IT industry favors the young
Jeff W Morris
stated: "Anyone who has been in the industry for more than 10 years should
know that the IT industry favors the young and that the demand for skills is
far less in management than in raw technology".
Is this statement true for the traditional Unisys environment with which
the participants of this Newsgroup are familiar (excluding ES7000/Microsoft
toys for boys)?
I believe we are a group of highly skilled professionals but we are an
ageing population and there is not enough being done to attract youngsters
to our exciting and complex world.
If we took a poll of participants in this Newsgroup, I believe that a large
percentage would be over 45 or even over 55. What are we doing to attract
young entrants (20-25 year-olds), bearing in mind that none of us are
immortal?
There have been cries for help in this Newsgroup recently by people who
admit that they are new to the environment. The exemplary help and
encouragement they get is praiseworthy. However, I would like to know how
many of these newcomers are 'young' (whatever the computer weekly writer
means by that) and how many are contemporaries of us old stagers, just
changing ships in midstream?
I ask anyone responding to this thread to include an indication of their age
and I will collate the results. To get the ball rolling, I am close to 60.
--
Dr Jeff W Morris: Pretoria Software Solutions
One of the largest LINC software houses in the Southern Hemisphere
Specialising in 'offshore' system development and maintenance worldwide
Jack Benny
Worked on Unisys (Burroughs) for 27 years. Small, Medium and
Large(A-Series).
On Tue, 22 Jan 2002 20:40:57 +0200, "Jeff W Morris" <p...@icon.co.za>
wrote:
>In a letter to the editor of a local computer weekly a writer recently
>stated: "Anyone who has been in the industry for more than 10 years should
>know that the IT industry favors the young and that the demand for skills is
>far less in management than in raw technology".
>
> Is this statement true for the traditional Unisys environment with which
>the participants of this Newsgroup are familiar (excluding ES7000/Microsoft
>toys for boys)?
>
> I believe we are a group of highly skilled professionals but we are an
>ageing population and there is not enough being done to attract youngsters
>to our exciting and complex world.
>
> If we took a poll of participants in this Newsgroup, I believe that a large
>percentage would be over 45 or even over 55. What are we doing to attract
>young entrants (20-25 year-olds), bearing in mind that none of us are
>immortal?
>
> There have been cries for help in this Newsgroup recently by people who
>admit that they are new to the environment. The exemplary help and
>encouragement they get is praiseworthy. However, I would like to know how
>many of these newcomers are 'young' (whatever the computer weekly writer
>means by that) and how many are contemporaries of us old stagers, just
>changing ships in midstream?
>
>I ask anyone responding to this thread to include an indication of their age
>and I will collate the results. To get the ball rolling, I am close to 60.
______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
With NINE Servers In California And Texas - The Worlds Uncensored News Source
Cairns Hardware CO PTY LTD
Working with Unisys BTOS/CTOS Hardware for the last 10years. Ngen&Sgen
series
"Jack Benny" <pkl...@nospamltdcommodities.com> wrote in message
news:tcgr4u87nqkgdnocn...@4ax.com...
Bryan Souster
news:a2kbn6$3c7$1...@ctb-nnrp2.saix.net...
Hello Jeff,
I am in the over-50 group and my background is MCP Servers for the last
20-odd years.
While I agree that in general the IT industry related to mature technologies
(this excludes MS and Unix btw) is for the main part populated with highly
skilled and experienced people that are typically in or near my agegroup,
the situation seems mostly created by the following:
- Our western world Universities are dominated by MS and Unix platforms.
Students therefore only get this exposure.
- Mature technologies such as those based on the mainframe, centralised,
typically transaction-processing technologies (ie MCP, OS2200, MVS etc) are
generally percieved by the young graduate as dated, dying technologies that
will inevitably eventually be replaced by more up-to-date Windows and/or
Linux) servers.
The problem is not so much that our part of the industry is dying but the
perception by the youngsters entering our industry that MS/Unix (with Linux
at home) is the place to be. This is borne partly from an ignorance of
mature technologies and partly the natural recoil-reaction of working in a
world full of old codgers like us.
Unix has had its day. MS are still so laughably short of ever producing an
operating system capable of centralised enterprise-wide services (which must
start with the entire elimination of virus/worm infections) that the next
decade will not see them get much beyond the desktop. Each release of all
their software will contain the current trend of 'greater reliability' and
'less vulnerabiliy'. Bill Gates recent pronouncements on trustworthy
computing (if I have understood what he said correctly) reveals just how
juvenile MS thinking is.
The ES7000 is an excellent large-scale thin-desktop-client server when sold
as an ES7000, but but the true future of CMP is with Clearpath Plus.
The problem was partly highlighted by year 2000 issues and will rear its
head again over the next decade as we retire and companies do NOT abandon
mature technologies. Action to fix this issue will only be taken when
individual employers face the skills loss and realise that the mature
technology they use is still the way forward for the forseeable future. Then
they will be forced to take action and recruiting youngsters will be the
inevitable result.
Well, enough from me - just my 2c worth.
Bryan.
Edward Reid
> If we took a poll of participants in this Newsgroup, I believe that a large
> percentage would be over 45 or even over 55.
53 next month. I fit the hypothesis to a T.
and wishing I could challenge those young'ns by retiring today ...
I always said Burroughs should have been giving away B6700s to
universities despite the cost. Instead, they pushed the universities so
hard that most of them left for other systems. I still say they should
be giving away LX100 licenses for academic use -- though without a
Fortran90 compiler they are at a distinct disadvantage even with free
software.
I do enjoy the people who come on here saying "I found this old old
machine and I want to make it run just for the hell of it, to
experience what it did". I gives me faith that mankind won't be
completely flattened by the dominant steamrollers.
Edward Reid
Steve J. Martin
to fitting in the hypothesized group. I've long thought that the IT
industry favors the young (done it myself!).
I'm still highly employable on the 2200 but, should I lose my
current job, only if I'm prepared to move hundreds of miles to the
next one. This makes it hard on 2200 employers, too, because
middle-aged people are less likely to pull up stakes and move far
away.
Were I 21 again, I wouldn't begin working on Unisys 2200 or A
series, not because they're bad environments (we all know better than
that!) but because installations are lightly spread across the map and
getting more so all the time. I'd be afraid to wind up an orphan at
40.
Regards,
Steve J. Martin
Halifax, Nova Scotia
Bob Nicol
snip
>I ask anyone responding to this thread to include an indication of their age
>and I will collate the results. To get the ball rolling, I am close to 60.
Jeff,
I'm 48, and have been working on MCP systems since 1976.
I attended the recent Unite conference in Phoenix , my first in several years,
and was dismayed at the lack of young people attending the technical streams.
Bob
Mike Kain
security.
First 3 years in UNIX, and get to work with Linux a bit here.
I totally agree with the poster who said that new graduates look to the
mainframe
environment as a dying breed, that Linux/UNIX is the place to be... I teach
at Drexel University as well, and my students constantly look at the "hot"
technology
to get a job, rather than look at overall skills and being able to learn
first. They
all look at the journals, see the average salarys in the $$$$ for some
certain
technology and think that that's the ticket. They give in to the hype of
the papers.
I think that it is due to the fact that they just want a job, and don't know
what the
skills are that will get them one.
It also can be attributed to what is in the universities. Drexel's CS
department has
Solaris machines and NT/Win2K workstations as well as some Macs... I also
agree
with the poster who said that we should have given away B6700s and now LX100
licenses to universities. Students don't know about the mainframe
environment today,
because their schools don't have them.
When I was an undergraduate at the University of Delaware, there were
replacing the
B7700 that they used for all of the records. Burroughs was bidding an
4xA15, but
lost to IBM due to the short sightedness of the department. Had Burroughs
kept
that there, more students there would have had exposure to that environment
as well.
But this is before PCs/Windows (1985).
Microsoft dominates the university landscape
because of the sweet deals that they give universities. As a faculty member
at Drexel,
they GIVE me Windows 2000 and Office XP for my home PC to use for school.
Nothing can beat that price. Unisys has to start to WANT to get into that
environment
with other than the ES7000 for students to know about the mainframe
environment
(MCP/OS2200). I assume that Solaris/Unix is there as well. Universities
don't have
big budgets for hardware/software, so they're looking for the best deals.
Linux being
FREE helps here as well.
Just my 2 cents.
Mike Kain
Unisys Tredyffrin
Tom
etc etc evnironment for 30years with brief dips into the MCP, Unix, and
Windows worlds for variety.
