Updating MV Family Tree
dawn
MultiValue Family Tree poster in the coming year. The 2002 version of
the pdf is found at
http://www.tincat-group.com/mv/familytree.html
>From what people have told me, the poster is very accurate, but I
believe there are some inaccuracies, I just don't know what they are.
If you have a chance to look it over and spot anything, however minor,
that appears inaccurate or misleading, I would appreciate the
information you have.
Also, there have been some changes in products or companies, including
at least jBASE and OpenQM, since this poster was developed. Any
information about the products and companies since early 2002 that
would help update this poster for today would be much appreciated.
Thanks in advance for any information you can give. You may either
e-mail me at dwolt at tincat-group dot com or respond to this posting,
especially if you want input from others on the accuracy of your input.
Thanks in advance. --dawn
Tony Gravagno
notes, my memory is real vague on all this and I never was a Pick
history buff. I did have more notes but scratched them the closer I
looked. :)
I see you have the hardware associated with a few platforms.
In addition to Honeywell for Ultimate there was the DEC (VAX)
Ultimate, which I believe was a firmware implementation - and
significant in its own right for that.
The three companies currently involved with the jBASE DBMS are jBASE
International, Temenos, and mPower1. This is as of 2002 when Mpower1
acquired rights for support and distribution. Ref:
http://www.jbase.com/about/ and elsewhere. "jBASE" is listed as
"Maintains an MV System in application". Technically that's not
accurate as jBASE does not have an application but Temenos does. I
can't suggest exactly how this should all be displayed but as it is
it's not 100% accurate.
ADP is mentioned in the Reality branch. From recent CDP discussions
and e-mail exchanges, I see they are still a player in the auto
dealership vertical market. I don't know if they just lock-down stock
DBMS releases or if they are still categorized as a licensee. If they
are a licensee, this could make them one of the longest running
licensee/resellers out there.
Noting the MultiValue logo, it's a shame that SMA isn't recognized
somewhere in the mix as having an impact on some of the cross-platform
standards that we still enjoy today.
Might be Pick-ayune, but I doubt whether the distinction between
Advanced Pick and AP Pro is worthy of mention. AP Protected mode was
just one step off from AP Native, which isn't mentioned there, and I
don't think Native is worthy either considering Pick was originally a
native platform in the first place. If AP PRO is in there then so
should be AP DOS (EOL 5.2.5) and AP Native (EOL 5.2.7). If not, and
this is my preference, I'd suggest scratching AP PRO and just leaving
Adv Pick.
Also under "ports", was Data General really a licensee or were MV
products merely ported over DG/UX?
Finally, we often hear Jim Idle say "jBASE is not Pick", and I know
that confuses people. We see in the tree that MDIS had impact on JAC
where jBASE was written, and the license was subsequently transferred
to jBASE International. But to understand that history I found a neat
article which explains that process _and_ really explains well how and
why jBASE was developed. This provides some compelling insight into
jBASE, and may help to explain why jBASE may be better positioned than
some other MV platforms for the open market. Ref:
http://www.ukpua.org/archives/articles/JAC.html
Regards,
T
Luke Webber
> Thanks for the effort Dawn. You'll need to get confirmation of these
> notes, my memory is real vague on all this and I never was a Pick
> history buff. I did have more notes but scratched them the closer I
> looked. :)
>
> I see you have the hardware associated with a few platforms.
> In addition to Honeywell for Ultimate there was the DEC (VAX)
> Ultimate, which I believe was a firmware implementation - and
> significant in its own right for that.
The Honeywell Level 6 and DPS6 implementations were firmware, as were
the earlier DEC QBus versions (the 1500, 2000 and 3000 series). But I
don't believe the VAX version was.
Luke
Jim Idle
view of the world. In reality there were a lot more beer mats involved
than this.
frosty
>> I see you have the hardware associated with a few platforms.
>> In addition to Honeywell for Ultimate there was the DEC (VAX)
>> Ultimate, which I believe was a firmware implementation - and
>> significant in its own right for that.
>
> The Honeywell Level 6 and DPS6 implementations were firmware, as were
> the earlier DEC QBus versions (the 1500, 2000 and 3000 series). But I
> don't believe the VAX version was.
Level 6 "firmware" was contained in Writable Control Store (WCS)
which was a little softer than firm; it was a RAM, not a ROM.
Also, not all the O/S was in the WCS; the equivalent of "monitor code"
was written in assembly language. In some of the earlier releases, a
few of the "Pick" virtual assembly language instructions were executed
in software, as well.
--
frosty
dawn
Tony Gravagno wrote:
> Thanks for the effort Dawn. You'll need to get confirmation of these
> notes, my memory is real vague on all this and I never was a Pick
> history buff. I did have more notes but scratched them the closer I
> looked. :)
>
> I see you have the hardware associated with a few platforms.
I put in some that I thought might be of historical significance.
Showing IBM at the start (at the same time Codd was at IBM researching
the relational model, by the way) as well as at the end of the history
seemed significant enough to point out.
> In addition to Honeywell for Ultimate there was the DEC (VAX)
> Ultimate, which I believe was a firmware implementation - and
> significant in its own right for that.
>
> The three companies currently involved with the jBASE DBMS are jBASE
> International, Temenos, and mPower1. This is as of 2002 when Mpower1
> acquired rights for support and distribution. Ref:
> http://www.jbase.com/about/ and elsewhere. "jBASE" is listed as
> "Maintains an MV System in application". Technically that's not
> accurate as jBASE does not have an application but Temenos does.
Yes, that happened after I published this version, so I will work on
straightening that out. I can only include so much info, so I don't
know if jbase.com is even relevant in this diagram. I should perhaps
take that question to the jbase list.
> I
> can't suggest exactly how this should all be displayed but as it is
> it's not 100% accurate.
>
> ADP is mentioned in the Reality branch. From recent CDP discussions
> and e-mail exchanges, I see they are still a player in the auto
> dealership vertical market. I don't know if they just lock-down stock
> DBMS releases or if they are still categorized as a licensee.
I thought they were maintaining it themselves, but maybe someone else
knows more about this. I did not get a contact for that implementation
to interview for this diagram and it would be good to have one.
> If they
> are a licensee, this could make them one of the longest running
> licensee/resellers out there.
Anyway you look at it, they are one of the longest running MultiValue
implementations, I would think. Datatel, my former employer, was
started in 1968 (making it one of the oldest existing application
software companies, I suspect) and is another long running MV VAR. I
don't know when they started with it, but they went from Microdata IIRC
to Prime Information to UniData.
> Noting the MultiValue logo, it's a shame that SMA isn't recognized
> somewhere in the mix as having an impact on some of the cross-platform
> standards that we still enjoy today.
Good point. I tried to come up with a way to recognize them since they
didn't fit the other categories and decided that using the logo was a
way to do that. I could put a URL to their site or the explanation of
the logo on the next version. I'll check with Gus.
>
> Might be Pick-ayune, but I doubt whether the distinction between
> Advanced Pick and AP Pro is worthy of mention.
That is helpful to know. I took info from many people and sources. I
never worked on the PICK side of the house, so I appreciate that type
of input.
> AP Protected mode was
> just one step off from AP Native, which isn't mentioned there, and I
> don't think Native is worthy either considering Pick was originally a
> native platform in the first place. If AP PRO is in there then so
> should be AP DOS (EOL 5.2.5) and AP Native (EOL 5.2.7). If not, and
> this is my preference, I'd suggest scratching AP PRO and just leaving
> Adv Pick.
That is what I'll do.
> Also under "ports", was Data General really a licensee or were MV
> products merely ported over DG/UX?
