Conference 7.286::digital

This is probably an idea that has been kicked around before, but I cant see 
it anywhere.

Problem Statement
-----------------
Field staff find it very hard to find the product manager and marketing 
managers for any particular product. Although they are ostensibly in VTX 
the speed of infrastructure movement and the shifts going on in locations 
make them still difficule to find. The weight of personal information 
arriving at product managers personal mail areas can make it difficult to 
sort out what is product related from what is not.

A Possible Solution
-----------------
2 nodenames: Prdmgr and Mktmgr or 1 called Prduct
Accounts for each product area.

For instance, to send product related mail to the X.400 product manager,
send to Prdmgr::X400.

If the manager changes, the new one gets the account. No one has to go 
chasing the new one or following the chain of "X used to look after that, 
didnt they? You phone X, I havent looked after that for yonks...try Y". 

A common address exists for internet mail for providing customer feedback 
as well.

An electronic history of interaction on the product set would also exist so 
that when the managers change over they can understand issues and who the 
prime participants from the field are. This is important when we consider 
the amount of change we have had recently. 

This would be a simple, no fuss way of providing basic supply chain 
reengineering without the convoluted schemes that we are currently seeing.


- Just a thought,
John

    How is this any better than the VTX system?  It is based on a
    database which is actively managed and is generally up to date.
    Your proposal is difficult to administer and doesn't seem to have
    any real advantage.
    
    					Steve

    re: -1...
    
    I don't know about the "generally up to date" part.  I've found 
    information in VTX (re: contact names, etc.) that is a year out of date, 
    sometimes more.  
    
    M.

VTX isnt always up to date. People move.

I cant use VTX when I'm on site in a remote location, I can send to the 
product manager using this suggestion via Internet etc. My customer can 
also send to the CURRENT product manager, without me having to come back to 
the office or dial in, look up VTX...They also dont need to know their name 
to pass feedback on in the first instance. In fact, the product manager's 
mail address for a product can be put on the feedback forms that come with 
the product and we know, if the product sits on the shelf for a while, when 
it is bought the opinions will still reach the target they intended.

The important point is that it becomes a relatively constant thing for 
everyone.

Its just an idea, I'd like to hear others that people have.

    Why not use the Employee Locator File (ELF). Use a keyword for a
    product name.
    As an example I entered "Exabyte" in one of my keywords.
    
    Test it with "find Exabyte" in ELF.
    
    Regards,
    Peter.

Again, that is only as good as the people who keep the information up to
date.  

					Steve

    
    	Actually, the idea is simple, straight-forward, very inexpensive to
    maintain and administer. I, for one, am at a complete loss as to why
    anyone would knock it - unless you are part of the problem.
    	If Digital does not drastically change its bureaucratic thinking, 
    Digital is going to be a memory in this industry.
    
    
    		the Greyhawk

    Re: previous
    
    The technology exists today to provide a solution along the lines
    suggested. It would be a no-brainer to implement but a monster to
    maintain. Keeping the information up to date is the largest part (rule
    of thumb - 80%) of the cost of ownership of such a system.
    
    What is the probability that field people would be any more amenable to
    keeping account responsibility information up to date than the folk in
    the puzzle palace in the Greater Maynard Area are to keeping marketing
    and product management assignments current? It's just as difficult for
    us here in GMA to find out who is responsible for Customer X as it is
    for field folk to find the contact for Product Y.

    
    I suspect that administration would be no better nor worse than the
    VTX approach -- the products that care will be up to date and
    reachable.
    
    The big benefits would be:
    
     -  you wouldn't need internal access to find out where to send the mail
    
     -  addresses would be more stable (for those who wrote them down) and
        more intuitive (for those who didn't)
    
    I like it.
    

    I absolutely agree with this thought.  It is a terrific idea.  Within
    ALL-IN-1, all the new manager has to do set an auto forward on the
    account, and all of the messages will be sent to his/her regular mail
    address, VAXmail or otherwise.  This would be completely transparent to
    employees, customers, and (in the long run) the product manager.
    
    Great idea.
    
    Anybody have other thoughts on how to improve service to our customers?
    
    M.
    

	the hole that I see in this is that the mapping between the
	*actual* product name and the mail account name would not always
	be intuitive...even in the example given, X.400->X400; would
	a customer and/or field person "guess" that? howz about
	PATHWORKS for OpenVMS (Netware)?? so now, on top of maintaining
	a mapping between mail accounts and product managers, one has to
	maintain a mapping between product names and mail accounts, and a
	list would need to be published....sounds like an awful mess
	to me.........

	_kelley

    Kindda sounds like VTX ATOZ ... which is kept current.
    
    (Sorry for the gramma, I was raised on da Sout side of Chicago)
    

    My vote for a simple idea to improve service ... eliminate voicemail in
    favor of increased use of current technolgy (ie. cellular phones,
    pagers, and INTERNET).

    I too favor the idea, while the VTX_IR is certainly reasonable approach
    it is cumbersome, not up to date (currency issue) and certainly not in
    tune with the world of Windows we are gravitating to in the field from
    a workstation/server perspective.
    
