Most of these facilities were sold to AST Research (later renamed
AST Computer), but by 1997, none of those facilities existed either,
as AST let their Texas-based employees go and the buildings were sold to
unrelated companies, some of which still stand empty today.
The employees of Tandy R&D that survived the layoffs of January 16th, 1993
either were unemployed on July 1st 1993, or were simply employees of AST.
(They sent you to one of two meeting rooms - either to be let go or to fill
out new insurance and employment paperwork.)
The few Tandy hardware engineers that made the transition to AST were let go
in late 1995, and all but a handfull of the now-AST software development and
test people were let go on June 13th, 1996.
July 1st, 2003 also marks the date when Tandy technically ceased to design or
anufacturing computers, although some designs already in process were
ompleted by the now-AST employees and sold under the Tandy name (the
Sensation II was the main system in this category), although arguably these
machines and any others released in the fall of 1993 were not manufactured
by Tandy Corporation. Tandy continued to buy computers from AST with the
Tandy logo for another two years and then discontinued the use of the Tandy
name on computers entirely.
The first computer produced by Tandy in 1977 was a Z-80 based system that ran
at just under 2.0MHz. The fastest computer Tandy made before ceasing
operations 16 years later in 1993 used a 486 DX2 50MHz processor.
Tandy Research and Development: Born August 1977, Died July 1993.
Cause of death: Microsoft anti-competitive activities (particularly on
the VIS system and against Deskmate) as well as Tandy's own massively
incompetent management.
Survivors: Over 40 computer and peripheral designs, some ground-breaking,
and some textbook examples of failure. Tandy-designed computers were sold
under the Radio Shack, Tandy, GRiD, Panasonic, Digital, AST, Olivetti,
Victor, Memorex, and other names. Several operating systems and dozens of
applications were also designed and produced by the software division of
Tandy R&D.
Frank Durda IV - only this address works:|"I'll huff, and I'll puff, and
<uhclem.aug03%nemesis.lonestar.org> | I'll get promoted."
This Anti-spam address expires Aug. 31st |
http://nemesis.lonestar.org | - Old Tandy Management Saying.
Copr. 2003, ask before reprinting.
Ken Harbit
kr...@cvip.net
> Tandy Research and Development: Born August 1977, Died July 1993.
>
> Cause of death: Microsoft anti-competitive activities (particularly on
> the VIS system and against Deskmate) as well as Tandy's own massively
> incompetent management.
Oh please. How can you possibly blame Microsoft for this? That
Tandy's management was "massively incompetent" is pretty well-known, but
what did MS have to do with things in Texas?
Jim
I was only a teenager at the time, but from what I've read and heard
Microsoft was a thorn in Tandy's side from the beginning. It seems that they
had a "theirs or nothing else" attitude about the software they were
licensing for pretty extraordinary fees. From what I understand alot of
Tandy projects were scrapped when Microsoft got wind of them and came out
with their own version like 'Xenix'. I'm sure I'm not too far off target
here. If Tandy was allowed to be more liberal with software they may have
stuck around a bit longer.
Mo
Microsoft did Many, Many, Many things over the years that lowered the quality
and the very offerings that Tandy sold. (It's unfortunate that no one
paid any attention to what had been going on for over a decade until
Microsoft went after Netscape and the DOJ jumped on Microsoft. Apparently
Andresson? got the right news reporter interested...)
Microsoft was throwing their weight around at Tandy as early as the fall of
1982, threatening that if Tandy went with UNOS instead of XENIX, Microsoft
would not only not do ports of BASIC and Multiplan for UNOS (certainly that
was childish but their choice), but Microsoft also said that they would
"delay" all other work for all other Tandy computers or never deliver at all
as punishment. In particular, no more BASICs, no more Multiplans would be
provided for Tandy computers if Tandy didn't go to XENIX. Tandy switched to
XENIX despite having the UNOS product virtually completed and days away
from QA testing.
Shortly after that, Microsoft hired away the Tandy Vice President of computer
merchandising (chief product picker/buyer), but not before this VP suddenly
canceled work on several in-house projects or those being comissioned from
third parties (including XENIX and some of the Small Computer Company stuff)
and ordered the replacements be purchased from - guess who - Microsoft
instead, whom was his new employer six weeks later.
Once at Microsoft, he promptly used his inner knowledge (and embarassing
photos) of Tandy management to get them to do exactly what Microsoft wanted,
and continued doing it until he retired from Microsoft as a zillionaire,
around 1991. By that time, Tandy was so used to having a ring through its
nose, Tandy didn't really care who at Microsoft was doing the leading from
that point on. At one point we wondered if Bellevue/Redmond needed to
approve bathroom breaks.