Tom Sherren
Ray Hampton
My stomach still turns everytime I see the IBM add with the kid in a big
room and the execs think someone has stolen their computers. Unisys should
have spotlighted that ad over a year ago with the ES7000; but once again
they dont know the meaning of the word "marketing".
Ray Hampton
"Jeff W Morris" <p...@icon.co.za> wrote in message
news:a2kbn6$3c7$1...@ctb-nnrp2.saix.net...
William J. Toner
on Unisys 2200 for the past 16 years.
Bill Toner
"Jeff W Morris" <p...@icon.co.za> wrote in
news:a2kbn6$3c7$1...@ctb-nnrp2.saix.net:
>>snip
>
> I believe we are a group of highly skilled professionals but we are an
> ageing population and there is not enough being done to attract
> youngsters to our exciting and complex world.
>
> If we took a poll of participants in this Newsgroup, I believe that a
> large
> percentage would be over 45 or even over 55. What are we doing to
> attract young entrants (20-25 year-olds), bearing in mind that none of
> us are immortal?
>>snip
pudge
layed off...from Unisys.
Pudge....59, and got the hell>>>>>>>>>>>>>OUT.
AN/FST2(1961) > NX5800 hardware.
(3000 hot vacuum tubes there)
(design 1956)
Jeff W Morris
"pudge" <1pu...@bellsouth.net> wrote in message
news:3C4F61B7...@bellsouth.net...
> And that's why my buddy with 23 years MCP + Linc + MS certification got
> layed off...from Unisys.
>
> Pudge....59, and got the hell>>>>>>>>>>>>>OUT.
> AN/FST2(1961) > NX5800 hardware.
> (3000 hot vacuum tubes there)
> (design 1956)
Out to pasture (retired) or pastures new? Just for interest sake, of course.
--
Dr Jeff W Morris: Pretoria Software Solutions
One of the largest LINC software houses in the Southern Hemisphere
Specialising in 'offshore' system development and maintenance Worldwide
Tom Duller
"Long time lurker, first time poster". Very impressed with the names I see
here.
Been there, did that, on most of the Burroughs side platforms for 25 years or
so. Others as well, but these PC's and LAN's just aren't stable or
secure enough yet IMO. I have to work with them anyway...
There's nothing like big iron in a locked room when it really counts.
==Tom==
Stephen Fuld
"Bryan Souster" <inv...@nogoog.invalid> wrote in message
news:6Cl38.2388$4i5.5...@news11-gui.server.ntli.net...
At the risk of being called a troll (I am not) let me make some comments on
this.
>
> While I agree that in general the IT industry related to mature
technologies
> (this excludes MS and Unix btw) is for the main part populated with highly
> skilled and experienced people that are typically in or near my agegroup,
Certainly agree. The facts seem unasailable.
> the situation seems mostly created by the following:
>
> - Our western world Universities are dominated by MS and Unix platforms.
> Students therefore only get this exposure.
True. But you have to look behind this to find out the reason. One is
cost. A university can buy a large number of PCs and run Windows or Linux
for a lot less money than running a comparible amount of computing power on
A series or 2200 Series. Plus, the bulk of employers want people with
Windows and Unix experience and the schools respond to that.
>
> - Mature technologies such as those based on the mainframe, centralised,
> typically transaction-processing technologies (ie MCP, OS2200, MVS etc)
are
> generally percieved by the young graduate as dated, dying technologies
that
> will inevitably eventually be replaced by more up-to-date Windows and/or
> Linux) servers.
It is not an inacurate perception. How many new name accounts has Unisys
gotten for either A or 2200 series in the last decade? I believe very few,
if any. We all know of some accounts that are leaving the A series or 2200
series fold. It may be a trickle, but if no new ones come in, the trend is
still there; we are only arguing about the rate of decline. Now we can
argue about the benefits of our favorite systems all we want, but that
doesn't change the "facts on the ground". Young graduates see comparitivly
few potential employers with these systems compared with the number who have
Unix, Windows, etc. Their choice is obvious.
> The problem is not so much that our part of the industry is dying but the
> perception by the youngsters entering our industry that MS/Unix (with
Linux
> at home) is the place to be.
Again, I respecfully disagree. As I pointed out above, there are virtually
no new name accounts. The number of people required to administer/support
the MIPS growth in existing accounts is pretty small. Form their
perspective, there are fewer potential jobs, and the number of jobs is not
growing (and may even be shrinking). This is close enough to "dying" that
the difference is mostly a semantic argument. Again, the rate is slow
enough that there will probably be "mainframe" jobs for about all the people
looking for them. It is a slow death. But that doesn't make it attractive
for a newcomer.
> This is borne partly from an ignorance of
> mature technologies and partly the natural recoil-reaction of working in a
> world full of old codgers like us.
I'm sure those are factors also. Together with the relative paucity of
employer choices (as mentioned by Steve Martin), and the steeper learning
curve because they have no experience in the area to start with.
> Unix has had its day.
This seems like whistling past the grave yard. Just look at the growth of
Linux, for example. Or look at the growth of Sun servers, or even HP-UX
systems. Their growth rates are what Unisys can only look at longingly.
Again, this is independent of the quality of the systems or their fitness
for the task, etc. Just look at the usage numbers.
> MS are still so laughably short of ever producing an
> operating system capable of centralised enterprise-wide services (which
must
> start with the entire elimination of virus/worm infections) that the next
> decade will not see them get much beyond the desktop.
Interesting prediction. Considering that they now have something like
40-50% of the server market. Are you saying that within the next decade all
of those NT server customers will switch to Unisys systems? Do you realize
how ridiculous that prediction is?
> Each release of all
> their software will contain the current trend of 'greater reliability' and
> 'less vulnerabiliy'. Bill Gates recent pronouncements on trustworthy
> computing (if I have understood what he said correctly) reveals just how
> juvenile MS thinking is.
You keep making the same mistake of conflating quality or suitability with
popularity. Read "The Innovator's Dilemma" by Clayton Christianson for a
counter view.
> The ES7000 is an excellent large-scale thin-desktop-client server when
sold
> as an ES7000, but but the true future of CMP is with Clearpath Plus.
That may be true - unfortunately for the Unisys stockholders.
> The problem was partly highlighted by year 2000 issues and will rear its
> head again over the next decade as we retire and companies do NOT abandon
> mature technologies. Action to fix this issue will only be taken when
> individual employers face the skills loss and realise that the mature
> technology they use is still the way forward for the forseeable future.
Then
> they will be forced to take action and recruiting youngsters will be the
> inevitable result.
There will be a race. Unix (and to a much lesser degree Windows) is getting
more "enterprise ready" with time. Some companies are abandoning the legacy
mainframe systems, there is certainly some training of new people and some
bringing in "old codgers" as consultants. But if you see a big resurgance
of traditional mainframe systems, please tell us all what substance you are
inhaling/injecting/eating, as it seems like real good stuff :-).
--
- Stephen Fuld
e-mail address disguised to prevent spam
Edward Reid
> At the risk of being called a troll (I am not) let me make some comments on
> this.
You are not a troll. You are trolling. There's a difference.
(For the record: ;-)
I read Bryan as more commenting on the situation than making
predictions. That's the nature of my comments too.
> It is not an inacurate perception. How many new name accounts has Unisys
> gotten for either A or 2200 series in the last decade? I believe very few,
> if any.
How many new name accounts has anyone obtained for traditional
applications? They don't exist. Every large account is already fully
computerized. New business comes only from new applications. In recent
years, that's mostly meant the web.
But that full computerization also means inertia. It's become steadily
more difficult for companies to jump from one platform to another.
Despite all the talk, there's no flood of applications being ported
from mainframes to Unix -- a few certainly, but not a flood.
Instead, companies are adding more, and more powerful, systems of the
same kind. You're right that this doesn't increase the personnel
requirements very much -- it may even decrease them, since it demands
more automation in the system administration. But it does provide real
growth in hardware sales. I don't have any recent reports, but the last
I heard -- a few years ago -- the NX systems were still jumping off the
shelves. All to existing customers as you point out, but nonetheless
real sales for real money.
> There will be a race. Unix (and to a much lesser degree Windows) is getting
> more "enterprise ready" with time.
Been hearing that for about a decade now ... maybe we should admit that
it's going to be very slow to happen and that there are some real
barriers. The very popularity of these systems has made it difficult
for them to stabilize to the extent needed for critical enterprise
work.
> Some companies are abandoning the legacy
> mainframe systems, there is certainly some training of new people and some
> bringing in "old codgers" as consultants.