    We have such awesome technology available to us in this company, why oh
    why don't we get it up to speed?
    
    Another idea, how about all product managers with pictures on a CD-ROM
    which sits in my workstation or on a server on the wire.  That way I
    have the info. all the spec.'s, and a contact name to get a hold of if 
    I need some further information.  And here's another idea, what if we
    put this up on the wire and let customers access it and our business
    partners use it to find information out on our products.
    
    As Arsenio would say, "Hmmmmmmmmm......."

   See note 3525.0 for how effective using  'NODE::GENERIC_NAME' or
   'generic-name@node.dec.com' is.

    I vote for the idea.  In fact, I would vote for spreading this out even
    further then Product Managers.  Why not use it for other positions as
    well.
    
    Example, I'm in Australia.  I'm driving a customer program that will see
    equipment sold and delivered in several PacRim countries (Oz, NZ, Thailand,
    Japan, Singapore . . .).  I'd like to get in contact with a Sales Rep
    in one of these countries and drop a Blue Bird on his/her lap.  (Please
    log this order on your local systems.)
    
    With a carefully thought out and published naming convention, I could
    simply send a mail message to, say, ABU_MANAGER_THAILAND and ask for
    someone to be assigned.  If, 6 months later, I need some other
    assistance from the ABU_MANAGER_THAILAND, I can simply send another mail
    message.  I don't really care if it is the same person or not.
    
    This whole idea of having 2 (or more) mail accounts is not new.  I
    remember reading about the idea 12 years ago (has it been THAT LONG)
    when I was doing my MBA at UofC.  I was actually shocked when I started
    at DEC and learned that DECies only had 1.
    
    Now, how do we get beyond talking about it and get someone to start
    taking action?
    
    Later,
    
    Bill

    
    RE: Mappings.
    
    We could have an account say FULLINDEX and INDEX.
    
    FULLINDEX would return a complete list of all products and product
    managers with contact names.  You could put this on the PC and take it
    whereever you go.
    
    Alternatively, if you don't want such a large? file, you could mail the
    INDEX account with a phrase in the title and returns those entries with 
    that word/phrase in em.
    
    e.g.
    PRTMGR::INDEX
    
    Title: X400
    
    Just a thought.
    
    Mike

	This is a great idea - particularly the point about external
	access and "intuitive" addressing.

	It needs no corporate approval - interested groups/product managers
	can start today if they wish.  I have just set this up for the
	ABU/SI activity "Electronic Commerce" and will be publishing the
	Internet address shortly.

	Our overworked friends in IM&T could probably help though with
	a couple of suggestions (N.B. "suggestions" not rules) on
	naming and addressing convention.  The suggestions that have
	been made to me so far are ;

	1)	Use directory assisted addresses.  We have the infrastructure
		in place to do this and it resolves all issues of hard-wired
		mailbox names, multiple mail accounts, different mail agents,
		misspelling of product names - In short, WE CAN DO IT!
		
			e.g.	Product_Name@mts.dec.com

	2)	Agree on and use a fairly consistent convention for the addressing e.g.
		underscores in place of spaces, no puntuation, etc.

	3)	IM&T could set up a reflector account (say Product_Management@mts.dec.com)
		which returns an auto-reply of all Product Management contacts
		registered and their exact addresses.

	4)	Delegate the maintenance to the Product Managers.  This is as 
		simple as incoming Product Manager thinking to inform IM&T
		to route general poduct keyword mail to his/her account.  (Same 
		way that they probably already do with switchboard and mailroom
		staff).

    I think this is a great idea. Just do it!
    
    Laurie.

Would it not be better for emails to be fed into a call handling system that
could do automatic escalation when queries did not get a response? The
escalation should go all the way to the top (Palmer), that way you would not get
very many blackholes in the process.

Dave.

        RE: .21
        
        This is the best idea yet.
        
        RE: most of the others
        
        The idea of mail accounts doesn't work unless there's a guarantee
        that someone will monitor the account.  This would mean no
        forwarding allowed and some way to monitor if a mail message has
        gone some period of time without being read.  If mail is not read
        within about 2 weeks, an automatic mail message is generated to
        the product managers superior.  This can be repeated as many times
        as necessary.
        
        When mail is fowrarded, the process falls apart if one link in the
        chain not available for some reason (say TFSO).  It's rare that 2
        links get lost at the same time.
        
        Dave Zaniewski

    The system will need to be very tolerant about how users address the
    function they are looking for. Things like ABU_MANAGER_THAILAND,
    although it sounds great, has the drawback that ABU can become ZPY
    tomorrow and then be managed by CUSTOMER_DELIVERY in SINGAPORE. Digital
    organizations have never been too stable. Product names and families
    change very often ...

    Perhaps using a more intelligent mail processor would help. Allow
    people to send simple structured messages to the mail processor with
    commands in subject line like HELP PC HARDWARE and receive a response
    with Digital ROLES associated with that. Brief catalogs of main
    categories available can be obtained by sending a LIST command, or
    for example: LIST PC. We could also use some primitive WAIS based indexing
    code/algorithm to allow people to type cleartext search terms in the
    body of the mail that will then be used to locate roles similar or
    close to what customer/user is looking for.

