jarij@tukki.jyu.fi (Jari Junikka) (04/10/90)
We are considering purchasing DECstation 5000. The sales person completely "forgot" to tell as about LMF. I would appreciete any information concerning how the system (fails) works in practise? How expensive it is to get licenses which in practise do not hinder system/program development? Jari (junikka@math.wisc.edu)
kaiser@cheese.enet.dec.com (04/10/90)
In article <4074@tukki.jyu.fi>, jarij@tukki.jyu.fi (Jari Junikka) writes... >I would appreciete any information concerning how >the system (fails) works in practise? How expensive it is >to get licenses which in practise do not hinder system/program >development? Is everyone clear that license management has no affect whatever on software YOU create? LMF is there for the system management side only of licensed products. This won't be adequately cross-referenced, but yesterday someone posted a note that pointed out, accurately, that LMF enables us (and other parties) to begin distributing Ultrix layered software on CDROM. Here's how it works for VMS, where we've been doing it for a while: there's something called "Consolidated Distribution" (CONDIST) which is a set of CDROMs, now two disks, containing the operating system and all our LMF-compliant layered products for VMS. A subscriber gets the distribution like clockwork (right!?) every two months; whenever a new set is made, everything that's on the shelf ready to go gets put on the CDROM. So subscribers always have this nifty, tiny little package of CDROMs containing (1.3GB?) of software that's guaranteed to be no more than two months old. Suppose you now want to use the software. You make a phone call and use your credit card or some approved method (I'm not a sales type -- ask them) to get a license key, and you can install the SW and license and run it immediately, rather than waiting for it to ship, etc. And of course you have the documentation already, on the Bookreader CDROM. No sweat. This also applies to temporary loans of products, of course, since a license can be set up to cover a limited period. Want to try COBOL (something of infinite interest to everyone in this group)? Great; here's a one-month license free. The crucial enabling technologies are LMF and CDROMs. CDROMs make it possible to ship enormous quantities of stuff cheaply, safely, and securely; and LMF makes it safe for the beancounters and capitalism. Besides my Unix systems I manage a VMS cluster -- perhaps I'm being punished for sins in an earlier life. And since I'm the guy who's always handled LMF on the cluster and the Unix systems, it's never been any problem at all. There was a complaint from someone that it was difficult; I've never found that to be true. (He said that the previous system manager had screwed things up before he had to take over. This seems to me a broader system management/personnel issue not limited to LMF, so why make LMF take the rap? My preceding system manager left things pretty disorganized too!) But to repeat: LMF doesn't apply to anything YOU do. Unless, of course, you want it to. And in that case you're probably talking to us already. ---Pete kaiser@cheese.enet.dec.com +1 508 480 4345 (machine: +1 617 641 3450)
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/10/90)
In article <10142@shlump.nac.dec.com>, kaiser@cheese.enet.dec.com writes: > >The crucial enabling technologies are LMF and CDROMs. CDROMs make it possible >to ship enormous quantities of stuff cheaply, safely, and securely; and LMF >makes it safe for the beancounters and capitalism. This sounds like a marketing ploy. Like you can't print multiple CD-ROMs for each product? Besides, if I want to evaluate a product, I wanna see the docs too, not just the executables. Of course, I could be strange by wanting to read the manual once in a while. >Besides my Unix systems I manage a VMS cluster -- perhaps I'm being punished for >sins in an earlier life. And since I'm the guy who's always handled LMF on the >cluster and the Unix systems, it's never been any problem at all. There was a >complaint from someone that it was difficult; I've never found that to be true. >(He said that the previous system manager had screwed things up before he had to >take over. This seems to me a broader system management/personnel issue not >limited to LMF, so why make LMF take the rap? My preceding system manager left >things pretty disorganized too!) Any other UNIX operating system (Sun OS, Berkie, AT&T, AIX) doesn't do this LMF stuff. Do they? Nope. Only Digital. It shows a lack of fundmental trust at some level. OH well. In a couple of decades from now, Digital will dump LMF, or we'll all be running Berkie or something, just on the principle of the thing.