I had that one on-again, off-again when I was developing it. My
criteria for inclusion on that list was not really tight. My interest
is less on licensing than on whether there was an implementation of the
languages that was separate from other implementations (a "port" to
that platform). But my notes are not clear enough for me to figure out
who gave me DG information so I could ask some follow-up questions.
Hopefully others will know. I think there are other items in this list
that might stand out as being of a different ilk too and would like to
identify those.
> Finally, we often hear Jim Idle say "jBASE is not Pick", and I know
> that confuses people. We see in the tree that MDIS had impact on JAC
> where jBASE was written, and the license was subsequently transferred
> to jBASE International. But to understand that history I found a neat
> article which explains that process _and_ really explains well how and
> why jBASE was developed. This provides some compelling insight into
> jBASE, and may help to explain why jBASE may be better positioned than
> some other MV platforms for the open market. Ref:
> http://www.ukpua.org/archives/articles/JAC.html
Yes, I believe that was one of my sources, although I did talk to Jim
too at some point, if I recall correctly.
> Regards,
> T
Thanks a bunch. --dawn
Nikolai Lukin
news:4381b...@news.melbourne.pipenetworks.com...
> Tony Gravagno wrote:
>> I see you have the hardware associated with a few platforms.
>> In addition to Honeywell for Ultimate there was the DEC (VAX)
>> Ultimate, which I believe was a firmware implementation - and
>> significant in its own right for that.
>
> The Honeywell Level 6 and DPS6 implementations were firmware, as were the
> earlier DEC QBus versions (the 1500, 2000 and 3000 series). But I don't
> believe the VAX version was.
In the late 80ies Ultimate (France/UK?) developed an Inmos Transputer based
ULT-Pick implementation. I had a chance to play around that board installed
into 286 PC. Technically the solution seemed quite promising those times.
Nick
dawn
Don't know why I thought of this, Jim ;-) but wouldn't a history of
litigation in the MultiValue community be a fun one to write up too?
Cheers! --dawn
Simon Verona
customised version of Reality/X at the time, which they developed to suit
themselves (they updated the compiler to assist in us porting our UK
application from MOE to ADP's Reality)... In addition, they would from
time to time retrofit items from the standard version of Reality/X into
their own..
As far as I am aware, whilst they have tried to replace Reality, I believe
that it is still used today unless somebody else knows better!
Out of interest, they also did some work with jBase several years ago but
decided not to move from Reality at the time, though we did move the UK
application from ADP Reality to jBASE running on an IBM RS/6000 platform.
Not sure if this info helps!
Regards
Simon
"dawn" <dawnwo...@gmail.com> wrote in message
news:1132594482.2...@f14g2000cwb.googlegroups.com...
Tony Gravagno
I remember now that the boot sequence off the front panel of a Level 6
prompted for the Ultimate boot tape and allowed for some diagnostics,
so some part of the monitor was in firmware. I worked on all three of
the DECs, 1500, 2000, 3000 so I know those for sure - and yes, VAX was
VAX. Thanks for the correction.
T
Tony Gravagno
>> Also under "ports", was Data General really a licensee or were MV
>> products merely ported over DG/UX?
>
>I had that one on-again, off-again when I was developing it. My
>criteria for inclusion on that list was not really tight. My interest
>is less on licensing than on whether there was an implementation of the
>languages that was separate from other implementations (a "port" to
>that platform). But my notes are not clear enough for me to figure out
>who gave me DG information so I could ask some follow-up questions.
>Hopefully others will know. I think there are other items in this list
>that might stand out as being of a different ilk too and would like to
>identify those.
Well, if you're talking "ports" then there was AP and D3 for DG/UX,
HP-UX, SCO, and Siemens-Nixdorf (different than Siemens' own port).
There was a rotten AT&T port which was associated with a spinoff
marketing company from Pick Systems called PickTel. (Is that an
accurate assessment of the relationships?) And I think Henry Vu (Pick
Systems) at some point at least got to a logon prompt over
Apple/Macintosh (!). Other DBMS products were similarly ported but I
don't think this is all truly significant. From a historical
perspective it's probably better to just focus on licensees than
"ports". Some porting, non-licensee, efforts were significant though,
like the SMI 370/4300 port. Tough calls...
As I write this I'm now reminded that there was a Data General system
back in the 80's that looked Pick-ish, but I never coded over it and
never did more than the most basic operations with it. Anyone have a
clue about that one?
I understand the focus and complexity of what's already there, but
it's sort of a shame to not see some of the milestones. Like:
- When did Dick first discuss porting Pick to IBM - and then decide he
didn't want to?
- Who had the first MV/Unix port?
- I believe AP was the first commercial database (MV or otherwise)
ported to Linux. Can someone contest that?
Just going on... Thanks again.
T
Luke Webber
> Luke Webber wrote:
>>The Honeywell Level 6 and DPS6 implementations were firmware, as were
>>the earlier DEC QBus versions (the 1500, 2000 and 3000 series). But I
>>don't believe the VAX version was.
>
> Level 6 "firmware" was contained in Writable Control Store (WCS)
> which was a little softer than firm; it was a RAM, not a ROM.
Well, yeah, but it operated very much the same as a firmware processor.
I think it might even have used the same bit-slice processor
architecture as other firmware boards. The fact that firmware was
written at boot time was really just a bonus.
The later Ultimate-specific processors wee also writeable, of course,
but much faster than the original WCS.
> Also, not all the O/S was in the WCS; the equivalent of "monitor code"
> was written in assembly language. In some of the earlier releases, a
> few of the "Pick" virtual assembly language instructions were executed
> in software, as well.
Not many, but some. They were just cross-assembled to MCAL instructions.
Luke
Peter McMurray
Hi
In Australia the Microdata REALITY was marketed by AWA and became the ROYALE
around 1978, the scuttlebut had it that the name change was because of a
royalty agreement arising from one of Dick's famous copyright stoushes.
In 1988 there was an Advanced Pick release on a piece of junk called EDGE.
I was the first and probably the last poor soul to use this. It was
launched with a stupendously expensive Superman fanfare. I was forced into
it instead of the established machines I preferred by the "experts" from a
big six accounting firm. How good was it? Suffice to say the response to a
tape read error was a hex core dump to printer and immediate halt on a 120
user system! Fantastic!
Keep up the good work Dawn
Peter McMurray>
Ross Ferris
I remember seeing your EDGE - 'redefining the desktop' is a term that
springs to mind :-)
dawn
Tony Gravagno wrote:
<snip>
> I understand the focus and complexity of what's already there, but
> it's sort of a shame to not see some of the milestones. Like:
> - When did Dick first discuss porting Pick to IBM - and then decide he
> didn't want to?
> - Who had the first MV/Unix port?
> - I believe AP was the first commercial database (MV or otherwise)
> ported to Linux. Can someone contest that?
>
> Just going on... Thanks again.
> T
I almost ready to shut the lid on this laptop and pack it up for a trip
back home for Thanksgiving (eager to see the kids, parents, and
everyone else), but a quick response on this.
I have a spreadsheet with a history of "stories" with some known dates
and other estimates. This is the research from which I developed the
poster, so there are plenty of stories and milestones that were out of
scope completely. I'm planning to get this on the web so that people
can submit stories and milestones for inclusion. I would like
something that permits people to add their names as "verifying" the
data. That way stories without multiple sources can still go out
there, but we can know they are not verified by anyone else. So, stay
tuned for that.
Happy Thanksgiving to the US folks and for those outside the US, feel
free to overeat while counting your blessings with the rest of us.
