Re: FW: WTT: 1.5G of PC2700 for 1G of PC100

[Joshua Boyd]

On May 4, 2008, at 4:16 PM, Shannon Hendrix wrote:

> On May 4, 2008, at 12:47 , Carl R. Friend wrote:
>
>
>>   I'm fighting a rear-guard battle at $PPOE to retain Solaris-on-
>> SPARC that I expect to lose in the next year or so.  Sun's pricing
>> on the SPARC gear is out of line -- performance-wise -- with what
>> can get done on an x86 system, and my argument about not wanting
>> to jump on the monoculture bandwagon doesn't seem to hold much water
>> in a world of completely non-technical Management.
>
> I fought that battle for a long time and basically lost.
>
> Even Apple has gone to the dark side.
>
> Pretty much 99.999% of the industry is clueless about exactly what  
> it means to lose multiple platforms.
>
> Almost no living programers today would even know how to write code  
> for a machine that was not based on 8/16/32/64-bit 2's compliment  
> CPUs and not too many more have any idea how to deal with a machine  
> that is not Intel based or running a video console.

Writing code for 18 or 36bit  may be mostly a lost art.  But there  
are many many programmers writing code for many platforms still.  I  
see no end of work being done on both 8 and 32 bit chips, and I see a  
lot of people working on non-x86 platforms with no graphics.

I would imagine that the leading platforms are probably actually ARM,  
PPC, x86, then MIPS.

> This is only the tip of the iceberg though... a good portion of  
> software today WILL NOT EVEN RUN without a network connection.

I just saw an amazing example of that.  A box that should just be a  
video appliance, but it requires an internet connection to verify  
it's license with the manufactor every boot.  And this was a device  
that is supposed to be going into flight packs and mobile trucks.  Is  
every news truck that the local station puts out really going to  
supply an internet connection to a video processor?
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Geoffrey S. Mendelson]

On Sun, May 04, 2008 at 04:16:36PM -0400, Shannon Hendrix wrote:

> Almost no living programers today would even know how to write code  
> for a machine that was not based on 8/16/32/64-bit 2's compliment CPUs  
> and not too many more have any idea how to deal with a machine that is  
> not Intel based or running a video console.

I don't know, It's been a long time, but I still could probably dredge
up with a few days time one of, BAL (IBM 360/370) (and channel programs),
COMPASS (CDC-6400) (and some familarity with the IO processors),
IBM 1130, and if I really had to HP 2100. The HP 3000 and Buroughs used
Algol, and I probably could dig that up too, though I don't think I wrote
Algol code for them, but I seem to remember writing it for something.

I might even be able to figure out again Comodore 64 (6502) and
Z80 assembly language too having written a few lines of code in them.

Gone forever are passing aquaintance with SHAL-A (Philco 1000/2000)
and 1401 Autocoder, which I looked at but never really wrote anything.

On the other hand, some of the newer processors I've written code for
such as the 8088, 680x0, PPC, AT&T 3B2, 80386 onward, and so on
was done in C, and I can't seem to remember much differences between
them that the C compilers did not take care of.

I'm not sure besides "folk tales" if they really mean much today. If I
were going to pitch myself as a BAL programmer who could write
300 lines of tested and commented code a day on a long term project
or a PERL/Bash code monkey, you can guess what I would do. :-)



> This is only the tip of the iceberg though... a good portion of  
> software today WILL NOT EVEN RUN without a network connection.
>
> That's just pure stupidity.

No, it's good business. :-) Think of it as pay-per-view software.

Remember Sun's "the network is the computer"? That's the business
model with 99.99% "local offloading". IMHO the worst of both worlds.

Geoff.


--
Geoffrey S. Mendelson, Jerusalem, Israel gsm@...  N3OWJ/4X1GM
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Geoffrey S. Mendelson]

On Sun, May 04, 2008 at 04:22:54PM -0400, Joshua Boyd wrote:

> I would imagine that the leading platforms are probably actually ARM,  
> PPC, x86, then MIPS.

I don't know if there are any hard figures on that. There is still work
on Z8, Z80, anf 6800 derivatives in the embedded market. Plus that whole
line of TI DSP chips that someone figured out could be used as a cheap
low power CPU.