fkittred@bbn.com (Fletcher Kittredge) (04/10/90)
In article <00934FDA.44D76CA0@KING.ENG.UMD.EDU> sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes: ... >Any other UNIX operating system (Sun OS, Berkie, AT&T, AIX) doesn't do this >LMF stuff. Do they? Nope. Only Digital. Your command of technical trends is as bad as your command of English. Most serious Unix Workstation vendors have announced such products. HP/Apollo has NLS (Network License Server) Sun has SunNet License. There are a variety of third party vendors selling Licensing solutions (Elan, et.al.). DEC just got to market a few months earlier by offering a limited product. What you ought to be doing is beating up DEC for offering a non-networking, non-standard product. fletcher Fletcher E. Kittredge fkittred@bbn.com
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/11/90)
In article <54695@bbn.COM>, fkittred@bbn.com (Fletcher Kittredge) writes: >>Any other UNIX operating system (Sun OS, Berkie, AT&T, AIX) doesn't do this >>LMF stuff. Do they? Nope. Only Digital. > I was referring to Digital. If you purchase a Sun box, you get SunOS with an unlimited number of users. If you purchase a DEC box, you get ULTRIX with a two user license. If you want more than two users/windows at a time, you have to pay to upgrade to 4, 8, and so forth. In the VMS world, I've yet to install a software product on our cluster which users LMF other than Digital's products.
michaud@decvax.dec.com (Jeff Michaud) (04/11/90)
> We are considering purchasing DECstation 5000. > The sales person completely "forgot" to tell > as about LMF. Remember LMF is new for V4.0, and V4.0 has only just been announced, the sales person probably hasn't heard of LMF yet either :-). Also, the only products to initially use LMF are ULTRIX User Upgrades and DECnet-ULTRIX. /--------------------------------------------------------------\ |Jeff Michaud michaud@decwrl.dec.com michaud@decvax.dec.com| |DECnet-ULTRIX #include <standard/disclaimer.h> | \--------------------------------------------------------------/
jg@max.crl.dec.com (Jim Gettys) (04/11/90)
Oh, so you want N CD roms every couple months, where N is large? The number of products is going up greatly with time, including 3RD party products, who insist on security in distribution. And then you'll complain about simple products costing lots of money since the distribution costs will dominate what we ought to charge for it... Basically, if Unix is to succeed (or VMS for that matter) we need a much cheaper distribution mechanism than we have right now. And floppies and tape don't cut it... They are manual, expensive, slow, small and less reliable than CD's. The model you suggest basically doesn't work; you should see the number of tapes we get internally, subscribing to all DEC products as we do here. Granted CD's are smaller and cheaper, just the time to put them in and out of the CDROM reader becomes a major hassle. In time, one would like to be able to press a button, and get an up-to-date system installed, without manual intervention, including layered and 3rd party products. Having in my deep dark past had to do serious system administration, such a scenario is more than slightly attractive.... The docs will also be on line (look at Bookreader on VMS). If it is a hassle, then we've failed. But don't damn it before you touch it. - Jim
meissner@osf.org (Michael Meissner) (04/11/90)
In article <00934FEF.79711AA0@KING.ENG.UMD.EDU> sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes: | In article <54695@bbn.COM>, fkittred@bbn.com (Fletcher Kittredge) writes: | >>Any other UNIX operating system (Sun OS, Berkie, AT&T, AIX) doesn't do this | >>LMF stuff. Do they? Nope. Only Digital. | > | I was referring to Digital. | | If you purchase a Sun box, you get SunOS with an unlimited number of users. | If you purchase a DEC box, you get ULTRIX with a two user license. If you | want more than two users/windows at a time, you have to pay to upgrade | to 4, 8, and so forth. | | In the VMS world, I've yet to install a software product on our cluster | which users LMF other than Digital's products. I believe the 'two user' problem is an artifact of the System V.2 license structure, which has such silliness in it. For those of you who never dealt with AT&T System V.2 licenses, there was a tiered scheme where a vendor selling a 1 user system had to send xx dollars to AT&T, and as the number of users increased, so did the dollars flowing back to AT&T. At one point when I worked for Data General, we had something like 32 different model numbers for UNIX (16 levels + USA/international). Since Ultrix is based on System V.2 [somewhat], and not on System V.3 which did away with different user systems except for 1/many users, I would imagine, DEC didn't want to raise prices in order to send the appropriate bucks to AT&T. -- Michael Meissner email: meissner@osf.org phone: 617-621-8861 Open Software Foundation, 11 Cambridge Center, Cambridge, MA Catproof is an oxymoron, Childproof is nearly so
