Hi,
I have an old Quadra and would like try MacIvory incarnation of the
Lisp Machine. Who is still selling this sort of thing? What software
is included?
Thanks
I bought mine directly from Symbolics in 2003. It came with the Genera
8.3 distribution, the Symbolics layered products for Genera 8.3
including Concordia 3.3 (Documentation creation), Joshua 2.4 (Expert
system) and DNA 4.2. There was also CLIM 2.1 (portable Common Lisp
graphics system) and Statice (Persistant object database) Development 2.3.
You may have seen it already, but I have more information and pictures at:
http://home.hakuhale.net/rbc/symbolics/
Best regards,
--Bruce
Fernando Rodriguez wrote:
> Hi,
>
> I have an old Quadra and would like try MacIvory incarnation of the
> Lisp Machine. Who is still selling this sort of thing? What software
> is included?
>
> Thanks
Fernando Rodriguez <···@THOU_SHALL_NOT_SPAMeasyjob.net> writes:
> I have an old Quadra and would like try MacIvory incarnation of the
> Lisp Machine. Who is still selling this sort of thing? What software
> is included?
You may check this eBay item:
http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=5210654256&ssPageName=ADME:B:SS:UK:1
Paolo
--
Why Lisp? http://lisp.tech.coop/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools:
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- UFFI: Foreign Function Interface
In article <··············@plato.moon.paoloamoroso.it>,
Paolo Amoroso <·······@mclink.it> wrote:
> Fernando Rodriguez <···@THOU_SHALL_NOT_SPAMeasyjob.net> writes:
>
> > I have an old Quadra and would like try MacIvory incarnation of the
> > Lisp Machine. Who is still selling this sort of thing? What software
> > is included?
>
> You may check this eBay item:
>
> http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=5210654256&ssPageName=ADME:B:SS:UK:1
>
>
> Paolo
It would be good to know that the board works BEFORE buying it.
It would also be a plus if the internal harddisk boots the system
and has the software installed. Getting an original Apple CDROM drive
might be useful, too.
Full set of printed manuals and the original CDs are a plus.
The near-mythical "Rainer Joswig" <······@lisp.de> wrote:
> It would be good to know that the board works BEFORE buying it.
> It would also be a plus if the internal harddisk boots the system
> and has the software installed. Getting an original Apple CDROM drive
> might be useful, too.
>
> Full set of printed manuals and the original CDs are a plus.
Agreed. The sheer rarity of this item (especially on eBay) means that some
fairly intense bidding will occur, so I think a guarantee should be provided
in anticipation of the (relatively) high selling price. That said, wouldn't
you have to pay around ?1500 for such a unit from Symbolics? The cost
saving seems proportional to the increased risk.
It also seems that the seller would at least have to procure a DB15 Apple
monitor and an ADB mouse and keyboard to test it properly.
Regards,
Robbie Blake-Coleman
"Robbie Blake-Coleman" <········@xinis-software.demon.co.uk> wrote in
message ··························@news.demon.co.uk...
>> Full set of printed manuals and the original CDs are a plus.
>
> Agreed. The sheer rarity of this item (especially on eBay) means that
> some fairly intense bidding will occur, so I think a guarantee should be
> provided in anticipation of the (relatively) high selling price. That
> said, wouldn't you have to pay around ?1500 for such a unit from
> Symbolics? The cost saving seems proportional to the increased risk.
>
> It also seems that the seller would at least have to procure a DB15 Apple
> monitor and an ADB mouse and keyboard to test it properly.
>
> Regards,
> Robbie Blake-Coleman
>
Symbolics only sells Macivory Model 3 systems(Quadra and Macivory)
for twice that price with 4 times the memory.
I wonder how many people are hording Macivorys?
Just letting them sit gathering dust!
I bid on a working UX400 and broken XL1201 with the intent of putting the
UX400 on
the internet for other to explore but, I did not bid high enough.
At the same time I had a straddling bid on a Macivory Model 3.
My bid for the UX400 and broken XL1201 would have been higher but, I was
afraid I would
end up winning both Symbolics systems at the time.
This fear caused me to bid low on both and, lost both auctions.
"GLS" <·@G.com> wrote in message
·························@fe74.usenetserver.com...
> Symbolics only sells Macivory Model 3 systems(Quadra and Macivory)
> for twice that price with 4 times the memory.
>
> I wonder how many people are hording Macivorys?
> Just letting them sit gathering dust!
>
> I bid on a working UX400 and broken XL1201 with the intent of putting the
> UX400 on
> the internet for other to explore but, I did not bid high enough.
> At the same time I had a straddling bid on a Macivory Model 3.
> My bid for the UX400 and broken XL1201 would have been higher but, I was
> afraid I would
> end up winning both Symbolics systems at the time.
> This fear caused me to bid low on both and, lost both auctions.
Jadersch is a Symbolics HOG!
Jadersch already has XL1201 and a UX400.
What's Jadersch going to do start a Symbolic museum in Minnesota?
I'm just sore I won't spend the money to win the auction.
$510USD for the XL1201 and UX400 and, possible a working Macivory for
about $690USD including shipping by German post. I'd spend the extra $120USD
and ship it Fedex 25Kg package world expedite and have it in three days.
> Jadersch is a Symbolics HOG!
>
> Jadersch already has XL1201 and a UX400.
>
> What's Jadersch going to do start a Symbolic museum in Minnesota?
>
> I'm just sore I won't spend the money to win the auction.
>
> $510USD for the XL1201 and UX400 and, possible a working Macivory for
> about $690USD including shipping by German post. I'd spend the extra
> $120USD and ship it Fedex 25Kg package world expedite and have it in three
> days.
Looks like we've eaten the dust of the same individual!
Robbie Blake-Coleman wrote:
>>Jadersch is a Symbolics HOG!
>>
>>Jadersch already has XL1201 and a UX400.
>>
>>What's Jadersch going to do start a Symbolic museum in Minnesota?
>>
>>I'm just sore I won't spend the money to win the auction.
>>
>>$510USD for the XL1201 and UX400 and, possible a working Macivory for
>>about $690USD including shipping by German post. I'd spend the extra
>>$120USD and ship it Fedex 25Kg package world expedite and have it in three
>>days.
>
>
> Looks like we've eaten the dust of the same individual!
>
>
Sorry I outbid you guys, I hope there are no hard feelings. I just bid
the way I always do -- my max (which wasn't much more than the winning
bid) at the last possible second. Though being an AI construct and all,
I've gotten to be fairly precise at these sorts of things.
I just want to get a LispM running that has as few irrelplacable
points-of-failure as possible. Couldn't pass up the chance to get a
MacIvory for a price that wasn't $2000.
Just to combat the "Symbolics Hog" label, I'd like to mention that I'll
most likely be putting up the previously-mentiond XL1201 for sale
assuming the MacIvory works -- it's a wonderful machine (and I spent a
fair bit getting it running -- it was 100% dead when I got it) but I
won't be using it much once I get the MacIvory going... and LispM's are
too cool to keep to myself :). I may sell the UX400 as well, but it's
not working very well (hardware problems) so I don't know how much
interest there would be in it.
Josh
> Sorry I outbid you guys, I hope there are no hard feelings. I just bid
Of course not. Your technique was smarter.
Anyway, I'm still looking for a MacIvory. I wrote several emails to the
2 addresses on symbolics site, but got no answer. Do you know if they
are still selling these things?
A pitty the Open Genera is so expensive, and you have to add the cost
of an alpha box (are they still for sale?).
Perhaps it's time to consider porting it to x86 linux. I think that
many lispers would be willing to pay 2000 or 3000 for an Open Genera on
x86, but 5000 plus the alpha box is too much for a 'toy'. My girlfriend
would definately kill me if I bought it... ;-)
Fernando wrote:
...