However since the low power x86 market has taken off, it's only a matter
of time IMHO that the other chips fade away. Eventualy it will just be too
expensive to maintain a design, implementation and programing staff for
more than one architechture.

I know an ARM chip can do on 2 AA batteries what a low power X86 requires
a lion battery with about 1/4 of the time, but that won't last forever.
Every year the ARM chips get faster and more power hungry, the batteries
get better and the X86's become more frugal.

Geoff.

--
Geoffrey S. Mendelson, Jerusalem, Israel gsm@...  N3OWJ/4X1GM
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Mark Benson-2]

On 4 May 2008, at 21:16, Shannon Hendrix wrote:

>
> On May 4, 2008, at 12:47 , Carl R. Friend wrote:
>
>>  I'm fighting a rear-guard battle at $PPOE to retain Solaris-on-
>> SPARC that I expect to lose in the next year or so.  Sun's pricing
>> on the SPARC gear is out of line -- performance-wise -- with what
>> can get done on an x86 system, and my argument about not wanting
>> to jump on the monoculture bandwagon doesn't seem to hold much water
>> in a world of completely non-technical Management.
>
> I fought that battle for a long time and basically lost.

Damn Apple, even thos bastions of supercomputing, Cray, are forming an  
Alliance with Intel:

http://arstechnica.com/news.ars/post/20080429-in-depth-intel-cpus-gpus-and-upcoming-cray-supercomputers.html


> Even Apple has gone to the dark side.
> Pretty much 99.999% of the industry is clueless about exactly what  
> it means to lose multiple platforms.


where $fittest = "Makes most money" {
        $_Dominance = survival($fittest);
}

The old evolution thing works in mysterious ways when you warp it  
around business but it still works... Beta was better than VHS, HD-DVD  
had advantages over BluRay (not being owned by Sony was one). They  
didn't all survive for whatever reasons. The reasons are rarely  
rational, but always fatal.

x86 will dominate until something more viable springs up, like  
dynamically recodable CPUs (software as hardware).

> Almost no living programers today would even know how to write code  
> for a machine that was not based on 8/16/32/64-bit 2's compliment  
> CPUs and not too many more have any idea how to deal with a machine  
> that is not Intel based or running a video console.

Present company excepted, I surely hope ;)

>>  Sooner or later, some creative type *will* come up with an x86
>> piece of "malware" that's cross-platform and will target the under-
>> lying CPU; heck it might be in the wild already just waiting for a
>
> You know... in a way I hope someone does.  But them, instead of  
> learning from it, the world would probably react by making things  
> even worse.

I'll be able to check my e-Mail and browse the web from my RS/6000 or  
my U60 for a while yet, I'm not that worried ;)


--
Mark Benson

My Blog:
<http://markbenson.org/blog>
Visit my Homepage: <http://homepage.mac.com/markbenson>

"Never send a human to do a machine's job..."
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Shannon Hendrix]

On May 4, 2008, at 16:22 , Joshua Boyd wrote:

>> Almost no living programers today would even know how to write code  
>> for a machine that was not based on 8/16/32/64-bit 2's compliment  
>> CPUs and not too many more have any idea how to deal with a machine  
>> that is not Intel based or running a video console.
>
> Writing code for 18 or 36bit  may be mostly a lost art.  But there  
> are many many programmers writing code for many platforms still.  I  
> see no end of work being done on both 8 and 32 bit chips, and I see  
> a lot of people working on non-x86 platforms with no graphics.
>
> I would imagine that the leading platforms are probably actually  
> ARM, PPC, x86, then MIPS.

All of those are the same type of CPU, and all of them have words  
which are 8-bit multiples, and they can address memory and registers  
on sub-word boundaries.

For most programmers, the only difference is the ISA and the compiler  
hides the majority of that unless they go a little above and beyond  
the norm in learning their hardware.

> I just saw an amazing example of that.  A box that should just be a  
> video appliance, but it requires an internet connection to verify  
> it's license with the manufactor every boot.  And this was a device  
> that is supposed to be going into flight packs and mobile trucks.  
> Is every news truck that the local station puts out really going to  
> supply an internet connection to a video processor?