    Unspecified mesages to be pre-processed by real Digital person at
    gateway. Someone would maintain the role directory, provided those who
    want to be reached provide the information. Tie to personnel files and
    automatically reassign person's records to supervisor if s/he leaves
    cost center so we don't end up with too many stale records.

    >Per

    
    	You know what is really nice about this whole idea?
    
    	It is packagable AND sellable outside our four walls (and one
    Ethernet).
    
    	I think it is task force, find sponsor, allocate resources, and
    do it time.
    
    			the Greyhawk

|    
|    	You know what is really nice about this whole idea?
|    
|    	It is packagable AND sellable outside our four walls (and one
|    Ethernet).

	Uhh .. lets forget it then .. typical digital: engineering things
	to perfection again .. invented customer requirements .. can't
	let that happen anymore :^)

	>Per

    Apparently folks don't understand that ELF provides such as service
    NOW!
    
    What makes it not terribly useful is that most people have not modified
    their "Position" attribute to give any hint of those areas for which
    they are responsible.
    
    No large project is required, no budget, no task force, no glory to
    uninvolved management, just each individual being committed to keep
    their own entry up to date.
    
    So the ball really is in the individual employees' court. This is the
    time to find out whether your resolve equals your rhetoric.
    
    You can modify your ELF entry from the $VTX ELF screen if you have (and
    remember) your ELF password. Passwords are obtained through your local
    MTS support group.
    
    Then to find the appropriate person from the $VTX ELF screen (it seems
    not to work from the command line interface):
    
    	FIND /POS=Product* 
    
    where product is the product name or area of responsibility on which
    you have a question. ELF only does trailing wild card (*) matches but
    you can abbreviate the keywords such as "position".
    
    Now doing FIND /POS=Alpha* finds a whole bunch of people involved in
    various aspects of the Alpha Program. In my case, since I am the
    MAILbus Product Manager, I will be found by FIND /POS=Mail*.
    
    Digital has it now. Whether Digital employees will "do what it takes"
    to make it work remains to be seen.

    > Digital has it now. 
    
    ... and it has long ago lost the battle for user acceptance within the
    Corporation, partly due to an antique UI and partly due to the general
    perception that it isn't available on all platforms.
    
    >           -< This is the challenge for all of you. >-
    > Whether Digital employees will "do what it takes" to make it work
    > remains to be seen.
    
    I disagree; I believe the intra-DIGITAL verdict on VTX has been in for
    a long while.  They haven't, so they won't.  VTX isn't fun to use,
    hence its infobases tend not to be frequented, so they tend not to be
    maintained.  Vicious circle.
    

    re .26
    
    What's a $VTX ELF screen?
    
    Signed, 
    perplexed Windows NT user in LJO.
    

    VTX Sucks. Any questions?
    
    Harry
    

    No one asked for an appraisal.

    From the VAX or Alpha perspective running Open VMS, VTX is a text
    retrieval program that has a large number of data sources.  ELF is the
    Employee Locator Facility and you can find the location, E-mail address
    and DTN of Digital employees using it.

    If from NT you can have a window open to a VMS system as a terminal,
    type VTX ELF at the "$" prompt.  I've include the first page of VTX
    help.

	mikeP

VTX

   The VTX command invokes the VAX VTX terminal control program that lets
   you access VTX infobases on your system or on other VAX systems that
   have VTX.  The VTX command also lets you access VTX information
   provider tools.

   There are three command formats for subscribers to use to run VTX:

   o  $ VTX [destination-page] [/LANGUAGE=language]
  	
   o  $ VTX/INQUIRE [destination-page] [/LANGUAGE=language] -- used
      only with DIGITAL terminals

   o  $ VTX/CLASSIC/DEVICE_TYPE=type [destination-page] [/PROTOCOL=protocol] -
      _$ [/LANGUAGE=language] -- used for all non-DIGITAL terminals

   There are three additional VTX commands for information providers
   to use to access the information provider tools:

   o  VTX/PAGINATE -- used to route documents to the VTX print queue,
      which paginates these documents for use in a VTX infobase

   o  VTX/UPDATE -- used to access the VTX Update Utility

   o  VTX/WORKBENCH -- used to access the VTX Workbench Utility



  Additional information available:

  Parameter  /USER_DATA /LANGUAGE  /INQUIRE   /INTERFACE /TRANSPORT /SERVER_NODE          /OBJECT_NUMBER        /PORT_NUMBER
  /FAILOVER_LIST        /CLASSIC   /PROTOCOL  /PAGINATE  /UPDATE    /WORKBENCH Examples

And the DECwindows Motif interface to VTX is much better than the character cell
version.  A version that would run on Windows NT would be nice, or is there one
already?

Paul

    
>   VTX Sucks. Any questions?
    
    Yes, one question. Any ideas on how to solve the problem?
    
