1 INFO-VAX	Sat, 11 Jun 2005	Volume 2005 : Issue 324       Contents:
 Re: a sad day 3 Re: Intel neuters Montvale, Itanic screams in alarm 3 Re: Intel neuters Montvale, Itanic screams in alarm 3 Re: Intel neuters Montvale, Itanic screams in alarm 3 Re: Intel neuters Montvale, Itanic screams in alarm 3 RE: Intel neuters Montvale, Itanic screams in alarm 3 Re: Intel neuters Montvale, Itanic screams in alarm 3 Re: Intel neuters Montvale, Itanic screams in alarm  Re: minimize VMS footprint Re: minimize VMS footprint5 Sueposium: Five days of peace and VMS. Thank you Sue! ( Re: What Terminal Servers are you using?( Re: What Terminal Servers are you using?( Re: What Terminal Servers are you using?( Re: What Terminal Servers are you using?( Re: What Terminal Servers are you using?# Re: why are alias names a bad idea? # Re: why are alias names a bad idea?   F ----------------------------------------------------------------------  % Date: Sat, 11 Jun 2005 21:13:44 +1200 $ From: "Lurker" <nowhere@nothing.com> Subject: Re: a sad day2 Message-ID: <eVxqe.6600$U4.949659@news.xtra.co.nz>  : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message& news:42A9FA15.C5AC33AF@teksavvy.com...  F > In the first case, the clinic gets paid by the government or chargesG > government approved rates for procedures and the patients get paid by  > the government.   D How benevolent of the goverment - it pays both the clinic and/or the patient.I Hold on a second though - were it not the patients who paid the goverment  for the priviledge first?    ------------------------------  % Date: Sat, 11 Jun 2005 02:20:45 -0400 ' From: Dave Froble <davef@tsoft-inc.com> < Subject: Re: Intel neuters Montvale, Itanic screams in alarm0 Message-ID: <11al0ivavnh1rd4@corp.supernews.com>   Main, Kerry wrote: >>-----Original Message-----1 >>From: David Mathog [mailto:mathog@caltech.edu]   >>Sent: June 10, 2005 11:49 AM >>To: Info-VAX@Mvb.Saic.Com > >>Subject: Re: Intel neuters Montvale, Itanic screams in alarm >> >>Main, Kerry wrote: >> >>> >>>However, other vendors that have also made recent positive  >> >>announcements  >>% >>>wrt to Itanium's future potential-  >>> 7 >>>- SGI Announces new Itanium Workstation: April, 2005  >>@ >>Now there's a vendor known for making wise business decisions! >> >><SNIP> >> >>? >>>And in case, anyone thinks Intel is hurting right now which   >> >>might make >>< >>>them change any strategies they have currently in place,  >> >>their business >>. >>>update today certainly indicates otherwise:C >>>http://www.intel.com/pressroom/archive/releases/20050609corp.htm ; >>>"SANTA CLARA, Calif., June 9, 2005 -- Intel Corporation   >> >>expects revenue  >>> >>>for the second quarter to be between $9.1 billion and $9.3  >>
 >>billion, as  >>B >>>compared to the previous range of $8.6 billion to $9.2 billion,D >>>primarily driven by ongoing strong demand for notebook products." >>: >>Even so, Itanium is a money pit and Intel knows it.  AMDF >>called Intel's Itanium bluff and Intel is in the process of throwing? >>in those cards. It's clear to all but the true believers that @ >>there finally is an "industry standard" architecture and it's 
 >>x86. Apple  E >>migrating to x86 just further confirms it.  (That doesn't mean that H >>Intel is the industry standard though, other than by marketing clout, = >>since there are also x86 processors from AMD and VIA.)  HP   >>will probably @ >>keep its head firmly in the sand and erase the remaining VMS,  >>HPUX, and ? >>even Tandem markets.  Maybe someday somebody at HP will ask:   >>what from < >>the CUSTOMER'S perspective is the advantage of running on  >>Itanium over   >>x86? >> >  > 3 > As I said earlier, the game has only just begun.   > H > Competition is a good thing and AMD is doing a great job in some areasI > right now. However, as the financial analysts and press has highlighted D > recently, their overall financials need to improve rather quickly. > H > Imho, Itanium is a way for Intel to both break out of the x86 back andH > forth race with AMD and to firmly position their future offerings over