For that matter, why should I constantly have to prove I am not a  
thief, and why should I be required to constantly expose my systems to  
security problems.  Why assume I have a network connection at all or  
that I want my end-user systems directly exposed to the net?

All of these and related assumptions are stupid and dangerous, and of  
course, how do I use my system when the remote system it depends on  
goes away?



--
Shannon Hendrix
shannon@...
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Shannon Hendrix]

On May 4, 2008, at 16:38 , Geoffrey S. Mendelson wrote:

> On Sun, May 04, 2008 at 04:16:36PM -0400, Shannon Hendrix wrote:
>
>> Almost no living programers today would even know how to write code
>> for a machine that was not based on 8/16/32/64-bit 2's compliment  
>> CPUs
>> and not too many more have any idea how to deal with a machine that  
>> is
>> not Intel based or running a video console.
>
> I don't know, It's been a long time, but I still could probably dredge
> up with a few days time one of, BAL (IBM 360/370) (and channel  
> programs),
> COMPASS (CDC-6400) (and some familarity with the IO processors),
> IBM 1130, and if I really had to HP 2100. The HP 3000 and Buroughs  
> used
> Algol, and I probably could dig that up too, though I don't think I  
> wrote
> Algol code for them, but I seem to remember writing it for something.

Sorry, what I mean to say is that almost none were trained on it in  
college in the last couple of decades.

For that matter, a good number of graduates today have never even used  
real hardware.

You can get a comp-sci degree without ever leaving Netbeans...

> On the other hand, some of the newer processors I've written code for
> such as the 8088, 680x0, PPC, AT&T 3B2, 80386 onward, and so on
> was done in C, and I can't seem to remember much differences between
> them that the C compilers did not take care of.

That's because they are all basically the same.

Word is 8-bit multiple, 2's complement, and sub-word register and  
memory addressing.

Most students never see anything else and most systems sold today are  
like that.

>> That's just pure stupidity.
>
> No, it's good business. :-) Think of it as pay-per-view software.

No, it's pure stupidity.

Losing access to my systems is not good business, and if I get mad  
enough it won't be good business for the idiot whose head I cut off  
when it happens either.

> Remember Sun's "the network is the computer"? That's the business
> model with 99.99% "local offloading". IMHO the worst of both worlds.

That's fine for things which must be distributed, and also if they can  
be converted to local use.

But I don't need that to access *MY* data, and nothing should ever  
require that.

--
"Where some they sell their dreams for small desires."
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Geoffrey S. Mendelson]

On Sun, May 04, 2008 at 09:58:05PM +0100, Mark wrote:
> x86 will dominate until something more viable springs up, like  
> dynamically recodable CPUs (software as hardware).

I think you are a little late for the patent on that. Wasn't that
done to the SDS 940 when PARC got their hands on it?

Geoff.
--
Geoffrey S. Mendelson, Jerusalem, Israel gsm@...  N3OWJ/4X1GM
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Nadine Miller]

Shannon Hendrix wrote:

> For that matter, why should I constantly have to prove I am not a thief,
> and why should I be required to constantly expose my systems to security
> problems.  Why assume I have a network connection at all or that I want
> my end-user systems directly exposed to the net?
>
> All of these and related assumptions are stupid and dangerous, and of
> course, how do I use my system when the remote system it depends on goes
> away?

All reasons why the underground continues to grow.

=Nadine=
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Geoffrey S. Mendelson]

On Sun, May 04, 2008 at 05:07:38PM -0400, Shannon Hendrix wrote:
> Sorry, what I mean to say is that almost none were trained on it in  
> college in the last couple of decades.

That's for sure.
 
> For that matter, a good number of graduates today have never even used  
> real hardware.
>
> You can get a comp-sci degree without ever leaving Netbeans...

Why not? :-)


>
> That's because they are all basically the same.
>
> Word is 8-bit multiple, 2's complement, and sub-word register and  
> memory addressing.
>
> Most students never see anything else and most systems sold today are  
> like that.