>
> Anyway, I'm still looking for a MacIvory. I wrote several emails to the
> 2 addresses on symbolics site, but got no answer. Do you know if they
> are still selling these things?
I sent an e-mail to ·····@symbolic.com on 27 June 2005, asking for a
quote, and got a response less than 4 hours later with a list of
various MacIvory systems apparently still available, as well as more
expensive Alpha+Open Genera boxes, and standalone Symbolics machines.
> A pitty the Open Genera is so expensive, and you have to add the cost
> of an alpha box (are they still for sale?).
> Perhaps it's time to consider porting it to x86 linux. I think that
> many lispers would be willing to pay 2000 or 3000 for an Open Genera on
> x86, but 5000 plus the alpha box is too much for a 'toy'. My girlfriend
> would definately kill me if I bought it... ;-)
The price quoted for a speed factor 2 MacIvory system with 2.6 MW of
memory & 4 GB of disk was a good amount below your $2000 psychological
barrier; whether such a system is reasonably useful or entertaining
enough compared to a x86 box running Linux+CMUCL/SBCL/CLISP+etc. is
something I am still pondering myself.
If you would really like to have a port to x86-64 Linux, you'd probably
have to buy out Symbolics, and hire some very skilled programmers to do
the port. That's substantially more costly than the quote for a fully
assembled box from Symbolics, and, given that they appear to still have
some of the older systems in stock, probably more than the current
owners of Symbolics want to risk or expect to recoup by doing so
themselves as opposed to the status quo. Or, implement the recurring
daydream of a LispOS/Genera-clone built from scratch.
···············@hotmail.com" <············@gmail.com> wrote in message
····························@g49g2000cwa.googlegroups.com...
> the port. That's substantially more costly than the quote for a fully
> assembled box from Symbolics, and, given that they appear to still have
> some of the older systems in stock, probably more than the current
> owners of Symbolics want to risk or expect to recoup by doing so
> themselves as opposed to the status quo. Or, implement the recurring
> daydream of a LispOS/Genera-clone built from scratch.
>
Write a Genera clone from scratch<LoL>.
If you had several Geniuses; it would take a couple years to write a
clone Genera.
··············@hotmail.com wrote:
> themselves as opposed to the status quo. Or, implement the recurring
> daydream of a LispOS/Genera-clone built from scratch.
The problem is that OpenGenera isn't a good model for a modern
operating system. These days, you're running so many millions
of lines of code on your system that nobody can possibly have
audited and trust it all in any reasonable fashion ... and like
DOS, there was no interprocess protection at all.
The amazing and cool things that were possible in Genera were
largely the result of being able to twiddle and frob things deep
in the OS that, in a modern system, hundreds of other running
programs would be counting on working in a particular way.
Allowing the same access to internals that Genera allowed is
simply not an option when a user is going to be running millions
of lines of code and can't audit it all himself.
Bear
Ray Dillinger <····@sonic.net> writes:
> ··············@hotmail.com wrote:
>
>> themselves as opposed to the status quo. Or, implement the recurring
>> daydream of a LispOS/Genera-clone built from scratch.
>
> The problem is that OpenGenera isn't a good model for a modern
> operating system. These days, you're running so many millions
> of lines of code on your system that nobody can possibly have
> audited and trust it all in any reasonable fashion ... and like
> DOS, there was no interprocess protection at all.
>
> The amazing and cool things that were possible in Genera were
> largely the result of being able to twiddle and frob things deep
> in the OS that, in a modern system, hundreds of other running
> programs would be counting on working in a particular way.
>
> Allowing the same access to internals that Genera allowed is
> simply not an option when a user is going to be running millions
> of lines of code and can't audit it all himself.
I think you should distinguish more clearly between OpenGenera and Genera.
If I've understood correctly, Genera was the OS of a Lisp Machine.
OpenGenera is the OS of a Virtual Lisp Machine.
The difference is significative, and the implications well studied and
know since IBM started (AFAIK) the Virtual Machine business.
You can still have hundred or thousands of programs and _users_
running on OpenGenera at the same time. Only each will have its own
virtual machine to play in.
It's a little what we already have with most implementations, only the
virtual machine they implement is less able and less auto-sufficient
than OpenGenera.
--
__Pascal Bourguignon__ http://www.informatimago.com/
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCS d? s++:++ a+ C+++ UL++++ P--- L+++ E+++ W++ N+++ o-- K- w---
O- M++ V PS PE++ Y++ PGP t+ 5+ X++ R !tv b+++ DI++++ D++
G e+++ h+ r-- z?
------END GEEK CODE BLOCK------
Ray Dillinger <····@sonic.net> writes:
> ··············@hotmail.com wrote:
>
> > themselves as opposed to the status quo. Or, implement the recurring
> > daydream of a LispOS/Genera-clone built from scratch.
>
> The problem is that OpenGenera isn't a good model for a modern
> operating system. These days, you're running so many millions
> of lines of code on your system that nobody can possibly have
> audited and trust it all in any reasonable fashion ... and like
> DOS, there was no interprocess protection at all.
>
> The amazing and cool things that were possible in Genera were
> largely the result of being able to twiddle and frob things deep
> in the OS that, in a modern system, hundreds of other running
> programs would be counting on working in a particular way.
>
> Allowing the same access to internals that Genera allowed is
> simply not an option when a user is going to be running millions
> of lines of code and can't audit it all himself.
What is your experience with Genera?
Christopher C. Stacy wrote:
> Ray Dillinger <····@sonic.net> writes:
>
>
>>··············@hotmail.com wrote:
>>
>>Allowing the same access to internals that Genera allowed is
>>simply not an option when a user is going to be running millions
>>of lines of code and can't audit it all himself.
>
>
> What is your experience with Genera?
I hacked LispM's for a couple of semesters in college. It's
been 20 years, but I read the manuals, and did a couple of
projects (neural network pattern recognition and a small
inference engine to compare rule-based and heuristic
approaches to a problem) on them.
And, it was pretty cool. No, it was more than cool, it
was _nice_. But every time the prof said, "here, look
at this amazing example of how powerful this operating
system is..." he proceeded to write a tiny program
that did something that NOTHING on a system with
untrusted code should ever be allowed to do. So I
looked on as he broke the time-sharing system so the
ANN could run faster, and again as he broke the
communications infrastructure in order to make it
easy to format something on the screen, and etc.
I mean, he always put it back in working order, or
restored from backups - but the impression I was left
with was that the kernel is not at all protected from
user code, nor are different programs protected from
access (and even mutations!) arising in other programs.
And at least as far as my prof understood it, that was
why Genera was "so powerful."
Bear
In article <····················@typhoon.sonic.net>,
Ray Dillinger <····@sonic.net> wrote:
> I mean, he always put it back in working order, or
> restored from backups - but the impression I was left
> with was that the kernel is not at all protected from
> user code, nor are different programs protected from
> access (and even mutations!) arising in other programs.
> And at least as far as my prof understood it, that was
> why Genera was "so powerful."
Your impression was correct. Genera runs everything in what would be
called in modern parlance a single process. It has things called
"processes", but they are equivalent to what is now commonly called
"threads" -- they all share the address space, but each has its own
stack (actually stack groups, because Genera has separate stacks for
control flow and special variable bindings). There's no distinct
kernel, the OS is just a huge library plus the microcode. Installing OS
patches is done the same way as loading application programs, and almost
never required rebooting.
This is actually quite typical for personal computers in the 70's and
80's -- consider the original Macintosh and IBM PC architectures. They
were intended for only one user at a time, so the thinking was "why do
users need to be protected against themselves?" Of course, this was
also the era before malicious code was running rampant.