Microsoft suckered Tandy into being one of its guniea pigs on their scheme
for OEMing MS-DOS for the 2000, their version of the PC-DOS product sold by
IBM. At this time, Microsoft refused to deliver the IO.SYS layer of MS-DOS,
and the graphics drivers for BASIC, and expected the OEM to write these based
on non-functional "reference code" Microsoft provided, which was written in
an obsolete Sperry mainframe assembly language, completely incompatible with
Intel, and full of logic bugs and omissions.
Microsoft also compounded the 2000 situation early on by insisting that
shiny-new MS-DOS and BIOS compatibility was all you needed, that hardware
compatibility was not important, urging Tandy to deviate and put high-end
but incompatible hardware in the design, the very point that later made the
2000 a textbook failure. After that debacle, none of the other OEMs were
ever put through that torture again and Microsoft provided full MS-DOS and
graphics layers for BASIC. Of course, by this point, Microsoft was no longer
afraid of IBM anymore, so they were no longer worried about IBM going after
Microsoft for selling MS-DOS to IBM clone makers. But Tandy was stuck with
tens of thousands of Model 2000s that were essentially doomed. Tandy got
burned by Microsoft, but had trouble remembering the lesson because of market
demand that forced them to go back to Microsoft for more.
It got worse in the late 1980s and early 1990s, with Microsoft routinely
threatening to revoke Tandys OEM rights to MS-DOS or delaying the delivery to
Tandy of fixes and new versions unless Tandy stopped carrying some competiting
spread sheet or word processing program (or what have you) and sold only
Excel and Word (or some other Microsoft thing). The same tactic eventually
got Deskmate removed and Windows put in its place. Scripsit for the generic
PC was released by Tandy and then was quietly removed from the market,
shortly after Microsoft complained about Tandy competiting against Word.
Microsoft even fought to get Tandy to stop using the Logitech mouse driver
and instead buy Microsofts mouse driver that cost three times as much.
In that day, we paid $15 per machine for MS-DOS and $3 for the Logitech mouse
driver, or $9 for the Microsoft mouse driver, regardless of quantity. That
money is just for the mouse driver, not the actual mouse hardware.
In fact, the $75 million disaster of late 1992 that directly led to the end
of Tandy being in the computer biz was a system called VIS, which Microsoft
was the OS partner on for 18 months. But in the months before delivery, a
growing dispute between the two companies over how the product would be
marketed and what software would be allowed to run on it boiled over.
The key points of the dispute were: (1) Microsoft wanted the Microsoft $9
mouse driver used so they would get the money - Tandy wanted the Logitech
driver embedded to save cost. (2) Microsoft wanted only Modular Windows
applications to be allowed to run on the system - Tandy wanted DOS apps
to run as well, mainly because there were more of them available, and only
a handful of Modular Windows apps that didn't stink. (3) Microsoft suddenly
wanted the machine to be branded with a Modular Windows logo (full screen
logo on boot-up which stayed there for 10 to 30 seconds while Windows
booted) and on the case - Tandy insisted on sticking with the VIS logo
(which displayed until the App or Modular Windows started loading), a point
that had been previously agreed between Roach and Gates. (The cases were
already being made when this latter issue popped-up, so already had the VIS
logo on them.)
There were perhaps a half-dozen smaller disputes like Tandy wanting to fix
Modular Windows to not read the SYSTEM.INI file off the Compact Disc 150
separate times on each boot (that made Modular Windows really slow) and
reading WIN.INI another 70+ times, but Microsoft insisting that this
re-reading was necessary or that fixing it would be hard, or Microsoft going
many many K over their ROM budget, forcing the removal of Tandy-provided
features and a constant re-packing of the Tandy parts of the ROM. That's
the main reasons that the VIS boot-up screen is so simple. Modular Windows
bloated up all the ROM space with useless code, including code to operate
peripherals that can't exist on the platform, like that Network printer
driver we found in there...
The Modular Windows logo in the code that appeared late in the game was
probably the key thing that caused the split, with Microsoft sneaking the
logo into the ROM at the release candidate phase of the project, getting
yelled at to remove it, which they claimed they did, but the next object
image, the new MS code was 20 odd bytes *larger* than the previous one with
the logo in it, and 20K larger than the the one before that which definitely
didn't have the logo. A disassembly of the latest Windows start-up code
showed that Microsoft had hidden the logo but it was still there, and coded
Modular Windows so that it displayed the logo if you appended an exclamation
point to the executable name you wanted to run, something Microsoft was
probably going to direct the third-party software writers to do the moment
Tandys back was turned.