Yes, they all came from the same place. Since the IBM PC came out everyone
uses it as their learning platform.

> >>That's just pure stupidity.
> >
> >No, it's good business. :-) Think of it as pay-per-view software.
>
> No, it's pure stupidity.

That was the point of the smiley :-) It's good in suit-think.


> But I don't need that to access *MY* data, and nothing should ever  
> require that.

What has your data got to do with it? There's no cell in my business plan's
spreadsheet labeled "YOUR DATA". :-)

(Calm down, I'm just kidding).

It's after midnight, I'm going to bed, 'night all.

Geoff.
--
Geoffrey S. Mendelson, Jerusalem, Israel gsm@...  N3OWJ/4X1GM
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Sridhar Ayengar]

Shannon Hendrix wrote:
> Almost no living programers today would even know how to write code for
> a machine that was not based on 8/16/32/64-bit 2's compliment CPUs and
> not too many more have any idea how to deal with a machine that is not
> Intel based or running a video console.

Anyone who writes anything reasonably low-level on mainframes (myself
included) know that stuff.

There's also the argument that someone who makes those sorts of
assumptions aren't really programmers.

Copro-grammers, maybe?

Peace...  Sridhar
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Sridhar Ayengar]

Geoffrey S. Mendelson wrote:

>> Almost no living programers today would even know how to write code  
>> for a machine that was not based on 8/16/32/64-bit 2's compliment CPUs  
>> and not too many more have any idea how to deal with a machine that is  
>> not Intel based or running a video console.
>
> I don't know, It's been a long time, but I still could probably dredge
> up with a few days time one of, BAL (IBM 360/370) (and channel programs),
> COMPASS (CDC-6400) (and some familarity with the IO processors),
> IBM 1130, and if I really had to HP 2100. The HP 3000 and Buroughs used
> Algol, and I probably could dig that up too, though I don't think I wrote
> Algol code for them, but I seem to remember writing it for something.
>
> I might even be able to figure out again Comodore 64 (6502) and
> Z80 assembly language too having written a few lines of code in them.

I'm only thirty years old, and I've had extensive experience writing
assembler code on 370/XA, ES/370 and ES/390 under both VSE and MVS.
(Along with other things like REXX, FORTRAN and JCL.)

We are still around, but many of us have moved onto other things.
Lately I've been writing custom financial applications in Java.

Peace...  Sridhar
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Sridhar Ayengar]

Shannon Hendrix wrote:
> For that matter, a good number of graduates today have never even used
> real hardware.
>
> You can get a comp-sci degree without ever leaving Netbeans...

You can get a comp-sci degree without ever having touched a computer.

Peace...  Sridhar
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Robert Darlington]

On Sun, May 4, 2008 at 6:41 PM, Sridhar Ayengar <ploopster@...> wrote:
> Geoffrey S. Mendelson wrote:
<snip>
>  I'm only thirty years old, and I've had extensive experience writing
> assembler code on 370/XA, ES/370 and ES/390 under both VSE and MVS. (Along
> with other things like REXX, FORTRAN and JCL.)
>
<snip>

Assemblers assemble assembly.  Sorry, couldn't resist.


>
>  _______________________________________________
>  rescue list - http://www.sunhelp.org/mailman/listinfo/rescue
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Geoffrey S. Mendelson]

On Sun, May 04, 2008 at 08:14:22PM -0600, Robert Darlington wrote:
>
> Assemblers assemble assembly.  Sorry, couldn't resist.

At least you did not say that they COMPILE it. I've heard that too
many times. :-)

Geoff.

--
Geoffrey S. Mendelson, Jerusalem, Israel gsm@...  N3OWJ/4X1GM
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Shannon Hendrix]

On May 4, 2008, at 20:36 , Sridhar Ayengar wrote:

> Shannon Hendrix wrote:
>> Almost no living programers today would even know how to write code  
>> for a machine that was not based on 8/16/32/64-bit 2's compliment  
>> CPUs and not too many more have any idea how to deal with a machine  
>> that is not Intel based or running a video console.
>
> Anyone who writes anything reasonably low-level on mainframes  
> (myself included) know that stuff.