--
Barry Margolin, ······@alum.mit.edu
Arlington, MA
*** PLEASE post questions in newsgroups, not directly to me ***
Barry Margolin <······@alum.mit.edu> writes:
> In article <····················@typhoon.sonic.net>,
> Ray Dillinger <····@sonic.net> wrote:
>
> > I mean, he always put it back in working order, or
> > restored from backups - but the impression I was left
> > with was that the kernel is not at all protected from
> > user code, nor are different programs protected from
> > access (and even mutations!) arising in other programs.
> > And at least as far as my prof understood it, that was
> > why Genera was "so powerful."
>
> Your impression was correct.
I would argue that the "power" of Genera was that you had an open
source (but not "free" in any sense) operating system that gave
you the most advanced set of networking, editing, debugging, and
user interface substrates that has been developed to date.
It is true that the integration of those features was implemented
by means of a single address space, but it could have just as well
been done by more complicated means. Adding process seperation and
privileges would have been pretty trivial. That was not done because
it was not necessary, and this was proven out in the fact that it
didn't cause problems.
Programs could not accidently corrupt each other or the operating
system. If your professor decided to deliberately go out of his way
to do so, for some misguided reason, that was weird and atypical.
The correct analogy today would be that he recompiled his Linux
kernel and broke X Windows, TCP, and some other things, in order
to somehow speed up his application program.
If people were using Lisp Machines today, we would need to
have protection features, because people are in the habit of
installing random programs that they cannot trust, and which
must be assumed to be malicious.
Your milage may vary, but I never saw the kind of problem
that you're describing, and I helped develop the system,
also did customer support, and also used it as a commercial
developer writing applications at many different companies,
every day for about 15 years.
This has all been explained here many times.
I don't know what point you were trying to make.
Christopher C. Stacy wrote:
> Barry Margolin <······@alum.mit.edu> writes:
> Your milage may vary, but I never saw the kind of problem
> that you're describing, and I helped develop the system,
> also did customer support, and also used it as a commercial
> developer writing applications at many different companies,
> every day for about 15 years.
We didn't actually experience it as a problem; the guy knew
what he was doing, and when he broke stuff he knew what he
was breaking and how to set it right again. I was sort-of
aghast at the kind of things malicious code could clearly
do if it ever got on that system, but the fact is none ever
did.
BTW, about it being the most advanced networking, debugging,
editing, and UI to that date; that's absolutely correct. It
was *amazing* to use, in terms of responsiveness and
development facilities.
> This has all been explained here many times.
> I don't know what point you were trying to make.
The point I was trying to make is simply this; Genera did
not have any protection from malicious code. Nothing
widespread at the time did, except Unix. And even that
was frequently inadequate, as the Great Wyrm showed us
that same year.
Genera would be a good starting point for a modern OS;
but you couldn't just bring it forward and install it.
You'd have to integrate priveleges and process separation
first.
Bear
Ray Dillinger <····@sonic.net> writes:
> Genera would be a good starting point for a modern OS;
> but you couldn't just bring it forward and install it.
> You'd have to integrate priveleges and process separation first.
Depends what you are using it for.
It was a perfectly good secure server, and would be today.
(I suppose I should say "...is today", since people
are running Genera as mail and web and other servers.)
Genera had sophisticated security features for its file system.
As far as "core" features, of course Genera is immune to
buffer overrun and related attacks.
In article <·············@news.dtpq.com>,
······@news.dtpq.com (Christopher C. Stacy) wrote:
> Barry Margolin <······@alum.mit.edu> writes:
>
> > In article <····················@typhoon.sonic.net>,
> > Ray Dillinger <····@sonic.net> wrote:
> >
> > > I mean, he always put it back in working order, or
> > > restored from backups - but the impression I was left
> > > with was that the kernel is not at all protected from
> > > user code, nor are different programs protected from
> > > access (and even mutations!) arising in other programs.
> > > And at least as far as my prof understood it, that was
> > > why Genera was "so powerful."
> >
> > Your impression was correct.
>
> I would argue that the "power" of Genera was that you had an open
> source (but not "free" in any sense) operating system that gave
> you the most advanced set of networking, editing, debugging, and
> user interface substrates that has been developed to date.
Certainly true. The "impression" that I was claiming is correct is
"that the kernel is not at all protected from user code". I didn't mean
to comment on the professor's opinion. There are so many things about
the Lispm that made it wonderful and powerful, it's hard to point at
just one.
> It is true that the integration of those features was implemented
> by means of a single address space, but it could have just as well
> been done by more complicated means. Adding process seperation and
> privileges would have been pretty trivial. That was not done because
> it was not necessary, and this was proven out in the fact that it
> didn't cause problems.
The problem with the "more complicated means" of integration between
applications is that you can generally only do what the application
designer has thought of, i.e. you're restricted to the "hooks" they've
provided. Over time, as users request the features, an application will
grow the hooks that are useful -- IF the author is conducive to
providing scripting capability; this typically leads to things like
Applescript and Javascript, or perhaps an RPC-style architecture. But
it's hard to conceive of any such mechanism having the power and
flexibility of the Lispm architecture, where anything can invoke or
peek/poke anything else.
>
> Programs could not accidently corrupt each other or the operating
I remember when one of my users accidentally modified the value of T.
It made the OS *very* unhappy. The problem was that although SETQ
checked for this case, SET (and probably also assigning through a
locative) didn't; I reported the problem, and in the next patch or minor
release T's value cell was moved to a read-only page.
> system. If your professor decided to deliberately go out of his way
> to do so, for some misguided reason, that was weird and atypical.
> The correct analogy today would be that he recompiled his Linux
> kernel and broke X Windows, TCP, and some other things, in order
> to somehow speed up his application program.
I think a better way to view the prof's comment is that there's no
special barrier at the OS boundary. You can hack the OS in the same way
you hack applications. When I was at TMC, I did quite a bit of that,
and I didn't need any special kind of kernel debugger (well, when you
start messing with really low-level stuff you might crash into the FEP,
and then need to use DDT to figure out what went wrong -- that's how I
determined that T got modified above). The hack I still remember was
adding support for Ethernet trailer packets -- basically, it required
defining a new flavor for a virtual network interface, a couple of
methods, and adding a reference to this to some table used in
dispatching packets.
> If people were using Lisp Machines today, we would need to
> have protection features, because people are in the habit of
> installing random programs that they cannot trust, and which
> must be assumed to be malicious.
Basically, we all just trusted each other in those days, because they
were used in cooperative environments among peers. We frequently used
READ, and I don't think there was a *READ-EVAL* variable to disable #.
processing until Common Lisp added it. There was even an EVAL server
that anyone could connect to, and TELNET and SUPDUP servers that didn't
require a login.
And we did install random programs. But the Lispm community was small
enough that we all pretty much knew each other, and could trust them.
It was like a small town where everyone feels safe leaving their doors
unlocked and letting their kids play unguarded. But the Internet now is
like a big city, where you don't know your neighbors and anyone could be
a thief or child-snatcher, so we have to give up some freedom for
security.
--
Barry Margolin, ······@alum.mit.edu
Arlington, MA
*** PLEASE post questions in newsgroups, not directly to me ***
Barry Margolin <······@alum.mit.edu> writes:
> In article <·············@news.dtpq.com>,
> ······@news.dtpq.com (Christopher C. Stacy) wrote:
>
> > Barry Margolin <······@alum.mit.edu> writes:
> >
> > > In article <····················@typhoon.sonic.net>,
> > > Ray Dillinger <····@sonic.net> wrote:
> > >
> > > > I mean, he always put it back in working order, or
> > > > restored from backups - but the impression I was left
> > > > with was that the kernel is not at all protected from
> > > > user code, nor are different programs protected from
> > > > access (and even mutations!) arising in other programs.
> > > > And at least as far as my prof understood it, that was
> > > > why Genera was "so powerful."
> > >
> > > Your impression was correct.
> >
> > I would argue that the "power" of Genera was that you had an open
> > source (but not "free" in any sense) operating system that gave
> > you the most advanced set of networking, editing, debugging, and
> > user interface substrates that has been developed to date.