Apart from the argument over re-branding of the machine, the logo was blowing
the ROM budget entirely (20K over when linked together) and the extra logo
simply had to go. We had a 1Meg ROM and Microsoft was using over 800K of
it for Modular Windows, well over the allocated budget, and leaving no
room for the BIOS, video, audio, keyboard/mouse/controller and CD-ROM drivers
that Tandy provided, and MS-DOS that had to be in there too.
This eventually got to John Roach calling Bill Gates personally and demanding
that Microsoft stick to the original agreement between the two from 18 months
earlier on what the box would be and how it would be branded. A new object
then appeared which I went over with a fine comb to make sure they hadn't
hidden the graphic somewhere else, and when combined with the Tandy code,
that was the version that went out. (I still have images of the offending
versions.)
Just 48 hours after the paperwork was signed ordering 30,000 ROMs with
that code in them, Microsoft slightly altered the Modular Windows OS and
shipped that altered version in their official OEM SDKs, so that the Tandy
machine would not be compatible with anything else using that Microsoft OS
(would not even start anymore). Then Microsoft warned all the software
developers not to develop anything for the VIS platform. A few months
later in front of the press, Microsoft even denied that they had ever
released such an OS, saying "Modular Windows is a concept, not a product."
All the final release of OEM SDKs Microsoft were now incompatible with the
VIS system. Microsoft and probably Bill himself, had made the decision to
torpedo the VIS platform, and began marketing the Modular Windows code
to Sony and Toshiba, trying to get them to make a virtually identical box to
the VIS (but slightly faster and with a bit more ROM/RAM), because Microsoft
still wanted in that market segment, but they wanted it on under the
"Modular Windows Compatible" brand, not "VIS" or any other brand.
Fortunately, both Sony and Toshiba declined to get involved, but it must have
sparked some interest in the set-top game box market segment at Sony as
they went on to develop their own set-top game system.
We found out about Microsofts breaking of Modular Windows compatibility when
they fed-exed the galleys of the SDK manuals and final copies of the discs
being sent for pressing. Immediately the execs at Tandy were calling all
over Microsoft to demand that this stop, but Microsoft simply responded that
this was all part of the evolution of Modular Windows and it was Tandys fault
for not following Microsofts guidelines? Huh? Microsoft arbitrarily changed
the command the software developer puts on the Compact Disc to run Modular
Windows and that somehow is Tandys fault? No, it's monopolistic (or
very childish) behavior once again.
Tandy was trapped and holding the investment bag. It was now too late to
stop the ROM fabrication and still have the units in the stores for
Christmas, so we had to go forward and try to convince the 150 or so
third-party software OEMs not to jump ship, while Microsoft was warning these
same companies to *not* write to the VIS platform and its incompatible
Modular Windows platform (which Microsoft made incompatible). We could have
patched the code we had to match the slight change Microsoft had done, but
the call was made to not miss Christmas and go with what we had.
Thanks to this very successful sabotage by Microsoft (and the fact that the
machine itself was marginal at best), Tandy ended up being stuck with 50,000
of these "concepts" in their stores, warehouses and factories. Zenith was
also going to sell the Tandy units under their name, but after a then-secret
meeting with Microsoft reps, Zenith pulled out of the deal, leaving Tandy
with even more parts than what would never be needed. Tandy pressed on
alone, quickly selling the units at $40 below manufacturing cost and
gradually bundling more and more third-party software at no extra cost, but
with Microsoft actively lobbying and threatening other software developers
to avoid writing software for the now-non-Modular Windows compatible
platform, the VIS Systems fate was sealed. Some of the most promising
titles abandoned their work for the platform under the pressure from
Microsoft. Tandy ended up writing off $75 million on the project,
including development, marketing and inventory and payments to third-party
software developers for guaranteeing minimum purchases of their titles
by the Radio Shack stores.
Yes, you can't blame 100% of the VIS failure on Microsoft. Certain members
of Tandy management pushed the thing through despite virtually every
non-management engineer saying that it was a plug-in disaster at least
nine months before release, while these determined management types went as
far as repeatedly lying to the Tandy CEO (who, sadly, could not spot
deception if it was painted purple and dancing on a piano), telling the CEO
over and over that the VIS was a "Sega killer".
The VIS could only kill a Sega if you dropped the VIS on the Sega from a
great height and the Sega might still work.
However, Microsofts sabotage of the third-party software titles for the
VIS platform and finally their own denial of participation in the project
was a devestating blow to increasing the number of good software titles,
and that rapidly caused the VIS systems failure.
By the way, Modular Windows reappeared two years later with a brand new
name but the same clunky code: "Windows CE". At least they changed the
branding logo. The original one looked like a rip-off of the "Wool" logo,
but in red, green and blue.
Suffice it to say that Microsoft was indeed guilty of all this and far more.
Frank Durda IV - only this address works:|"I tell you again! Windows has no
<uhclem.aug03%nemesis.lonestar.org> | bugs! There is no proof that any
This Anti-spam address expires Aug. 31st | any one said that we are a monopoly!
http://nemesis.lonestar.org | - Spokesman for Microsoft's
Copr. 2003, ask before reprinting. | Ministry of Information
We even had a UNOS system to start writing Profile 16 on, and were then
told they were going with Xenix instead. As I recall, there were even
little scraps of UNOS left over on Tandy's Xenix. The software that read
TRS-DOS disks and copied files to Xenix was called something like "tx"
for TRSDOS-to-Xenix. (I'm sure someone will give the correct name if I'm
wrong.) Apparently, this was originally called "tu" for TRSDOS-to-UNOS.
In reading the comments in the script, it looked like someone had made
the "tu" into "tx" by simply substituting "x" for "u". (Or maybe a
global search-and-replace of "tu" to "tx.) This had the side effect of
changing other words in the comments as well, making it obvious (to
anyone who looked at it) what had happened.
> Shortly after that, Microsoft hired away the Tandy Vice President of computer
> merchandising (chief product picker/buyer), but not before this VP suddenly
> canceled work on several in-house projects or those being comissioned from
> third parties (including XENIX and some of the Small Computer Company stuff)
> and ordered the replacements be purchased from - guess who - Microsoft
> instead, whom was his new employer six weeks later.
First time I've heard SCC mentioned in this context. Any chance that you
can elaborate on that part?
[...]
--
+---------+----------------------------------+-----------------------------+
| Kenneth | kenbrody at spamcop.net | "The opinions expressed |
| J. | http://www.hvcomputer.com | herein are not necessarily |
| Brody | http://www.fptech.com | those of fP Technologies." |
+---------+----------------------------------+-----------------------------+
> I still use Deskmate, on Windows98 ... I still think RS made the best
> computers, my 1000 fell off a desk when a desk leg broke. We picked it
> up, checked it out, it still worked fine.
Hell, we used to demonstrate the durability of Mod Ones by unplugging them
from the monitor, tape deck and power supply (later, EI as well), throwing
the keyboard/CPU into the far corner of the room in a bank shot, then
plugging them back in and booting them up. Lost a left shift keycap once
in about 200 throws.
--
Ward Griffiths wdg...@comcast.net
Guess what they said was true, flying a spaceship was like having sex
on a bicycle: a difficult skill to learn and not easily forgotten.
Jack Hopkins, _Satellite Night News_
Xenix was a Microsoft product. :)
Barry
"ward" <wdg...@comcast.net> wrote in message
news:G4-dnX3Os-t...@comcast.com...
> Ken Harbit wrote:
>
> > I still use Deskmate, on Windows98 ...
The last version, 3.5, I suppose ?
> > I still think RS made the best
> > computers, my 1000 fell off a desk when a desk leg broke. We picked it
> > up, checked it out, it still worked fine.
>
> Hell, we used to demonstrate the durability of Mod Ones by unplugging them
> from the monitor, tape deck and power supply (later, EI as well), throwing
> the keyboard/CPU into the far corner of the room in a bank shot,
I once had my model 4P set up onto a very narrow table. It was - for the
extra tilt - put ON the lid/cover. Suddenly, it slid off the lid and dropped
to the floor. I hung onto the keyboard! And my 4P was there, dangling
halfway towards the floor. I held on in pure panic!
Where do you find a computer that you can suspend by its keyboard ? My model
4P did!
> Ward Griffiths wdg...@comcast.net
Greetings from the TyRannoSaurus
Jan-80
Yes and no. Microsoft couldn't make up their minds whether they wanted to
rule the multi-user world or not and constantly wavered on whether they wanted
to do XENIX or not. Tandy took over their own development on V7-derived
XENIX when Microsoft got bored with it and shifted future efforts to SCO.
SCO delivered the System III 68000 base (a port also meant for the Lisa)
that Tandy eventually also had to take over and maintain on our own, because
the 68000 platform fell out of fashion at SCO. It was already out at
Microsoft.
When someone figured out how to do System III XENIX on a 286, control of XENIX
shifted back to Microsoft, cutting SCO largely out of the picture for a while.
(a pity, as SCO was a lot easier to work with than Microsoft, who met OEM
release deadlines by shipping blank diskettes and blaming the unreadability
of the media on Fed-Ex). Control stayed at Microsoft until spring 1986 or so
when System III XENIX 286 was shipping at IBM and Tandy was trying to beat
IBM by getting System V XENIX 286 out first on the 3000, then Microsoft
suddenly got bored again with the XENIX world but this time would not turn
the source over to anybody, letting the 286 tree twist in the wind. SCO
eventually got some access to the 286 work, but the bulk of 286 activity was
over by then.