    Personally I don't think that the user interface or database format is 
    the big problem. Not that I would mind if we moved to WWW or something 
    over time. But it would not solve the (IMHO) more important problem. 
    How does Digital (that's all of us) get the disipline(sp?) to keep all 
    that information up to date. I don't think that new UI's or database 
    formats does much  by them selfs to help solve those problems. 
    
    To me Digital always been a organization were the individual has had to 
    go and find the information he/she needs by him/her self. This isn't 
    possible to the same extent anymore, as many (most?) simply lack the time 
    needed to search lots of possible sources. And your colleges(sp?) don't 
    have as much time to share what they know. This means that we must rely
    much more than earlier on the information _that is_ available. And if
    that information is incorrect or not up to date, lots of people has to
    try to find the information elsewere. The problem has always been there
    it's just much more visible now.
    
    /Kjell

    >    Yes, one question. Any ideas on how to solve the problem?
    
    The Internet and the World-Wide Web
    
    Any questions?
    
    :)
    

    > Any questions?
    
    A good list of available converters from/to different sources of
    material would be useful to understand, how much better WWW would be.
    
    Minimizing manpower, which is required to update material and also
    the possibility to allow updates directly from information sources
    would certainly improve the quality of material.
    
    --Ritva
    

    > Yes, one question. Any ideas on how to solve the problem?
    
    Let's begin by losing the "Hit FORM ENTER" to continue BS. What is the
    FORM ENTER key combo on a generic VT emulator, running on a laptop
    anyway? I should be able to use the arrow keys to move around and do
    what I want. VTX assumes that I'm sitting at a VT 320. Bad assumption
    these days. 
    
    Now, what was the name of the database that I was looking for? Damn!
    Someone has changed the main menu page and NONE of the options seems to
    be the one that I'm looking for. What now? Oh, I know, I'll select
    EVERY ONE of them, one at a time, hoping that one of the submenus will
    point me in the right direction. Damn! The VTX page I selected is not
    available. Let's see, hit PEE-EFF-THREE to go back one screen. Damn! It
    took me all the way back to the opening menu! That's not where I wanted
    to be!
    
    VTX. The user interface is pathetic. The info is not indexed and
    incredibly difficult to search. Keywords often don't find the info,
    even though I KNOW that it's there. I guess that unless the keywords
    are defined, I'm outa luck. 
    
    I was one of the many that railed against the conversion of ELF to the
    VTX database a few years ago. I didn't like the VTX interface then and
    I don't like it now. It seems totally archaic in these days of point
    and click computing.
    
    Harry
    

    "-< Start from scratch >-"  ... no need to go that far back ...
    

    RE: .37-
    
    
    >Let's begin by losing the "Hit FORM ENTER" to continue BS. What is the
    >FORM ENTER key combo on a generic VT emulator, running on a laptop
    >anyway? I should be able to use the arrow keys to move around and do
    >what I want. VTX assumes that I'm sitting at a VT 320. Bad assumption
    >these days. 
    
    	Amen. I don't care how cool the "other" interfaces are- this a
    P.I.T.A. that has to go. Most Digits have none other than character
    cell terminals I'd venture. So, where is the FORM ENTER key(s) on these,
    huh?
    
    Phil

    >If from NT you can have a window open to a VMS system as a terminal,
    
    Now why would I do that?
    
    "SET HOST" as a method of connecting systems was the way things
     were done in the 1970s.  I don't have the patience for such 
     things, just like I no longer remember how to read Hollerith 
     code.

    You had HOLLERITH Code?  *L*U*X*U*R*Y*!!!!
    
    (Sorry, wrong file.)
    
    

    	The correct solution for NT would be an NT look-and-feel client for
    VTX. It can still talk the same protocol on the net and use the same
    database and database servers. Maybe such a client would not only ease
    our own use of ELF, it might even help us to sell NT systems.
    
    	Now which ISV are we going to try to persuade to write that for us?

    ... I'm not saying that an existing VTX customer wouldn't value a
    native VTX client for the NT platform... but why have yet another
    application written _de novo_ for NT in these days when a far more
    universal strategy exists?  The WWW paradigm of a universal info client
    has been extended to all major platforms, and needs only gateways to
    connect its users to legacy applications.  One exists for VTX.  This
    opens VTX infobases to perusal by Web clients, makes it unnecessary for
    another protocol to be spoken between client and gateway, and (most
    importantly imho) infuses new life into legacy VTX infobases by
    bringing UI-frustrated users back.  I'm not saying that Web clients and
    VTX gateways will prolong the life of VTX infobases indefinitely, but
    imho they allow a graceful transition period to a more web-centric
    future.  
    
    >  Now which ISV are we going to try to persuade to write that for us?
    
    ... Netscape Communications Corporation?  :-)
    

The WWW clients that i've seen are far superior to VTX (which is pretty
pathetic), but the same problem seems to exist of locating information
if you don't know the URL (as with the keyword in VTX).  It will be
interesting to see how people make use of the vast information resources
on WWW when something more efficient than "surfing" mode is required.

- paul

FORM ENTER?

VTX no longer requires this.  Many infobases, however, still require it even
though VTX has used the RETURN key (I believe most laptops have one of those
;-)) for a couple of years now.