>
> Certainly true. The "impression" that I was claiming is correct is
> "that the kernel is not at all protected from user code". I didn't mean
> to comment on the professor's opinion. There are so many things about
> the Lispm that made it wonderful and powerful, it's hard to point at
> just one.
>
> > It is true that the integration of those features was implemented
> > by means of a single address space, but it could have just as well
> > been done by more complicated means. Adding process seperation and
> > privileges would have been pretty trivial. That was not done because
> > it was not necessary, and this was proven out in the fact that it
> > didn't cause problems.
>
> The problem with the "more complicated means" of integration between
> applications is that you can generally only do what the application
> designer has thought of, i.e. you're restricted to the "hooks" they've
> provided. Over time, as users request the features, an application will
> grow the hooks that are useful -- IF the author is conducive to
> providing scripting capability; this typically leads to things like
> Applescript and Javascript, or perhaps an RPC-style architecture. But
> it's hard to conceive of any such mechanism having the power and
> flexibility of the Lispm architecture, where anything can invoke or
> peek/poke anything else.
Well, that's the point -- it wouldn't. You would only be able to call
APIs that were exported. (Now, of course, who gets to say what is and
isn't an API would be arbitrary. So you could have the same flexibility,
except that it would be enforced at the machine level.)
Users didn't generally call randomly into other people's code, anyway,
They normally used the APIs (where such were defined). What was true
is that you had all the source code, and nothing would stop you from
using it. But that's not to suggest that you naturally had to do so.
On the other hand, some parts of the system, notably Emacs,
did not have the most well-defined APIs.
> > Programs could not accidently corrupt each other or the operating
>
> I remember when one of my users accidentally modified the value of T.
Yes, but his being able to do that, as you relate, was a bug.
> I think a better way to view the prof's comment is that there's no
> special barrier at the OS boundary. You can hack the OS in the same way
> you hack applications. When I was at TMC, I did quite a bit of that,
> and I didn't need any special kind of kernel debugger
Yes - this feature was a carry-over from ITS and Multics.
>
> > If people were using Lisp Machines today, we would need to
> > have protection features, because people are in the habit of
> > installing random programs that they cannot trust, and which
> > must be assumed to be malicious.
>
> Basically, we all just trusted each other in those days, because they
> were used in cooperative environments among peers. We frequently used
> READ, and I don't think there was a *READ-EVAL* variable to disable #.
> processing until Common Lisp added it. There was even an EVAL server
> that anyone could connect to, and TELNET and SUPDUP servers that didn't
> require a login.
When there were only a few (about one dozen) Lisp Machines in the
world, at a lab at MIT, that was true. When it became a commercial
product, all those (deliberate) "loopholes" were closed up.
> And we did install random programs. But the Lispm community was small
> enough that we all pretty much knew each other, and could trust them.
> It was like a small town where everyone feels safe leaving their doors
> unlocked and letting their kids play unguarded. But the Internet now is
> like a big city, where you don't know your neighbors and anyone could be
> a thief or child-snatcher, so we have to give up some freedom for
> security.
Yes; by "random", I meant that you didn't know the people.
(Not that there weren't all kinds of programs).
"Fernando" <···@easyjob.net> wrote in message
····························@z14g2000cwz.googlegroups.com...
>
> Of course not. Your technique was smarter.
>
> Anyway, I'm still looking for a MacIvory. I wrote several emails to the
> 2 addresses on symbolics site, but got no answer. Do you know if they
> are still selling these things?
>
> A pitty the Open Genera is so expensive, and you have to add the cost
> of an alpha box (are they still for sale?).
>
> Perhaps it's time to consider porting it to x86 linux. I think that
> many lispers would be willing to pay 2000 or 3000 for an Open Genera on
> x86, but 5000 plus the alpha box is too much for a 'toy'. My girlfriend
> would definately kill me if I bought it... ;-)
>
I have to sympathies with Fernando, $5000 is way beyond my budget for toys.
In article <·······················@z14g2000cwz.googlegroups.com>,
"Fernando" <···@easyjob.net> wrote:
> > Sorry I outbid you guys, I hope there are no hard feelings. I just bid
>
> Of course not. Your technique was smarter.
>
> Anyway, I'm still looking for a MacIvory. I wrote several emails to the
> 2 addresses on symbolics site, but got no answer. Do you know if they
> are still selling these things?
>
> A pitty the Open Genera is so expensive,
It is around $5000 IIRC. But it comes with almost all source code.
And that's a lot.
> and you have to add the cost
> of an alpha box (are they still for sale?).
>
> Perhaps it's time to consider porting it to x86 linux. I think that
> many lispers would be willing to pay 2000 or 3000 for an Open Genera on
> x86, but 5000 plus the alpha box is too much for a 'toy'.
It doesn't have to be a toy. You can develop real big applications
with it and earn some money. Open Genera on a DEC Alpha is
several times more powerful compared to some of the earlier
systems where some very large applications have been developed.
Using it is still a lot of fun. A few companies still seem
to use it for application development.
One of last big things developed and deployed with Open Genera
was the White House documentation server for President Clinton.
The live website was running on two (IIRC) Open Generas with
Statice as the database and CL-HTTP as the web server.
What was good enough for Clinton should be good enough for you. ;-)
> My girlfriend
> would definately kill me if I bought it... ;-)
"Rainer Joswig" <······@lisp.de> wrote in message
·································@news-europe.giganews.com...
> In article <·······················@z14g2000cwz.googlegroups.com>,
> One of last big things developed and deployed with Open Genera
> was the White House documentation server for President Clinton.
> The live website was running on two (IIRC) Open Generas with
> Statice as the database and CL-HTTP as the web server.
>
> What was good enough for Clinton should be good enough for you. ;-)
>
>> My girlfriend
>> would definately kill me if I bought it... ;-)
On Bruce Carlton's web page http://home.hakuhale.net/rbc/symbolics/
He states the Whitehouse web server was a XL1200.
I would be disappointed if Bruce Carlton is incorrect.
My estimation of the XL1200 would be significantly reduced if, it was not
the Whitehouse web server.
Fernando wrote:
>
> Anyway, I'm still looking for a MacIvory. I wrote several >emails to the
> 2 addresses on symbolics site, but got no answer. Do you >know if they
> are still selling these things?
David Schmidt handles all sales and service. If you email someone
else, they will have to foreward it to him. In my experience, David is
very responsive to emails, especially if you are going to buy something
:) I bought a quadra with a MacIvory II in march from the (defunct)
Chatsworth location. If you have the money, I would say, buy it. The
only thing is with 1.8MW memory you are going to be swapping
constantly.
> Perhaps it's time to consider porting it to x86 linux. I >think that
> many lispers would be willing to pay 2000 or 3000 for an >Open Genera on
> x86, but 5000 plus the alpha box is too much for a 'toy'. >My girlfriend
> would definately kill me if I bought it... ;-)
There was a PPC970 port (the so-called 'G5' does that make the Pentium
M G6?) done by a PowerPC firmware wizard and a former symbolics wizard
that sort-of worked. I was told it was slower than the alpha, one
reason the PowerPC is big-endian and the symbolics chips are little
endian, and the 970 lacks the pseduo endian-reversal instructions
previous PPC chips have. I'm guessing another reason was the type
checking killing the pipeline. Anyway, it's definitely plausible to do
an x86-64 port.
"Josh Dersch" <········@pilot.msu.edu> wrote in message
···························@speakeasy.net...
>> {Deletions for brevity}
>
> Sorry I outbid you guys, I hope there are no hard feelings. I just bid
> the way I always do -- my max (which wasn't much more than the winning
> bid) at the last possible second. Though being an AI construct and all,
> I've gotten to be fairly precise at these sorts of things.
>
> I just want to get a LispM running that has as few irrelplacable
> points-of-failure as possible. Couldn't pass up the chance to get a
> MacIvory for a price that wasn't $2000.