A year or more later, SCO brought out the System V for the 386 and by that
time, Microsoft had disappeared from the picture for XENIX and stayed gone.
The only remaining specter of Microsoft gumming up the XENIX world was the
10% stake in SCO they bought and may still hold. However, Microsoft was
now pushing Tandy to dump SCO XENIX entirely - now that it was no longer a
Microsoft product - and switch to this Windows thing they were working on.
To the rescue came one of the Tandy bonehead computer merchies, who agreed,
and killed the Tandy XENIX development group in early 1988, laying some
programmers off, and sending the remaining people to work on Deskmate
DOS-based applications.
Also, to be really blunt, when XENIX was really coming from Microsoft, it was
little more than a port of whatever AT&T released, plus as many non-network
parts of UCB BSD UNIX that they could cram in. (As it came from Microsoft,
the finger command would happily tell you which dorm at the UCB campus
your telephone number was in, as Microsoft forgot to rip this code out before
they put their copyright notice in place of UCBs.)
In XENIX, Microsofts real contribution was marketing (something AT&T was not
allowed to do at that point and would have done badly - just look at the
license costs), and the processor specific parts of the OS for the 68000
and 286 that Microsoft couldn't just steal elsewhere (before 1986 or so,
AT&T did not ship UNIX ready to run on any Intel or Motorola processor).
The platform dependent stuff (like drivers and boot stubs) were typically
done by the OEM, not Microsoft or SCO.
Tandy ended-up having to trash virtually all of the Microsoft-provided
hardware drivers for 68000 and for 286 XENIX (which only came with a couple
that worked) and then replace them with ones written at Tandy. Microsoft
simply did not seem to understand hardware, not even in the PC-compatible
era, or Microsoft would not address the flaws in their drivers and not
provide the source code to allow them to be fixed by the OEMs, so the Microsoft
drivers were routinely dumped and re-written.
Frank Durda IV - only this address works:|"We see the potential opportunity
<uhclem.aug03%nemesis.lonestar.org> | of small business. At Microsoft,
This Anti-spam address expires Aug. 31st | we want to encourage small business,
http://nemesis.lonestar.org | so that later we can suck them dry."
> Tandy ended-up having to trash virtually all of the
Microsoft-provided
> hardware drivers for 68000 and for 286 XENIX (which only came with
a couple
> that worked) and then replace them with ones written at Tandy.
Microsoft
> simply did not seem to understand hardware, not even in the
PC-compatible
> era, or Microsoft would not address the flaws in their drivers and
not
> provide the source code to allow them to be fixed by the OEMs, so
the Microsoft
> drivers were routinely dumped and re-written.
Let us know when your book comes out - I'll buy a copy! Perhaps
Amazon will pair it with Tracy Kidder's book, "The Soul of a New
Machine".
"Those who like the law or sausages should never watch either being
made"!
Neil
I bet that straightened the cord out!
Mo
Being from Santa Cruz originally myself, I knew several programmers for SCO and
one of them claimed to be an original designer of the XENIX system for SCO.
Maybe he was just full of John Bull or something, I don't know.
If anyone could shine a bit more light on the subject for me it'd be great.
Thanks!
Louis
AFAIK, Microsoft ported Unix to a few different machines and called
it Xenix, as you couldn't use the name Unix at the time. Eventually
they lost interest and sold it to SCO, who first sold Xenix and later
licensed and used the name Unix.
Neil
--
+----------------------------------------------------------------+
| Charles and Francis Richmond richmond at plano dot net |
+----------------------------------------------------------------+
Extremely unlikely. Any code Microsoft generated during its period
of interest in XENIX would have been washed away by the time SCO
went to a fresh System V AT&T tree, AND when they definitely threw away
the Microsoft swapping implementation code in favor of real paging code
(likely hardly touched from the SYSV base) when the 386 versions appeared.
Microsoft was already out of the XENIX development picture by then.
Perhaps some MS port work remains in the odd application or in the
compiler in the last versions of XENIX, but I doubt it, and further
cleansing would have occurred when they went to UNIX/OpenDeskTop(X) as a
base.
Frank Durda IV - only this address works:|"The Knights who say "LETNi"
<uhclem.aug03%nemesis.lonestar.org> | demand... A SEGMENT REGISTER!!!"