Paul

    The scope of data available within the DIGITAL firewall is mind-
    boggling, but it's dwarfed by orders of magnitude by the external info,
    and the Web is still in its infancy.  So yes, the problem of locating
    info is real.  However, since the Internet is populated not only by
    brilliant starving CS students, but increasingly by brilliant
    entrepreneurs, the problem is well on the way to being brought under
    control.  There are many "directory services" and "Web-crawlers" and
    library services and knowbots abroad in Cyberia.  Closer to home, there
    are some excellent exemplars of all of those being developed at DEC. 
    
    Nearer-term, various groups are beginning to fund the creation of "Info
    Centers" maintained by both wetware and software agents, which serve as
    reference points for various end-user populations, such as engineers
    and marketeers. 
    
    In amplification & closing, here's a pleasant Vision of the Future ...
=====
From PC Magazine, 20 December 1994, p. 152, excerpted from "Gazing
into Cyberspace" by Bill Machrone & Robin Raskin:

...

"...by the end of 1995, you'll begin to see a new breed of worker: 
librarians and researchers who will help you get to the information
you're looking for.  Most of them will be human; others will be
software agents, and some will be cybernetic -- data robots guided by
humans.  People will specialize in styling and presenting data, just
as artists put a face on your published materials.  Programs that can
read, extract meaning, and create excerpts will become commonplace.
..."         

    I like Mosaic style user interface and I do believe, that it will
    be the correct approach for the next 10 years.
    
    I would, however, be glad to use ANY user interface, which would give
    me an access to a reliable, up to date, no nonsence information base, which
    would allow me to save valuable time in my work. As one of the previous
    noters pointed out, people in the field don't have time to surf around,
    be it Web or Notes or VTX, if we need to make profit with current
    product margins.
    
    Therefore, the tools, which make database updates a natural part of the
    imformation production are probably even more important, than the UI.
    That would support best the timeliness of information. We don't have too 
    many of these tools yet.
    
    I typically get information from a product
    
    1) through readers choice
    2) through different partner chains
    3) through product support organisations (often the slowest, unless
       it comes directly from the source
    4) from notes
    5) from SPDs
    6) from Sales Updates
    etc
    
    Typically, the message from these different sources is not the same,
    and none of them gives the whole picture. In cases, where the sources
    complement each other, it's (almost) ok. Just takes too much time to
    build the picture. Often the information is misleading or in conflict
    with the information with other sources. Building one more database
    doesn't solve anything, unless it's cordinated better, than the
    previous ones.
    
    --Ritva
     

re: .46

I agree that it will take some entrepreneurial folks to make sense out of
the information glut on WWW.  Also, .47's point is germane...stale and
incomplete (or even inaccurate!) information is a very real problem in
all on-line scenarios (not to mention information that's been tampered
with by hackers...<shudder>).

But the nature of the global network being very much bottom-up vs. top-down,
i don't think you'll ever get an overarching registry of sources (and
purging such a registry of stale or obsolete references would probably be
a mind-boggling task).  Maybe we (or some entrepreneur) could establish an
"information-bonding" service that people could submit their information
pages to (i mean bonding like Moody's or S&P rating company debt), with
some type of annual renewal required.

- paul

    >> Maybe we (or some entrepreneur) could establish an "information-
    bonding" service that people could submit their information pages to (I
    mean bonding like Moody's or S&P rating company debt), with some type
    of annual renewal required. <<
    
    Undoubtedly true.  There will be access-for-revenue services and
    freely-available services...  And there will be services in the middle
    ground -- where folks pay to be listed and verified, and users don't
    have to pay.  Only time will tell which model(s) will survive.  Within
    a matter of months, secure hypertext transfer protocol (SHTTP) and
    various add-on layers will provide means for actual commerce on the
    Web.  When that happens, the currently "wild-west-Web" atmosphere will
    be slightly more domesticated, and the current rate of growth should
    even accelerate!  Those big players who are now getting established
    will reap the benefits of preparation and familiarization.
    
    But we (mostly I) digress from the intent of the original discussion. 
    Pardon the rathole. :-)       
    

>    I like Mosaic style user interface and I do believe, that it will
>    be the correct approach for the next 10 years.

Given the pace of change tends to accelerate (particularly given that this
medium tends to cross-fertilize ideas), I suspect "information access UIs"
will probably evolve fast enough to make Mosaic look like a very early low
resolution bit-mapped workstation - a pale foreshadow of things to come.
One big difference this time around, however, is that many, many people
already have the requisite hardware for bigger/better things - this is
going to be a largely software revolution.

(Mosaic, while very good at being able to provide conceptual links, is
still primarily a user-driven locating mechanism - including having
to have the user do things essentially by hand.  One obvious extension -
particularly with 100mip+ desktops and increasing local storage - would
be a much smarter browser that could automatically navigate a lot more
and provide a list of "constructed links" of likely relevant information
before the user even asks.  Such a "intellegent browser" would increasingly
look like an "intelligent personal information assistant", and would probably
start to apply to more things rather than just "the web" - e.g. personal
"mail" (voice/data/fax), etc.)

    re: .48
    
>But the nature of the global network being very much bottom-up vs. top-down,
>i don't think you'll ever get an overarching registry of sources (and
>purging such a registry of stale or obsolete references would probably be
>a mind-boggling task).  Maybe we (or some entrepreneur) could establish an
>"information-bonding" service that people could submit their information
>pages to (i mean bonding like Moody's or S&P rating company debt), with
>some type of annual renewal required.
    