palowoda@fiver.UUCP (Bob Palowoda) (04/11/90)
From article <54695@bbn.COM>, by fkittred@bbn.com (Fletcher Kittredge): > In article <00934FDA.44D76CA0@KING.ENG.UMD.EDU> sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes: > ... [discussion about LMF set aside] > just got to market a few months earlier by offering a limited product. What > you ought to be doing is beating up DEC for offering a non-networking, > non-standard product. Hmm, now this sounds interesting. How does DEC networking products get along with other networks such as SUN's, HP's and maybe UNIX 3.2 on 386 PC's? What is this "non-networking" or "non-standard"? Dosn't DEC have a standard TCP/IP implementation with the ULTRIX product? Are there any problems useing DEC's networks when working with X-windows to another OS say's SUN's? ---Bob -- Bob Palowoda palowoda@fiver | *Home of Fiver BBS* Home {sun}!ys2!fiver!palowoda | 415-623-8809 1200/2400 {pacbell}!indetech!fiver!palowoda | An XBBS System Work {sun,pyramid,decwrl}!megatest!palowoda| 415-623-8806 1200/2400/19.2k TB+
barnett@grymoire.crd.ge.com (Bruce Barnett) (04/11/90)
In article <MEISSNER.90Apr10172729@curley.osf.org> meissner@osf.org (Michael Meissner) writes: | I believe the 'two user' problem is an artifact of the System V.2 | license structure, which has such silliness in it. As I understand it, the term multi-user depends on how many terminals are on a machine. Network connections do not count, as no real terminal is physically attached when a remote login occurs. Therefore a Sun with two serial ports is a two-user machine. You can have 100 people log on over the network. and it won't kick you off with "too many users". If you add a 16-port serial interface card, you should purchase a multi-user upgrade. For some reason DEC considers a network connect a new user, while Sun does not. Don't ask me why. -- Bruce G. Barnett barnett@crd.ge.com uunet!crdgw1!barnett
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/11/90)
In article <5085@crltrx.crl.dec.com>, jg@max.crl.dec.com (Jim Gettys) writes: >Oh, so you want N CD roms every couple months, where N is large? >And then you'll complain about simple products costing lots of >money since the distribution costs will dominate what we ought >to charge for it... Producing a single CD-ROM probably runs between $5-7 dollars, once you amoritize all the capital costs of equipment. I don't work for Tower or Astiria records, so I can't tell you the exact price. Of course, equating the distribution system of major record companies to a major software distributor is probably quite silly. They're geared for economies of scale. Of course, I came across a blurb in Infoworld where you can get 500 CD-ROMs done for some low price ( Under $5K). >Basically, if Unix is to succeed (or VMS for that matter) we need >a much cheaper distribution mechanism than we have right now. >And floppies and tape don't cut it... They are manual, expensive, slow, >small and less reliable than CD's. I think the success or failure of OSes is not dependent upon the distribution mechanism. VMS and UNIX seem to have survived this long without CD ROM. Tape is 2-3 times as expensive, mediawise, than CD-ROM. At $15 vs $5-7 dollars, I don't think you'll break anyone there, even if you have 2-3 tapes. >The model you suggest basically doesn't work; you should see the number >of tapes we get internally, subscribing to all DEC products >as we do here. >Granted CD's are smaller and cheaper, just the >time to put them in and out of the CDROM reader becomes a major >hassle. Make a fortune. Take a 5 or 6 disk commecial CD player mechanism and convert it to data reading. Sell zillions to those people who want to have access to 5-6 CDs, but not all at once. >The docs will also be on line (look at Bookreader on VMS). What happens when the machine crashes and you can't get Bookreader? >If it is a hassle, then we've failed. But don't damn it before >you touch it. Having done LMF on VMS, it is a hassle. Since I have a VAX cluster with 20 machines, how many little LMF paks do I have to type in? Twenty. Of course, since the VAXstation IIs were set up differently than the VAXstation 3100s, I had to backwards-engineer what happened in the first place, licensewise.... Doug
meissner@osf.org (Michael Meissner) (04/11/90)
In article <BARNETT.90Apr11084318@grymoire.crd.ge.com> barnett@grymoire.crd.ge.com (Bruce Barnett) writes: | Therefore a Sun with two serial ports is a two-user machine. You can | have 100 people log on over the network. and it won't kick you off with | "too many users". If you add a 16-port serial interface card, you should | purchase a multi-user upgrade. | | For some reason DEC considers a network connect a new user, while Sun | does not. Don't ask me why. Yeah, when I was at DG we tried to get AT&T to say one way or another whether network users (and batch jobs on the layered UNIX port) were considered users or not...... Fortunately for us at that time, AT&T made the issue moot by removing all of the user classifications (and lowering the per CPU cost, though raising the initial cost). Another fun issue is whether a cluster type of computers is one computer or many...... -- Michael Meissner email: meissner@osf.org phone: 617-621-8861 Open Software Foundation, 11 Cambridge Center, Cambridge, MA Catproof is an oxymoron, Childproof is nearly so
fkittred@bbn.com (Fletcher Kittredge) (04/11/90)
In article <1040@fiver.UUCP> palowoda@fiver.UUCP (Bob Palowoda) writes: >From article <54695@bbn.COM>, by fkittred@bbn.com (Fletcher Kittredge): >> In article <00934FDA.44D76CA0@KING.ENG.UMD.EDU> sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes: > Hmm, now this sounds interesting. How does DEC networking products >get along with other networks such as SUN's, HP's and maybe UNIX 3.2 >on 386 PC's? What is this "non-networking" or "non-standard"? >Dosn't DEC have a standard TCP/IP implementation with the ULTRIX product? >Are there any problems useing DEC's networks when working with X-windows >to another OS say's SUN's? > No, no, no! My complaint is with LMF, not DEC's networks. Presently, LMF is only of use on DEC systems. In a distributed, heterogeneous environment, this greatly reduces the usefulness of LMF. If DEC decides to develop a product for this environment, either they will have to produce an LMF for each of their competitors systems, or they will have to adhere to some standard for network licensing. One choice makes life easy for customers and ISVs; one makes it more difficult. regards, fletcher > >-- >Bob Palowoda palowoda@fiver | *Home of Fiver BBS* >Home {sun}!ys2!fiver!palowoda | 415-623-8809 1200/2400 > {pacbell}!indetech!fiver!palowoda | An XBBS System >Work {sun,pyramid,decwrl}!megatest!palowoda| 415-623-8806 1200/2400/19.2k TB+ Fletcher E. Kittredge fkittred@bbn.com
jg@max.crl.dec.com (Jim Gettys) (04/12/90)
By most standards, both Unix and VMS have failed in the mass market to date. For each system of either type, there are more than 10 PC's. This looks like failure to me. And as these machines have/are becoming the same price, as PC's, we better have a software distribution mechanism that is at least as economical, despite the fact the software is often much larger. By this standard, we aren't surviving..... If the costs of media were where the costs of software distribution were, your points would make more sense. Most of the costs however are in order entry, order processing, stocking, and shipping. This is where the (overwealmingly large) hidden costs are; and guess what, we have to price software accordingly. Other hidden costs include delay to the customer in this loop. If the bits are already in your hands, the whole transaction can be handled over the phone, for example, with billing via account or even credit card. Only one initial subscription order needs to be processed, stocking costs (like what you do with obsolete media) evaporate, almost all the shipping costs are gone, etc. You get instant gratification, and the ability to "sample" software before buying, as well as easier software installation. And at this point in time, we (the workstation market) sure doesn't have the retail distribution channel that the PC market has. I haven't played with LMF on VMS (or on Ultrix, for that matter), so can't comment on the hassle quotient. As I said before, if it is a pain, then we are failing. - Jim
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/12/90)
In article <54736@bbn.COM>, fkittred@bbn.com (Fletcher Kittredge) writes: > >No, no, no! My complaint is with LMF, not DEC's networks. Presently, LMF >is only of use on DEC systems. In a distributed, heterogeneous environment, >this greatly reduces the usefulness of LMF. Useful? <supressive comments on> > If DEC decides to develop a >product for this environment, either they will have to produce an LMF for >each of their competitors systems, or they will have to adhere to some >standard for network licensing. DEC to develop a product which doesn't run on DEC-stamped hardware? Mmmmmm. Everyone is going to do their "own thing" in network licensing, unless OSF or someone else develops A Standard.