This Anti-spam address expires Aug. 31st |"A what?"
http://nemesis.lonestar.org |"LETNi! LETNi! LETNi!" - 1983
>Charles Richmond <rich...@ev1.net> wrote:
>: There
>: was also an agreement that a certain amount of Mi$uck code
>: would stay in Xenix and Mi$uck would get some royalty for it.
>Extremely unlikely. Any code Microsoft generated during its period
>of interest in XENIX would have been washed away by the time SCO
>went to a fresh System V AT&T tree [...]
I inferred that they are deeming some percentage for royalty purposes
even if the actual quantity of M$ code is zero.
Nick.
--
http://www.nick-andrew.net/ http://aus.news-admin.org/
Do not send me email copies of postings. Keep it in USENET please.
> Neil wrote:
> >
> > " [snip...] [snip...] [snip...]
> >
> > AFAIK, Microsoft ported Unix to a few different machines and called
> > it Xenix, as you couldn't use the name Unix at the time. Eventually
> > they lost interest and sold it to SCO, who first sold Xenix and later
> > licensed and used the name Unix.
> >
> IIRC, Mi$uck owned as much as 20% of SCO at one time. There
> was also an agreement that a certain amount of Mi$uck code
> would stay in Xenix and Mi$uck would get some royalty for it.
Mi$uck? Mi$uck?
Grow up already.
Jim
"Jim" <nu...@business.net> wrote
> Charles Richmond <rich...@ev1.net> wrote:
> > IIRC, Mi$uck owned as much as 20% of SCO at one time.
>
> Mi$uck? Mi$uck?
Small&limp, then ?
> Grow up already.
>
> Jim
Bill Gates to Divine Brown (of Hugh Grant fame), "I know why they
call you Divine".
Divine Brown to Bill Gates, "I know why you called it MicroSoft"!
N
Would I be too anoying if I requested a list of Tandy designed
products sold under other names?
I'll have to try that some time.
> Suddenly, it slid off the lid and dropped to the floor.
Or maybe not...
Probably no-one knows now. I used to suspect that when some far East
factory came out with a loser product, they always figured, "Hey, we
can always sell them to Tandy!" Now they have to sell them to RCA/GE.
Neil
I doubt a complete list exists simply because of the number of models,
and what lists I have are the internal project numbers, which don't
mean much on the retail side of things.
When Tandy was building computers for DEC (1988-1992), there was a period
when they had a custom-build program. Order one or more in a given config,
and Tandy would have it built and delivered in 72 hours. DEC never actually
touched these computers. It was assembled in Fort Worth and shipped directly
to the customers address. There were several thousand possible system
build combinations alone in that program alone.
Further assume Tandy produced four to six stock designs per year for
Digital (in several variations for different countries), and at least that
number sold under the GRiD line for desktops, plus anywhere from three to
eight portable GRiD models each year. Oh, and some sold under the Tandy
name at Radio Shack stores as well. Victor probably had two or three per
year. Olivetti and Panasonic were brief partners, and probably had
two or three models at most.
By the time the Tandy computer factory became ASTs, it wasn't surprising to
see forty different system builds coming off the line over a six month
period, with most sold only to Radio Shack, or Best Buy, Computer City,
Circuit City, etc. This was done in order to protect each store from the
others guaranteed lowest price policy, buy never building the exact same
configuration for two store chains. The store could always argue that
their computer came with a slightly different amount of RAM or a different
FAX software package, or brand of printer or speakers, anything to make
them not the same. Sneaky. It's still done today by the remaining PC
makers. Unfortunately, it's also exactly what resulted in an enormous
inventory that AST was unable to dispose of, and put the company into
decline.
Frank Durda IV - only this address works:|"I picked up a Magic 8-Ball the
<uhclem.aug03%nemesis.lonestar.org> | other day and it said 'Outlook
This Anti-spam address expires Aug. 31st | not so good'. I said 'Sure,
http://nemesis.lonestar.org | but Microsoft still ships it.'"
> ...
> By the time the Tandy computer factory became ASTs, it wasn't
surprising to
> see forty different system builds coming off the line over a six
month
> period, with most sold only to Radio Shack, or Best Buy, Computer
City,
> Circuit City, etc. This was done in order to protect each store
from the
> others guaranteed lowest price policy, buy never building the exact
same
> configuration for two store chains. The store could always argue
that
> their computer came with a slightly different amount of RAM or a
different
> FAX software package, or brand of printer or speakers, anything to
make
> them not the same. Sneaky. It's still done today by the
remaining PC
> makers....
And many others like mattress makers. They offer lowest price
guarantees, but you can't compare because of different make and model
names, although in fact the mattresses may be the same.
Neil
Not a problem if the very narrow table is close enough to the wall and
properly anchored if it weighs less than 26 lbs.