Hey, the critical word there is "old". But as long as they pay us as
consultants, what me worry? We are in a field that's about as old as I
am. Ten years is an eternity. Who predicted, ten years ago, what kinds
of work we would be doing today -- particularly the web focus?
Similarly, I'm less worried about predicting what will be happening ten
years hence than in keeping the options open, both personally and for
the field at large. I'm less concerned about the survival of particular
systems than about the trend toward narrowing the field, with one or
two systems and frameworks pushing out the others. Computing just isn't
sufficiently mature to narrow our options. That's what bothers me about
the dominance of one or two platforms -- much less the weaknesses of
those platforms than the loss of choices.
Edward Reid
WDS
[snip]
>I ask anyone responding to this thread to include an indication of their age
>and I will collate the results. [...]
I am 44.
While I am not particularly aged, my Unisys platform (V-Series) is.
Stephen Fuld
"Edward Reid" <edw...@paleo.org> wrote in message
news:01HW.B8758AB50...@news-east.usenetserver.com...
> On Thu, 24 Jan 2002 1:41:09 -0500, Stephen Fuld wrote
> > At the risk of being called a troll (I am not) let me make some comments
on
> > this.
>
> You are not a troll. You are trolling. There's a difference.
>
> (For the record: ;-)
I don't disagree with most of what you said - except that I was trolling :-)
I was responding to the comments that young people entering our industry
perceive the kind of systems we work on as "dead ends" and not very
attractive for long term career consideration. I believe that those
perceptions are accurate - for them. That does not mean they are bad for us
(older folk) or that they are going away in the next 5 years, etc.
> I read Bryan as more commenting on the situation than making
> predictions. That's the nature of my comments too.
I agree. I was looking at it from the perspective of a new grad and showing
how what Bryan said was a perception was, from his/her perspective, the
truth.
> > It is not an inacurate perception. How many new name accounts has
Unisys
> > gotten for either A or 2200 series in the last decade? I believe very
few,
> > if any.
>
> How many new name accounts has anyone obtained for traditional
> applications? They don't exist. Every large account is already fully
> computerized. New business comes only from new applications. In recent
> years, that's mostly meant the web.
Yes - my point exactly. If you were a new grad, you wouldn't take, as a
first choice, a job where there weren't any/many new applications to develop
and where you weren't working with the newest technology and where you
couldn't take advantage of what you learned in school. That seems to me to
be a very rational decision. That is why there are so few young people in
our area of the computing world.
> But that full computerization also means inertia. It's become steadily
> more difficult for companies to jump from one platform to another.
> Despite all the talk, there's no flood of applications being ported
> from mainframes to Unix -- a few certainly, but not a flood.
Absolutely. But given that trickle out and no new ones comming in, the
eventual result is inevitable. I don't think we disagree (see below) about
the time frame.
> Instead, companies are adding more, and more powerful, systems of the
> same kind. You're right that this doesn't increase the personnel
> requirements very much -- it may even decrease them, since it demands
> more automation in the system administration. But it does provide real
> growth in hardware sales. I don't have any recent reports, but the last
> I heard -- a few years ago -- the NX systems were still jumping off the
> shelves. All to existing customers as you point out, but nonetheless
> real sales for real money.
Yes - but Brian's original post was about the paucity of young people in the
Unisys world because they perceived it as being "out of vogue", or the
equivalent. My point was that their perceptions were acurate. I don't
think we disagree on that.
>
> > There will be a race. Unix (and to a much lesser degree Windows) is
getting
> > more "enterprise ready" with time.
>
> Been hearing that for about a decade now ... maybe we should admit that
> it's going to be very slow to happen and that there are some real
> barriers.
Oh Absolutely! To quote Mark Twain "The rumors of my death have been
greatly exagerated." It will take a long time (perhaps even multiple
decades) but for a youngster, why start out on something that has a downward
trend at all?
> The very popularity of these systems has made it difficult
> for them to stabilize to the extent needed for critical enterprise
> work.
Interesting point. What makes you think that is so? I attribute it to
changing technology (low cost makes it OK to have one system per
application) and to a less demanding customer base (people accept say 10-15
second response time on a complex web page load.)
> > Some companies are abandoning the legacy
> > mainframe systems, there is certainly some training of new people and
some
> > bringing in "old codgers" as consultants.
>
> Hey, the critical word there is "old". But as long as they pay us as
> consultants, what me worry?
Yes - it apears there will be sufficient jobs for all of us :-) Just don't
expect to be surrounded with young coworkers.
> We are in a field that's about as old as I
> am. Ten years is an eternity. Who predicted, ten years ago, what kinds
> of work we would be doing today -- particularly the web focus?
Agreed.
>
> Similarly, I'm less worried about predicting what will be happening ten
> years hence than in keeping the options open, both personally and for
> the field at large. I'm less concerned about the survival of particular
> systems than about the trend toward narrowing the field, with one or
> two systems and frameworks pushing out the others. Computing just isn't
> sufficiently mature to narrow our options. That's what bothers me about
> the dominance of one or two platforms -- much less the weaknesses of
> those platforms than the loss of choices.
I agree. Things are moving so fast that people are not taking the time to
learn lessons that were hard won by their predecessors. I think we need some
way of codifying the things that were done in older/mainframe systems (both
good and bad - avoiding mistakes is as important as copying good ideas.)
that may still be relevant in such a way that young people can learn about
them easily and make use of them to inform their work on new systems. But
it is hard to get this done. No one wants to read old manuals and the
usefull information there (I mean usefull in its applicability to other
systems) isn't very dense. We need a better way, but I haven't come up with
one.
Bryan Souster
news:01HW.B8758AB50...@news-east.usenetserver.com...
> > At the risk of being called a troll (I am not) let me make some comments
on
> > this.
>
> You are not a troll. You are trolling. There's a difference.
>
> (For the record: ;-)
>
> I read Bryan as more commenting on the situation than making
> predictions. That's the nature of my comments too.
>
> > It is not an inacurate perception. How many new name accounts has
Unisys
> > gotten for either A or 2200 series in the last decade? I believe very
few,
> > if any.
>
> How many new name accounts has anyone obtained for traditional
> applications? They don't exist. Every large account is already fully
> computerized. New business comes only from new applications. In recent
> years, that's mostly meant the web.
>
I have read both Stephen's and Edward's posts - responding here.
Everything I said was based on my personal knowledge and experience - the
fact that in both NZ and the UK in recent times few if any using Clearpath
and predecessor platforms have moved these applications away. I did not
suggest that this technology was expanding - merely holding. The people
skilled to use and support it, in all aspects, are diminishing because of
age and will (in the near future) need to be replaced. The influx of the
young will be slow because of perceptions but it will need to happen so it
will probably happen.
One day, in my hopefull opinion, Unisys will realise that Clearpath has a
competitive edge over WINTEL and Unix platforms.
Bryan.
Colin Zealley
Colin
"Edward Reid" <edw...@paleo.org> wrote in message
news:01HW.B8739E540...@news-east.usenetserver.com...
Bert Hyman
Jeff W Morris <p...@icon.co.za> wrote:
>In a letter to the editor of a local computer weekly a writer recently
>stated: "Anyone who has been in the industry for more than 10 years should
>know that the IT industry favors the young and that the demand for skills is
>far less in management than in raw technology".
>
> Is this statement true for the traditional Unisys environment with which
>the participants of this Newsgroup are familiar (excluding ES7000/Microsoft
>toys for boys)?
>I ask anyone responding to this thread to include an indication of their age
>and I will collate the results. To get the ball rolling, I am close to 60.
I don't think the industry favors "the young", it favors people with current
skills. There's not much demand for people who can design, build or repair
automobile carburetors anymore either.
Your exclusion of "ES7000/Microsoft toys" makes it a little difficult to
comment further, since that's where the major emerging markets are at the
moment. Or, at least where they appear to be. Or, maybe where people think
they might be. [Was that hedged enough?]
I'm 54. I started programming on a Univac 1005 with SAAL (Single Address
Assembly Language) in 1968. Until about 1985, I programmed almost exclusively
in 1100/2200 assembly language (ASM or MASM), using FORTRAN when I needed
a high-level language.
I now do almost all my work in C or C++ for various flavors of Windows, but
occasionally get called on to write code for OS2200 applications, which
I also do in C. The Windows code I write is all written to ultimately
talk to or support OS-2200 applications.
In the interim, I did work on various flavors of UNIX (BSD and SVR4 based)
including X and Motif work. I've programmed in Algol, SNOBOL, APL, Modula-II,
Lisp, MAD, JOVIAL or anything else that looked like it might be fun.