michaud@decvax.dec.com (Jeff Michaud) (04/12/90)
> Tape is 2-3 times as expensive, mediawise, than CD-ROM. At $15 vs $5-7 > dollars, I don't think you'll break anyone there, even if you have 2-3 tapes. As you said, that's only media cost. You have to factor in the labor also. I believe it costs a lot more to have a bunch of people cutting tapes than it is to have a single person sitting around watching CD's get pressed. /--------------------------------------------------------------\ |Jeff Michaud michaud@decwrl.dec.com michaud@decvax.dec.com| |DECnet-ULTRIX #include <standard/disclaimer.h> | \--------------------------------------------------------------/
avolio@decuac.dec.com (Frederick M. Avolio) (04/12/90)
Anyone want to back me up and declare a Rat Hole on this? How about we all wait and see ULTRIX 4.0 delivered and decide whether LMF rots. We could also see what happens in the standards area around networked license management standards. Haven't we speculated enough on the unknown? Or is it getting fun and am I getting old? Fred
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/12/90)
In article <5087@crltrx.crl.dec.com>, jg@max.crl.dec.com (Jim Gettys) writes: >By most standards, both Unix and VMS have failed in the mass market to >date. For each system of either type, there are more than 10 PC's. >This looks like failure to me. Mostly for the price of one UNIX or VMS box, you can get 3-4 PCs, on average and the entry-level price point for a PC has traditionally been something which an Average Joe can afford. Besides, supporting a PC is MUCH cheaper than a UNIX or VMS box. Besides, you buy them to do different things. I wouldn't buy a miniVAN if all I have to do is drive myself to work, eh? Most people aren't going to buy UNIX and VMS boxes. >And as these machines have/are becoming >the same price, as PC's, we better have a software distribution mechanism >that is at least as economical, despite the fact the software is often >much larger. By this standard, we aren't surviving..... You merely charge more per copy, since you are going to sell fewer copies. Besides, supporting multiuser systems is always going to be more expensive than supporting a single user PC. >If the costs of media were where the costs of software distribution were, >your points would make more sense. > >Most of the costs however are in order entry, order processing, stocking, >and shipping. This is where the (overwealmingly large) hidden costs are; >and guess what, we have to price software accordingly. A) You're never going to rid of tape (how else are you going to fix the bugs on the CD-ROMS? :-) Besides the installed user base is too high). B) If you're telling me that the largest cost of software is in DISTRIBUTION, I think DEC outta farm out distribution to someone else. Let's play "what if?" Let's say that distribution of package XYZ costs $100. Improved distribution via CD ROM cuts that to $10. So, on a $500 package, you're still at $410 bucks. But you still have to keep tape around for eveyrone who owns a TK 50 or 9 track. So how do you account for the other $410? Development, programmers salaries, advertising and your Internet link. Going from tape to CD ROM won't drop the cost of software to $100 per pop, even if you do get "more efficient." You have to produce more copies of your product to drop the price per copy. Besides, you still need to keep manuals around. >Other hidden costs include delay to the customer in this loop. Cost to customer, not to you. This doesn't affect your cost, which is the premise of your arguement.
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/12/90)
In article <10195@shlump.nac.dec.com>, michaud@decvax.dec.com (Jeff Michaud) writes: > > As you said, that's only media cost. You have to factor in > the labor also. I believe it costs a lot more to have a bunch > of people cutting tapes than it is to have a single person > sitting around watching CD's get pressed. If you can provide the cost breakdown for how much it takes DEC to distribute software, I'd greately appreciate it. How much labor does it take to hang a tape? You don't need a Ph.D or 3 hours to do it. Maybe a couple of bucks. It is a small cost compared to to the total value of the product...
