Conference 7.286::digital

   I guess I would like to discuss how this wonderful marvel called
   DECwindows will find its way into the daily lives of the many Digital
   employees.  Of course, DECwindows is merely a platform from which many
   products can spring.  If indeed DECwindows is as the first paragraph stated
   "is considered to be the most expansive commitment that Digital has
   ever undertaken" then certainly it will eventually have an impact on
   the way Digital employees do their computing.

   If I were able to hazard a guess, the average 'user in the field'
   will not be using DECwindows-based applications for a number of years due to
   these reasons:
      
   - Budget cutbacks means less bucks for ordering new equipment,
     especially equipment that will obsolete existing equipment.  The
     buzzword today in the field today is ROI. If you can't prove it will
     save a bunch of money then you don't buy it. Sorry, DECwindows won't
     run on hand-me-down VT52's and 100's Mr./Ms. Field Manager.
      
   - Internal applications won't migrate to DECwindows becuase the
     equipment base in the field won't be ready- and many application
     groups will be reluctant to support applications on two different
     platforms (character cell (VTs) and DECwindows).
   
   - Managers will be reluctant to order equipment to support
     applications that aren't ready yet (this and the item above will
     create a chicken-and-egg syndrome and neither will occur unless
     someone steps in with a large sum of $ to pay for it).

   - Networks will require beefing up as if a more distributed approach
     is to be taken with applications. I feel, for the most part, we are
     growing towards centralized computing instead of distributed.
   
   - Support for existing applications is often inadequate. Support
     staffs would need to undergo retraining and/or expansion to support the
     new equipment and applications.
   
.0>This undertaking reflects a change that keeps Digital in step with 
.0>the computer industry as a whole.  This change centers on the 
.0>support of distributed applications, of heterogeneous environments,  
.0>common graphical user interface, and development of open systems as 
.0>a standard.

   A debateable point. Is windows and/or distrubited applications the
   direction the computer industry is going? If so, do we need to go that
   direction with our internal computing? Would it benefit Digital to
   move the bulk of its applications from a centralized, character-cell
   based scheme to a distributed, DECwindows based system? Do we have
   management with enough vision and clout to steer us in that direction?
   Would this help us do a better job of selling and supporting
   DECwindows and distributed computing to our customer base? Or would it be
   better to plan on staying with the character-cell/centralized computing
   because it is better understood?
   
   I really don't have any answers, and I'm sure that there are a 1000
   more questions. I guess I feel that Digital is attempting to found a
   new way of computing- enterprise wide, networked, distributed,
   workstation/windows based. Digital would like for customers to embrace
   this new way and let Digital be their partner in making it happen.
   Digital seems to have the tools to accomplish it. When will the
   internal Digital world partake of this new way of computing?
   
   Just trying to start some discussion.
   
   Lee

    There is nothing new about graphical interfaces. People with Macs
    and IBM PCs have been doing it for quite a while now. The product
    set in that space is quite mature now. Many products are up to V3
    or V4. The platforms that they run on are mostly less than $10k.
    
    On the other hand, we do not have a software product in this arena.
    DECwindows is in FT2 and the Enterprise (nee Epic) product suite has
    not reached internal field test yet. The platform is somewhat more
    expensive when you consider that to really do anything you need an 8800
    to serve your 6 meg VS2000 with a network in between. Besides that you
    need Jack Smith's signature to get any of this stuff internally. 
    
    What the message should say is that we're quite far behind and realy
    need to get off the pot if we are going to be a seriuos player in
    this market. We have the hardware, we've had it for 4 or 5 years
    now. We just don't have any software to use on it. There are a lot
    of VS's around the company. They just don't seem to be in a place
    that makes sense. 
    
    

    
    
    I'd love to be able to use DECwindows internally, for my own selfish
    reasons, as well as to be able to justify to customers that we use
    what we say they should use.
    
    However, we would need an intelligent display (something more than
    a VT) on every desk, and at least in the field I don't see that
    happening for years.
    

    This note is a few weeks old.. but I'd like to say that I'm ENTRANCED
    by DECwindows and the EPIC tools !  As a Worksystems Specialist,
    I was able to receive my very OWN VS2000.  And this is in Arizona.
    
    However !  The VS2000 has only 4MB of memory, and the latest 
    version of VMS DECwindows I've received literally WILL NOT RUN more 
    than 3 open windows.  The previous version slowed down, this one
    halts the system.
    
    NO ONE seems to know how to get these 4Mb VS2000s upgraded to 6Mb.
    
    Advice ??
    
    thanks, karl moeller principal specialist TUO

	Check DEC-direct or the Systems Catalog for VAX.  I think you can 
order a new memory board that will either replace the 4Mb one, or add 2Mb.  If
you can get the 80% discount on an internal order, and the total cost is below
$1000, then your CC manager doesn't have to capitalize it.  Unfortunately, I
don't recall the price of the upgrade option. 

	Of course, you know what our priority for internal DELIVERY is, don't 
you?  Right after we fill all single unit, bare-bones orders from customers!

        On page 2.13 of the 1988 October-December Systems and Options
        catalog, it lists the MicroVAX 2000 and VAXstation 2000 memory
        upgrade options.  The parts are:
        
        	MS400-BA	4 Mbyte parity expansion memory
        
        	MS400-CA	12 Mbyte parity expansion memory
        
        There's 2 Mbyte on the system board, and 2 Mbyte on the expander
        card in the box (which you can throw away), so the -BA gives
        you a 6Mb system, and the -CA gives you a 14Mb system.
        
        As for actually getting these, that's a problem I'll leave
        it to you to solve.
        
        Tom

Yeah, but the problem is that he can't order w/o his CC manager's approval,
and that's highly unlikely in the field as a general rule.

	But in this case, I bet the CC manager would realize that his job 
can't succeed without this resource, and since Country (?) is funding these 
positions anyway, the expense at 20% would be in the noise.

	But, as stated before, the problem won't be money, but availability.

Of course, you could always buy the memory from Clearpoint
whos  memory add-ins give a maximum of 16MB (ours is 14MB) 
and their delivery is much sooner.

_bill

Who has a 16MB VAXstation 2000 somewhere

        The 4MB boards are available if you know where to look (I have
        it figured out, but I'm not telling).  The 12MB boards aren't
        readily available, as far as I can tell.  Bill's suggestion
        about buying the Clearpoint board is perhaps reasonable, if
        you can (a) get permission to spend money outside, and (b)
        actually get the thing delivered, and (c) get FS to support
        it.  And the difference between 14MB and 16MB in performance
        is likely to be impossible to measure.  (The difference between
        6MB and 14MB, however, is probably very real.)
        
        Tom

    re .4
    
    You might want to run Autogen again, or otherwise check your
    workstation tuning.
    
    I'm currently running DECwindows on a standalone 4-meg VS2000. It
    ain't zippy, but it does run, and I tend to have more than 3 windows
    open (I run banner, which I have configured as 4 small windows
    across the bottom of the screen, plus 3 or more DECterm windows and
    the occasional Mail or FileView window). With a system this starved
    for memory, the only way for it to work reasonably is to make sure
    DECwindows is installed on a faster machine (e.g. your local cluster
    if any) and run significant applications like Mail or Calendar (or
    Epic/Write, which is a joke on a 4-meg system) on the cluster with
    their display output redirected to the VS2000. 

    Interesting discussion.
    
    One thing of importance, though, which was mentioned briefly by
    Lee in .2, is SUPPORT!  I've had a VAXmate running MS-Windows
    (not DECwindows, I know, but I think it's the same type of problem),
    and only one person in the building knows how to support the software
    required to keep this thing on the LAN.  If he's out and some component
    of our LAN goes down, our VAXmates are next to useless.  
    
    All the talk about getting the money and the equipment is certainly
    valid and important to the discussion.  But another question to ask
    is:  Once I get it, can I use it?  Machines running MS-Windows and
    DECwindows really need knowledgeable support resources in order for
    them to work right.
    
    
    

       RE: .12
       
       	People like that are very hard to come by.. They are usually of
       the system management type and GOOD system managers are REALLY hard
       to find! (We've had reqs open for a while!)
       
        These type of jobs also don't seem to some management types to offer
       the biggest payback cuz it's hard to quantify the work done..
       "Well, I helped so-an-so.." "But did you get whatever done??"
       "Uuuuhh"  That's unfortunate..
       
       						mike

    Re: .13
    
    TELL ME ABOUT IT!!!
    
    My last three reviews rated me low on "task completion". Guess what
    I spent most of my time doing? Servicing calls. Helping people.
    
    Oh well, can't win them all. And in the end it call comes back to
    you.
    
    							Warren
    

    Interesting, Mike.  When I moved up here to Engineering I spent a
    little bit of time on doing system management for the GPXes in my 
    group (and helping our cluster manager) so people would use them.
    Guess being from the field there was something repulsive about fully
    loaded GPXes just sitting there while people used VT100's 'cause
    they were easier.
    
    My manager's comment?  Real engineers don't do system management. 
    Oddly enough, when I look around I see that attitude all over the
    place.  Yet, without good system management engineers can't get their
    work done.  I've worked with good system managers and with bad ones.
    You notice the difference from day one onwards.

       {Sorry to rathole here folks}
       
       	Agreed Jim, what is needed is a team approach between the
       engineers writing the code and the system managers assisting
       in providing an ideal environment to the engineers. And of course,
       this not only applies to engineers/system managers but to 
       whomever/system managers.
       
       	Thankfully in some departments, helping the other guy out with
       a problem, possibly at the expense of 10 things sitting on your
       desk and 25 mail message waiting for you, is encouraged..
       
       					mike