Conference 7.286::digital

    Try the office of the Chief Engineer.  The first function listed
    is "Consulting on New Product Nomenclature."  The manager is June
    Payne at 223-2912..........Hope this helps.

    Thanks for the hint.  I'll give her a try.  Success or failure will
    be posted here.
    
    Mark

    
       Is there really a naming standard in use?
    
       I can see where VT = Video Terminal, and LP = Line Printer, but
    what about LA100, or TU80, TS11, etcetera? Some don't seem to make
    much sense at all.
    
    
    Frank
    

    That's why I asked!  As a field person, like yourself, I wondered
    if there was any method to this madness.

If memory serves correctly, this is expalined in grusome detail in 
LDP::DEC_HISTORY (KP7 and all that...).

Regards,
BobW

    The history of the HW lettering noemclature may be in DEC_History, but
    I just looked at all the topic titles there, and it certainly isn't
    obvious just where it might be!

    
    	   Actually, I enjoy the diversity of naming conventions that
    	come into play, and make a game out of understanding how to 
    	translate them.  To draw upon previously mentioned examples,
    
    	   TU80 = Tape Unit 80
    	   TS11 = Tape Subsystem 11
    
    	   Whether or not these are valid remains to be seen, but this
    	is how I remember them.  I never did figure out why adding an
    	HSC interface to a TU made it a TA... that one still bugs me.
    	I figure it means Tape Access, but that's just a guess.  The 
    	storage media group has to be the worst when it comes to naming
    	conventions.  
    
    	   Except for the cables group... what on eart could be meant 
    	by the name "BNE3K"?  "BC16E" seems deceptively simple, until
    	you try to figure some significance for the 16... is it the
    	resistance of the line?  Perhaps the bandwidth?  Maximum
    	voltage?  Wire guage?  What the heck does the 16 mean?
    
    	   Networking products are a lot easier to understand.  Their
    	names are acronyms, almost always starting with DE (Digital
    	Ethernet).  The notable exception is the terminal server line,
    	which are DS's... I suppose DETS was not considered an appealing
    	product name.
    
    	   Are you looking for a real mental challenge?  Try finding any
    	five pieces of software in the price books... it's absolute
    	insanity!  Q-numbers out the wazzoo, split across two books, in
    	no obvious logical order.  
    
    	   Try using AQS!
    
    	   What a company... I love the challenges (excuse me,
    	opportunities).
    
    	- Greg

    Re: l7
    
    Pogo *also* said "we are surrounded by insurmountable 
    opportunities"  :-)
    
    Karen
    

    There *are* conventions:
    Anything beginning with "K" pertains to a cpu.
    Anything beginning with "M" pertains to memory.(core or mos)
    Anything beginning with "D" pertains to data comm.
    Anything beginning with "T" pertains to tapes.
    Anything beginning with "R" pertains to disks.
    
    Anyway,these are the conventions that prevailed 15 years ago...there
    are exceptions now.
    BTW: The second letter of an option designation used to mean something
    but now it is arbitrary.(or so it seems)

    Given the explosion in our product line in the past few years, it
    is next to impossible to have meaningful 2 character letter
    combinations.
    
    I must admit, our Ethernet folks seem to have to most fun naming
    things.  I'm waiting for our news Unibus Ethernet controller to
    come out:
    
    	The DUMPER
    
    it also gets rid of excess Ethernet traffic :-)
    
    Bob

re: < Note 742.7 by HSSWS1::GREG "The Texas Chainsaw" >
 >                          -< Product names, and such >-
 >    	   TS11 = Tape Subsystem 11
 >

Well, that could be, Greg, but in all the years that I was with the RSTS/E
Engineering group we SWORE that it actually stood for "Tape Stretcher 11"!

-Jack

        Like Bob Withers, I'm sure this topic has been raised, and beaten
        to death, several times in several conferences in the past.
        
        However, other than here (I haven't checked, but I wouldn't be
        surprised to find it here already), I believe it has been asked
        and answered (without, of course, a definitive answer) in one of
        the issues of ASKENET. 
        
        If you *do* find a definitive answer, and the information can be
        found on-line, please DO NOT POST IT HERE.  Post a pointer to
        where you found the information.  (Unless, of course, it's a page
        or less.  I'll be amazed if it's that little.) 
        
        Tom

    RE: .12
    	I *was* referring to the *REAL* names.  If we added all the
    DECrynisms, we could probably fill up an RA90.
    
    RE: .13
    	'Twas exactly my intent, Tom.  If a pointer exists, I would
    have provided same (so far, my discovery process hasn't come up
    with anything).
    	If it is "large" and unwieldy for this conference, I'll keep
    it locally, and recommend that anyone wanting a copy send me mail,
    and I'll FWD the list back.
    
    Mark

    While I was working in Corporate Product Operations, I attended
    one of the strategy meetings that produced the 62XX series of VAX'es.
    
    The chairman (Don't think he'd appreciate his name being published
    here...),  got up and explained the process, which 
    he likened to mixing up a bowl of oatmeal, throwing some around, and 
    going with whatever stuck to the walls.

    Discussion ensued, oatmeal flew and the 6210 stuck.
    
    Cheers,
    
    Gil
    
    

    What makes you think it was oatmeal?

    Only because one can't get vulgar in these conferences.
    
    Remember, big brother (in this case Tom Blinn) is watching over
    us!

    Other than the first letter naming convention, things get pretty
    mixed up. There are 'guidelines' in the product family groups,
    but some things occur only because the marketing and engineering
    folks want to DIS-associate the product from a dud predecessor.
    
    The tapes family is fairly traditional, but some of the old style
    units persist as the 'exceptions'.
    
    	T=tapes (general family)
    	TA= tape assembly and is the general family for reel-to-reel
    	TB= one that's coming, but I'm not clear about the 'B=' what yet
    	TC= tape control and is a TCU if a seperate piece
    	TU= tape drive(transport), couldn't use TT since thats a teletype
    Therefore, a TA90 consists of a TC90 and at least one TU90. Simple huh?
    
    	TK is for the cartridge tape units.
    	Initially, the number referred to the form factor EG TK50= 1/2 rack
    and 'full' height while the TK30 would have been '1/2' height but then
    came higher density in the TK70 that was the same size as the 50 8-).
    
    Then we get into the controllers, TUK for Ubus, TQK for Qbus but then
    came the SCSI controller and TSK was out of the question for several
    reasons, so it became the TZK.
    
    Bob Mc

    LA as in LA120, 180 etc. stands for  "Line Asynchronous"
    There is logic to all of this; simply not known to us.
    I suspect the logic is a product of Dick Best.
    

    Here in NaC, where we build boxes that sit on the Ethernet,
    the defacto rule seems to be:
    
    	5 letters starting with DE.
    
    Thus the DECnet/SNA Gateway-CT hardware is the DESNA, because "SNA" 
    seemed vaguely appropriate, although maybe GCT would have fitted in
    better with the software name, but we didn't have it at the time.
    If we ever build a follow-on, we'd probably call it a DESNB, using
    the old increment rule.  If we ever built a follow-on to that, though,
    we'd be out of luck since someone else has claimed DESNC for the
    eNCryption box.
    
    Algorithm: think of a name that fits in with other things you've
    seen.  Apply to the office of the chief engineer to claim it for
    your own.   Other people will then use your name as a de facto
    standard, and before you know it you're part of a convention.
    

    There were articles in the Engineering Newsletter last June and September
    describing the numbering scheme for option level parts.  The articles
    were submitted by Marcia Russell of the Chief Engineer's Office. 
    Probably worth a look.  I can photocopy the articles for you if you
    can't get them elsewhere.
    
    -willie
    

    I remember hearing about an ill fated project to produce a DES
    encryption board that sat on the unibus.At the time the scheme for
    numbering things was AAA11, as opposed to the older AA11A schema.  The
    first two letters were "KG" indicating that it was an auxilliary
    processor.  Since there was already a KG11A, it was to be called the
    KGB11.  Fitting.  I heard the government didn't take too kindly to the
    name.
    
    Larry

    Re: .22
    
    Then the government can't be too pleased with VMS which has various
    symbols used to interface to the security features that have prefixes
    such as KGB$, NSA$ and CIA$....
    
    				Steve

    
    At an IS Symposium I attended awhile back, one of the presenters
    swore that the HSC of the HSCXX stood for Hot "doo-doo" Controller.
    
    They actually had to retrofit the name Hierarchical (sp) Storage
    Controller.  
    
    May be true, Maybe not...
    
    Brent

    It finally came to mind. The name of the person who created Digital's
    "option designation list" in 1964 was Authur Hall. He was not a "CPU
    engineer", so he relegated the CPU options to the letters he had left
    over after assigning reasonable letters to the things he was familiar
    with. CPUs, even today, have an option number beginning with the letter
    K; this isn't for processor "chip" names, but rather for the modules
    the chips go on.
    
    Dick Best, as Digital's official "chief engineer", still heads the
    office that assigns all the engineering option numbers. 

Anyone who has read Bell & Newell's classic text on computer architecture will
recall that in the Processor/Memory/Switch (PMS) notation, "K" was used for
CPUs.  If the K convention dates from 1964, then I suspect that Bell took that
with him.

    Gordon Bell was well aware of the option-number convention in 1964.
    Somehow I doubt that was the source of the "K" in the PMS notation, but
    I don't have any good reason for the doubt. Next time I see Gordon,
    I'll ask him. 

    re: .26
    
> Anyone who has read Bell & Newell's classic text on computer architecture will
> recall that in the Processor/Memory/Switch (PMS) notation, "K" was used for
> CPUs.  

    If I'm not mistaken "P" was used for processor in PMS (hence the P in
    PMS) and "K" was used for link or link/switch (control) elements.
    
    -willie

    re .28:
    
>   If I'm not mistaken "P" was used for processor in PMS (hence the P in
>   PMS) and "K" was used for link or link/switch (control) elements.
    
    You are indeed correct: the designers of the HSC50 followed that convention
    and hence the names "P.io" and "K.sdi".

    -Jim