the Danish support.  G OK, documentation of OSS software sucks and so does the quality of the  E less poppular programs, but you can look into the source and in most  H cases it is not that difficult to get your questions answered that way. I   And most software bugs can be fixed or circumvented with minor changes   to the programs.  C Then there is Windows.  From my point of view as an IT person that  H platform really combine the problems of closed source software with the F problems of open source software.  If you want real support, then you ? have to pay real money.  Otherwise you are stuck with a poorly  F documented OS, were you can use very long time to troubleshoot simple H problems simply because you do not know what is going on inside Windows.  F Further, software for software development is NOT free for any of the H closed source platforms.  E.g. on Windows you have to subscribe to MSDN C and buy Visual Studio and a database if you what to do any serious   programming.  H I do not want to go back to a model, where a single vendor controls the D platform.  Before Microsoft became the hate object of everybody not G liking monopolies it was IBM that had status.  I would prefer that IBM  E does not get that position again, and thus I would be tempted to buy  6 from other vendors should IBM try to monopolies Linux.   ------------------------------  % Date: Sat, 12 Mar 2005 20:57:29 -0800 ( From: Jeff Cameron <roktsci@comcast.net>Y Subject: Re: Transferring Executables to a VAX by a tortuous route - help please pleasepl / Message-ID: <BE590839.920A%roktsci@comcast.net>    On 3/12/05 5:33 AM, in articleC 1110634430.022792.252980@z14g2000cwz.googlegroups.com, "Alan Greig"  <greigaln@netscape.net> wrote:   > Jeff Cameron wrote:  > C >> If you don't have an IMAGE transfer file type, you can still use  > binary and! >> then enter in the VMS command:  >>  < >> $SET FILE/ATTRIBUTE=(RFM:FIX,LRL:512) executable-file.EXE > G > Unfortunately not with Kermit-32 as it used 510 byte RMS records when I > storing in binary mode. Of course you can convert the record format but ) > just changing the attribute won't work.  > -- > Alan Greig > K I don't know exactly what you mean? I use Kermit-32, and I have transferred L VAX and Alpha Executable images using both these methods for years now, with no problem.  Jeff   ------------------------------   End of INFO-VAX 2005.143 ************************                                                                                                
            Trace((stderr, "do_wild:  match() succeeds\n"));
            /* remove trailing dot */
            fnamestart += strlen(fnamestart) - 1;
            if (*fnamestart == '.')
                *fnamestart = '\0';
            return matchname;
        }
    }

    closedir(wild_dir);     /* have read at least one entry; nothing left */
    wild_dir = (DIR *)NULL;
    notfirstcall = FALSE;   /* reset for new wildspec */
    if (have_dirname)
        free(dirname);
    return (char *)NULL;

} /* end function do_wild() */

#endif /* !SFX */



/**********************/
/* Function mapattr() */
/**********************/

int mapattr(__G)
    __GDEF
{
    /* set archive bit (file is not backed up): */
    G.pInfo->file_attr = (unsigned)(G.crec.external_file_attributes & 7) | 32;
    return 0;
}



/**********************/
/* Function mapname() */
/**********************/

int mapname(__G__ renamed)
    __GDEF
    int renamed;
/*
 * returns:
 *  MPN_OK          - no problem detected
 *  MPN_INF_TRUNC   - caution (truncated filename)
 *  MPN_INF_SKIP    - info "skip entry" (dir doesn't exist)
 *  MPN_ERR_SKIP    - error -> skip entry
 *  MPN_ERR_TOOLONG - error -> path is too long
 *  MPN_NOMEM       - error (memory allocation failed) -> skip entry
 *  [also MPN_VOL_LABEL, MPN_CREATED_DIR]
 */
{
    char pathcomp[FILNAMSIZ];      /* path-component buffer */
    char *pp, *cp=(char *)NULL;    /* character pointers */
    char *lastsemi=(char *)NULL;   /* pointer to last semi-colon in pathcomp */
    char *last_dot=(char *)NULL;   /* last dot not converted to underscore */
    int dotname = FALSE;           /* path component begins with dot? */
    int killed_ddot = FALSE;       /* is set when skipping "../" pathcomp */
    int error = MPN_OK;
    register unsigned workch;      /*              hold the character being tested */


    if (G.pInfo->vollabel)
        return MPN_VOL_LABEL;   /* Cannot set disk volume labels in FlexOS */

/*---------------------------------------------------------------------------
  