 > Powerx.   G The itanic isn't competition for Power, except in the same way x86 and  H such were competition for VAX and Alpha.  There will always be the many B that can get along with 'good enough', and then there will be the 7 serious people who will demand more than 'good enough'.   ? IBM has got to be very pleased with Intel.  One of the serious  A competitors to Power was Alpha.  Intel has managed to reduce the   competition to Power.   G 'More than good enough' will never be a large segment of the computing  @ world, but it will persist, and IBM seems to be the only player  interested in such business.  I I do enjoy your classification of AMD.  Good way of saying that they are  - only lucky and won't remain so.  Yeah, right.   I Technically, AMD is kicking Intel's ass, and with products that will run  H the world's choice of software.  They have a lead, and unless they give F it up, or Intel comes up with something new and unexpected, they will I continue to eat into Intel's business.  Not overnight, but slowly.  With  I enough time, if they continue to excel, Intel's financials won't look as   well as they do now.   --  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: Sat, 11 Jun 2005 02:49:04 -0400 ( From: Bill Todd <billtodd@metrocast.net>< Subject: Re: Intel neuters Montvale, Itanic screams in alarm= Message-ID: <GMCdna0mWdn8FDffRVn-2A@metrocastcablevision.com>    Rob Young wrote:   ...   C > 	Paul DeMone has a good read on IPF , few dispute his projections $ > 	as they aren't wildly optimistic.  I Horseshit, Rob:  anyone at all familiar with RWT interchanges knows that  B people (far more than just me) dispute Paul's ridiculously biased L projections about Itanic's future market acceptance on a very regular basis.  G I recently had the occasion to remind him of one of his more laughable   ones (   http://groups-beta.google.com/group/comp.arch/browse_thread/thread/bdcaced0c6f62efd/c18ab6223ffba9c2?q=mckinley+sales+author:demone#c18ab6223ffba9c2  C - about 1/3 of the way down the page dated Jun 3, 2002, 7:51 A.M.):   H "I think the odds are better than even that within a year of commercial H release (probably later this month) McKinley will drive IA64 unit share F to second place behind SPARC in the 64 bit high end MPU market place."  H Well, McKinley didn't quite manage that during its commercial year-long H reign as Supreme Itanic Overlord (or is that 'overload'?).  And Madison I didn't quite manage it either over the course of well over an additional  G year in that capacity.  And Madison II hasn't come anywhere near doing   it, either.   G You may crow, "But just wait for Montecito!" (just as you crowed "Just  C wait for McKinley!" and "Just wait for Madison!"), but even in the  D highly unlikely event that Montecito *did* manage to fulfill Paul's G market-penetration projection he would still have been off by a rather  I significant factor ('within a year' becoming 'after 4+ years') and hence  E could hardly be held up as a paragon of objectivity and conservatism.    - bill   ------------------------------  % Date: Sat, 11 Jun 2005 02:58:48 -0400 ( From: Bill Todd <billtodd@metrocast.net>< Subject: Re: Intel neuters Montvale, Itanic screams in alarm= Message-ID: <DOudnTeuFOQ0FjffRVn-gg@metrocastcablevision.com>   F Whoops - I got so involved with finding that 2002 quote that I forgot  this far more recent howler:   Rob Young wrote (quoting Paul):    ...   P > For the sake of argument let's say these various scenarios pan out and in 2007N > IPF has exceeded 110k systems worth $5B a year in sales. That is roughly the@ > space peviously occupied by PA-RISC, Alpha, and MIPS combined.  < Well, perhaps for sufficiently coarse values of 'roughly'...  G Prior to the Alphacide, we have it direct from Compaq (and for the VMS  G part also direct from the mouth of Rich Marcello) that VMS, Tru64, and  F Tandem systems brought in respectively $4 billion, $3 billion, and $2 I billion annually, for a total of $9 billion within Compaq systems alone.  H   My vague recollection is that HP-UX was about 3x the size of Tru64 in I sales back then, so figure something like another $9 billion there.  And  D of course there was MPE on PA-RISC (not entirely insignificant back E then) and SGI's MIPS products, but I'm guessing that we don't really  I need to throw them into the pot to make it clear just how far off Paul's  H statement was - even if you discount close to half that total as likely / being service revenue rather than sales per se.    - bill   ------------------------------  % Date: Sat, 11 Jun 2005 12:27:52 +0200 + From: Karsten Nyblad <nospam@nospam.nospam> < Subject: Re: Intel neuters Montvale, Itanic screams in alarm= Message-ID: <42aabcb0$0$78284$157c6196@dreader1.cybercity.dk>    Rob Young wrote:m > In article <42a95ead$0$67263$157c6196@dreader2.cybercity.dk>, Karsten Nyblad <nospam@nospam.nospam> writes:  >  >>Rob Young wrote: >>_ >>>In article <42A8C6E7.DF1918EC@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: - >>>	No more than the port of HP/UX and NSK to A >>>	8086 is inevitable.  Enterprise architectures aren't switched : >>>	out whimsically - it costs way too much money and your >>>	customer base gets antsy.  >>D >>True.  However, how many companies do you think has a strategy of J >>betting their business on Itanic?  It would be lightheaded to have such F >>a strategy before Itanic has demonstrated that it can generate more H >>profit to Intel than what it costs in ongoing development, marketing, H >>etc.  Thus HP can move away from Itanic now, but the longer they wait # >>the more customers will be angry.  >  > C > 	Paul DeMone has a good read on IPF , few dispute his projections > > 	as they aren't wildly optimistic.  Here is a recent read of > 	his:  > l > http://www.realworldtech.com/forums/index.cfm?action=detail&PostNum=3180&Thread=78&roomID=11&entryID=47485 > I >  In 2004 over 150k I2s shipped in around 30k systems worth about $1.5B.  > 	 > [snip]]  > Q > By HP's predictions within two years it will be shipping roughly three times as P > much IPF gear or roughly 75k systems worth over $3B a year. In two years it isN > quite credible that SGI could be selling 10k to 15k systems worth over $1B aM > year. So HP and SGI alone could bring IPF to about 25% RISC market share by  > 2007.   G Futures, futures, futures.  You have been selling futures as long as I  H have known.  Besides, it might be that those building the machines make F money, but my argument was that Intel is selling so few chips, that I D doubt their profit can pay for developing future versions of Itanic.  M > What is harder to predict is how well the rest of the IPF OEMs will do. The P > best prospects are NEC, Fujitsu, Hitachi, Bull, and Unisys. Dell is a bit of aO > dark horse. Montecito will make its 8870 based systems a lot more competitive Q > but it isn't apparent that either potential IPF customers want to buy from Dell Q > or that Dell really wants to sell IPF gear to them. A best case scenario IMO is Q > that within 2 years the rest of the IPF pack would be selling about 20k systems  > a year worth about $1B.  > P > For the sake of argument let's say these various scenarios pan out and in 2007N > IPF has exceeded 110k systems worth $5B a year in sales. That is roughly theO > space peviously occupied by PA-RISC, Alpha, and MIPS combined. For Intel this N > scenario translates to sales of over 500k IPF MPUs worth about $1.0-1.2B andH > profits of over $0.5B ($1.0B - $400m in R&D cost - $60m silicon cost). >  > ---  > E > 	By Paul's estimation (see another RWT thread) Itanium has reached  F > 	break-even.  We'll see how well it does profit-wise other the next D > 	few years.  But the PA-RISC , Alpha and MIPS lifeboats have been G > 	burned - Itanium has a pretty lucrative captive market so it should   > 	do more than ok.   F A sizable fraction of the market you are describing is in Windows and H Linux.  In fact, I have a hunch that that fraction is more than half of E the sale.  That sale will be gone in a heartbeat should x86 start to  G scale beyond 4 processor systems.  Do you remember how fast Windows NT  I for Alpha disappeared when Pentium III came to market, and Alpha did not  G have any performance advantage to x86 any more?  Captive market?  Yes,   sure...    ------------------------------  % Date: Sat, 11 Jun 2005 11:01:03 -0400 ' From: "Main, Kerry" <kerry.main@hp.com> < Subject: RE: Intel neuters Montvale, Itanic screams in alarmR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB65099E@tayexc19.americas.cpqcorp.net>   > -----Original Message-----3 > From: Dave Froble [mailto:davef@tsoft-inc.com]=20  > Sent: June 11, 2005 2:21 AM  > To: Info-VAX@Mvb.Saic.Com > > Subject: Re: Intel neuters Montvale, Itanic screams in alarm >=20  	 [snip ..]      > >> > >=20 > >=207 > > As I said earlier, the game has only just begun.=20  > >=20B > > Competition is a good thing and AMD is doing a great job in=20 > some areasB > > right now. However, as the financial analysts and press has=20