Cheers! --dawn
Ross Ferris
stories Dawn
Excalibur
I would have thought boat anchor would have been a better description. It
filled my long wheel base LandRover and took several removalists to shift
out of my office.
Of course it also caused some extremely "interesting" moves in the marketing
area from a certain Lionel Singer and Pyramid. In relation to Dawn's
request there are a couple of names for her history
Lionel Singer - Prime + Primos - Pyramid - Universe (I think).
A gentleman accountant from Canberra with many talents. I believe his first
foray into the IT field was with Wang.
He certainly made a big impression for a little guy.
"Ross Ferris" <ro...@stamina.com.au> wrote in message
news:1132620278.7...@o13g2000cwo.googlegroups.com...
Proco_Mark
At Tina Turner show in the Sydney Hilton in 1984, ten or so of us sat
down and the first thing Lionel did was take off his shoe and put it on
the table; it had a phone in it.
Bruce Nichol
>Quite an impression.
>At Tina Turner show in the Sydney Hilton in 1984, ten or so of us sat
>down and the first thing Lionel did was take off his shoe and put it on
>the table; it had a phone in it.
But, the best thing - besides Tina's legs - was the doorman wouldn't
let Molly Meldrum in.... "Private function" ... "Um...Um...Um ..Um"
>
>Excalibur wrote:
>> Hi Ross
>> I would have thought boat anchor would have been a better description. It
>> filled my long wheel base LandRover and took several removalists to shift
>> out of my office.
>> Of course it also caused some extremely "interesting" moves in the marketing
>> area from a certain Lionel Singer and Pyramid. In relation to Dawn's
>> request there are a couple of names for her history
>> Lionel Singer - Prime + Primos - Pyramid - Universe (I think).
>> A gentleman accountant from Canberra with many talents. I believe his first
>> foray into the IT field was with Wang.
>> He certainly made a big impression for a little guy.
>>
>> "Ross Ferris" <ro...@stamina.com.au> wrote in message
>> news:1132620278.7...@o13g2000cwo.googlegroups.com...
>> > Hey Peter,
>> >
>> > I remember seeing your EDGE - 'redefining the desktop' is a term that
>> > springs to mind :-)
>> >
Regards,
Bruce Nichol
Talon Computer Services
ALBURY NSW Australia
If it ain't broke, fix it until it is....
dawn
dawn wrote:
> I am planning to make changes to the electronic version of the
> MultiValue Family Tree poster in the coming year. The 2002 version of
> the pdf is found at
>
> http://www.tincat-group.com/mv/familytree.html
>
I received some information from Alan Pritchard that his publications
had a complete list of all implementations of Pick he 't have access to
copies right now. Does anyone have copies of the Pick Resources Guide
(1986) and Pick Resources Guide/International (1987 and 1990)? He said
the information was under the hardware sections. Thanks in advance for
any help in finding these.
He also wrote "There was at least one microfilm retrieval system
(Kodak, maybe) that used an embedded Pick system in it." Does anyone
recall that?
Thanks. --dawn
Steve Lancour
> dawn wrote:
>
<snip>
>
> He also wrote "There was at least one microfilm retrieval system
> (Kodak, maybe) that used an embedded Pick system in it." Does anyone
> recall that?
>
> Thanks. --dawn
>
Dawn,
I once salvaged data for a customer from a Kodak KAR 4000 system. KAR
4000 was an ADDS machine running Mentor and held a database providing an
index to documents stored on microfilm.
Hope this helps.
Steve Lancour
dawn
Very cool. I think that means it is an application on ADDS Mentor
rather than a separate port. Although there are some great apps and a
better historian than I might be able to track a lot of apps down
sometime, I think this means that the Kodak implementation doesn't make
it to the poster. If anyone disagrees, let me know. Thanks a bunch!
--dawn
Steve Lancour
It was not a Kodak-specific port. MOE was well-hidden from the user but
it was definitely MOE.
dawn
If you recall an approximate date, I can include it and other
applications in the varied timeline even if not in the poster. Thanks,
Steve. --dawn
Bruce Nichol
T'was installed at Kodak in Melbourne mid '80's... Based on ADDS
Mentor - they were neighbours in Hauppauge (sp??), NY...
>
>Thanks. --dawn
Mark Brown
company called Scan Optics(?). I think they were aimed at medical imaging,
but they did a lot of testing with Zippy comics. I belive they were an OA
shop. Frosty, just before he came to work for you?
Mark Brown
"dawn" <dawnwo...@gmail.com> wrote in message
news:1133119232....@f14g2000cwb.googlegroups.com...
EPPICK77
It was from Kodak and it used an ADDS machine.
Eugene
"dawn" <dawnwo...@gmail.com> wrote in message
news:1133119232....@f14g2000cwb.googlegroups.com...
>
Steve Lancour
I worked on the customer's KAR 4000 in about 1996 and it was already
pretty old by then. A Google search for "KAR 4000" found this:
http://www.kodak.com/global/en/corp/historyOfKodak/1980.jhtml?pq-path=2217/2687/2695/2702
Steve
dawn
Good find, Steve. Thanks! --dawn
> Steve
csig...@hotmail.com
> He also wrote "There was at least one microfilm retrieval system
> (Kodak, maybe) that used an embedded Pick system in it." Does anyone
> recall that?
>
> Thanks. --dawn
In 1974 I left the microfilm (hence the name Microdyne) and IT business
to concentrate on manufacturing ergonomic chairs. Sometime after that
I saw a demo of that Kodak system and was very impressed with its
computer indexing and retrieval system. In 1978, when I started
looking for a comprehensive software solution for the chair company, I
recalled that Kodak system and an old friend at my erstwhile biggest
competitor revealed the secret that led me to ADDS. At about the same
time, with the help of an article about Pick in one of the computer
magazines I was able to put the Pick, Microdata and ADDS pieces
together. There are of course many things that I love about Pick but
it was especially that portability, the fact that the same system ran
on hardware from different manufacturers, that made Pick attractive to
a start-up company with big ambitions. The other thing that sticks in
my mind as a deciding factor was the Pick internal date that made the
system Y2K proof but that Y2K term was, of course, not used at that
time. Our first machine was a Microdata, mostly because the Microdata
salesman showed up. In 1984 we switched to ADDS because NCR provided
local service. To this day I still use MentorPro to look up historical
data for the chair company that I sold at the end of 1989.
Henry Keultjes
Microdyne Company
Mansfield Ohio USA
(latimerp)
very fault tolerant, very expensive. Ultimate also ran on the IBM
9370. It was ok but the (Hyphas?) cables could make connectivity
a big pain.
Patrick, <;=)
dawn wrote:
> I am planning to make changes to the electronic version of the
> MultiValue Family Tree poster in the coming year. The 2002 version of
> the pdf is found at
>
> http://www.tincat-group.com/mv/familytree.html
>
>>From what people have told me, the poster is very accurate, but I
> believe there are some inaccuracies, I just don't know what they are.
> If you have a chance to look it over and spot anything, however minor,
> that appears inaccurate or misleading, I would appreciate the
> information you have.
>
> Also, there have been some changes in products or companies, including
> at least jBASE and OpenQM, since this poster was developed. Any
> information about the products and companies since early 2002 that
> would help update this poster for today would be much appreciated.
>
> Thanks in advance for any information you can give. You may either
> e-mail me at dwolt at tincat-group dot com or respond to this posting,
> especially if you want input from others on the accuracy of your input.
>
>
> Thanks in advance. --dawn
>
Tony Gravagno
>I worked on the customer's KAR 4000 in about 1996 and it was already
>pretty old by then.
I trained a couple of the KAR developers for ADDS development back
around 1990. It was definitely just an app.