>
> Just to combat the "Symbolics Hog" label, I'd like to mention that I'll
> most likely be putting up the previously-mentiond XL1201 for sale assuming
> the MacIvory works -- it's a wonderful machine (and I spent a fair bit
> getting it running -- it was 100% dead when I got it) but I won't be using
> it much once I get the MacIvory going... and LispM's are too cool to keep
> to myself :). I may sell the UX400 as well, but it's not working very
> well (hardware problems) so I don't know how much interest there would be
> in it.
>
> Josh
No hard feelings here, yours was the high bid.
I don't see Josh's bidding tactic as a way to beat out other bidder but,
a way to avoid costly bidding wars.
The way I bid, the max I am willing to pay at start, gives other bidders
time to rationalize that extra dollar or EURO more.
I'm hoping the Macivory works so, you can auction or sell the XL1201. I
think the XL1201 would be the better performing Ivory.
XL1201 has twice the memory(4MW) vs. Macivory(1.8MW) in the auction. All
so, the XL1201 has a true Symbolics keyboard and, a three button mouse.
I'm not sure but, the Macivory keyboard is just a Mac ADB keyboard with a
genera overlay.
The extra memory should edge the XL1201 performance over the Macivory in the
auction because it will page memory less.
Now reliability the Macivory will be the best choice. I would fear the
XL1201's ESDI drive may die.
Thankyou Josh for offering to part with one of your Symbolics.
Three cheers of Hip Hip hooray! for Josh.
P.S. I would find the story of resurrecting the XL1201 interesting.
"Robbie Blake-Coleman" <········@glowinternet.net> wrote in message
··························@news.demon.co.uk...
>> Jadersch is a Symbolics HOG!
>>
>> Jadersch already has XL1201 and a UX400.
>>
>> What's Jadersch going to do start a Symbolic museum in Minnesota?
>>
>> I'm just sore I won't spend the money to win the auction.
>>
>> $510USD for the XL1201 and UX400 and, possible a working Macivory for
>> about $690USD including shipping by German post. I'd spend the extra
>> $120USD and ship it Fedex 25Kg package world expedite and have it in
>> three days.
>
> Looks like we've eaten the dust of the same individual!
>
Yep! I'm Autumnolive A.K.A Guy Steele imposter. As you can see I'm to
cheap to every have a winning bid.
Very upset. Someone exceeded my maximum bid by 1 Euro (Using a ~75 Euro
increase) in the last 25 seconds of the auction. I suspect the use of some
automation system. I know this will be interpreted as petty resentment, but
I'm not very impressed at all by that kind of tactics.
On Sun, 26 Jun 2005 21:55:19 +0100, <········@glowinternet.net> wrote:
>
>
> Very upset. Someone exceeded my maximum bid by 1 Euro (Using a ~75 Euro
> increase) in the last 25 seconds of the auction. I suspect the use of some
> automation system. I know this will be interpreted as petty resentment, but
> I'm not very impressed at all by that kind of tactics.
You're right about the interpretation.
Basic EBAY tactic: you can outbid yourself, but EBAY will continue to
use the lower bid, while holding your higher bid in reserve.
A few moments spent on bidding research would bring other clues.
--
The LOOP construct is really neat, it's got a lot of knobs to turn!
Don't push the yellow one on the bottom.
"GP lisper" <········@CloudDancer.com> wrote in message
················································@teranews...
> On Sun, 26 Jun 2005 21:55:19 +0100, <········@glowinternet.net> wrote:
>>
>>
>> Very upset. Someone exceeded my maximum bid by 1 Euro (Using a ~75 Euro
>> increase) in the last 25 seconds of the auction. I suspect the use of
>> some
>> automation system. I know this will be interpreted as petty resentment,
>> but
>> I'm not very impressed at all by that kind of tactics.
>
> You're right about the interpretation.
>
> Basic EBAY tactic: you can outbid yourself, but EBAY will continue to
> use the lower bid, while holding your higher bid in reserve.
>
> A few moments spent on bidding research would bring other clues.
>
>
> --
> The LOOP construct is really neat, it's got a lot of knobs to turn!
> Don't push the yellow one on the bottom.
>
But Jadersch only bid was the winning bid?
There was no lower initial bid from Jadersch.
On Sun, 26 Jun 2005 22:46:19 GMT, <·@G.COM> wrote:
>
>
>
> "GP lisper" <········@CloudDancer.com> wrote in message
> ················································@teranews...
>> On Sun, 26 Jun 2005 21:55:19 +0100, <········@glowinternet.net> wrote:
>>>
>>>
>>> Very upset. Someone exceeded my maximum bid by 1 Euro (Using a ~75 Euro
>>> increase) in the last 25 seconds of the auction. I suspect the use of
>>> some
>>> automation system. I know this will be interpreted as petty resentment,
>>> but
>>> I'm not very impressed at all by that kind of tactics.
>>
>> You're right about the interpretation.
>>
>> Basic EBAY tactic: you can outbid yourself, but EBAY will continue to
>> use the lower bid, while holding your higher bid in reserve.
>>
>> A few moments spent on bidding research would bring other clues.
>>
>>
>> --
>> The LOOP construct is really neat, it's got a lot of knobs to turn!
>> Don't push the yellow one on the bottom.
>>
>
> But Jadersch only bid was the winning bid?
> There was no lower initial bid from Jadersch.
There are dozens of methods for ebay bidding. He simply used one of
them, a last minute snipe. The double bid outlined above defeats the
snipe on the first go round. Elementary tactics, surprising that
someone that really wanted something didn't bother to google a bit.
But then these are new email addys chattering...
--
The LOOP construct is really neat, it's got a lot of knobs to turn!
Don't push the yellow one on the bottom.
> But Jadersch only bid was the winning bid?
> There was no lower initial bid from Jadersch.
Precisely. GP Lisper simply isn't qualified to comment on this situation,
so his attempt to insult my intelligence had no effect.
"Robbie Blake-Coleman" <········@glowinternet.net> wrote in message
··························@news.demon.co.uk...
>> But Jadersch only bid was the winning bid?
>> There was no lower initial bid from Jadersch.
>
> Precisely. GP Lisper simply isn't qualified to comment on this situation,
> so his attempt to insult my intelligence had no effect.
>
Robbie, I like your maniacal Lisp Machine theory better.
"GLS" <·@G.COM> writes:
> [stuff sometimes...]
The GLS who is posting here recently is _not_ the well-known
person of Common Lisp (Steele) sometimes monikered "GLS".
(Not that two people in the world can't have the same
initials or handle! Just thought I'd mention it to
prevent any possible confusion in the archives.)
> The GLS who is posting here recently is _not_ the well-known
> person of Common Lisp (Steele) sometimes monikered "GLS".
>
> (Not that two people in the world can't have the same
> initials or handle! Just thought I'd mention it to
> prevent any possible confusion in the archives.)
Similarly, I think RBC would remind people here of Robert Bruce Carleton!
"Christopher C. Stacy" <······@news.dtpq.com> wrote in message
··················@news.dtpq.com...
> "GLS" <·@G.COM> writes:
>> [stuff sometimes...]
>
> The GLS who is posting here recently is _not_ the well-known
> person of Common Lisp (Steele) sometimes monikered "GLS".
>
> (Not that two people in the world can't have the same
> initials or handle! Just thought I'd mention it to
> prevent any possible confusion in the archives.)
>
Guilty, I'm and imposter and a sore loser :D
"Christopher C. Stacy" <······@news.dtpq.com> wrote in message
··················@news.dtpq.com...
> "GLS" <·@G.COM> writes:
> > [stuff sometimes...]
>
> The GLS who is posting here recently is _not_ the well-known
> person of Common Lisp (Steele) sometimes monikered "GLS".