 > highlighted F > > recently, their overall financials need to improve rather quickly. > >=20@ > > Imho, Itanium is a way for Intel to both break out of the=20 > x86 back and> > > forth race with AMD and to firmly position their future=20 > offerings over > > Powerx.=20 >=20? > The itanic isn't competition for Power, except in the same=20  > way x86 and=20@ > such were competition for VAX and Alpha.  There will always=20 > be the many=20F > that can get along with 'good enough', and then there will be the=209 > serious people who will demand more than 'good enough'.  >=20C > IBM has got to be very pleased with Intel.  One of the serious=20 E > competitors to Power was Alpha.  Intel has managed to reduce the=20  > competition to Power.  >=20A > 'More than good enough' will never be a large segment of the=20  > computing=20D > world, but it will persist, and IBM seems to be the only player=20 > interested in such business. >=20? > I do enjoy your classification of AMD.  Good way of saying=20  > that they are=20/ > only lucky and won't remain so.  Yeah, right.  >=20  	 [snip ..]   D Luck has nothing to do with it. They have made some good advances in3 specific area's and one has to give them credit.=20   H On the other hand, there is the practical reality of making profits fromH an overall company perspective (what shareholders and analysts look at).  
 Reference:=20    April 13, 2005> http://www.eetimes.com/showArticle.jhtml?articleID=3D1607024784 "AMD posts loss in Q1, blames flash-memory business"   January 18, 2005+ http://www.theinquirer.net/?article=3D20783  "AMD turns in $30 million loss"    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: Sat, 11 Jun 2005 08:16:20 -0700 # From: "Tom Linden" <tom@kednos.com> < Subject: Re: Intel neuters Montvale, Itanic screams in alarm( Message-ID: <opsr7opix5zgicya@hyrrokkin>  J On Sat, 11 Jun 2005 11:01:03 -0400, Main, Kerry <kerry.main@hp.com> wrote:   >  >> -----Original Message----- 1 >> From: Dave Froble [mailto:davef@tsoft-inc.com]  >> Sent: June 11, 2005 2:21 AM >> To: Info-VAX@Mvb.Saic.Com? >> Subject: Re: Intel neuters Montvale, Itanic screams in alarm  >> >  > [snip ..]  >  >  >> >>  >> > >> >5 >> > As I said earlier, the game has only just begun.  >> >@ >> > Competition is a good thing and AMD is doing a great job in
 >> some areas @ >> > right now. However, as the financial analysts and press has >> highlightedG >> > recently, their overall financials need to improve rather quickly.  >> >> >> > Imho, Itanium is a way for Intel to both break out of the >> x86 back and < >> > forth race with AMD and to firmly position their future >> offerings over  >> > Powerx. >>= >> The itanic isn't competition for Power, except in the same  >> way x86 and> >> such were competition for VAX and Alpha.  There will always >> be the manyD >> that can get along with 'good enough', and then there will be the: >> serious people who will demand more than 'good enough'. >>A >> IBM has got to be very pleased with Intel.  One of the serious C >> competitors to Power was Alpha.  Intel has managed to reduce the  >> competition to Power. >>? >> 'More than good enough' will never be a large segment of the  >> computingB >> world, but it will persist, and IBM seems to be the only player >> interested in such business.  >>= >> I do enjoy your classification of AMD.  Good way of saying  >> that they are0 >> only lucky and won't remain so.  Yeah, right. >> >  > [snip ..]  > F > Luck has nothing to do with it. They have made some good advances in2 > specific area's and one has to give them credit. > J > On the other hand, there is the practical reality of making profits fromJ > an overall company perspective (what shareholders and analysts look at). >  > Reference: >  > April 13, 2005> > http://www.eetimes.com/showArticle.jhtml?articleID=1607024786 > "AMD posts loss in Q1, blames flash-memory business" >  > January 18, 2005+ > http://www.theinquirer.net/?article=20783 ! > "AMD turns in $30 million loss"    Here's another referenceD http://www.eet.com/news/latest/showArticle.jhtml?articleID=164302189  # AMD bets future on 64-bit processor    	  Spencer Chin 	 	   Page 1 of 2    	  EE Times (06/10/2005 1:52 PM EDT) 	   	 A MANHASSET, N.Y. — AMD Inc. told analysts Friday (June 10) the   I microprocessor maker's future success hinges primarily on improving the   G power handling and processing capabilities of its 64-bit x86 platform   G while seeking to leverage the platform for a wider range of computing   