T
Bruce Nichol
The ADDS box (2020 or 2040?) at Kodak in Melbourne in '84/'85 was a
"standard" box - no "Kodak" stickers .... and IIRC the Kodak people
involved called themselves "ADDS distributors" which got up the nose
of Clegg Driscoll (*the* Australian distributors) a bit...... but, as
I said before, Kodak and ADDS were neighbours in NY and it all seemed
to germinate form there......
Simon Verona
machine about 18" square footprint and approx 3 1/2 feet high).... IIRC
this used a slow Zilog Z8000 processor and had a massive 512kb of ram.
Connectivity was via a serial port board that slotted into the back of the
machine.
If this is the machine in question, then it was my first contact with
"Pick". I have fond memories of those machines - particularly using
different combinations of the 3 push buttons on the case to access the
monitor mode to reload the operating system or perform diagnostics..
It always amazes me that we could run 32 dumb screens off one of those
boxes! I used to have one sitting in our hallway - they make great
plant stands!
Regards
Simon
"Bruce Nichol" <revers...@taloncs.com.au> wrote in message
news:8tspo1l1rfljukn5i...@4ax.com...
murthi
news:ELednUTPX8i...@comcast.com...
> Don't forget the Stratus Pick systems running over VOS. Very nice,
> very fault tolerant, very expensive. Ultimate also ran on the IBM
> 9370. It was ok but the (Hyphas?) cables could make connectivity
> a big pain.
expect to get it transmitted to the cpu and echoed correctly... but I guess
for the dump-entire-screen IBM folk it was like a miracle... Chandru
dawn
(latimerp) wrote:
> Don't forget the Stratus Pick systems running over VOS. Very nice,
> very fault tolerant, very expensive. Ultimate also ran on the IBM
> 9370. It was ok but the (Hyphas?) cables could make connectivity
> a big pain.
>
> Patrick, <;=)
Do you have any recollection of what Stratus named their flavors of
BASIC and the query language? I don't know how to find such
information, so if anyone has good recall or old documentation for any
of these ports or new implementations, that would be most helpful.
Thanks! --dawn
(latimerp)
> (latimerp) wrote:
>
>>Don't forget the Stratus Pick systems running over VOS. Very nice,
>>very fault tolerant, very expensive. Ultimate also ran on the IBM
>>9370. It was ok but the (Hyphas?) cables could make connectivity
>>a big pain.
>>
>>Patrick, <;=)
>
>
> Do you have any recollection of what Stratus named their flavors of
> BASIC and the query language? I don't know how to find such
> information, so if anyone has good recall or old documentation for any
> of these ports or new implementations, that would be most helpful.
> Thanks! --dawn
http://ftp.stratus.com/vos/srbs/r12.0_srb.memo
Patrick <;=)
>
>
<snip>
dawn
Looking over the list of manuals in that document, I don't see anything
about either the query language or database. It sounds like VOS was
the O/S and from what you indicated, that Pick was running on top of
that. VOS BASIC, however, appears to be a BASIC compiler for the VOS
O/S. I don't see anything in their list of documents that gives hints
at the other Pick components, so I'm not sure if VOS BASIC is a
DataBASIC flavor or not, but I'm guessing not. I don't see any hint of
a database or query language, for example. Do you think they ported
the Pick BASIC p-code compiler into a native VOS compiler? --dawn
> Patrick <;=)
> >
> >
> <snip>
Rob Allen
> I worked for ADP for several years (albeit 7 years ago).. They ran a
> customised version of Reality/X at the time, which they developed to suit
> themselves (they updated the compiler to assist in us porting our UK
> application from MOE to ADP's Reality)... In addition, they would from
> time to time retrofit items from the standard version of Reality/X into
> their own..
>
> As far as I am aware, whilst they have tried to replace Reality, I believe
> that it is still used today unless somebody else knows better!
>
> Out of interest, they also did some work with jBase several years ago but
> decided not to move from Reality at the time, though we did move the UK
> application from ADP Reality to jBASE running on an IBM RS/6000 platform.
I work at ADP now, and we do indeed still use our own version of
Reality, running on Red Hat Linux. There hasn't been a serious attempt
to replace Reality since they stopped the jBASE project in 1996, and
there have been few if any changes made to our version of Reality since
then except for the web interface.
Our last machines that ran Reality as the OS went away in the
preparations for Y2K - we decided we weren't going to update the
software and we made the clients move to Unix. The last ones were a few
McDonnell-Douglas 'Spirit' (Motorola 68000) machines and some of our
homegrown Micro-1000 and Micro-2000 boxes - Intel 8086 machines running
Reality. Those are the only computers that ADP ever produced ourselves,
and probably the only instance of the Reality OS running on Intel
processors.The near-legendary HP 3000 Pick implementation never made it
out of the lab. Dick raved about how good it was for years, but I've
been told that it was comparable to others of its time. Dick knew that
no one would ever see it, so he could say anything he wanted.
Rob Allen
ADP Dealer Services
Portland, OR
dawn
> Simon Verona wrote:
> > I worked for ADP for several years (albeit 7 years ago).. They ran a
> > customised version of Reality/X at the time, which they developed to suit
> > themselves (they updated the compiler to assist in us porting our UK
> > application from MOE to ADP's Reality)... In addition, they would from
> > time to time retrofit items from the standard version of Reality/X into
> > their own..
> >
> > As far as I am aware, whilst they have tried to replace Reality, I believe
> > that it is still used today unless somebody else knows better!
> >
> > Out of interest, they also did some work with jBase several years ago but
> > decided not to move from Reality at the time, though we did move the UK
> > application from ADP Reality to jBASE running on an IBM RS/6000 platform.
>
> I work at ADP now, and we do indeed still use our own version of
> Reality, running on Red Hat Linux.
Do you call it "Reality"? I wasn't sure, so I simply have the company
name and not the product name in the poster. What do you call basic
and query?
> There hasn't been a serious attempt
> to replace Reality since they stopped the jBASE project in 1996,
any stories related to that project?
> and
> there have been few if any changes made to our version of Reality since
> then except for the web interface.
>
> Our last machines that ran Reality as the OS went away in the
> preparations for Y2K -
Wow. I didn't think of people still running pick as the OS that late.
I wonder if anyone out there is still running Pick as an OS.
> we decided we weren't going to update the
> software and we made the clients move to Unix. The last ones were a few
> McDonnell-Douglas 'Spirit' (Motorola 68000) machines and some of our
> homegrown Micro-1000 and Micro-2000 boxes - Intel 8086 machines running
> Reality. Those are the only computers that ADP ever produced ourselves,
> and probably the only instance of the Reality OS running on Intel
> processors.The near-legendary HP 3000 Pick implementation never made it
> out of the lab. Dick raved about how good it was for years, but I've
> been told that it was comparable to others of its time.
That squares with my information. I'm not much of a hardware man
myself, but I included a few pieces of hardware in the poster: IBM
7090 that I believe had the first installation, the XEROX Sigma 7 where
Pick worked at UC Irvine and on which he & Simms encountered the
p-machine, Microdata's hardware which included the Intertechnique
something or other (need to consult notes), Intertechnique, which is
how Pick could market the product after splitting with Microdata, and
Pr1me because it has a place in my heart, I mean, because Prime
Information was the start of the third stream of Pick languages, the
way I have identified it.
I considered others, such as the first Intel port, but that was muddied
by the board that could be slipped into an AT. It was an Intel port,
but not a DOS port and I didn't know how to handle it. At this point,
I forget whether the first product running on Intel was from Pick
Systems or elsewhere. I also don't know who did the first DOS port -
Revelation, perhaps?