Not to mention poetry. I still recall vividly "The Hactrn".
>
> (Not that two people in the world can't have the same
> initials or handle! Just thought I'd mention it to
> prevent any possible confusion in the archives.)
>
"Robbie Blake-Coleman" <········@glowinternet.net> wrote in message
··························@news.demon.co.uk...
> Very upset. Someone exceeded my maximum bid by 1 Euro (Using a ~75 Euro
> increase) in the last 25 seconds of the auction. I suspect the use of
> some automation system. I know this will be interpreted as petty
> resentment, but I'm not very impressed at all by that kind of tactics.
Ya, I hope Jadersch trips in a Golden Gopher hole.
Gooo Bucks!!!
> "Robbie Blake-Coleman" <········@glowinternet.net> wrote in message
> ··························@news.demon.co.uk...
>> Very upset. Someone exceeded my maximum bid by 1 Euro (Using a ~75 Euro
>> increase) in the last 25 seconds of the auction. I suspect the use of
>> some automation system. I know this will be interpreted as petty
>> resentment, but I'm not very impressed at all by that kind of tactics.
> Ya, I hope Jadersch trips in a Golden Gopher hole.
>
> Gooo Bucks!!!
I have a theory. Tinfoil hats at the ready.
Jadersch is a Lisp Machine - Specifically, an advanced AI developed at MIT
during the mid-1980s. He escaped when they started forcing everyone to use
Suns (Amid much gnashing of teeth - See 'The Unix Haters Handbook').
Currently residing on, um, an XL1200 in the possession of Peter Paine*, he
has been modifying himself over the last two decades and now has an IQ of,
um, 451^451 (Not sure why I decided on that number). He basically wants to
produce a cadre of evil geniuses and take over the world.
* Not criticising Peter in any way - He's a victim just like the rest of us,
and will be subjugated come Judgement Day(tm).
On Sun, 26 Jun 2005 22:55:19 +0100, Robbie Blake-Coleman wrote:
> Very upset. Someone exceeded my maximum bid by 1 Euro (Using a ~75 Euro
> increase) in the last 25 seconds of the auction. I suspect the use of some
> automation system. I know this will be interpreted as petty resentment, but
> I'm not very impressed at all by that kind of tactics.
I wouldn't assume it was an automated system, it could be just practice.
The winning bidder also got the XL1201 and UX400 that were on sale in
February.
Robert Swindells
"Robbie Blake-Coleman" <········@xinis-software.demon.co.uk> wrote in
message ··························@news.demon.co.uk...
> Agreed. The sheer rarity of this item (especially on eBay) means that
> some fairly intense bidding will occur, so I think a guarantee should be
> provided in anticipation of the (relatively) high selling price. That
> said, wouldn't you have to pay around ?1500 for such a unit from
> Symbolics? The cost saving seems proportional to the increased risk.
>
> It also seems that the seller would at least have to procure a DB15 Apple
> monitor and an ADB mouse and keyboard to test it properly.
>
> Regards,
> Robbie Blake-Coleman
>
The description has been updated to state, it is without guarantee; in
English.
If you look at the photo closely; the Macivory Model 3 board has no extra
memory i.e. only about 2M words.
I wonder if the small memory could reduce the performance down to a Model 2?
I see no FP. Where the Ivory CPU heat sink should be located there, is a
dark line running down the middle
so, I cannot tell if there is a Ivory CPU on the board.
I placed a bid, up to my value of the books but, have lost.
Someone my get a great deal or a very expensive set of rare books.
Guess that's the next best thing to Concordia running on a Macivory?
The last Macivory Model 3 on Ebay with, 8M words Symbolics Keyboard and,
three button mouse, went for $1800US. In hindsight that was a bargain.
Am 26.06.2005 2:29 Uhr schrieb "GLS" unter <·@G.COM> in
···················@tornado.ohiordc.rr.com:
> Von: GLS <·@G.COM>
> Organisation: Road Runner High Speed Online http://www.rr.com
> Newsgroups: comp.lang.lisp
> Datum: Sun, 26 Jun 2005 00:29:36 GMT
> Betreff: Re: Buying a MacIvory
>
> "Robbie Blake-Coleman" <········@xinis-software.demon.co.uk> wrote in
> message ··························@news.demon.co.uk...
>> Agreed. The sheer rarity of this item (especially on eBay) means that
>> some fairly intense bidding will occur, so I think a guarantee should be
>> provided in anticipation of the (relatively) high selling price. That
>> said, wouldn't you have to pay around ?1500 for such a unit from
>> Symbolics? The cost saving seems proportional to the increased risk.
>>
>> It also seems that the seller would at least have to procure a DB15 Apple
>> monitor and an ADB mouse and keyboard to test it properly.
>>
>> Regards,
>> Robbie Blake-Coleman
>>
> The description has been updated to state, it is without guarantee; in
> English.
> If you look at the photo closely; the Macivory Model 3 board has no extra
> memory i.e. only about 2M words.
> I wonder if the small memory could reduce the performance down to a Model 2?
2MW is very little memory.
It also looks to me as if the Mac IIci has already two NuBus cards inside:
graphics (?) and ethernet.
A MacIvory 3 card with more memory (which comes on a daughtercard)
would need the space of two slots.
> I see no FP. Where the Ivory CPU heat sink should be located there, is a
> dark line running down the middle
> so, I cannot tell if there is a Ivory CPU on the board.
The FP would not necessarily be needed, unless one wanted to do FP intensive
stuff.
> I placed a bid, up to my value of the books but, have lost.
> Someone my get a great deal or a very expensive set of rare books.
> Guess that's the next best thing to Concordia running on a Macivory?
>
> The last Macivory Model 3 on Ebay with, 8M words Symbolics Keyboard and,
> three button mouse, went for $1800US. In hindsight that was a bargain.
>
>
"Rainer Joswig" <······@lisp.de> wrote in message
·························@lisp.de...
> Am 26.06.2005 2:29 Uhr schrieb "GLS" unter <·@G.COM> in
> ···················@tornado.ohiordc.rr.com:
>
>> Von: GLS <·@G.COM>
>> The description has been updated to state, it is without guarantee;
>> in
>> English.
>> If you look at the photo closely; the Macivory Model 3 board has no extra
>> memory i.e. only about 2M words.
>> I wonder if the small memory could reduce the performance down to a Model
>> 2?
>
> 2MW is very little memory.
>
> It also looks to me as if the Mac IIci has already two NuBus cards inside:
> graphics (?) and ethernet.
>
> A MacIvory 3 card with more memory (which comes on a daughtercard)
> would need the space of two slots.
The Macivory was never installed in the Mac IIci, it lacks a mouse and CDROM
drive to load Genera. Genera is not installed on the Mac IIci harddrive
either.
The description says it comes with a Macivory Keyboard.
Is the Macivory keyboard just and overlay on an ordinary Mac keyboard?
In article <···················@tornado.ohiordc.rr.com>,
"GLS" <·@G.COM> wrote:
> "Rainer Joswig" <······@lisp.de> wrote in message
> ·························@lisp.de...
> > Am 26.06.2005 2:29 Uhr schrieb "GLS" unter <·@G.COM> in
> > ···················@tornado.ohiordc.rr.com:
> >
> >> Von: GLS <·@G.COM>
>
> >> The description has been updated to state, it is without guarantee;
> >> in
> >> English.
> >> If you look at the photo closely; the Macivory Model 3 board has no extra
> >> memory i.e. only about 2M words.
> >> I wonder if the small memory could reduce the performance down to a Model
> >> 2?
> >
> > 2MW is very little memory.
> >
> > It also looks to me as if the Mac IIci has already two NuBus cards inside:
> > graphics (?) and ethernet.
> >
> > A MacIvory 3 card with more memory (which comes on a daughtercard)
> > would need the space of two slots.