 applications.   K At an analysts' conference in New York largely devoid of controversy, AMD   I President and CEO Hector Ruiz emphasized the company was sharpening its   K focus on the microprocessor market. "We are committed to one platform —   K the x86," Ruiz told analysts. "Our purpose is to reinvent the dynamics of    the microprocessor industry."  ...    > 	 > Regards  >  > Kerry Main > Senior Consultant  > HP Services Canada > Voice: 613-592-4660  > Fax: 613-591-4477  > kerryDOTmainAThpDOTcom > (remove the DOT's and AT)  > 6 > OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Sat, 11 Jun 2005 12:10:54 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> < Subject: Re: Intel neuters Montvale, Itanic screams in alarm; Message-ID: <z2Eqe.20071$_n2.1398651@news20.bellglobal.com>   5 "Bill Todd" <billtodd@metrocast.net> wrote in message 7 news:0bednXc-RaQxYjXfRVn-jw@metrocastcablevision.com...  [...snip...] > L > This means that whatever advantage Itanic may have enjoyed due to the lackL > of credible high-end x86 systems will tend to vanish by that point, thoughL > in fact with IBM's currently-shipping Hurricane Xeon chipset (which is hotL > on the heels of the top-of-the-line Itanic TPC-C score at 4 processors andL > has just edged ahead of it at 8 processors - it supports up to 32 sockets,L > but they are apparently working their way up the benchmark slowly) and theK > imminent appearance of the Horus Opteron chipset (also up to 32 sockets), I > plus whatever larger Opteron facilities Sun has in its back pocket that H > Andy B. has been working on for years, one might suggest that any such9 > advantage is being seriously eroded starting right now.  >   L I don't think x86 (high-end or otherwise) will ever vanish. I saw an articleK the other day at Technology Review that told how Intel has plans to replace 5 copper interconnects with silicon-based laser diodes. M http://www.technologyreview.com/articles/05/07/issue/feature_intel.asp?trk=nl I Couple this with the fact that Intel (as well as other vendors) intend to K "comply" with Moore's law by doubling the core count every two years or so, G and you're going to have some very powerful computing technology in the  hands of the common consumer.   G If I was running things at HP, I'd start a skunk works to begin porting # OpenVMS to extended x86 technology.   	 * * * * *   H On a related note, when Compaq decided to kill Alpha, I don't think theyL gave any thought to how this action would affect Cray (who depended on AlphaH as their primary CPU technology). Like wise, if Intel decides that IntelI shareholders would like to see Itanium die, they won't give any though to  how this will affect HP.  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.! http://www3.sympatico.ca/n.rieck/    ------------------------------    Date: 11 Jun 2005 08:09:12 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)# Subject: Re: minimize VMS footprint , Message-ID: <42aa9c28$1@news.langstoeger.at>  Z In article <d8d4o2$ud3$1@news01.intel.com>, Ken Fairfield <my.full.name@intel.com> writes:H >> You must not forget that *.c can be a part of vmsinstalable images... > B >Eberhard, what are you talking about here???  While VMS certainlyC >allows one to rename MY_GREAT_PROGRAM.EXE to DONT_LOOK_IN_THERE.C, D >I've never heard of doing an INSTALL of a C-language source file...  F He is talking about a kit for installation by SYS$UPDATE:VMSINSTAL.COMD which usually consists of kitname.A, kitname.B, kitname.C, and so on  / No C-language source file, only a file named .C   < You did know it and I hope you're not too frustrated now ;-)   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Sat, 11 Jun 2005 06:44:26 -0700 # From: "Tom Linden" <tom@kednos.com> # Subject: Re: minimize VMS footprint ( Message-ID: <opsr7kgcvzzgicya@hyrrokkin>  : On 11 Jun 2005 08:09:12 +0100, Peter 'EPLAN' LANGSTOEGER   <peter@langstoeger.at> wrote:   = > In article <d8d4o2$ud3$1@news01.intel.com>, Ken Fairfield   " > <my.full.name@intel.com> writes:I >>> You must not forget that *.c can be a part of vmsinstalable images...  >>D >> Eberhard, what are you talking about here???  