michaud@decvax.dec.com (Jeff Michaud) (04/13/90)
> By most standards, both Unix and VMS have failed in the mass market to > date. For each system of either type, there are more than 10 PC's. > This looks like failure to me. Depends on how you are counting. Counting success by the number of machines is like saying Mercedez Benz (sp?) has failed compared to the number of datsuns or toyotas there are on the road. Does anyone have numbers for installed "user count" for UNIX, VMS, PC's? /--------------------------------------------------------------\ |Jeff Michaud michaud@decwrl.dec.com michaud@decvax.dec.com| |DECnet-ULTRIX #include <standard/disclaimer.h> | \--------------------------------------------------------------/
kaiser@cheese.enet.dec.com (04/13/90)
In article <0093516D.B9040540@KING.ENG.UMD.EDU>, sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes... >How much labor does it take to hang >a tape? You don't need a Ph.D or 3 hours to do it. Maybe a couple of bucks. >It is a small cost compared to to the total value of the product... Think of it not as the cost just of the product (in an increasingly competitive marketplace) but as overall cost. There are what, over 100 products on the VMS Consolidated Distribution CDROM set? Now for each product calculate the cost of -- the tape itself -- changing the tape (at "a couple of bucks") to make it -- changing the tape by the user (or the user's operator) -- watching the machine for when it's time to change the tape VS the cost of idle time for the machine -- magnetic media that are erased or go bad through accident or mishandling -- storing the tape (the cost of shelf space plus the cost of climate- conditioning for magnetic media) -- the delicacy (and maintenance) of all tape drives -- etc. (I must have missed something) Contrast that with stamping CDROMs containing the same information. NOW how do the costs come out? ---Pete kaiser@cheese.enet.dec.com +1 508 480 4345 (machine: +1 617 641 3450)
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/13/90)
A) The licensing fee is going to be the same B) I'm going to have to pay for bug-fixes and upgrades and that's going to come on tape, unless you wish to make me wait for the next CD-ROM stamping. C) I would want to pay for paper documentation so my programmers don't go blind or CRT-crazy. D) At best, I have enough money for both tape and CD-ROM.
braun@drivax.UUCP (Kral) (04/14/90)
In article <00934FDA.44D76CA0@KING.ENG.UMD.EDU> sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes: > >Any other UNIX operating system (Sun OS, Berkie, AT&T, AIX) doesn't do this >LMF stuff. Do they? Nope. Only Digital. > >It shows a lack of fundmental trust at some level. From a user perspective, I don't mind LMF. Here's why: Clustering technology makes it real easy for sites to breach their CPU licenses. No copy necessary, you just execute from shared disks on multiple CPUs, all for a single CPU price. If your vendor finds out, you suddenly owe them (potentially) a lot of money. And it's real easy to accidentally get setup this way. Other programs may prevent this by checking CPU serial IDs. Do you consider this a " lack of fundmental [sic] trust at some level"? -- kral 408/647-6112 ...amdahl!drivax!braun The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (04/15/90)
In article <JKLH2S1@drivax.UUCP>, braun@drivax.UUCP (Kral) writes: > >From a user perspective, I don't mind LMF. Here's why: Clustering technology >makes it real easy for sites to breach their CPU licenses. No copy necessary, >you just execute from shared disks on multiple CPUs, all for a single CPU >price. If your vendor finds out, you suddenly owe them (potentially) a lot of >money. And it's real easy to accidentally get setup this way. For what products? LMF is used (mostly) for DEC products in the VMS world and soon to be inflicted on ULTRIX. If I run an ultrix cluster of 24 machines, I'm going to have to type in 24 LMF packs, and activate them for all 24 CPUS. ANNOYING. If you purchase 24 machines, but only one "right to upgrade", the vendor is going to come talk with you real quick. >Other programs may prevent this by checking CPU serial IDs. Do you consider >this a " lack of fundmental [sic] trust at some level"? It depends on how this process is implemented, but we're talking about the OPERATING SYSTEM here, bunkie! You only have one (or two) sources to get that from, and few sources of support. For applications, I have no problem at all at going to fetch the CPU serial number. I do have a problem with companies OTD (other than Digital) which do MORE inane things, such as have me type in a 40 line by 60 character matrix of coded random lettering in order to get the product to work across our VAXcluster, instead of supplying it on a TK-50 or floppy disk....