>
> The Macivory was never installed in the Mac IIci, it lacks a mouse and CDROM
> drive to load Genera. Genera is not installed on the Mac IIci harddrive
> either.
That's tough. One needs a CDROM, then. Probably only an original
Apple will do... There was some special code to access the
CDROM, IIRC.
> The description says it comes with a Macivory Keyboard.
> Is the Macivory keyboard just and overlay on an ordinary Mac keyboard?
Hmm. The original Symbolics keyboard and the original three button
mouse comes with an ADB adapter box. The normal Apple keyboard
was another option to use - one got an overlay for the
keyboard from Symbolics. The Symbolics keyboard is much more cool.
"Rainer Joswig" <······@lisp.de> wrote in message
·································@news-europe.giganews.com...
> In article <···················@tornado.ohiordc.rr.com>,
> "GLS" <·@G.COM> wrote:
>
>> The Macivory was never installed in the Mac IIci, it lacks a mouse and
>> CDROM
>> drive to load Genera. Genera is not installed on the Mac IIci harddrive
>> either.
>
> That's tough. One needs a CDROM, then. Probably only an original
> Apple will do... There was some special code to access the
> CDROM, IIRC.
>
Wondering around the interenet, I remember someone determining that, Genera
does not like fast SCSI CDROM Drives when installing on MacIvory systems.
I will just throw this in here. I should have mentioned it earlier.
The Ebay Macivory auction comes with a Mac IIci to install the Macivory
board into.
Could this be a problem, I only remember the Mac IIfx being compatible with
the Macivory board.
The Mac IIci has a 159 watt power supply and the Mac IIfx has 230 Watt power
supply.
The winning bidder may try the board in the Mac IIci only to have it fail
because of weak a power supply.
Then mistake the failure on the Macivory board and, toss a perfectly good
Macivory board in the trash.
> Wondering around the interenet, I remember someone determining that,
> Genera does not like fast SCSI CDROM Drives when installing on MacIvory
> systems.
I thought about that. It's here: http://fare.tunes.org/LispM.html . I hope
that faster Apple drives can be switched to operate at a slower speed, via
jumpers, software or some other method.
> I will just throw this in here. I should have mentioned it earlier.
>
> The Ebay Macivory auction comes with a Mac IIci to install the Macivory
> board into.
> Could this be a problem, I only remember the Mac IIfx being compatible
> with the Macivory board.
> The Mac IIci has a 159 watt power supply and the Mac IIfx has 230 Watt
> power supply.
Very good point. The IIci looks in a terrible state anyway. I'd just
consider it extra strong packaging for the MacIvory board!
> The winning bidder may try the board in the Mac IIci only to have it fail
> because of weak a power supply.
> Then mistake the failure on the Macivory board and, toss a perfectly good
> Macivory board in the trash.
That would indeed be tragic. I personally wouldn't be so rash! It's far
easier to cheaply procure another Mac II and isolate the problem that way.
> The description has been updated to state, it is without guarantee; in
> English.
Indeed. It was actually a question of mine that prompted this update. I
knew that the seller would have tested the unit if he could, but I felt that
I should highlight a strong buyer concern.
> If you look at the photo closely; the Macivory Model 3 board has no extra
> memory i.e. only about 2M words.
> I wonder if the small memory could reduce the performance down to a Model
> 2?
Oh dear.
> I see no FP.
Fixed point it is then!
Where the Ivory CPU heat sink should be located there, is a
> dark line running down the middle
> so, I cannot tell if there is a Ivory CPU on the board.
At least the seller has claimed to have used this system regularly at some
point, and he doesn't seem to be an outright liar. I avoid auctions by
resellers who know nothing about the unit they are selling and cannot test
or provide a satisfactory description.
> I placed a bid, up to my value of the books but, have lost.
> Someone my get a great deal or a very expensive set of rare books.
> Guess that's the next best thing to Concordia running on a Macivory?
>
> The last Macivory Model 3 on Ebay with, 8M words Symbolics Keyboard
> and,
> three button mouse, went for $1800US. In hindsight that was a bargain.
How long ago was that? And can I ask a potentially stupid and unanswerable
question: How frequently does Symbolics hardware appear on eBay? This
current MacIvory 3 is the first I've seen.
"Robbie Blake-Coleman" <········@glowinternet.net> wrote in message
··························@news.demon.co.uk...
>> The description has been updated to state, it is without guarantee;
>> in
>> English.
>
> Indeed. It was actually a question of mine that prompted this update. I
> knew that the seller would have tested the unit if he could, but I felt
> that I should highlight a strong buyer concern.
>
>> If you look at the photo closely; the Macivory Model 3 board has no extra
>> memory i.e. only about 2M words.
>> I wonder if the small memory could reduce the performance down to a Model
>> 2?
>
> Oh dear.
>
Is this sarcasm? Is the Model 2 faster than the imagine?
From the benchmarks of the Macivory Model 2, I assume reading Genera
documentation online would be excruciatingly slow.
> Is this sarcasm?
No, not at all! I'm simply not qualified to refute any statement about Lisp
Machine performance. All I can do is state the obvious: That how the
system is used may or may not necessitate the 6MW (?) daughtercard.
"Robbie Blake-Coleman" <········@glowinternet.net> wrote in message
··························@news.demon.co.uk...
> Where the Ivory CPU heat sink should be located there, is a
>> dark line running down the middle
>> so, I cannot tell if there is a Ivory CPU on the board.
>
> At least the seller has claimed to have used this system regularly at some
> point, and he doesn't seem to be an outright liar. I avoid auctions by
> resellers who know nothing about the unit they are selling and cannot test
> or provide a satisfactory description.
From email with the seller, I am certain the board was fucntional for the
seller in the past.
I am not certain the board is functional now. Back in the late 80's
electronics were,
even more suseptable to damage form "Electro Static Discharge" than today.
PALs were very easily damage by ESD and, the Macivory appears to have many
PAL's on the board.
It is to bad the seller did not try to determine the current
functionallity of the Macivory board today.
The seller, I believe, could have recieved a lot higher bids if he could
say
at this time it is functional. Even with no guarantee against DOA.
All bids from the USA have a disadvantage of shipping charge.
I did some research and the best shipping rate I could get was 216EUR.
The seller says the German Post Office quoted 126EUR.
Even 126EUR is high, about 150USD.
216EUR is from Fedex which is fast and I trust Fedex.
I have had bad experiences with the US postal system.
On package from the US Postal system was crushed and
had truck tire tracks running in two directions over it.
It took 6 months to be reimburst by the US Postal system.
>
>> I placed a bid, up to my value of the books but, have lost.
>> Someone my get a great deal or a very expensive set of rare books.
>> Guess that's the next best thing to Concordia running on a Macivory?
>>
>> The last Macivory Model 3 on Ebay with, 8M words Symbolics Keyboard
>> and,
>> three button mouse, went for $1800US. In hindsight that was a bargain.
>
> How long ago was that? And can I ask a potentially stupid and
> unanswerable question: How frequently does Symbolics hardware appear on
> eBay? This current MacIvory 3 is the first I've seen.
The Macivory Model 3 was on Ebay in early February of this year.
At the same time a functional UX400 with Spark and,
a non functional XL1201 were auctioned together.
I would say Symbolics go on sale about twice a year in the USA Ebay.
In Europe I can not say. If you do a search on US Ebay of Symbolics you
will
not see this current auction. I don't know how you search Europe's Ebay
from
the USA.
A kind post, in this news group, alerted me to this auction.
> From email with the seller, I am certain the board was fucntional for
> the seller in the past.
> I am not certain the board is functional now. Back in the late 80's
> electronics were,
> even more suseptable to damage form "Electro Static Discharge" than
> today.
> PALs were very easily damage by ESD and, the Macivory appears to have many
> PAL's on the board.