--
Ward Griffiths wdg...@comcast.net
What also irks me are vendor's 'technical' people (whose qualifications
would be almost sufficient get them an extra role in a remake of
"Deliverance") who are at the conference to give you the 'expert opinion
straight from the horse's mouth'. True, there's a part of the horse's
anatomy involved, but it's not the mouth...
The BOFH (Simon Travaglia)
At least find a publisher writing a book on how Microsoft has been
anti-competetive and include your chapter.
Call Andersson (?) at Sun - I'm sure he'd fund it.
-d
You've replied to a thread that dates from July of 2003.
Just because google has now made the error of allowing replies to old
threads, in an archive that goes back a quarter century, is not reason
to reply to old messages. The conversation has long moved on, and in
most cases the people posting aren't around either. google is an archive
of Usenet messages, but Usenet is not google. Only people who read
the newsgroup at google will see it in the form of a webpage, and
see the previous messages on that page.
To make matters even worse, you didn't even bother to quote the message
you were replying to.
Michael
Well you have to admit, that without quoting it was confusing.. my
first thought was also 'wtf'...
We are getting this problem now in other groups.. people replying from
posts 5 years ago.. Long after the people are since gone, and not
quoting enough of the orginal for anyone to make sence out of it..
Just more bandwidth wasting.. increasing the s/n ratios..
One good thing is that AOL is stopping support for Internet News. So most
lamers will soon be gone all by themselves.
PeterV
"Mike" <a...@def.ghi> wrote
> > > Frank,
> > > Definately should write a book.
Well, he's been asked several times already.
> > > At least find a publisher writing a book on how Microsoft has been
> > > anti-competetive and include your chapter.
> > > Call Andersson (?) at Sun - I'm sure he'd fund it.
Yep, and he'll be sued by BG for slander. Even if it is the truth, there's
no proof anymore.
> > > -d
> >
> > You've replied to a thread that dates from July of 2003.
>
> I wasn't aware that (1) you were the Usenet Thread Police, or that (2)
> there is a time limit to replies.
True. The only thing I consider when replying to a message is if it's
on-topic. This one is. No problem.
Furthermore, I was there in July 2003, and a lot of people that are on the
list now, are still here too.
So, MB, I don't see a problem.
> Get over yourself.
>
> Mike
>One good thing is that AOL is stopping support for Internet News. So most
>lamers will soon be gone all by themselves.
Except that AOLS "suggestion" to those AOL users who want usenet access is
to use....guess what?.....<drum roll>.....Google ;-)
Dave
--
xtals...@yahoo.com is a valid reply-to address but I don't check it every day.
> By the way, Modular Windows reappeared two years later with a brand
> new name but the same clunky code: "Windows CE". At least they
> changed the branding logo. The original one looked like a rip-off of
> the "Wool" logo, but in red, green and blue.
>
As best as I recall, in the same timeframe that Tandy was playing with
"Modular Windows", HP also had a pre-CE ROM-based Windows on the HP300
Omnibook.
Any relation between the two projects or just coincidence ? I don't recall
ever hearing the term "modular windows" vis-a-vis the HP, but was that just
marketing terminology ?
- FM -
[1]Fred Mau <fred-d...@comcast.net> wrote, following up two years later:
[1]As best as I recall, in the same timeframe that Tandy was playing with
[1]"Modular Windows", HP also had a pre-CE ROM-based Windows on the HP300
[1]Omnibook.
[1]Any relation between the two projects or just coincidence ?
Modular Windows was pulled as product from the official Microsoft history
no later than Q1 1993 (although they started making the "It's not a product,
it's a concept" statements from Microsoft in Q4 1992), so you would have to
be more specific about when the HP product existed. (Tandy had tens of
thousands of those "concepts" sitting in stores and warehouses, along with
hundreds of thousands of software packages for that platform.)
Note that there were some earlier attempts to put pre-Windows 3.x Windows
into ROM, some done by OEMs for specific devices. Perhaps this is what this
HP device had. Even Deskmate and other early GUI systems made it into
ROMs.
In 1991/1992, Microsofts vision for Modular Windows was to put it into every
possible consumer device, including set-top game boxes, cable (and later,
satellite) TV boxes, kitchen and laundry appliances, automobiles, even
vacuum cleaners (now that last one would be fitting). At Tandy, we all
laughed at Microsofts vision of a Modular Windows-equipped car running
everything including the engine, a car that would not function for two
minutes after you turned the key because Modular Windows was still doing
some mysterious and unexplained initialization, like checking to see if the
network printer was online and had paper. Plus each Modular windows device
was expected to have a 640x480x256 display that could add hundreds of
dollars of cost to the device.