    We mostly solved this problem in DECgenisys V1 (prior to being
    connected to the Integrated Repository), by requiring all information
    that was inserted into our Knowledge Base have a responsible party and
    lifetime.  Thirty days before information was due to expire, it's name,
    responsible party, and expiration date appeared on a report which was
    used to notify information providers of the impending demise of their
    information.  (We could have automated the notification process, but
    didn't since we knew the IR was coming).  This eliminated stale
    information in the Knowledge Base, but didn't ensure that
    complete/replacement information was added.
    
    Now with our tie into the IR, we are at the mercy of the people
    responsible for keeping the IR current.
    
    Bob
    
    

     > will probably evolve fast enough to make Mosaic look like a very early
     > low resolution bit-mapped workstation
    
        That's why I said Mosaic style... ;^).
    
        Before we can have really a lot better UIs, telecommunication
        costs/bit must come down to make more complex tools, which usually
        mean larger transfer volumes, feasible.
    
        What comes to the information content, I don't expect that Internet
        at large becomes more organized, but parts of Internet can. I believe,
        that the current anarchy has some real value in itself by allowing
        unlimited innovation within the worldwide community to flourish. I
        hope though, that Digital can build a nucleus of information, where the
        content is under continuous check to garantee a source of essential,
        up to date information. I also believe, that that's mandatory, if
        we want that resellers willingly sell our products and that they are 
        successful in it.
    
    --Ritva
    

    re: the last set of notes concerning WWW and current information
    
    Work is underway to connect the IR with the Web, which should help with
    the problem of finding up-to-date information from the responsible
    person/group.
    
    There are also Web-based services that address the problem of how to
    find information:
    
    	1. the Digital Announcement Directory is an online catalog of over
    	   900 Web servers, both internal and external, with descriptions
    	   and links to the server. Anyone can announce to the Announcement
    	   Directory, and it is kept updated constantly. It is at
    
    		http://lgp30.mso.dec.com/announce/pa-toc.html
    
    	2. The Web Pages are a common set of pages categorized by major
    	   topics with updated lists of servers. The categories include
    	   internal and external servers, competitors, financial, 
    	   reference, government, education and other servers. It includes
    	   search tools, technical information about the Web, and a Web
    	   Toolbox. And there is a Feedback form where you can ask for
    	   a specific type of information or service. The Web Pages help
    	   users (especially new users) deal with the overwhleming amount
    	   of information available online via the Web.

	   The Web Pages are at:

		http://www-iu.mro.dec.com/public/front.html

    
    	   Various groups have set up their own focused Web pages (e.g.
    	   Seminconductors has one, there are several Sales related pages),
    	   again to help people find classes of information.
    
    These are not complete solutions, but the good news is work is underway
    to make the WWW more easily applied to the work we do.
    
    /Tim Horgan

	<FLAME ON>

    I cannot resist but to comment on the more recent VTX bashing notes. I
    find them arrogant and misinformed. All this complaining about FORM
    ENTER in VTX and that the Web/Mosaic will cure the worlds problems. 

    There are at least 5 problems:

	1. Features of the information delivery vehicle (capabilities,
	   platform coverage, UI, etc).

	2. Management of information backbone (presenting a coherent set
	   of information repositories)

	3. Dissemination of information access clients/servers to ensure
	   parity for the masses.

	4. Creation of homogenous repositories of information.

	5. Maintenance of repositories

    The Web and Mosaic can only excell in one single category today: #1,
    in particular if you compare it to VTX of 1989, like most of you seem to
    do. And yes, receiving substatially more funding and attention, Mosaic
    and the Web will forever remain ahead of VTX, in category #1.

    Because VTX has been one of our centerpieces for delivery of reference
    information a lot of the problems that you have seen with it in recent
    years are a direct result from exactly the same things that can make
    Mosaic and the Web look like sh*t:

	1. Inadequate staffing to manage the information backbone. Lack of
	   staffing to establish and enforce look and feel guidelines.
	   This includes training the information providers. We all think
	   we're expert providers of information..

	2. Lack of resources to ensure that there is indeed a Mosaic client
	   on everyone's desktop and assurances that new information is
	   developed for it, and not something else. For Web/Mosaic at Digital
	   this also includes a program to phase in Mosaic capable desktops
	   for all employees. A hacky VT100 Mosaic client from some univeristy
	   won't do, it will only create unwarranted friction between the
	   providers and the consumers.

	3. It takes people to create appealing information repositories
	   that follow the guidelines. This is in particular true for
	   information intended for a large audience. It can't be done as
	   a midnight hack, by an overworked and overcommitted engineer.