Thanks a bunch. --dawn
Rob Allen
> Do you call it "Reality"? I wasn't sure, so I simply have the company
> name and not the product name in the poster. What do you call basic
> and query?
Well, it isn't really a product that we sell, it's just the environment
where our core applications run. Internally, we call it Reality or
CoRA; basic is Data/Basic, and the query language is English. I think
"CoRA" originated as McDonnell-Douglas' internal project name for the
Reality-on-Unix product, later named RealityX and now just Reality.
It's an acronym for "Common Reality Architecture" or something like
that.
>
> > There hasn't been a serious attempt
> > to replace Reality since they stopped the jBASE project in 1996,
>
> any stories related to that project?
It was known internally as "the CoRA Bridge" that would take us over to
the Unix world. When we first looked at jBASE, they had the ability to
take Data/Basic programs and turn them into C programs. At our
insistence, they added the option to go to C++ also. Then management
decided that the payoff wasn't there, so they pulled the plug on the
project and we stayed with CoRA.
> I considered others, such as the first Intel port, but that was muddied
> by the board that could be slipped into an AT. It was an Intel port,
> but not a DOS port and I didn't know how to handle it. At this point,
> I forget whether the first product running on Intel was from Pick
> Systems or elsewhere. I also don't know who did the first DOS port -
> Revelation, perhaps?
Revelation and CDI-1000 were almost simultaneous DOS ports. The latter
disappeared quickly. I think CDI is the company that also did the IBM
Series/1 port. Here's another little-remembered product: in the early
80s, McDonnell-Douglas licensed Revelation, tweaked it to resemble
Reality, and sold it on PCs from Convergent Technologies. They named
it the "M-1000" system. The base OS was Convergent's CTOS, which had a
DOS emulation in which they ran Revelation, which was a Pick emulation.
Lots of chances for things to go wrong... and they did.
The first native Intel implementation I know of was the Altos 586,
which was done around 1983-84. We resold them in the now-defunct ADP
Accounting Services division; Dealer Services never sold anything but
Reality. Accounting Services had previously sold ADDS Mentors too. A
couple of years later, Pick Systems came out with their PC-XT version.
We got a copy to evaluate but didn't see it fitting our needs.
Rob Allen
dawn
Interesting. I think Pr1me also bought Revelation or some derivative
thereof and turned it into PI/Open. IBM dropped support for PI/Open
earlier this year (or perhaps last year).
>
> The first native Intel implementation I know of was the Altos 586,
> which was done around 1983-84. We resold them in the now-defunct ADP
> Accounting Services division; Dealer Services never sold anything but
> Reality. Accounting Services had previously sold ADDS Mentors too. A
> couple of years later, Pick Systems came out with their PC-XT version.
> We got a copy to evaluate but didn't see it fitting our needs.
>
> Rob Allen
Thanks, Rob. Is ADP married to Reality for the forseeable future or
has the relational database trend over the past couple of decades
reared its head within ADP directions at all? --dawn
Simon Verona
Reality (or CoRA as they called it)... The Operating system (Motorola based
SVR4 when I was there but looks like RedHat now) was pretty tied in with
CoRA and with the application.
I do seem to recall a couple of attempts of migrating from Reality whilst I
was at ADP - one was with jBASE - which is how I got into jBASE - I worked
for a company in the UK that was purchased in 1992 by ADP and seeing as
jBASE was UK based (Hemel Hempstead) I was chosen to "evaluate" the product.
It wasn't much of a product in those days - pretty much just a filesystem
and data-basic compiler - but the potential was obvious, and the enthusiasm
for the product from the likes of Jim (Idle), Clive (Ketteridge) and Greg
(Cooper) was infectious.
Needless to say, today I run on jBase and have done for several years..
I have to say that porting our UK product from MOE (ADDS Mentor on NCR UNIX
SVR4 Intel machines) to ADP's CoRA was one of the hardest things I've had to
do in my professional career to date. It was obviously the "wrong" thing to
do - we had already done a succesful port to jBase but politics stopped us
releasing - CoRA was finely tuned to support the ADP applications which made
life very difficult. However, ADP had (and probably still do) some *very*
talented people who worked with CoRA - they updated the compiler to support
some of the data-basic syntax in Mentor (particularly using stuff like
EXECUTE rather than PERFORM).
I seem to recall a project that ADP had to port to a Delphi front end with
an SQL based database - this I believe was not dissimilar to the project
that Reynolds and Reynolds recently canned. I also believe that this
project was still-born. I won't say that they chose the wrong products, but
maybe had the wrong people trying to implement!
I'm secretly pleased to head that a massive organisation like ADP is still
highly tied to a multi-value platform!! It kind of vindicates my position
to stay MV!
All this reminising is starting to make me feel old.... Funny thing is
that I'm not really old enough to remember back as far as I do.... It's
just that I started working with Mentor back in 1984 at the tender age of 13
(my father had started a software company which used Mentor)...
"Pick" - it just seems to get under the skin.. once you've got it, it seems
to stay with you!
Regards
Simon
"dawn" <dawnwo...@gmail.com> wrote in message
news:1133555722.2...@g47g2000cwa.googlegroups.com...
dawn
Simon Verona wrote:
<snip>
> All this reminising is starting to make me feel old.... Funny thing is
> that I'm not really old enough to remember back as far as I do.... It's
> just that I started working with Mentor back in 1984 at the tender age of 13
> (my father had started a software company which used Mentor)...
Oh yeah -- that's how I worked on a Pr1me in the 70's -- as a mere
child. After I watched Mr. Roger's Neighborhood, then I coded ;-)
>
> "Pick" - it just seems to get under the skin.. once you've got it, it seems
> to stay with you!
I'm a convert. I didn't see Prime Informaiton until the late 80's.
After considerable IMS / CICS / COBOL I was amazed at how productive a
team of P/I developers could be. I am not at all stuck with any
products and could jumpt to XML databases, MySQL, or anything else
right now, but I can see that nothing SQL-based is going to be as good
from the perspectives I care about (flexibility,
big-bang-for-the-buck). I'm planning to join the vast numbers of
bloggers in January. I haven't been a writer before, so we shall see
how that goes.
Cheers! --dawn
Joe
news:1133555722.2...@g47g2000cwa.googlegroups.com:
AFAIK, ADP owns their own "version" of Reality, which I hear gets
tweaked every now and then up in Portland. As Rob said, all new boxes
are Linux, and the application is very solid and runs very fast. On
the down side, there is still the old 32k limit, so you're limited to a
certain size when you compile/catlog.
Regards,
Joe
Mark Brown
me to lunch and we talked for hours.
They wanted someone to write a "windows" style interface to the extent that
they wanted the system to "remember" what was on the screen and repaint it
on demand.
I brought code to show it could be done and after an hour at a keyboard, had
the basic save-the-data/display-the-data going. But I was asking for $40K
and they balked.
SB+ and several others have "windows" that either pop-up or fill a screen,
but when they repaint, they repaint EVERYTHING. As far as I know I'm still
the only one with "pixel" level repaint capabilities. But with the increase
of real Windows, it's pretty much "bringing you yesterdays technolgy
tomorrow."
Mark
"Joe" <avoidi...@nospam.com> wrote in message
news:Zj5kf.34246$i7.2...@bignews2.bellsouth.net...
> "dawn" <dawnwo...@gmail.com> wrote in
> news:1133555722.2...@g47g2000cwa.googlegroups.com:
>
>>
dmm
Mark Brown wrote:
> When CDI laid me off in 1985, I interviewed with ADP in Portland. They took
> me to lunch and we talked for hours.