While VMS certainlyE >> allows one to rename MY_GREAT_PROGRAM.EXE to DONT_LOOK_IN_THERE.C, F >> I've never heard of doing an INSTALL of a C-language source file... > H > He is talking about a kit for installation by SYS$UPDATE:VMSINSTAL.COMF > which usually consists of kitname.A, kitname.B, kitname.C, and so on > 1 > No C-language source file, only a file named .C  > > > You did know it and I hope you're not too frustrated now ;-)* Starlet might be built during install, no? >    ------------------------------  % Date: Sat, 11 Jun 2005 12:34:14 +0200 - From: Didier Morandi <prenom.nom@freesurf.fr> > Subject: Sueposium: Five days of peace and VMS. Thank you Sue!4 Message-ID: <42aabe26$0$28797$626a14ce@news.free.fr>   All,  G Now that the 2005 OpenVMS BootCamp in Nashua, NH, USA is over, it is -  I to me - time to say thank you to Sue Skonetski, who organized everything  H and took care of us so well, to all the folks who did their job to make H it happen and to HP-III who seems to start to like our beloved platform.  I I will not comment any contents as all presentations were protected by a  H non disclosure agreement, but I will only say that since 1998 it is the I first time that I found such happiness and interactivity during the show.   0 This really reminded me good old DECUS Symposia.  % Again, merci beaucoup Sue, on t'aime.    D.  $ PS: "Sueposium" is (c) 2005 Hoff :-)   ------------------------------  # Date: Sat, 11 Jun 2005 07:55:09 GMT   From: John Santos <john@egh.com>1 Subject: Re: What Terminal Servers are you using? ( Message-ID: <xNwqe.356$lb5.336@trnddc04>   Rich Alderson wrote:1 > Kilgallen@SpamCop.net (Larry Kilgallen) writes:  >  > = >>I use DECserver 200s because I do not want to depend on IP.  >  > N > Since I have never been a regular user, much less administrator, of a DECnetN > system, I find this statement puzzling.  I have administered TCP/IP terminalP > servers since 1984, and have never found them to be undependable; on the otherN > hand, the LAT device we had to set up for some reason (I didn't have to deal9 > with it directly) was constantly having to be rebooted.   D Gee, I've had DECserver 90's, 200's, 550's and 700's stay up for 2-3F years with no need to reboot.  (My only DS100 died with a power supplyF problem several years ago and I haven't bothered to try to fix it, but+ I don't recall it being unreliable either.)    > , > So why would you not want to depend on IP? >   B TCP/IP-capable DECservers are not dependent on TCP/IP.  They do itB in addition to LAT, not instead of.  Lots more flexible.  When youA configure, you can restrict port-by-port (or all ports) to LAT or B TCP/IP (telnet) if you want.  You can also restrict which services@ they can connect to, set up dedicated ports, inbound-only ports,? outbound-only ports (for dialout modems, printers, null-modemed / system console ports, etc.), just like DS200's.    > Rich Alderson 1 > Last LOTS Tops-20 Systems Programmer, 1984-1991 / > (the academic computing facility at Stanford)    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  + Date: Sat, 11 Jun 2005 13:45:08 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) 1 Subject: Re: What Terminal Servers are you using? ( Message-ID: <d8ept4$3sj$1@pcls4.std.com>  " John Santos <john@egh.com> writes:   >Rich Alderson wrote: 2 >> Kilgallen@SpamCop.net (Larry Kilgallen) writes: >>  > >>>I use DECserver 200s because I do not want to depend on IP. >>  O >> Since I have never been a regular user, much less administrator, of a DECnet O >> system, I find this statement puzzling.  I have administered TCP/IP terminal Q >> servers since 1984, and have never found them to be undependable; on the other O >> hand, the LAT device we had to set up for some reason (I didn't have to deal : >> with it directly) was constantly having to be rebooted.  E >Gee, I've had DECserver 90's, 200's, 550's and 700's stay up for 2-3 G >years with no need to reboot.  (My only DS100 died with a power supply G >problem several years ago and I haven't bothered to try to fix it, but , >I don't recall it being unreliable either.)  H These conflicting reliability reports are probably due to the fact that C LAT lives on the edge somewhat and is rather sensitive to slightly  I misconfigured networks (such as one too many repeaters on a baseband net) D and gives up quickly.  TCP/IP on the other hand is often too slow to  notice a connection is dropped.    ------------------------------    Date: 11 Jun 2005 09:34:52 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) 1 Subject: Re: What Terminal Servers are you using? 