	4. It takes continued support from dedicated resources to keep the
	   repositories up to date. All these so wonderful and flashy demos
	   tend to get stale when the originator runs out of midnight oil.

    All of these problems have been in particular noticable with VTX when
    we've laid off "overhead types" and substantially cut back engineering.
    BTW, I'm sure that the recent dumping of documentation people isn't
    exactly helping us to accelerate the creation of usable Web/Mosaic
    based repositories. 

    Also, this attitude of "don't develop anything more for VTX cuz its
    dead" is very shortsighted. If the customer wants a VAX or if a VAX is
    the right solution for the customer, don't insist on selling an Alpha.
    This holds true for VTX versus Mosaic too. We need every dollar we can
    generate.

    I am not suggesting that VTX is the right vehicle for what is being
    discussed in this string of notes. But do not lose sight of the
    complete scope of the information delivery system that we are talking
    about here. VTX and MAIL is on everyone's desktop, Mosaic isn't even
    close.

	<FLAME OFF>

    >Per

    Kits that may work with windows nt are at
    \\vtxpce\public service vtxkits vtx214 disk1   like setup
    
    For OpenVMS on Alpha vtxa061.a and .b at welkin::
    
    As for form enter, "Press HERE to submit, or HERE to reset", yes
    soooo much better. :-)  Ya need something to Send, you can have
    it anywhere on your screen with http, on the bar in vtx.  
    Wanna bet they also move that to the bar? :-)
    
    Web have inline images, maps.  VTX has session management.
    Outside of those differences, (and the client re-scaling the
    font) presentations can be made VERY similar either way.
    
    P.S.  VTX information can be made available on the Internet,
    Compuserve, other nets and dial-ups.  So you can start from scratch,
    or keep evolving or set the newest fad;  keep the info with
    vtx, you'll be able to connect (most cases). :-)

> And the DECwindows Motif interface to VTX is much better than the
> character cell version.  A version that would run on Windows NT
> would be nice, or is there one already?

    In case any of the recent replies went clear enough,

    There IS a VTX PC/Windows client that runs under Windows 3.1, WFWG
    3.11 and Windows NT. It's located on a file server called...

    \\VTXPCE\PUBLIC\VTXKITS\VTX214\DISK1
    Run SETUP.EXE

    Choose a transport,  

    a. Decnet Network Support or 
    b  for TCP/IP, use WinSockets Network Support.

    Charly

    Is VTX PMG being kept current ?
    
    
    

    You'd have to contact the person listed on the PMG menu: Colleen Mears,
    DTN 640-7112
    
    Gita

    The problem is more general than "who is the product manager for XXX".
    There is an age old problem of finding the right person (sorry, human
    resource) to solve a problem in Digital.  
    
    In "the good old days" this was solved using the  old boys network. 
    Everyone built up their own database of contacts who knew someone who
    knew someone who would know what the answer was.  But those days are
    gone.... if one of the contacts in the chain is TFSO'd, the whole
    process breaks down.  The result?  Customers left hanging on waiting
    for an answer, and eventually going to a competitor.
    
    It is almost impossible to get a general answer to the simple question:
    
    	Who is responsible for XXX?
    
    where XXX might be a hardware or software product, a customer account, 
    a technology, another employee, a cost centre, etc.
    
    Every day we all get mails with the title "Announcement...."
    The information is rarely held centrally so that anyone can get hold of
    it when they need it.  Sometimes it might get onto VTX and is then left
    there until it is out of date and useless.
    
    WHY?
    
    The problem is the middle word of the question: RESPONSIBLE.
    If no-one is responsible for making sure the information is available
    to everyone, then Digital's organisation will continue to remain
    shrouded in mystery.  If you make someone responsible (e.g. for putting
    it into VTX), then what happens when the responsible person leaves?
    
    .6 and .7 and .26 are right - ELF provides a limited capability to solve the
    problem, but few people keep their ELF entry up to date, because they 
    aren't FORCED to be responsible for it. 
    .27 etc show why - people don't like the interface to change their details.
    
    So what's the answer?
    
    .35 is partly right - the Web provides the sort of capability we need,
    but there's more to it than that. It needs:
    
    1 - A centralised repository of information about employees, the
    	organisation, etc.  (a sort of ELF++)
    2 - Every employee is MADE to be responsible for keeping their
    	own "CV" up to date.
    3 - Every manager of a group is MADE to be responsible for keeping
    	the information on the group up to date.
    4 - All the information is indexed and tools are available to allow
    	searching on any criteria.
    5 - It is easy to update the information by a method which is
    	independent of the user's method of access.
    
    1,4 and 5 are the easy bits.  I've set up a prototype for the UK
    employee population using the WorldWide Web.  All a UK employee has to
    do to update their details is send a mail.  With a few days work,
    access to the appropriate data, and a bit more disk space and CPU
    horsepower it could be made  a worldwide database. It automatically
    updates the reporting structure and allows "groups" of people to be
    defined which are independent of reporting structure.  
    
    2 and 3 are the heart of the problem.  The ONLY way to ensure that a
    database with employee details is kept up to date is to make it 
    MANDATORY for employees to keep their own details current.  
    To do this requires that each employee sends maybe one mail a quarter.
    Is that too much to ask?  Any other way requires a group of people
    dedicated to keeping things current, and that the group stays in place
    come hell or high water.
    
    If you want to look at my ideas, have a look at
    http://midoff.bbp.dec.com/digital_uk/digital_uk.html
    
    It's far from perfect, but shows what can be done with relatively
    little effort and almost zero support.
    
    -- Rob

>     The problem is the middle word of the question: RESPONSIBLE.
>     If no-one is responsible for making sure the information is available
>     to everyone, then Digital's organisation will continue to remain
>     shrouded in mystery.  If you make someone responsible (e.g. for putting
>     it into VTX), then what happens when the responsible person leaves?

    EXACTLY! This IS the issue here, RESONSIBLITY! Not only for
    creating and making the data available but for MAINTAINING that
    data! Be it in VTX, WWW, a Mailing List... where ever.

    VTX has been around for a while, so there is plenty of stale
    information, dead-end links and non existent contacts there. 

    The same thing will happen with the WWW and to an even greater
    extent! Just give it time. I'm already seeing the beginnings of it
    and it will only get worse. 

    The real issues here are not the technologies used but rather the
    data managment and data resonsibility.

    Charly

    I'm not denying the possibility, and even the existence of the
    beginnings of, the stale-info problem on the Web.  But I think it's
    relevant to point out that by using that environment and the
    proliferating tools that are "coming out of the woodwork" worldwide,
    both shareware and buyware, we are considerably relieved of the task of
    creating and maintaining the tools to manage and index that info.  VTX
    is/was an INTERNALLY PRODUCED environment, thus we bear/bore FULL
    RESPONSIBILITY for all the tools needed to create/maintain/index/browse
    it.  Not so with the Web.  
    
    It won't make the problem go away -- but imho it should be far easier
    to manage than the VTX environment.  Plus, using the Internet/Web
    environment and toolset will put us in closer touch with our customers'
    common practices in internet computing.

    I personally have nothing against using Web as THE information
    provider. What I'm against is, is to use it as ONE MORE partial
    information base. Don't sell it as a tool, which takes care of
    itself. Building efficient sales and support info is a serious job
    and mandatory for the survival of the company.
    
    What comes to Web servers in Digital, most of them seem to be just
    "me too" experiments with no real additional value.
    
    And don't missunderstan me, I like Web.
    
    --Ritva
    

WWW access banned?  By whom?  The Sales Prevention Office?  Someone needs a
whack on the side of the head.

Paul

    RE: .63
    
    What??????  Why don't we all switch off all our computers to save
    electricity.  Back to paper and pencils, guys.
    
    Laurie, if you want access to the Web, you can have it from my system.
    From your account on PLAYER, SET HOST (or Connect to System if it's A1)
    to node MIDOFF, and you should get straight in.  If you want access
    from any other nodes, let me know (if PALYER is a cluster, I'll need
    all the nodenames of the cluster).
    
    -- Rob
                                                                 

    Anyone else who agrees, kindly chime in as to why those who instituted
    that policy at that site should reverse that decision.
    

re .63,

I'd be interested in knowing what resources are supposedly being
wasted; AFAIK, the current batch of Web browsers and servers are very
resource friendly:

Browser connects to server and makes request

Server satifies request and disconnects from browser

which means that the network connection is active only until the
current request is satisfied, as opposed to applications like VTX and
NNTP, which maintain a client<->server connection for the duration of
a session...

I try to _encourage_ WWW usage, not stifle it.

/harley

    When decisions are so off-base as this one is they get reversed,
    subverted or ignored as this one will be and the decision taker
    will end up claiming that his/her words were "misinterpreted" :-)
    and in any case it wasn't really a "decision" and gets to save 
    his/her face.
    
    More importantly, what I don't understand is how someone with 
    apparently enough seniority to be able to "take" these decisions can be
    so completely ignorant of the importance of the Web in a company
    that not only claims to be Internet proficient but even Internet
    leading edge.
    
    In this sense I guess all of us are probably at least partially 
    to blame :-(
    
    re roelof
    
    

    Good note, roelof.  You've obviously "been around the block" many times
    in the "DIGITAL Culture." (note:  I didn't say the DEC culture.)  :-(
    
    Any other testimonials?  Speak up out there, let's help .63 ! 
    

    And if there's still anyone out there with even the slightest 
    doubts of the commercial importance of Internet (and access to 
    Internet for anyone within Digital, this excellent independent 
    case study of Digital and the Internet has just come on line: 
    
	http://colette.ogsm.vanderbilt.edu/dec/internet.html

    (Oh, did I forget to mention that you need Internet/WWW access to read
    this commercially valuable case study? Sorry...:-)
    
    re roelof

Re: .70

>	http://colette.ogsm.vanderbilt.edu/dec/internet.html
>
>    (Oh, did I forget to mention that you need Internet/WWW access to read
>    this commercially valuable case study? Sorry...:-)

For those of you who DON'T have Internet/WWW access, the ASCII version is
available at the above location.

-- Shel