As the VIS project approached completion, Microsoft realized that their
minimum system hardware platform for Modular Windows made its bloat far too
slow for consumer acceptance for just about any application (although the
Modular Windows Tombstone product idea that could show you information on
the deceased was probably still practical since the dead would probably
be willing to wait), so Microsoft bumped the minimum from 12MHz 286 to
16MHz 386SX and doubled the RAM and ROM requirements to 2Meg each, which
effectively doubled the hardware cost for the processor and associated
components at that point in time, making Modular Windows even less
attractive.
Microsoft made these changes based on seeing how their creature crawled on
the actual speed hardware they were specifying (the VIS samples built to
their original specs, but actually about 33% faster), not just what
they saw on the fastest-available PC computers that their developers were
using at their desks (a common mistake in the programming industry).
However, Microsofts raising of the minimum requirements for Modular Windows
came in August/September of 1992, far too late to change the initial VIS
hardware platform. Some of the gate array semiconductors were already in
fabrication at that point and all the parts for the machines were already
arriving.
The never-built VIS-2 was to meet these improved specs if the price
problems could have been solved, but computations preducted it would have
still taken two or more minutes for Kings Quest V to reach the question
"Have you played this game before?", unless something was done about the
Modular Windows bloat, something Microsoft consistently refused to do
anything about.
It was bad. On the original Modular Windows platform, it took five full
minutes of staring at the Windows hourglass before this very first question
appeared. People had time to go to the store and buy a Sega and come back
before some Modular Windows games could even start. Microsoft blamed the
slowness on the applications' software developers ineptidtue (even though
their own built-in audio CD-player brought a new meaning to the word
"sluggish"), while the vendors complained that Microsoft claimed that they
would not have to change their code at all "and it would run great",
which is why these vendors agreed to do VIS versions in the first place.
However, the claim of no-change code was proven false months earlier,
when vendors found that they had to contend with the hand controls, very
limited RAM and interlaced video that made it virtually impossible to take
any existing Windows application and run it "as is" on Modular Windows.
Many DOS based games ran great on the VIS with some only needing small
shim drivers to translate the VIS hand controls into looking like a more
common type, something that infuriated Microsoft, who wanted Tandy to ship
only Windows games, and no DOS applications at all. The Sherlock Holmes
games were displaying video within 15 seconds, while Modular Windows had
barely managed to get the hourglass on the screen in that amount of time,
and was nowhere near even finding out what application was to be run.
[1]I don't
[1]recall ever hearing the term "modular windows" vis-a-vis the HP, but
[1]was that just marketing terminology ?
Marketing *is* Microsoft. Without marketing, their products do not exist
or certainly fail to reach the public. Modular Windows was planned and
approved at all levels of Microsoft. It had SDKs, OAKs, manuals, official
product logo (sneaked into the product at the last minute, blowing the ROM
budget completely), forums and presentations were provided to developers to
teach them its requirements and constraints, etc. With the exception of
the Gryphon/VIS materials produced by myself and others at Tandy, all the
Haiku/Modular Windows materials came from Microsoft, their building 6 if
I recall correctly.
I still have the Modular Windows pressed CD-ROMs with four-color graphics,
the white cover manuals, with four-color cover printing (that was standard
for all Microsoft products of the day). The manuals for Modular Windows
showed the Modular Windows logo, complete with Microsoft Corp. assembly and
part numbers, all shipped in a nice white cardboard box with black and white
Modular Windows logo on the outside sticker, and inside the materials came
complete with the Modular Windows shrink-wrap license agreement. Hundreds
of copies were shipped to the software developers working on the VIS project.
In theory the Modular Windows SDK could be purchased by anyone. Tandy
would sell the Gryphon/VIS hardware specific documentation and even make
available the emulation platform to software vendors interested in writing
software for the VIS.
Modular Windows definitely existed. It certainly killed a full year of
my life, along with the time and efforts of many other hardware and
software people at Tandy, and its failure was probably the biggest of
the three (or four, depending on how you count) events that led to the
destruction of Tandy as a maker of computers.
Like how our government leadership operates today, those who caused the
disaster by their poor decisions, lack of decisions, or outright lied
to corporate leadership about what was happening were not fired (some were
rewarded), while many who had no control over what was done were sacked.
I will have to photograph the final release of the Modular Windows products
from Microsoft and put them up on the web site.
Frank Durda IV - only this address works:|"The Pentagon announced today that
<uhclemLOSE.sep05%nemesis.lonestar.org> | US forces will invade and occupy
You must remove the "LOSE" to mail me. | West Africa. A spokesman said that
http://nemesis.lonestar.org | the military will fight hurricanes
Copr. 2005, ask before reprinting. | there before they can get here."