>
> They wanted someone to write a "windows" style interface to the extent that
> they wanted the system to "remember" what was on the screen and repaint it
> on demand.
>
> I brought code to show it could be done and after an hour at a keyboard, had
> the basic save-the-data/display-the-data going. But I was asking for $40K
> and they balked.
>
> SB+ and several others have "windows" that either pop-up or fill a screen,
> but when they repaint, they repaint EVERYTHING. As far as I know I'm still
> the only one with "pixel" level repaint capabilities. But with the increase
> of real Windows, it's pretty much "bringing you yesterdays technolgy
> tomorrow."
>
> Mark
[snip]
If we are talking screen refreshing on standard terminals, the 'green
screen' version of SB+ from about version 2.0/2.1(approx 1993/94) IIRC
had a 'smart refresh' option that would repaint _only_ the portion of
the screen that was 'damaged' (plus any function key row). For example,
if a new 'window' was displayed (regardless of size), only the screen
area consumed by that window was re-painted when that window was
destroyed.
For R83 systems (quite a few in those days), there was even an Assember
version to improve performance. This took up around 6 ABS frames, and
was installed by the user if required. There was another option that
would perform full screen refreshes, but once the 'smart refresh'
option was made available it was rather obsolete. Perhaps this is what
you are refering to.
I do not recall if SBClient did full screen refreshes, but on a PC
based windows system where easy mechanisms exist to do this kind of
thing, it probably made sense to do so.
I also think that Doug Dumitru sold/is-selling a windowing system for
R83 (and others?) that does partial screen refreshes, and has done
since (at least) the late 80's.
dave
Simon Verona
web server somehow and have written a web front end to their applications...
I'd guess that this is using home-grown middleware and extensions to
reality.
What was always ironic, in my humble opinion, is that whilst the guys at
Portland were extremely capable both as an individual and a group, ADP may
have been better served in buying off the shelf components/middleware rather
than doing it themselves. In my day, they were spending lots of time
reintegrating (or re-writing) features into their version of Reality that
was in the standard Reality release. The version of Reality that they
used was totally ADP unique in that is wasn't based solely on any version of
MDIS's Reality code but was a mish-mash of code from differing releases!
Regards
Simon
"Mark Brown" <mbr...@drexelmgt.com> wrote in message
news:u09kf.42$ka...@tornado.socal.rr.com...
Joe
news:4391bf70$0$63063$ed2e...@ptn-nntp-reader04.plus.net:
> I'm guessing (from their web site) that they've now interface reality
> with a web server somehow and have written a web front end to their
> applications...
Correct. The web services generally run on the same box the application
and database sits. There is now a browser-based GUI (IE-specific AFAIK)
that ties directly into the database.
> I'd guess that this is using home-grown middleware and extensions to
> reality.
Basicallly correct.
> What was always ironic, in my humble opinion, is that whilst the guys
> at Portland were extremely capable both as an individual and a group,
> ADP may have been better served in buying off the shelf
> components/middleware rather than doing it themselves. In my day,
> they were spending lots of time reintegrating (or re-writing)
> features into their version of Reality that was in the standard
> Reality release. The version of Reality that they used was
> totally ADP unique in that is wasn't based solely on any version of
> MDIS's Reality code but was a mish-mash of code from differing
> releases!
After so many years of tweaking, everything's proprietary in their
environment to the point where off-the-shelf products aren't nearly as
efficient as the "homegrown" solutions.
Regards,
Joe
Simon Verona
home-grown middleware... Had mv.net or something equivalent been around a
few years back, then I would be using that rather than my own middleware -
but now I've written it, it's not worth migrating!
Regards
Simon
"Joe" <avoidi...@nospam.com> wrote in message
news:_epkf.14731$wi2....@bignews1.bellsouth.net...
Jim Idle
I guess it should have been the quarter pounder in the US then!
John Travolta.
rog
news:1133671219.4...@g47g2000cwa.googlegroups.com...
Way back when Microdata brought out 3.0, wasn't the name of the 'new'
system Royale? Along with those gold cover manuals?
Roger
Tony Gravagno
>"Pick" - it just seems to get under the skin.. once you've got it, it seems
>to stay with you!
Reminds me of a pink t-shirt I have with a Pick Systems logo on the
back, a heart one of the sleeves, and this quote on the front:
PICK ... the life long love affair, not just a passing fancy!
Hey Dawn a couple days ago I was thinking about geeking out and
turning the family tree into an iron-on transfer for a t-shirt. Glen
wanted to do some shirts for Spectrum last year to raise funds to
support mvDevCentral and PickSource - mvDC is gone but this might be a
good idea for PS. I'm picturing the image on the front and URLs on
the back for PS, your company, maybe others.
T
Tony Gravagno
>I considered others, such as the first Intel port, but that was muddied
>by the board that could be slipped into an AT. It was an Intel port,
>but not a DOS port and I didn't know how to handle it.
You mean UPBoard? I don't see that in the image but it certainly
should be there somewhere. Wasn't that the same as Pick AT from a
look and feel standpoint?
Side items:
I see a line that ADDS "Code transfers to" R83. Is that accurate?
I've never touched R77 but I do know ADDS in later incarnations and it
would be interesting to have some idea of what innovations R83
acquired rather than what was built-in.
There is a line that mvPRO "Influences the spec of" D3. I dunno about
that. GA products like the Zebra line and R91 certainly influenced D3
with the $options compatibility mode but to my knowledge mvPRO has
never been referenced or acknowledged in AP or D3 development. If
mvPRO is to be considered in this way then other products like
Ultimate had a similar influence. It's not so much that these
products influenced D3 development but that from a competitive
standpoint, product compatibility was developed into D3 to facilitate
migrations. It's a fine line but I wouldn't call that an influence on
the spec as we see in other such relationships.
(Could be completely off.)
T
Tony Gravagno
an Ultimate running with a bunch of "programmable" Wyse60+ terminals
which had some page memory. I don't state this to profile that
project but in a humorous way to cite yet another example of major
projects that some of us worked on so many years ago that we've
completely forgotten about until some mention like this jogs the old
brain cells. How many features are in these products, and how many of
these neat home-grown solutions have been written, that few or none of
us know about simply because the people who wrote them are too
forgetful or no longer with us to remind us? I guess this is one of
the things that makes this a "living" market.
T
Mark Brown
built: one I believe was by a team headed by Wally Hobart, called "Vulture"
that was supposed to go into the IBM S/1; and another called the UpBoard
that I believe was developed by GA. I believe they we both attempts to
re-create the Ultimate processor that understood Pick Virtual Assembler.
No matter how much you tweek PVA, it still takes X-native instructions to do
1 PVA instruction. If there was a board that understood PVA, it was be at
least 15-20 faster than stand-alone Pick.
I'm not aware of any of these products actually seeing the light of day.
Mark
"Tony Gravagno" <g6q3x9...@sneakemail.com.invalid> wrote in message
news:2819p1hnp3i9pdad0...@4ax.com...
Ed Sheehan
Computing, they had a client on Upboard. I always though it was cute how it
was a 68xxx-based cpu running inside an Intel box.
It worked well, but was doomed my the price and lack of updates my the
manufacturer.
Ed
"Mark Brown" <mbr...@drexelmgt.com> wrote in message
news:Tj2lf.4373$hI1....@tornado.socal.rr.com...
Tony Gravagno
board that hung off the back of the AT, and I believe it had some MV
smarts in it. Maybe I'm confusing that hardware aspect of it with the
Digi-like multiport serial interface. I remember working with one
briefly around '86. Now my curiosity is up.