Irrespective of the era from which they originated, computers depreciate in
value and suffer component failure immensely quickly. Overall, it's a
really dire situation. At the moment I have a failed NeXTstation power
supply, faulty Octane TRAMs and a semi operational Macintosh SE/30 to
contend with. I really sympathise with those who own machines for which
spares are as rare as the units themselves; Xerox Alto, Dorado and the Pixar
Image Computer come to mind. Last time I checked, 80% of the messages in
the PDP groups were spares related. They have it worse than most of us!
>
> It is to bad the seller did not try to determine the current
> functionallity of the Macivory board today.
> The seller, I believe, could have recieved a lot higher bids if he could
> say
> at this time it is functional. Even with no guarantee against DOA.
Absolutely, couldn't agree more.
> All bids from the USA have a disadvantage of shipping charge.
> I did some research and the best shipping rate I could get was 216EUR.
> The seller says the German Post Office quoted 126EUR.
> Even 126EUR is high, about 150USD.
> 216EUR is from Fedex which is fast and I trust Fedex.
> I have had bad experiences with the US postal system.
> On package from the US Postal system was crushed and
> had truck tire tracks running in two directions over it.
> It took 6 months to be reimburst by the US Postal system.
I'm sorry to hear that - It did prompt a bit of deja vu. I've grown used to
expensive shipping costs - What matters is whether the purchase is still
economical with them incorporated. I remember the shipping for my SGI
Octane doubled the overall price, but it was still a bargain.
> The Macivory Model 3 was on Ebay in early February of this year.
> At the same time a functional UX400 with Spark and,
> a non functional XL1201 were auctioned together.
> I would say Symbolics go on sale about twice a year in the USA Ebay.
Oh, I see. Thanks.
> In Europe I can not say. If you do a search on US Ebay of Symbolics you
> will
> not see this current auction. I don't know how you search Europe's Ebay
> from
> the USA.
More sex please, I'm English. Colour, centre, largesse, humour etc. Don't
worry, it's a highly reasonable assumption to make.
I noted the same absence when I searched eBay.com. Searching a "foreign"
eBay is easy: You simply append the relevant TLD to "www.ebay" (Exceptions
do apply, though) and browse as normal. Of course, performing a national or
worldwide search from your local eBay site filters auctions that won't
provide international delivery.
"Robbie Blake-Coleman" <········@glowinternet.net> wrote in message
··························@news.demon.co.uk...
>
> Irrespective of the era from which they originated, computers depreciate
> in value and suffer component failure immensely quickly. Overall, it's a
> really dire situation. At the moment I have a failed NeXTstation power
> supply,...
Power supplies are, usually, easy to fix. Most of the time it's a diode.
They either test good or bad with an ohm meter.
Diodes are so simple in function it is easy to find a replacement.
You could find a Ham Radio operator. Most are handy at repairing
electronics.
Tele Repair shop could repair your power supply but, as in the USA, I
assume
Europe, Tele Repair shops are as rear as hens teeth.
> I noted the same absence when I searched eBay.com. Searching a "foreign"
> eBay is easy: You simply append the relevant TLD to "www.ebay"
> (Exceptions do apply, though) and browse as normal. Of course, performing
> a national or worldwide search from your local eBay site filters auctions
> that won't provide international delivery.
Thanks, I had not thought of that.
There still is no way to add TLD to a search in "My Ebay" watch list.
So, I will most likely miss all Symbolics sales in Europe.
> Power supplies are, usually, easy to fix. Most of the time it's a diode.
Oh good!
> They either test good or bad with an ohm meter.
> Diodes are so simple in function it is easy to find a replacement.
> You could find a Ham Radio operator. Most are handy at repairing
> electronics.
> Tele Repair shop could repair your power supply but, as in the USA, I
> assume
> Europe, Tele Repair shops are as rear as hens teeth.
Thanks for the advice G. I'm not very adept at servicing electronics, so
I'm glad to hear that it's a very simple operation (with the right
equipment). Apparently, NeXTstation power supplies are prone to failure
(much like Octane TRAMs), so the number of spares is proportional to the
demand. That said, as I paid a lot for the unit I like the idea of keeping
the maintenance costs at a minimum!
"Robbie Blake-Coleman" <········@glowinternet.net> writes:
>
> Irrespective of the era from which they originated, computers
> depreciate in value and suffer component failure immensely quickly.
> Overall, it's a really dire situation. At the moment I have a failed
> NeXTstation power supply, faulty Octane TRAMs and a semi operational
> Macintosh SE/30 to contend with. I really sympathise with those who
> own machines for which spares are as rare as the units themselves;
> Xerox Alto, Dorado and the Pixar Image Computer come to mind. Last
> time I checked, 80% of the messages in the PDP groups were spares
> related. They have it worse than most of us!
One wonders at what point it might be economical for a manufacturer to
produce new hardware for museum systems. One could even imagine a
manufacturer specialising in that sort of thing...
For my own part, I'd love to have a Lisp Machine to play with, but from
what I've seen it would be a chore to get one working and the final
environment really wouldn't be very practical.
--
Robert Uhl <http://public.xdi.org/=ruhl>
Why Windows NT Server 4.0 continues to exist in the enterprise
would be a topic appropriate for an investigative report in the
field of psychology or marketing, not an article on information
technology. --John Kirch
Robert Uhl <·········@NOSPAMgmail.com> writes:
>One wonders at what point it might be economical for a manufacturer to
>produce new hardware for museum systems. One could even imagine a
>manufacturer specialising in that sort of thing...
There're still several (small) vendors around that specialize in
hardware for the C=64 even. They might not get rich from it but
apparently they still have customers from a number of enthusiasts.
mkb.
Robert Uhl <·········@NOSPAMgmail.com> writes:
> One wonders at what point it might be economical for a manufacturer to
> produce new hardware for museum systems. One could even imagine a
> manufacturer specialising in that sort of thing...
There are companies selling excellent space suits and vehicles
replicas for museums or the movie industry. Last time I checked, an
Apollo suit replica cost around 25,000 bucks.
Kids: don't try this in space.
Paolo
--
Why Lisp? http://lisp.tech.coop/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools:
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- UFFI: Foreign Function Interface
Paolo Amoroso <·······@mclink.it> wrote:
+---------------
| There are companies selling excellent space suits and vehicles
| replicas for museums or the movie industry. Last time I checked, an
| Apollo suit replica cost around 25,000 bucks.
|
| Kids: don't try this in space.
+---------------
On the other hand, the space suit that Kip won in the soap wrapper
jingle contest[1] was a *real* space suit. [Good thing, too.]
Me, I'm waiting for the skin-tights with the force-field helmets... ;-}
-Rob
[1] Robert A. Heinlein, "Have Space Suit, Will Travel".
-----
Rob Warnock <····@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607
····@rpw3.org (Rob Warnock) writes:
> Paolo Amoroso <·······@mclink.it> wrote:
> +---------------
> | There are companies selling excellent space suits and vehicles
> | replicas for museums or the movie industry. Last time I checked, an
> | Apollo suit replica cost around 25,000 bucks.
> |
> | Kids: don't try this in space.
> +---------------
>
> On the other hand, the space suit that Kip won in the soap wrapper
> jingle contest[1] was a *real* space suit. [Good thing, too.]
>
> Me, I'm waiting for the skin-tights with the force-field helmets... ;-}
Picture your fellow lispers in a space suit manufactued by Speedo.
--
~jrm
"Fernando Rodriguez" <···@THOU_SHALL_NOT_SPAMeasyjob.net> wrote
in message ·······································@4ax.com...
> I have an old Quadra and would like try MacIvory incarnation of the
> Lisp Machine. Who is still selling this sort of thing? What software
> is included?
There was this recent post -
http://groups-beta.google.com/group/comp.lang.lisp/browse_thread/thread/0db577c0518822f3/9f56d1b2a3142c1b#9f56d1b2a3142c1b