The trick, I think, is to stay flexible and keep current.
It's all code.
--
--
Bert Hyman | St. Paul, MN | be...@visi.com
Erwin Richard
news:F4_38.7721$Wf1.2...@ruti.visi.com...
...
> In the interim, I did work on various flavors of UNIX (BSD and SVR4 based)
> including X and Motif work. I've programmed in Algol, SNOBOL, APL,
Modula-II,
> Lisp, MAD, JOVIAL or anything else that looked like it might be fun.
>
> The trick, I think, is to stay flexible and keep current.
>
> It's all code.
>
Over here it looks like there will be a shortage in experienced OS2200
administrators/programmers
to support the platform in the next 5 to 10 years. While new development is
certainly moving away
from OS2200 (mainly due to lack of efficient development tools and
experienced developers), there
are still some rather large sites with their core business applications on
2200 and unlikely to being able
to migrate within the next decade. People with OS2200 expertise and a fair
knowledge of Unix or NT and
supporting middleware such as TUXEDO and MQ Series (and able to communicate
with the Unix / NT guys)
will be in high demand.
Erwin
Larry Ramsey
45 years old, 23 years Unisys/Burroughs
Went to school on 5500 and 6700, came to work in '78 under the keen eye of
Dale Brenner on a big "medium" to "large" systems conversion. Later
introduced to B1000s and fell in love. Reacquainted myself with MCP on
"large" systems round the time of the 5900 and after the merger fell into
LINC as part of the "Power Team" training. Along the way learnied my share
of BTOS/CTOS, PC's , Unix, and Windows, even got myself certified in the
great Certification Drive of '99. Even learned to do some Mapper Run
writing, but don't tell anyone!
Anyway, it was a great ride, and I'm sad that they decided to do without me.
The irony of ironies: My last severance payment is on my hiring anniversary.
How sick is that?
ler
No matter where you go, there you are
"pudge" <1pu...@bellsouth.net> wrote in message
news:3C4F61B7...@bellsouth.net...
Jeff W Morris
> I don't think the industry favors "the young", it favors people with
current
> skills. There's not much demand for people who can design, build or repair
> automobile carburetors anymore either.
>
Can we look one level deeper? There must be engineers beavering away on new
releases of MCP, DMSII, OS2200, Mapper!, etc to make them integrate with
this newfandangled technology and operate on new chipsets. Extremely
talented and highly skilled people. Is there sufficient personnel renewal in
those teams to allow for another 10 years of life for these platforms?
--
Dr Jeff W Morris: Pretoria Software Solutions
One of the largest LINC software houses in the Southern Hemisphere
Specialising in 'offshore' system development and maintenance Worldwide
Edward Reid
>> The very popularity of these systems has made it difficult
>> for them to stabilize to the extent needed for critical enterprise
>> work.
>
> Interesting point. What makes you think that is so? I attribute it to
> changing technology (low cost makes it OK to have one system per
> application) and to a less demanding customer base (people accept say 10-15
> second response time on a complex web page load.)
What you say, and also less demanding in terms of reliability and
availability, but very demanding in terms of capabilities. Those may be
even more important in keeping the systems unstable.
> Yes - it apears there will be sufficient jobs for all of us :-) Just don't
> expect to be surrounded with young coworkers.
That's the hardest part to take ... no attractive young people of the
appropriate sex surrounding us middle-aged drones ... er, consultants.
Edward Reid
Jimi
Looks like I'm the odd one out in all these postings but here goes anyway!
I'm 23 and have been programming in 2200 Mapper (or BIS now!) for the last
4.5 years. I have CoolICE and NT experience too, and am currently part of a
project tasked with moving part of our 2200 system onto NT. I have to say
right off that being in a field with mostly older people is not particularly
easy. Many immediately take what us youngsters have to say in a bad light
purely because they don't believe we have the combination of knowledge and
experience. While this is true to a ceratin degree, it also means we haven't
as many filters clouding our view and hindering us from considering new
approaches. Also, some of the "aging population" don't take constructive
criticism too well, and so some aging ideas never get refreshed. Unisys
itself, though, seems very receptive to new ideas and comments.
After all of that whining though, I'll admit that at this stage of my life
I'm wondering whether this thread means I'm part of a dying breed, or a man
in demand? Should I stay or should I go now...? :)
James
"Ray Hampton" <rha...@bellsouth.net> wrote in message
news:i4D38.46615$Y17.3...@e3500-atl2.usenetserver.com...
Bill McDonald
oh, for the record
49 y/o started in 1970 on B3500's & 2500's
Tim Krantz
tell me it isn't so! "decided to do without me"?????
I worked with (or in the same vicinity with Larry for a number of years).
BTW, I am not far from these stats either. 44 years old, 22 years in the
Burroughs/Unisys environment. Also worked with B1000 and "large systems" and
currently work for the "keen eyed" Dale Brenner.
I left Unisys shortly after Unisys decided they could "do without my father"
in 1991. At the time I had about 12 1/2 half year with Burroughs/Unisys and
dad had over 28 years.
Sorry to hear it Larry.
Tim Krantz
In article <aL348.74489$n81.18...@typhoon.southeast.rr.com>, "Larry Ramsey"
Charles M Rader
years).
Ages in my group seem about evenly split between 30-to-40 and over-40,. We
actually have some 30-somethings learning ALGOL.
--
Charles Rader
A Series Technical Services, Eagan Service Center
Unisys Global Outsourcing
Eagan, Minnesota
*** Text contains my personal opinions ***
Edward Reid
> Can we look one level deeper? There must be engineers beavering away on new
> releases of MCP, DMSII, OS2200, Mapper!, etc to make them integrate with
> this newfandangled technology and operate on new chipsets. Extremely
> talented and highly skilled people. Is there sufficient personnel renewal in
> those teams to allow for another 10 years of life for these platforms?
I'm on the outside so I don't know for sure, but no insiders have
answered yet ... I think the answer is yes, there is plenty in the way
of human resources within Unisys. Probably there are many more who
would like such positions than Unisys can hire. The turnover is low in
these areas, certainly by industry standards and I think probably by
any US standards, and the work is good. As a result, the fact that they
come in knowing little about the systems is less important at this
level than it is on the outside. Heck, even 30 years ago Burroughs
hired beginners ... and back then you could tell by looking at the code
which programs had been written by the trainees.
Edward Reid
Hans Habers
I can do better.
49 years and 28 years in the MCp business and still having fun.
Hans Habers
"Charles M Rader" <Charle...@unisys.com> schreef in bericht
news:a2sm18$2pm$1...@si05.rsvl.unisys.com...
Juha Veijalainen
> This could have been me...
> 45 years old, 23 years Unisys/Burroughs
18 years with Unisys/Burroghs with me. From what I've read here it seems
that the latest "elimination of positions" at Unisys hit the 40+ years old
hardest (or maybe we are the most active group here?)
> Anyway, it was a great ride, and I'm sad that they decided to do without
> me. The irony of ironies: My last severance payment is on my hiring
> anniversary. How sick is that?
It's sick. But unfortunately it probably affects the stock holder value
positively. And only that matters to Unisys. And you have to keep up that
value every quarter. That is called "long term planning" :(
--
Juha Veijalainen, Helsinki, Finland
PGP key available, http://www.iki.fi/juhave
(Mielipiteet omiani - Opinions personal, facts suspect)
pudge
Pudge
pudge
George Ewins
around for a couple more years. I have been working on OS22OO systems since
1981 and prior to that some CDC (remember them?) mainframes.
I think the IT industry favors those that have the skills that are currently
needed. With the increasing capability of Windoze and Unix systems we have a
lot of people migrating away from the big mainframes and thus a need for
those that can work with the "toy" systems. In our organization the target
environment has some large IBM mainframes existing as basically database
servers with the actual applications running on "toys". Thus we will need a
few that can maintain the databases and an army to maintain the applications
on the "toys".
--
---------------------------
George Ewins
geo...@ewins.org
---------------------------
On 1/22/02 1:40 PM, in article a2kbn6$3c7$1...@ctb-nnrp2.saix.net, "Jeff W
Marc Wilson
<B87959E8.3B4%geo...@ewins.org>::
|I am 55 and just became eligible for retirement but I will probably stick
|around for a couple more years. I have been working on OS22OO systems since
|1981 and prior to that some CDC (remember them?) mainframes.
|
|I think the IT industry favors those that have the skills that are currently
|needed. With the increasing capability of Windoze and Unix systems we have a
|lot of people migrating away from the big mainframes and thus a need for
|those that can work with the "toy" systems. In our organization the target
|environment has some large IBM mainframes existing as basically database
|servers with the actual applications running on "toys". Thus we will need a
|few that can maintain the databases and an army to maintain the applications
|on the "toys".
44, worked on OS1100/2200 for 22 years. Initially Exec 8 and CMS in MASM,
but mostly MAPPER, some CoolICE in recent years.
A word of caution about the "toy" systems: while I'm no fan of the MS "2
billion beta testers and counting.." approach to systems, the general trend
seems to be that the OS is getting better- compare Win2K with NT, for
instance. When it's been going as long as Exec 8, it may be a damn good
system. And the other thing is- technical merit doesn't win the arguments,
just look at VHS/BetaMax.
--
Marc Wilson
Cleopatra Consultants Limited - IT Consultants - CoolICE Partner - MAPPER Associate
2, The Grange, Cricklade Street, Old Town, Swindon SN1 3HG
Tel: (44/0) 70-500-15051 Fax: (44/0) 870-164-0054
Mail: in...@cleopatra.co.uk
___________________________________________________________________
MAPPER User Group mailing list: send *SUBSCRIBE to M...@cleopatra.co.uk
Jeff W Morris
"Marc Wilson" <ma...@cleopatra.co.uk> wrote in message
news:vZVUPCTtK0kfKg...@4ax.com...
> In comp.sys.unisys, George Ewins <geo...@ewins.org> (George Ewins) wrote
in
> <B87959E8.3B4%geo...@ewins.org>::
,'''' the general trend
> seems to be that the OS is getting better- compare Win2K with NT, for
> instance. When it's been going as long as Exec 8, it may be a damn good
> system
end snip...
Exec 8, MCP, etc were damn good from their initial release. It didn't take
20 years and billions of beta testers to get a stable, reliable product in
the good old days. The big difference is the exponential growth in
complexity - but that is no excuse, looking at the profits paid out by MS
instead of being ploughed back into solid engineering.
--
Dr Jeff W Morris: Pretoria Software Solutions
One of the largest LINC software houses in the Southern Hemisphere
Specialising in 'offshore' system development and maintenance Worldwide
Stephen Fuld
"Jeff W Morris" <p...@icon.co.za> wrote in message
news:a32lba$asn$1...@ctb-nnrp2.saix.net...
>
> "Marc Wilson" <ma...@cleopatra.co.uk> wrote in message
> news:vZVUPCTtK0kfKg...@4ax.com...
> > In comp.sys.unisys, George Ewins <geo...@ewins.org> (George Ewins)
wrote
> in
> > <B87959E8.3B4%geo...@ewins.org>::
> > snip, snip
>
> ,'''' the general trend
> > seems to be that the OS is getting better- compare Win2K with NT, for
> > instance. When it's been going as long as Exec 8, it may be a damn good
> > system
>
> end snip...
>
> Exec 8, MCP, etc were damn good from their initial release. It didn't take
> 20 years and billions of beta testers to get a stable, reliable product in
> the good old days.
I can't speak to early MCP at all and my first Exec 8 release was 27, but
there were certainly stability problems in early Exec releases. The site I
was at considered it a good day when we didn't have a crash. We were a very
early user of TIP (after Air Canada) when it was delivered on a separate PCF
tape. Half the time it would even gen without errors, much less run.
Things have improved a lot. Yes, it didn't take "billions" of beta testers,
but it took quite a few, and damn close to 20 years. I remember several
"initiatives" within Roseville to improve stability (such as the base
release concept). It took a lot of hard work and a culture change to make
it happen. I hope that Gates' latest e-mail to the troops is the start of
the culture change within Microsoft, but I reserve judgement till I see some
effect.
> The big difference is the exponential growth in
> complexity
Yes. The Execs I worked with were about 350-400 thousand LOC. Win NT is
something like 10 million.
> - but that is no excuse, looking at the profits paid out by MS
> instead of being ploughed back into solid engineering.
>
>
Clearly they considered adding features more important that
stability/security, etc. We can hope that this is changing, but, as I said
above, I am not convinced.
Peter Kane
Juha Veijalainen wrote in message ...
>18 years with Unisys/Burroghs with me. From what I've read here it seems
>that the latest "elimination of positions" at Unisys hit the 40+ years old
>hardest (or maybe we are the most active group here?)
Speaking as one such, I'd go along with that !
Cheers,
Pete K
Louis Krupp
<snip>
> I can't speak to early MCP at all and my first Exec 8 release was 27, but
> there were certainly stability problems in early Exec releases...
The MCP had its share of problems in the early 80's IIRC;
the transition from ESPOL to NEWP (3.0 to 3.1?) was difficult
in places. The formatter, which had used ESPOL's low-level
features extensively to improve performance, had a lot of bugs.
> ... The Execs I worked with were about 350-400 thousand LOC. Win NT is
> something like 10 million.
As I recall, the last MCP release whose listing fit in a single
paper box was 2.6.
I'm 50. I used the B5500 off and on from about late 1965 to 1972,
and then the 6700/6800/5900 to 1982, when my employer at the time,
a university, switched to VAXes. I did a bit of A-Series stuff in
the late 80's.
I even worked for Burroughs once or twice.
I mostly use LINUX and C and Perl now at work. It hasn't made me
look any younger.
I read this newsgroup out of nostalgia.
I think today's students miss a lot by not being exposed to the
A-Series (or NX or whatever it's called now). It's a different
paradigm, and in a lot of ways, I think it's superior.
I had a gig porting C++ code at a place where people hadn't figured
out, for example, that you couldn't call malloc() ad infinitum, that
resources were really finite.
One program declared huge local arrays on the assumption that the
stack went on forever. Now, on the dinosaurs we grew up with, the
system would notice a stack overflow, and either expand the stack
or terminate the program. Not this modern UNIX workstation.
The stack grew toward higher memory addresses (just like our
dinosaurs), and just above the allocated stack area was a
*writable* page with pointers to system calls like read() and
write(). The program did its thing, the stack grew and overflowed,
and read() calls suddenly failed in ways that made no sense.
The B5500 detected stack overflows back in 1965. That was
probably before the people who designed that workstation were
born.
My reply address is spam-trapped.
Louis Krupp
Juha Veijalainen
Yeah, Unisys getting rid of the 40+ ,expensive - consultants and
specialists.
Please comment about the latest "elimination of positions" by e-mail to me
- I'm just gathering statistics, but your comments will help me...
John Bailey
This is NOT a new strategy on the part of Unisys. I was "let go" in
1995, and the bias against anyone over 40 years of age was well-known
back then. I was 44 at the time so at least I got a proper redundancy
payment - there were many aged 50 or over who got enforced early
retirement with a much smaller lump sum payment.
The other strategy in the lay-offs during the 1990s could best be
described as "We don't have products any more, only services.
Therefore we don't need product specialists."
John Bailey
A Series MCP environment since 1977 and still a specialist !
Dan Young
Age: 46
Years with Burroughs/Unisys: 19
Worked for Queen's University at Kingston for about 3 years prior to joining
Burroughs.
Currently working for Brock University; we have an LXE6500 production
machine and several LX100 development systems.
--
Dan Young
IT Architect
Information Technology Services
Brock University
(905)688-5550 x 4794
dan....@brocku.ca
> <snip>
Phil Wilson
slightly better picture was irrelevant compared to the fact that you
couldn't record anything on the first machines, the prices for the tapes and
the machines were too high, and probably most importantly, they didn't have
enough capacity to hold a movie. If technical merit was all consumers cared
about, we'd all be driving BMWs (or technically excellent brand of your
choice).
"Marc Wilson" <ma...@cleopatra.co.uk> wrote in message
news:vZVUPCTtK0kfKg...@4ax.com...
Randall Bart
as "Phil Wilson" <phil....@unisys.com> uttered:
>The Betamax vs VHS issue has been studied to death. The fact that it had a
>slightly better picture was irrelevant compared to the fact that you
>couldn't record anything on the first machines, the prices for the tapes and
>the machines were too high, and probably most importantly, they didn't have
>enough capacity to hold a movie.
Capacity wasn't the issue; the early VHS and Beta were both only one hour,
then they both extended it. And by the time that was critical, recorders
were available. The primary issue was price.
At one time Beta had the lead in market share. When the masses bought
machines, they often had to save up the money for a machine, or they waited
for the price to come down to a certain price point. Buying any machine now
was better than waiting to buy something better later. Sony was keeping the
monopoly on Beta, which meant Beta machines cost about 50% more. As more
and more people bought machines, their first purchase was VHS, and once you
bought one machine, you built up a library of tapes and you were hooked.