T
Ross Ferris
board itself was manufactured by Sanyo Icon. In essence not unlike the
little USB devices being marketed now that run Linux out of a thumb
drive & use the PC as an I/O port!
Dick....@gmail.com
couple of the little buggers. It was a Motorola 68020 processor IIRC
on a card that plugged into an A/T expansion slot. You would fire up
MS-DOS then fire up Pick. It ran basically R83. The development was
done by Seattle Lab. Compared to any PC based Pick at the time it was
stinking fast!
dawn
Dick....@gmail.com wrote:
> Yes, the UpBoard was marketed by Sanyo/ICON. In fact, I still have a
> couple of the little buggers.
Any chance you could e-mail me a picture? dwolt at tincat-group dot
com
> It was a Motorola 68020 processor IIRC
> on a card that plugged into an A/T expansion slot. You would fire up
> MS-DOS then fire up Pick. It ran basically R83. The development was
> done by Seattle Lab. Compared to any PC based Pick at the time it was
> stinking fast!
And Seattle Lab's relationship to Seattle OS was that Seattle OS became
Seattle Labs (any story there?) before Pick Blue, right?
Thanks. --dawn
Thiot
It was named after his dog but I can't remember the dog's name. As for
the relationship, as best as I can remember, the only relationship was
people. Jim started a new company with a couple of other guys sometime
after the buyout. I don't know or remember enough of the story beyond
that. As for a picture, I will send you a photo, no problem.
dawn
Good info. I had "Seattle Labs" on the poster. I'm trying to decide
if Seattle Lab belongs there at all now. I had them in the ports
category. I think that both CDI and Seattle OS had one or more ports.
Did Seattle Lab have a Pick license and have their own unique
implementation of Pick on some box or another? I would think that they
did not retain the RT port that Pick Systems named Pick Blue.
Also, did Seattle OS do both the RT nd RS/6000 ports or just the RT?
Thanks. --dawn
Glen B
Heh. I'm definately not a grafx designer, but I still have the PSD files
for the transfer prints. If you want 'em, I'll e-mail them to ya. I didn't
get enough takers on the free and fee shirts that year to justify the costs
involved in handing them out for publicity @ Spectrum. I was hoping to let
them be handed out with registrations, but things never seem to go my way
when I want to give out stuff for free. ;) I may try again.... I was
thinking about sponsoring an MV T-shirt design contest. The winner would get
their name/company in the design for a specific number of shirts. If anyone
is interested in that, or if you guys decide on a shirt graphic and want to
make it available, let me know. I will try to arrange getting shirts printed
and made available for sale.
Glen
--
Posted via NewsDemon.com - Premium Uncensored Newsgroup Service
------->>>>>>http://www.NewsDemon.com<<<<<<------
Unlimited Access, Anonymous Accounts, Uncensored Broadband Access
dawn
Chat with me offline re any ideas. I'm planning to host the image on
the web, but I'm open to any other uses of the image including others
using it to promote their own companies. I'm also going to learn
more graphic design and use Photoshop this time. The current poster
was prepared entirely in PowerPoint. Scary, eh? --dawn
>
> T
Mark Brown
RS/6000 was done internally at Pick Systems. Tim Rude and Rolando Ruiz I
believe are the only ones left from the original team.
After the law suit, there's no way Dick would have let Jim Whelen near a
Pick license.
Mark Brown
"dawn" <dawnwo...@gmail.com> wrote in message
news:1133958398....@g49g2000cwa.googlegroups.com...
dawn
Mark Brown wrote:
> Seattle OS did the RT port. Period.
>
> RS/6000 was done internally at Pick Systems. Tim Rude and Rolando Ruiz I
> believe are the only ones left from the original team.
Was that port done under the name Pick Blue? I'm trying to figure out
whether the name "Pick Blue" should be mentioned at all in the poster.
>
> After the law suit, there's no way Dick would have let Jim Whelen near a
> Pick license.
Make sense. I'll remove Seattle Lab from the MV Family Tree, although
that name is surely still part of the history. Thanks. --dawn
>
> Mark Brown
Jim Idle
Rob Allen wrote:
> dawn wrote:
>
> It was known internally as "the CoRA Bridge" that would take us over to
> the Unix world.
It was in fact done in Hemel with ADP as a main target for it.
> When we first looked at jBASE, they had the ability to
> take Data/Basic programs and turn them into C programs. At our
> insistence, they added the option to go to C++ also. Then management
> decided that the payoff wasn't there, so they pulled the plug on the
> project and we stayed with CoRA.
That's one way to put it ;-). I seem to remember a conversation like
this:
Jim> I don't think you want to convert to C++, you basically get a
bunch of uncommented BASIC code turned into a bunch of uncommented C++
code that won't perform as well. C++ is still realtively young... and
you don't employ any C++ programmers;
ADP Programmers> We agree with Jim - it would be better to start
writing new code in C/C++ and integrate with the BASIC compiled into C;
ADP Management> We don't agree - show us why;
...2 days elapse...
Jim> OK - here are all the reasons it's a bad idea...
ADP Management> Well, we accept all that but we think that at the end
of the day, if we do it, there will be some really good reasons for
doing it that will turn up later.
Dilbert> ??!!??
Charlie Brown> Good Grief.
.. time elapses...
ADP Management> Run this benchmark that we use. It has to be twice as
fast.
Dilbert> Well, we have code profiling in jBASE, it shows that all the
benchmark does is take 0.34% of the time doing a bunch of set up then
enters a loop which does a READ from a single file. I can't make the
hardware deliver the data from disk any faster than anything else
unless I cheat. However It is taking about 72% of the time that Cora
does.
ADP Management> Oh,well if you can't make it twice as fast then we are
not doing it - goodbye.
Dilbert> But don't you see that this benchmark has nothing to do with
your applications and that you have all these interfacing capabilities
in jBASE and so on? Also, you don't seem to understand the arithmetic.
If we take out the physical IO time, which we can't really do anything
about because you are making me do this on Motorola UNIX that doesn't
support memory mapping, then jBASE is about 4 times faster;
Pointey haired boss> No, sorry that sounds like excuses to me - this
number here isn't half that number so it isn't twice as fast!
However, this is an age old story in any large corporation - I don't
blame them particularly. If they had used the skills they had instead
of dreaming about C++ and avoided the internal politics (no chance of
course) then it would have been fine, as it was for everyone that came
afterwards! The internal ADP support group went on to use it for about
6 years or something before being forced to move to an Oracle based
package.
Jim
Thiot
was referred to by the Wicat folks as Wicat Australia whether that was
the company name or not. Lionel as I understand it was the hold of the
Pick license for Wicat and then he had an agreement with the hardware
manufacturer, Wicat to market Pick on Wicat everywhere but Australia.
Gus Giobbi can confirm this but it was Lionel who coordinated Tina
Turner for that Spectrum Show and that was right before her
re-emergence on to the musical scene. Lionel also had "golden
toothpicks" at that show that he referred to as "Pick Gold". Wicat was
the first commercially available computer with a Motorola 68000
processor and beat the market by 8 months. That year they went from $3
mil to $24 mil in sales. Their first O/S was proprietary mainly
oriented towards their authoring applications but they also licensed
Unix and Pick.
If anyone can add more light into all of this please jump in.
Dick
Simon Verona
You are reminding me of my time with ADP.
As I was kind of local to jBASE (120 miles) rather than 8000 miles, I spent
quite a bit of time looking at and working with jBase. Our UK application
was running on Mentor at the time.
We wanted to migrate from Mentor to jBase... We actually did the port, and
proved it worked.
