Re: Counting users

Al,
A good summary of why user-based pricing doesn't work, although it would be
nice if it did, because machine based pricing just causes under-sized boxes
and performance problems.
Clare

----- Original Message -----
From: <MacWheel99@aol.com>
To: <MIDRANGE-L@midrange.com>
Sent: Tuesday, July 31, 2001 5:55 PM
Subject: Re: Counting users


> Take your pick ... my full name is Alister Wm Macintyre
> to save hassles with phone callers (how do you spell pronounce ... is that
a
> boy's name or a girls name ... I often go by Al Macintyre
> around the office work place my moniker is Al Mac
>
> Do you have a naming policy for your auto configuration?
> We let auto config do its thing with DSP99 or PRT07 then we change the
name
> to one based on location
> PRTSHIP means printer in the Shipping Dept
> anything without a "P" in front is a work station
> BUY9 means it is in the Purchasing Dept
>
> In the Textual description area we identify the building location
> )e.g. NW corner of EVA factory)
> or the person desk it is on or office it is in
> (e.g. RECEIVING of LAW factory)
> and how many sessions that work station is capable of supporting
>
> This serves a multitude of functions.
> When we got error messages we not need to be saying
> "Where the heck is DSP27?" because many people know our naming conventions
&
> recognize what stuff is in accounting area, engineering areia, sales area,
> etc, & a few know how to look it up in the configuration.
>
> Many people meet variations on this question when it comes to software
> licenses & what's honest & fair when the methodology of license
enforcement
> is heavily based on our integrity, and it also comes up in sizing for
> performance.
>
> In a small office you can answer the question with a walk around.
> WHO is at that device, and how many are unattended with someone allegedly
> signed on & DSPLOG will tell you WHEN they signed on, and when you find
the
> USER of that user-id we can ASK them.
>
> You may know someone at a remote office who you can trust to do the walk
> around to diplomatically locate the 5 sessions that currently signed on
with
> same user-id & get you a spot check inventory ... while actually have
someone
> standing or seated at that address, and if no one home, ask nearest work
> station user if they remember who was last person to use that device.
>
> I do not believe DSPLOG can go to an *OUTFILE to chart on/off by user id
to
> locate overlaps, but if you have your suspicions about some pattern, you
can
> do DSPLOG F4 & search for specific types of actifity & in fact I have a CL
> that regularly does a variation on that.
>
> We do have some supervisory personnel who need to go to different places
in
> the company to consult with people on wide ranges of topics & while there,
> use whatever device is convenient to look up stuff, and invariably leave
it
> signed on, and hature of person job is they need high security, and I sure
> don't want to be the one to undermine their productivity.
>
> I am aware of the practice.  I know the persons who doing it.
> I have no authority to stop it.
> When people ask me how come this or that data got corrupted,
> I include this security risk as one of the possibilities.
>
> In a larger enterprise where you not know everyone, and there are remote
> offices, you may need a tool like http://www.precosis.com.au/rv1.htm to
spy
> on WHAT are those users DOING
>
> We have twinax work stations scattered across our factory floor.
> We have a very limited inquiry only sign on with a public password - by
using
> it the people can only get to certain menus, certain inquiry options into
our
> inventory & production & engineering information.
> On a typical day, this same identical public sign on is being used at half
a
> dozen different work stations in each of two factories for random look up
of
> stuff by people to whom we have not issued individual user profiles.
>
> This is a corporate policy that was suggested, discussed, approved.
> It is well known throughout the company.
>
> I could see that in some enterprises something like this might be going on
> without everyone concerned being aware of it, knowing why it is being
done,
> or even approving of it.
>
> Sometimes when a new person is hired & taught how to do their job, somehow
> MIS is not informed that we need to issue a new user profile, and they are
> using the sign-on & password of a co-worker.
>
> We have people who are the only people with security access to some
important
> stuff.  They go on vacation.  Their work gets done while they are away.  I
> have my suspicions how this is accomplished.
>
> I personnally have 5 user-ids.  I use different ones for production vs.
test
> environments, regular programming work v.s  high security stuff.  I am not
> the only person in this reality.  A count of user-ids is not a count of
real
> person users.
>
> We also have some dummy user profiles created to be depositories of shared
> messages, that started out as a different purpose
>
> DSPMSG SALES to see messages that the sales dept needs to be aware of
> Originally user SALES was intended for random inquiry by people who were
not
> going to be heavy BPCS users.
>
> The idea was ... issue user profiles to people doing heavy duty work & who
> needed different mix of security access, while each department have a
limited
> security public sign on for the folks who needed occasional access to some
> inquiries ... this simplified training & we could have index card on work
> stations with sign-on & password which was setup so there was no way to
spoof
> past that into stuff they not allowed to access, but then they started
> putting that on PCs & I had to pull that plug because of how BPCS 405
> "security" structured.
>
> I think that perhaps software could be developed & may even exist to track
> the kind of information that performance tools get ... which of these
> interactive sessions are active at any one moment.  With a person signed
onto
> multiple sessions of one terminal, theoretically they only active on one
side
> AT A TIME.  With SAME person signed onto multiple terminals, theoretically
> they only active at one AT A TIME but some other user might come along &
> start using the signed on terminal, without first signing off the earlier
> user & signing on themselves.
>
> > From: Ron@cpumms.com
> >
> >  MacWheel99@aol.com (Alister Wm Macintyre) (Al Mac)
> >  Not sure what name to use here....
> >
> >  This is exactly what we're trying to discover....
> >
> >
> >  << I suspect the reality is not really
> >  1 person signed on at 5 terminals vs.
> >  1 person signed onto 5 sides, but that the
> >  1 person signed on at 5 terminals is really
> >  5 people using the same sign-on,
> >  while the 1 person with 5 sides is
> >  really one person at a time at that terminal. >>
> >
> >
> >  Ron Hawkins
>
>
> MacWheel99@aol.com (Alister Wm Macintyre) (Al Mac)
> +---
> | This is the Midrange System Mailing List!
> | To submit a new message, send your mail to MIDRANGE-L@midrange.com.
> | To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
> | To unsubscribe from this list send email to
MIDRANGE-L-UNSUB@midrange.com.
> | Questions should be directed to the list owner/operator:
david@midrange.com
> +---

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---