--
RB |\ © Randall Bart
aa |/ ad...@RandallBart.spam.com Bart...@att.spam.net
nr |\ Please reply without spam 1-917-715-0831
dt ||\ Here I am: http://RandallBart.com/ I LOVE YOU
a |/ Giggle: http://ggroup.4r.st/?selm=35DCEAD5.515F%40att.spam.net
l |\ DOT-HS-808-065 MSMSMSMSMSMSMS=6/28/107 Joel 3:9-10
l |/ Stars and stripes: http://RandallBart.com/USFlag
Thomas M. Schaefer
and an A10H). I now work with NT and Linux systems integrating our telephone
systems in our ACD platform. The skills I learned from system internals on
the MCP, COMS, and DMSII still guide my work today. I design high-volume
messaging systems and many lessons I learned about how big something can
get and how important uptime are have caused many a discussion with my
co-workers. I was "raised" in an environment where uptime was assumed - in a
large part due to the realized uptime of our systems. When I worked at Home
Shopping and SWIFT, uptime was naturally assumed because we were running on
an A Series. Granted people could write bad code on that system, but there
was less of a tendency because you could not hide behind the system. If it
went down, odds were it WAS your fault, not some Blue Screen or a panic that
you just chalked up to fate.
The move from monolithic systems to client/server, then back to large
centralized systems (web services) has taken us full circle. if anyone
thinks they are not running a large central system with a SOAP based web
service, they are kidding themselves. granted there can be a web farm
accessing a database, but that is just a logistical issue. 100 separate
servers accessing an SQL server through a messaging system like MQ series.
or MS transaction server is just like 100 programs running in COMS servicing
an input queue. I did not see the difference then, and I still do not. But I
have digressed way off course.
Regards,
Tom Schaefer
Salt Lake City
Sid Hale
I got a little chuckle from the fact that the first respondent in this
thread was "Jack Benny".
Anyway, my youngest daughter informed me last week that I am 52 (I had
forgotten). Began working in Burroughs environment (dual B3500 w/ shared
disk) in 1971. Worked for Burroughs from 1973-78 (B2/3 and B17/18).
Transitioned to B6700 in 1980. Wrote first LINC system on B1000 in 1983.
LINC on everything since.
My mouth is running on... you didn't ask for my life story.
Like I said - 52
Sid Hale
Chesapeake Software Services, Ltd.
Publisher, LINCweb Technology Newsletter
"Jeff W Morris" <p...@icon.co.za> wrote in message
news:a2kbn6$3c7$1...@ctb-nnrp2.saix.net...
> --
> Dr Jeff W Morris: Pretoria Software Solutions
> One of the largest LINC software houses in the Southern Hemisphere
> Specialising in 'offshore' system development and maintenance worldwide
>
>
>
>
Donald Gregory
because the VHS tapes gave me 2 hours of recording time, while the
Beta tapes only gave 1 hour. 1 hour was just insufficient for any
practical use. No other thought was put into the decision.
I agree with Randall. This analogy has been beaten to death.
Moreover, it is totally irrelevant. You can run data processing on a
mixture of MCP, Unix, and Windoze platforms. You could not record
on a Beta tape and play it back on a VHS machine or vice versa.
You can read and write Windows and Unix files on an MCP machine.
The chief reasons why the MCP machines don't sell more are, IMHO:
1. Unisys management does not understand the value or applications
of the machine.
2. There is not now nor has ever been any attempt to market these
machines. They are the "best kept secret of the industry."
3. Burroughs strongly discouraged the development of a third party
add on market in the late 70's/early 80's period, when that was
crucial. (When they did relent, it was too late.)
4. Burroughs failed to indoctrinate the Universities and Colleges
into the MCP architecture, thereby letting the Unix abomination
gain a foothold.
5. After the merger, several ex-Sperry sales personnel discouraged
customers from buying A-Series machines. (Remember University of
Oklahoma or the Florida U.S. Fisheries?)
6. Management has made several very bad choices in product
offerings which helped discourage customers, such as:
a) Office Management System II (OMS II), which couldn't run on
the hardware available at the time;
b) InfoAccess over Transit ODBC;
c) Investing so much into LINC when a much better, more modern
solution is needed.
If management today would learn from some of these past mistakes,
I believe they could regenerate interest in the A-Series product
line. The A-Series could address a great many of the Internet
security issues. (For example, you can't get a buffer overflow
on an A-Series and gain control of the system.) The new security
demands of governments and corporations could be easily addressed
using the A-Series. But so far I haven't seen any effort on
Unisys' part to even attempt to play upon these opportunities.
Perhaps it is because they still don't realize what they have.
(Note: By "A-Series" I mean any flavor of MCP-based machine,
including the ClearPath NX and LX lines.)
Don Gregory
--
--------------------------------------
Donald J. Gregory
Gregory Publishing Company
See our A-Series Documentation Library
at: www.gregpub.com
E-Mail: don...@gregpub.com
---------------------------------------
Jeff W Morris
<31 years old 2
<36 1
<41 1
<46 9
<51 5
<56 9
<61 2
There were no responses from the gurus like Alex, Bob, Don (this morning, no
age given), Dr LINC, Jim and many, many more.
Are they all out to pasture now or are they writing DLLs for Bill Gates?
This was not a scientific sample of the population and no deductions can be
made
other than that the active participants in this NG are mostly beteen 41 and
55.
Many thanks for the lively and interesting debate!
Lets call it a day.
--
Dr Jeff W Morris: Pretoria Software Solutions
One of the largest LINC software houses in the Southern Hemisphere
Specialising in 'offshore' system development and maintenance Worldwide
Gordon DeGrandis
I am 49 going on to 50 in September.
After reading this news group I got to thinking. This year will be my 30th year in the IT industry and with Burroughs/Unisys equipment! I started on a B6700 in September 1972!!
I have done all in my career support, DBA for DMSII, Middleware, project manager, analysis, I even ddid some firmware work on the old DCDLPs. I worked in Canada for Burroughs/Unisys and others and now for the last 10 years I have been in Belgium at a LARGE Unisys A Series site where I did a whole bunch of different things.
I am now a technical Architect for systems. I seem to be moving away from A Series but I still keep in touch in fact I am working with a Unisys customer to architect their middleware.
Most of the people I work with were not born in 1972, but I know as much as they do including the MS and Unix/Linux kids. I intend to keep on going and if I find work with A Series I will take it.
I agree with the comments in the NG about the B6700's and following machines to universities. They should have given them away! Considering how the MCP has evolved over the years other OSes are still trying to catch-up. I remember back in 1975 working on a 2x2 B6700 some OSes still have problems supporting more than one CPU.
Well that's all for now
Gordon DeGrandis ;-)
-----------------------------------------------
Gordon DeGrandis - Contraste Europe SA
HTTP://www.contraste.com
e-mail: Gordon.D...@Contraste.com
Tel:+32 (2) 730.79.80
Have a nice day
-- ----------------------------------------------- Gordon DeGrandis - Contraste Europe SA HTTP://www.contraste.com e-mail: Gordon.D...@Contraste.com Tel:+32 (2) 730.79.80 Have a nice day
Donald Gregory
Sorry about that. I turn 50 on March 28.
Regards,
Don
--
Edward Reid
> I can't speak to early MCP at all and my first Exec 8 release was 27, but
> there were certainly stability problems in early Exec releases. The site I
> was at considered it a good day when we didn't have a crash.
The MCP was pretty much the same. I remember back, oh, probably around
1976, our B6700 went for 17 days with only one halt/load (reboot). That
was pretty astounding in those days. Then the field engineers had to
make some hardware adjustments, and we never did it again.
Of course, the software stability about matched the hardware stability,
so there would not have been a lot of value in making the software any
more stable at that time.
> I remember several
> "initiatives" within Roseville to improve stability (such as the base
> release concept). It took a lot of hard work and a culture change to make
> it happen.
Again, much the same for the MCP. One of the keys was when marketing
insisted that they had to support customers who didn't read source. The
classic B6700/B7700 site had at least a couple of people who could read
the MCP source to track down problems, answer questions, and so on. The
availability of the source was and is a very strong part of MCP
systems, but to reach a broader market it was necessary to remove it as
a *requirement*. That was another culture change.
Edward Reid
John Keiser
Edward Reid <edw...@paleo.org> wrote in message
> The MCP was pretty much the same. I remember back, oh, probably around
> 1976, our B6700 went for 17 days with only one halt/load (reboot). That
> was pretty astounding in those days. Then the field engineers had to
> make some hardware adjustments, and we never did it again.