We were about to go ahead with the port to jBase when we were visited by the
world-wide head of the ADP division that we were in.... We were basically
told to stop messing around with a small company like jBase who wouldn't be
able to support us properly, and we should therefore port to ADP Reality and
be supported by the massive team in Portland.
To be fair, the guys in Portland, did their best, but we were smal fry 8000
miles away, compared to the local developers who needed support in the next
office...
We did the port, but ADP Reality was pathetically slow and machine hungry.
We ended up porting to jBase 4 years later, and spending a fortune on free
upgrades to get customers onto a working platform...
I thought it was politics gone mad then, and it was this that made my mind
up that I wanted out of the "big corporation" and be my own boss - no
politics to worry about!!!
Regards
Simon
"Jim Idle" <ji...@temporal-wave.com> wrote in message
news:1134060887.9...@g43g2000cwa.googlegroups.com...
Luke Webber
Lionel also had the good sense to steal John Perrin from AWA. John went
to the US to do the 68K port with the Pick Systems guys.
I remember the Tina Turner show. There was a lot of money sloshing
around in MV in those days. How things have changed.
Luke
csig...@hotmail.com
I don't recall any 68K port that Pick Systems itself did and was
available in the USA.
Did Pick Systems do that port for Wicat specifically, rather than Wicat
doing that themselves like ADDS did here in the USA.
As to the first port on the 68K, I am reasonably sure that was not
true. At the first ever Comdex in Anaheim, the infamous one in the
tents at 100F+ , there was vendor with a very small, for that time, 68K
system running Pick. I can still kind of see in my mind the guy who
run a company, a guy with a PhD, kind of short Italian or Eastern
European looking.
Anyone else able to fill in those details?
Henry Keultjes
Database Scientifics Project http://www.ncolug.org/ppc.htm
Microdyne Company
Mansfield Ohio USA
sdavmor
<rimshot>
--
Cheers,
SDM -- a 21st century schizoid man
Systems Theory internet music project links:
soundclick <www.soundclick.com/systemstheory>
garageband <http://www.garageband.com/artist/systemstheory>
"Soundtracks For Imaginary Movies" CD released Dec 2004
"Codetalkers" CD coming end of 2005
NP: nothing
Bill Cooke
>
> Please reply to hbkeultjes at earthlink dot net
>
> I don't recall any 68K port that Pick Systems itself did and was
> available in the USA.
>
> Did Pick Systems do that port for Wicat specifically, rather than Wicat
> doing that themselves like ADDS did here in the USA.
>
> As to the first port on the 68K, I am reasonably sure that was not
> true. At the first ever Comdex in Anaheim, the infamous one in the
> tents at 100F+ , there was vendor with a very small, for that time, 68K
> system running Pick. I can still kind of see in my mind the guy who
> run a company, a guy with a PhD, kind of short Italian or Eastern
> European looking.
I recall walking into a loading dock in Irvine or Santa Ana, handing
over a check, and carrying the Pertec under my arm back to my car. What
a difference from Microdata or Ultimate! But then I also remember "The
Computer Store" in Santa Monica, the "world's only computer store".
Bill Cooke
sdavmor
> csig...@hotmail.com wrote:
>> Please reply to hbkeultjes at earthlink dot net
>>
>> I don't recall any 68K port that Pick Systems itself did and was
>> available in the USA.
>>
>> Did Pick Systems do that port for Wicat specifically, rather than Wicat
>> doing that themselves like ADDS did here in the USA.
>>
>> As to the first port on the 68K, I am reasonably sure that was not
>> true. At the first ever Comdex in Anaheim, the infamous one in the
>> tents at 100F+ , there was vendor with a very small, for that time, 68K
>> system running Pick. I can still kind of see in my mind the guy who
>> run a company, a guy with a PhD, kind of short Italian or Eastern
>> European looking.
>
> I recall walking into a loading dock in Irvine or Santa Ana, handing
> over a check, and carrying the Pertec under my arm back to my car. What
> a difference from Microdata or Ultimate! But then I also remember "The
> Computer Store" in Santa Monica, the "world's only computer store".
I remember in late 1981 sitting in a Scripps Ranch area coffee shop
with two executives from the company I was with at the time (Monitor
Labs). They wanted my professional opinion about "personal computers".
Was there any future in them, and did I think a personal computer
store might be a good business? I said "yes" and "yes", but I hedged
my bets, and I also told them I thought Apple was the smart way to go.
I did manage to tell them that Atari wouldn't be the winner. Good
guesswork there! They did alright, despite me, eventually being bought
out by Businessland IIRC.
--
Cheers,
SDM -- a 21st century schizoid man
Systems Theory internet music project links:
soundclick <www.soundclick.com/systemstheory>
garageband <http://www.garageband.com/artist/systemstheory>
"Soundtracks For Imaginary Movies" CD released Dec 2004
"Codetalkers" CD coming Spring 2006
NP: The Australian Pink Floyd Show "Live In Liverpool" DVD
Luke Webber
> Please reply to hbkeultjes at earthlink dot net
Why not keep it in CDP?
> I don't recall any 68K port that Pick Systems itself did and was
> available in the USA.
>
> Did Pick Systems do that port for Wicat specifically, rather than Wicat
> doing that themselves like ADDS did here in the USA.
>
> As to the first port on the 68K, I am reasonably sure that was not
> true. At the first ever Comdex in Anaheim, the infamous one in the
> tents at 100F+ , there was vendor with a very small, for that time, 68K
> system running Pick. I can still kind of see in my mind the guy who
> run a company, a guy with a PhD, kind of short Italian or Eastern
> European looking.
Thiot didn't actually say it was the first 68K port, but just that the
Wicat was the first commercially available 68K-based system. Not the
first 68K Pick system. I can't personally speak for either claim,
because those memories have gone with most of my high school maths. It
ran another OS as well, which was used for computer-based training. In
fact, WICAT stood for something very like "World Institute for Computer
Aided Training".
Luke
dawn
Bill Cooke wrote:
> csig...@hotmail.com wrote:
> >
> > Please reply to hbkeultjes at earthlink dot net
> >
> > I don't recall any 68K port that Pick Systems itself did and was
> > available in the USA.
> >
> > Did Pick Systems do that port for Wicat specifically, rather than Wicat
> > doing that themselves like ADDS did here in the USA.
> >
> > As to the first port on the 68K, I am reasonably sure that was not
> > true. At the first ever Comdex in Anaheim, the infamous one in the
> > tents at 100F+ , there was vendor with a very small, for that time, 68K
> > system running Pick. I can still kind of see in my mind the guy who
> > run a company, a guy with a PhD, kind of short Italian or Eastern
> > European looking.
>
> I recall walking into a loading dock in Irvine or Santa Ana, handing
> over a check, and carrying the Pertec under my arm back to my car. What
> a difference from Microdata or Ultimate! But then I also remember "The
> Computer Store" in Santa Monica, the "world's only computer store".
>
computer store? Was the Pick OS loaded on anything? If not, what OS's
were? When did it exist?
Thanks. --dawn
Bill Cooke
http://www.mactech.com/articles/mactech/Vol.02/02.04/Apr86History/
This is a story by Dick Heiser, Founder, The Computer Store, opened in
July 1975.
The Computer Store sold, primarily, Altairs and their ilk, plus parts
and accessories, books, and enthusiasm. It's where I bought an apple.
Definitely no Pick. Actually I was buying time at pick & assoc right
about then.
The Pertec was much later. The connection was simply my experience of
walking in, paying, and walking out literally with a computer under your
arm. It was added into a Pick configuration, doubling the system power
with only a serial connection, and a single shopping experience.
~ ~ Bill
> Thanks. --dawn