3 Message-ID: <ZgJzlYQa21nG@eisner.encompasserve.org>   b In article <d8ept4$3sj$1@pcls4.std.com>, moroney@world.std.spaamtrap.com (Michael Moroney) writes:  J > These conflicting reliability reports are probably due to the fact that E > LAT lives on the edge somewhat and is rather sensitive to slightly  K > misconfigured networks (such as one too many repeaters on a baseband net)  > and gives up quickly.   B Actually, LAT stands for _Local_ Area Transport.  It is for use on@ _Local_ Area Networks.  Faced with something else, like a potato/ field, there is no reason to expect it to work.   A There is a firm definition of how many repeaters are required for 	 Ethernet.   = I believe the Cobol compiler does a bad job of compiling Ada. , I do not attempt to use it for that purpose.   ------------------------------  + Date: Sat, 11 Jun 2005 15:52:45 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) 1 Subject: Re: What Terminal Servers are you using? ( Message-ID: <d8f1cd$a9m$1@pcls4.std.com>  / Kilgallen@SpamCop.net (Larry Kilgallen) writes:   c >In article <d8ept4$3sj$1@pcls4.std.com>, moroney@world.std.spaamtrap.com (Michael Moroney) writes:   K >> These conflicting reliability reports are probably due to the fact that  F >> LAT lives on the edge somewhat and is rather sensitive to slightly L >> misconfigured networks (such as one too many repeaters on a baseband net) >> and gives up quickly.  C >Actually, LAT stands for _Local_ Area Transport.  It is for use on A >_Local_ Area Networks.  Faced with something else, like a potato 0 >field, there is no reason to expect it to work.  & Very true.  I never claimed otherwise.  B >There is a firm definition of how many repeaters are required for
 >Ethernet.  B Yes.  My point was that the person with the LAT problem may have aE configuration problem on his network but it is minor enough that only A LAT is annoyed by it.  Breaking the old "two repeater rule" was a < typical example of a violation that often bothered only LAT.   ------------------------------  + Date: Sat, 11 Jun 2005 16:36:25 +0000 (UTC) 1 From: Jefferson Humber <matrix01@globalnet.co.uk> 1 Subject: Re: What Terminal Servers are you using? ? Message-ID: <d8f3u9$h2d$1@nwrdmz02.dmz.ncs.ea.ibs-infra.bt.com>   H I 2nd the Xyplex/MRV units.  We have been using them for about 12yrs in G a 24/5 (sometimes 7) production environment, and can't complain.  Most  F of the Xyplex's have lasted in excess of 8yrs of active service, with C the new replacements being MRV models.  We use them with LAT comms.    Jeff     Jeff Goodwin wrote: I > We recently replaced our 10 year old Xyplex units with MRV.  The Xyplex N > units were fine, but the MTBF was starting to hurt us.  MRV now produces theJ > latest units formally known as Xyplex.  There are some traditional unitsJ > based on the old DECserver command set and there are newer ones based on> > Linux.  We got the traditional ones and are happy with them. > K > If you're not picky about how old the units are, look around for some old  > DECserver 700s.  >  > -Jeff  >  > 5 > "Thomas Wirt" <twnews@kittles.com> wrote in message A > news:6f6a9$42a9b163$4367aba2$26820@msgid.meganewsservers.com...  > D >>I have a project coming up that will require us to get some new IPI >>capable Terminal Servers.  I plan on looking for refurbed Decserver 90m C >>units, since this is what we have had in the past and we have the D >>backplane/hub in which the 90 series can go.  Is anyone buying anyI >>alternatives to the 90m?  How much are you paying for how many terminal  >>ports? >> >>TIA  >> >>--  
 >>Thomas Wirt  >>Systems Manager  >>Kittle's Home Furnishings  >>Indianapolis, IN >  >  >    ------------------------------  + Date: Sat, 11 Jun 2005 07:29:49 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply), Subject: Re: why are alias names a bad idea?$ Message-ID: <d8e3td$lc8$2@online.de>  B In article <j5pqe.180$kj5.55@trnddc03>, John Santos <john@egh.com> writes:   C > I think startup may be the issue.  Converting the host name to IP D > address of the router requires sending a packet to the DNS server,@ > but if your default DNS server isn't local, it can't get thereD > until the IP stack knows the IP address of the gateway.  Catch-22.  ? This is why nameservers are stored as addresses in the on-disk  E configuration.  I was talking of the SET ROUTE command.  At startup,  F nameservers are OK since they are stored as addresses, so the name of  the default can be resolved.  @ > If the host name or alias of the gateway is in the local hosts@ > database, then you should be okay, but you'll have to manually@ > change the local database if you change the name or address of > the gateway.   Right.  @ > If you specify the gateway by IP address, you'll still have toB > update it if you change the gateway, but it is just one thing toA > change, and seems easier to do.  If I have a working DNS server @ > under my control, I try to use that and minimize what's in theA > local database, because redundancy can only lead to trouble and  > confusion later.  I My idea is to speed up manual intervention in the case of a problem with  G the DSL router of my hobbyist cluster.  Since changing IP addresses on  I VMS is easier than doing so in the router, I would change the IP address  I of the default route (it is stored as an address) and switch the alias I JD use for the on-disk configuration of the permanent route to another  router.t   ------------------------------  # Date: Sat, 11 Jun 2005 08:25:13 GMTc  From: John Santos <john@egh.com>, Subject: Re: why are alias names a bad idea?( Message-ID: <Jdxqe.675$lb5.542@trnddc04>  / Phillip Helbig---remove CLOTHES to reply wrote: D > In article <j5pqe.180$kj5.55@trnddc03>, John Santos <john@egh.com>
 > writes:  >  > C >>I think startup may be the issue.  Converting the host name to IPSD >>address of the router requires sending a packet to the DNS server,@ >>but if your default DNS server isn't local, it can't get thereD >>until the IP stack knows the IP address of the gateway.  Catch-22. >  > A > This is why nameservers are stored as addresses in the on-disk  G > configuration.  I was talking of the SET ROUTE command.  At startup,  H > nameservers are OK since they are stored as addresses, so the name of  > the default can be resolved. >   & I know we are talking about SET ROUTE.  C But if your nameserver isn't on your LAN, and your default route isnA specified by name, you'll have to query the nameserver to get thee@ IP address of the router.  The DNS inquiry has to go through theC router to get to the DNS server, but it doesn't know the address ofa< the router until the name (of the router) has been resolved.   > @ >>If the host name or alias of the gateway is in the local hosts@ >>database, then you should be okay, but you'll have to manually@ >>change the local database if you change the name or address of >>the gateway. >  >  > Right. >  > @ >>If you specify the gateway by IP address, you'll still have toB >>update it if you change the gateway, but it is just one thing toA >>change, and seems easier to do.  If I have a working DNS server>@ >>under my control, I try to use that and minimize what's in theA >>local database, because redundancy can only lead to trouble andi >>confusion later. >  > K > My idea is to speed up manual intervention in the case of a problem with  I > the DSL router of my hobbyist cluster.  Since changing IP addresses on nK > VMS is easier than doing so in the router, I would change the IP address sK > of the default route (it is stored as an address) and switch the alias I mF > use for the on-disk configuration of the permanent route to another 	 > router.   D The question is can the alias be resolved *before* the default routeF is known?  If the DNS server is the local host or on the LAN or if the@ alias is stored in the local database (UCX SET HOST...) then youB should be okay.  If the DNS server is off the LAN, you can't queryD it until the default route has been defined (or at least an explicitC route to the DNS server, even if you know the IP address of the DNS  server.l  C Are you sure that ucx set route/perm/gate=... stores the IP address A of the gateway in the permanent database rather than whatever you > specify?  If so, I don't see why it makes the slightest bit ofA difference whether you specify an address, a locally-defined hosteC name, a DNS-defined host name, or an alias.  It just resolves it toe> its current address and stores the address.  Maybe the warningB pertains to an initial configuration, when the network hasn't been) started yet, and nothing can be resolved?s   -- d John Santosd Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------   End of INFO-VAX 2005.324 ************************