And I remember the time when an MCP based machine first went almost 30 days
without a halt/load. We knew it was the very first time because it uncovered
a bug involving clock values and field overflow that caused a control state
loop.
I'm pushing 60 and have been deep in the bowels of the MCP since 1970.
Currently on long term disability leave from Unisys.
John Keiser
Timo Saari
"Donald Gregory" <don...@gregpub.com> wrote in message
news:3C5F3E9A...@gregpub.com...
> Donald J. Gregory
> Gregory Publishing Company
> See our A-Series Documentation Library
> at: www.gregpub.com
> E-Mail: don...@gregpub.com
> ---------------------------------------
57 and started working with a B6500(!) in 1970 and been doing it with
A/NX-series ever since.
Don's point about indoctrinating Universities and Colleges into MCP
architecture was at least to some degree validated in Finland. University of
Helsinki, for which I worked from 1970 to 1985 did use MCP systems and
people from that environment did help either as decision makers or as system
support personnell to achieve the modest success those systems had here .
Timo Saari
Unisys / Finland
Robert E. Sargent
1967, B6500 which became a B6700, Some medium system B3500 etc, B7900
and Several A Series.
I am 68.
Bob Sargent
Bryan Souster
> Jeff W Morris wrote:
> >
> > Here is a summary of respondent's ages as of today:
> >
> > <31 years old 2
> > <36 1
> > <41 1
> > <46 9
> > <51 5
> > <56 9
> > <61 2
> >
> > There were no responses from the gurus like Alex, Bob, Don (this
morning, no
> > age given), Dr LINC, Jim and many, many more.
> > Are they all out to pasture now or are they writing DLLs for Bill Gates?
Nothing from Dr LINC in ages.
Bryan.
Stephen Fuld
"John Keiser" <Spamfree...@worldnet.att.net> wrote in message
news:oq288.2002$CM1.2...@bgtnsc04-news.ops.worldnet.att.net...
OK, so we have pretty much established that in the early days of both Exec 8
and MCP there were stability problems, etc. Through a perception that
curtomers were demanding more "industrial strength", both companies put
forth lots of effort and, over time, stability, reliability, etc. increased
to the very high levels they are now.
So, the $64K question. If Microsoft perceives that customers are demanding
"industrial strength", is there any reason why, with time and effort, they
couldn't do the same thing to Windows? I note that there is some evidence
that they are going to pay more attention to at least security (Gates'
memo). There is no magic about the technology. If they perceive it is in
their interests, they will expend the effort and, over time, it will
improve, perhaps even rivaling mainframe OSs.
This perception of "toys", followed by a slow erosion of market share
starting from the bottom is the scenario portayed in the Innovator's Dilema,
and I predict it will happen with Unix (already has to some degree) and
later with Windows (has but to a lesser degree).
--
- Stephen Fuld
e-mail address disguised to prevent spam
>
> John Keiser
>
>
>
Martyn R Jones
perversion of 1986, pursued by a crazed, unscrupulous and clueless
Blumenthal, egged on by a bunch of shameless bankers and fellow
alocholics, and blessed by a prematurely senile Reagan - then things
just went from bad to worse.
But, it's probably not a good idea to dwell on the past, and if I
started to think of the detail behind the takeover my comments would
probably get nasty.
Cheers,
Martyn Jones
Dave
times, starting in 1972 with medium & large system supporting COFIRS
(which ran at a couple of colleges or universities, I think), then
migrated it to B1700s. Worked with MCP systems on & off through the
years. Current customer is migrating off LX-6150 (with home-grown RPG
apps, if you can believe that) to Oracle financials. The number one
complaint from the user groups is "Man, this new system is so slow". I
guess size does matter.
David W. Schroth
>
> "John Keiser" <Spamfree...@worldnet.att.net> wrote in message
> news:oq288.2002$CM1.2...@bgtnsc04-news.ops.worldnet.att.net...
> >
> > Edward Reid <edw...@paleo.org> wrote in message
> >
> > > The MCP was pretty much the same. I remember back, oh, probably around
> > > 1976, our B6700 went for 17 days with only one halt/load (reboot). That
> > > was pretty astounding in those days. Then the field engineers had to
> > > make some hardware adjustments, and we never did it again.
> >
> > And I remember the time when an MCP based machine first went almost 30
> days
> > without a halt/load. We knew it was the very first time because it
> uncovered
> > a bug involving clock values and field overflow that caused a control
> state
> > loop.
> >
> > I'm pushing 60 and have been deep in the bowels of the MCP since 1970.
> > Currently on long term disability leave from Unisys.
>
> OK, so we have pretty much established that in the early days of both Exec 8
> and MCP there were stability problems, etc. Through a perception that
> curtomers were demanding more "industrial strength", both companies put
> forth lots of effort and, over time, stability, reliability, etc. increased
> to the very high levels they are now.
>
At some level, I have to regard this as an apples to oranges comparison
(between OS1100/MCP and Windows). I think one can reasonably state that
in the early days of Exec 8 and MCP that both OSs could be regarded as
pushing the state of the art - there were no previously extant general
purpose Operating Systems that the developers could learn from. I think
one could also reasonably state that both Exec 8 and MCP were working
with much less reliable hardware, and that a significant fraction of the
stability problems were a result of running on not-very-reliable
hardware. Windows, on the other hand, certainly had decades of prior OS
research and experience available to the developers - most of which
appears to have been ignored. And whether or not one cares for the
IA-32 architecture, there is absolutely no doubt in my mind that the
hardware is much more reliable than, say, the 1110 hardware (to choose a
particularly bad example).
> So, the $64K question. If Microsoft perceives that customers are demanding
> "industrial strength", is there any reason why, with time and effort, they
> couldn't do the same thing to Windows? I note that there is some evidence
> that they are going to pay more attention to at least security (Gates'
> memo). There is no magic about the technology. If they perceive it is in
> their interests, they will expend the effort and, over time, it will
> improve, perhaps even rivaling mainframe OSs.
>
I see no reason why they couldn't do the same thing to Windows. I also
see no reason why they would - they have a monopoly, and there certainly
doesn't seem to be any "industrial strength" competitors around to
encourage them to do so. As far as the evidence that they are planning
to pay more attention to security - well, they made a big deal about how
they were going to automatically detect and resolve buffer overflow
problems. Something you stopped hearing much about when the latest
Windows XP buffer overflow problems surfaced.
I think Microsoft could indeed make Windows "Industrial Strength" if
they are willing to invest enough time and energy to change the
Microsoft culture so that quality is, indeed, a priority (rather than a
slogan). As you noted above, changing the culture at Univac/Burroughs
took time, and continuous effort. It *still* takes time and effort.
> This perception of "toys", followed by a slow erosion of market share
> starting from the bottom is the scenario portayed in the Innovator's Dilema,
> and I predict it will happen with Unix (already has to some degree) and
> later with Windows (has but to a lesser degree).
>
I'm not totally sure what you're trying to say with this paragraph. I
get the Innovator's Dilemma part, but I'm not clear on what you're
predicting for Unix and Windows.
And to touch, however briefly, on the original topic for this thread,
I'm 49, and have been using 1100 systems since the mid-70s, and have
been working on them since 1977.
Regards,
David W. Schroth
Edward Reid
> So, the $64K question. If Microsoft perceives that customers are demanding
> "industrial strength", is there any reason why, with time and effort, they
> couldn't do the same thing to Windows?
No reason they can't, but I see three impediments.
First, they don't control the architecture. While not an insurmountable
problem, this does make the task more difficult.
Second, they are still pushing in new features at a much faster rate
than either Exec or MCP did. As a practical matter, this makes it much
more difficult to attain stability, which you need for security.
Third, the current payment and support model is detrimental to
stability. Burroughs (and I assume Sperry, as it was the dominant model
of the day) charged a recurring license fee, which included support and
new releases for a specific period of time. There's a trend toward
subscription support in the Microsoft market, but AFAIK the dominant
model is still pay-for-upgrade. This means that the financial incentive
is for Microsoft to add features which are sufficiently compelling to
induce users to license new releases. As long as this is the model,
features will have a higher priority at Microsoft than stability. No
matter how much users want stability, it's mighty hard to demo, and so
doesn't sell well. That's those wonderful market forces for you.
Edward Reid
Stephen Fuld
"David W. Schroth" <David....@unisys.com> wrote in message
news:3C62C1E3...@unisys.com...
> Stephen Fuld wrote:
snip
> >
> > OK, so we have pretty much established that in the early days of both
Exec 8
> > and MCP there were stability problems, etc. Through a perception that
> > curtomers were demanding more "industrial strength", both companies put
> > forth lots of effort and, over time, stability, reliability, etc.
increased
> > to the very high levels they are now.
> >
>
> At some level, I have to regard this as an apples to oranges comparison
> (between OS1100/MCP and Windows). I think one can reasonably state that
> in the early days of Exec 8 and MCP that both OSs could be regarded as
> pushing the state of the art - there were no previously extant general
> purpose Operating Systems that the developers could learn from. I think
> one could also reasonably state that both Exec 8 and MCP were working
> with much less reliable hardware, and that a significant fraction of the
> stability problems were a result of running on not-very-reliable
> hardware. Windows, on the other hand, certainly had decades of prior OS
> research and experience available to the developers - most of which
> appears to have been ignored.
That is a good point. However, let's distinguish betweenthe Win 9X code base
and the Win NT code base. Clearly the 9X code base had fundamental
architecture problems (principly no use of the hardware protection afforded
in the chips), but that code base is how dead ended and its useage will
decline. So let;s concentrate on the NT code base. The supposed architect
of the system came out of VMS, which was clearly a "modern" OS at the time.
They did some things in strange ways, but I don't see evidence that all the
prior research and experience has been ignored. Can you cite some
specifics?
> And whether or not one cares for the
> IA-32 architecture, there is absolutely no doubt in my mind that the
> hardware is much more reliable than, say, the 1110 hardware (to choose a
> particularly bad example).
Ouch! (I had to support an 1110 for several years - plated wire memory and
all. :-( )
But sure, current IA 32 systems are much more hardware reliable than then
current mainframes, but so what? That just shows that inatention to certain
design goals in developing complex software can make even stable hardware
look bad. Now that both types of hardware can be more stable (I would
venture to say that current mainframe hardware is more stable than current
IA-32 hardware, but hardware problems are lost in the noise for IA-32), it
just allows more progress for Windows systems.
>
> > So, the $64K question. If Microsoft perceives that customers are
demanding
> > "industrial strength", is there any reason why, with time and effort,
they
> > couldn't do the same thing to Windows? I note that there is some
evidence
> > that they are going to pay more attention to at least security (Gates'
> > memo). There is no magic about the technology. If they perceive it is
in
> > their interests, they will expend the effort and, over time, it will
> > improve, perhaps even rivaling mainframe OSs.
> >
>
> I see no reason why they couldn't do the same thing to Windows. I also
> see no reason why they would - they have a monopoly, and there certainly
> doesn't seem to be any "industrial strength" competitors around to
> encourage them to do so.
Let's differreniate the desktop from servers. In server land, they
certainly have a substantial competitor in Linux. From what I understand,
WNT etc. has about 40% market share, Linux about 20% with lot of
othersystems making up the rest. And Linux usage in servers is growing
pretty fast. That might encourage them. In desktops, I agree with you.
The only real hope (aside from some miraculous growth in Linus or something
else entirely) is that the changes that Microsoft will be forced to put into
the server code will be applicable to the desktop code, now that they are
the same code base, and that desktop stability will improve from that. I
don't know if it will.
> As far as the evidence that they are planning
> to pay more attention to security - well, they made a big deal about how
> they were going to automatically detect and resolve buffer overflow
> problems. Something you stopped hearing much about when the latest
> Windows XP buffer overflow problems surfaced.
I agree their committment, beyond talk, is suspect.
> I think Microsoft could indeed make Windows "Industrial Strength" if
> they are willing to invest enough time and energy to change the
> Microsoft culture so that quality is, indeed, a priority (rather than a
> slogan). As you noted above, changing the culture at Univac/Burroughs
> took time, and continuous effort. It *still* takes time and effort.
On that we absolutely agree!
> > This perception of "toys", followed by a slow erosion of market share
> > starting from the bottom is the scenario portayed in the Innovator's
Dilema,
> > and I predict it will happen with Unix (already has to some degree) and
> > later with Windows (has but to a lesser degree).
> >
>
> I'm not totally sure what you're trying to say with this paragraph. I
> get the Innovator's Dilemma part, but I'm not clear on what you're
> predicting for Unix and Windows.
I am predicting a slow but steady erosion "from the bottom" of traditional
mainframe systems as Unix and Windows get more robust. And that the people
who disparage them as "toys" etc., will have to eat their words over time.
That is why I am not optimistic over the future of mainframe systems.
Again, let me state that I am not predicting the demise of A series or 2200
Series next year, or even within a few years. But I think the trend is
inevitable.
>
> And to touch, however briefly, on the original topic for this thread,
> I'm 49, and have been using 1100 systems since the mid-70s, and have
> been working on them since 1977.
I never responded to the original intent, so I suppose I should. I am 52
and am basically retired (though I do some consulting to prevent brain
atrophy). I started working on 1100 systems in 1972. I did some work with
A series from the late 80s to the early 90s. I basically stopped working
with both of them in 1993. (Yes - I am here primarily for nostalga.)
I have said that the perception of Unisys people being "older folk" is
accurate. And I suspect that the layoffs, etc. primarily affecting older
people is that (due to things like hiring freezes, etc in the preceeding
years) there were relativly few young people there to affect.
Stephen Fuld
> On Wed, 6 Feb 2002 19:06:20 -0500, Stephen Fuld wrote
>
> > So, the $64K question. If Microsoft perceives that customers are
demanding
> > "industrial strength", is there any reason why, with time and effort,
they
> > couldn't do the same thing to Windows?
>
> No reason they can't, but I see three impediments.
>
> First, they don't control the architecture. While not an insurmountable
> problem, this does make the task more difficult.
True. But what is worse is that they don't control the implementation.
Windows is expected to work with just about any combination of parts from
whatever vendors some guy in a garage chooses to put together.
> Second, they are still pushing in new features at a much faster rate
> than either Exec or MCP did. As a practical matter, this makes it much
> more difficult to attain stability, which you need for security.
Again, agreed. That is one of the things taht will have to ahange if they
are serious about stability.
> Third, the current payment and support model is detrimental to
> stability. Burroughs (and I assume Sperry, as it was the dominant model
> of the day) charged a recurring license fee, which included support and
> new releases for a specific period of time.
Later. When I started, the software was free when you bought the hardware.
> There's a trend toward
> subscription support in the Microsoft market, but AFAIK the dominant
> model is still pay-for-upgrade. This means that the financial incentive
> is for Microsoft to add features which are sufficiently compelling to
> induce users to license new releases. As long as this is the model,
> features will have a higher priority at Microsoft than stability. No
> matter how much users want stability, it's mighty hard to demo, and so
> doesn't sell well. That's those wonderful market forces for you.
Once again, you make good points. I suspect that the model will have to
shift more towards some kind of pay for support in order for the "industrial
strength" program to be successful. Fortunately, the big companies that
have non Microsoft systems on their servers are used to that kind of model,
so perhaps Microsoft will have less trouble implementing it. We shall see.
Edward Reid
>> Third, the current payment and support model is detrimental to
>> stability. Burroughs (and I assume Sperry, as it was the dominant model
>> of the day) charged a recurring license fee, which included support and
>> new releases for a specific period of time.
>
> Later. When I started, the software was free when you bought the hardware.
Oh, right ... how short is memory. Burroughs too of course.
Anyway, the economic argument is the same. You paid a fixed price for
hardware, software and support for the life of the system, which was
limited by technological advances just as today. So the vendor had to
provide features to woo new customers, but for existing customers could
concentrate more on stability.
Edward Reid
Leopold Kuipers
> Are they all out to pasture now or are they writing DLLs for Bill Gates?
Hi Jeff,
Thanks for stirring up this topic.
Just for your statistics.
54
Worked with Univac 1005, 1108 :-), 9300, 90/30,
Worked with Burroughs 3500,3700,1985,
Worked with Unisys A4,A6,A14,
Working with MS, writing dll's as you say, using the Linc :-)
programming language and Lion :-)) to make it all happen.
In my opinion MCP has been a wonder (in its time).
The new toys (as some people call it) are the new wonder!
Todays pc's deliver more performance than almost any of the machines
above.
Combining the legacy programs with this new technology creates a
miracle.
Fast, reliable, available, flexible, with the end user in control.
I believe in IT.
Leopold Kuipers
Colin Zealley
Started on Univac 1108, nowadays primarily a LINCster.
Colin
"Jeff W Morris" <p...@icon.co.za> wrote in message
news:a3noq0$p5h$1...@ctb-nnrp2.saix.net...