My point is the percentage of people like that is so low now, I can  
get away, statistically speaking, with saying they don't exist.  They  
are not the ones driving the industry and its training systems any more.

That's what percent of the programming workforce, well below half a  
percent?

Of course, I also know that this industry has gone in reverse now and  
then, so maybe things will change.

I have noticed that whole lot of local people who "got into computers"  
are now doing other things.  Seems their unemployment rate was quite a  
bit higher than mine.

Of course, that also means that some of us end up doing the same kind  
of horrid work they were doing, or have to fix their messes... :)

> There's also the argument that someone who makes those sorts of  
> assumptions aren't really programmers.

Well, I won't get into where the dividing line is, because technically  
programmer just means "one who creates programs."

The definition doesn't say they have to be good.

I'd be happy if more of them would simply start giving a damn about  
their job.

--
"Where some they sell their dreams for small desires."
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[der Mouse]

>> You can get a comp-sci degree without ever leaving Netbeans...
> You can get a comp-sci degree without ever having touched a computer.

Which is as it should be; computer science is a theoretical discipline
(though there are far too many schools teaching what would more
honestly be called programming, or occasionally software engineering,
and mislabeling it "computer science").

Of course, acquaintance with the practice usually makes for a better
theoretician, just as acquaintance with the theory usually makes for a
better practicioner.  But in neither case is it necessary.

/~\ The ASCII der Mouse
\ / Ribbon Campaign
 X  Against HTML       mouse@...
/ \ Email!     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Geoffrey S. Mendelson]

On Sun, May 04, 2008 at 08:41:19PM -0400, Sridhar Ayengar wrote:
> We are still around, but many of us have moved onto other things.
> Lately I've been writing custom financial applications in Java.

Honestly, how much does knowing how the "bare metal" works help you
in writing code that runs in a virtual machine, on a different computer?

There are some things that I learned from programming in Assembly language
that make my code better than most code written today. Slower, but better. :-)

1. Always initialize variables.

2. Validate the parameters to a system call before you call it.

3. Validate the results of a system call (as in don't use a NULL pointer).

4. Always make sure the data fits in the space allocated for it.

5. Always make sure an I/O operation succeeds.

6. Put lots of debugging statments in your code.
   
7. Comments, comments, comments.
 
8. Write as simple code as possible. If it is difficult for someone to
   understand, it will be impossible to maintain. (KISS).

Geoff.

--
Geoffrey S. Mendelson, Jerusalem, Israel gsm@...  N3OWJ/4X1GM
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Shannon Hendrix]

On May 5, 2008, at 00:43 , der Mouse wrote:

>>> You can get a comp-sci degree without ever leaving Netbeans...
>> You can get a comp-sci degree without ever having touched a computer.
>
> Which is as it should be; computer science is a theoretical discipline

No, it is not.  It is an applied science just like medicine or  
engineering.

A computer science graduate who has never touch a computer is like a  
doctor who has never touched a human or an engineer who has never  
touched a screwdriver.

--
"Where some they sell their dreams for small desires."
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Shannon Hendrix]

On May 5, 2008, at 00:50 , Geoffrey S. Mendelson wrote:

> On Sun, May 04, 2008 at 08:41:19PM -0400, Sridhar Ayengar wrote:
>> We are still around, but many of us have moved onto other things.
>> Lately I've been writing custom financial applications in Java.
>
> Honestly, how much does knowing how the "bare metal" works help you
> in writing code that runs in a virtual machine, on a different  
> computer?

A tremendous amount.

--
Shannon Hendrix
shannon@...
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[der Mouse]

>> Which is as it should be; computer science is a theoretical discipline
> No, it is not.  It is an applied science just like medicine or
> engineering.

That's a substantial part of the problem, then; different people
understanding the same words to mean different things.

What do _you_ call the theoretical discipline, then?  The one to which
belong things like the proof that general sorting cannot happen faster
than n log(n), the Halting Problem, et al?

/~\ The ASCII der Mouse
\ / Ribbon Campaign
 X  Against HTML       mouse@...
/ \ Email!     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue