[OT] Stupid Apple flamewar (was: SCSI)

Bill McGonigle bill at bfccomputing.com
Tue Nov 29 02:57:00 EST 2005


On Nov 29, 2005, at 01:12, Benjamin Scott wrote:

>   In 1982, Torx was a "security screw driver".

Well, Torx was developed in the mid-60's for the automation market.  
It's designed not to cam-out which is essential for factory assembly.  
Philips is designed to cam-out and standard is too hard to align for a 
robot.  The Mac 128 was among the first computers to be built on an 
automated assembly line.  But the lack of a Torx bit in the home 
toolbox wasn't the motivation.

>  Apple said you had to use a special (and surprisingly complicated) 
> tool.  Since Apple said it, it must be true, right?  Or does Apple 
> lie?  ;-)

And you change your oil every 3000 miles like GM says (or does GM lie)? 
  The Apple tool was designed to prevent marring of the case.  They 
sucked to use, but allowed the machine to return without the gouges 
mine always got.

>> Just so we know we're complaining about 1987-era stuff, under 
>> since-ousted leadership.
>
>   Riiiiight.  "Meet the new boss, same as the old boss."

Wow, Scully ran the company into the ground ... would you contend that 
what's happening now?  Is IBM still peddling Big Iron and OS/2 on the 
PS/2 or did Lou Gerstner show up and make his Elephants Dance?

>   I'm talking about those weird DIN ports.

As I understood it, Mini-DIN 8 was one of the connectors specified in 
the RS-422 spec.  I've only read that reported - I haven't read the 
spec myself.

>   Whack job funky square metal connector that needs a dongle for all 
> three of those.  Had a symbol that looked like this (IIRC):
> 	<----->

Right - mini-AUI - as one of the first PC motherboards to include 
Ethernet standard (1991 - I paid $300 for an el-cheapo NE-2000 clone 
the same year) they weren't going to make 3 computer motherboards/part 
numbers for each of the then-popular physical layers.  So they made a 
superset connector which could be broken out into any of the 3 (if you 
even needed Ethernet).  Other possible choices: include all 3 on the 
motherboard ($,fat), make people install a network card (thus negating 
the cost savings of the onboard chip) - decide to standardize on one of 
the connectors and eschew the other customers (10Base-T wasn't the 
clear winner yet).

>   There were slots that only accepted certain classes of cards in some 
> Mac models, so you had to check obscure compatibility lists.

Didn't we just play "3.3V/5V/33Mhz/66MHz/32-bit/64-bit" PCI Jeopardy 
here last week? ;)

>   Nobody in DC passed any laws requiring Apple to sue anybody.  Apple 
> filed the lawsuits purely of their own volition.  I don't sue you 
> under the DMCA for quoting my message text[1], because I'm not an 
> asshole[2].  Apple does.

Do you sue people who infringe on your patents or do you allow them to 
infringe on them?  That's how life is in the patent business.  Or do 
you prefer submarine patents?  Notably, they didn't sue the author of 
the article in Dr. Dobb's Journal reverse-engineering the iPod database 
format.  So it's not about reverse engineering it's about patent 
infringement.

>> Absolutely.  Still doesn't make it altruism.  :)

"There are three constants in the Universe - Matter, Energy, and 
Enlightened Self Interest."  Are we insisting on altruism from 
everybody now as the only acceptable motivation?

>> Yes, but sometimes you have to buy something ...
>   Not the point under discussion.  :-)

Well, I've long since forgotten what that might have been.  But you go 
ahead and run linux on your imaginary hardware. ;)  I'll buy mine from 
companies who plow millions of dollars a year into Open Source 
development.

>> So, should Apple have given away Firewire?
>> It's hard to encourage innovation with that model ...
>   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>   *eyes bug out*
>   *head twitch*
>   Did I wander into comp.os.windows.advocacy by mistake?
>   What happened to game theory?
>   Are you sure you're not standing within range of an Reality 
> Distortion Field?

Wait - are we advocating hardware companies give everything away too?  
Where's the 'service' angle - give away computers and charge for 
technicians to fix them?  Hardware has a non-zero cost of distribution 
and duplication, it doesn't fit the mold.  I thought we were still 
under the "we need hardware to run things on" impression.  Are we also 
pretending that patent licensing isn't an integral part of the computer 
hardware business?

>   Yah, just like PCI failed because it wasn't backwards-compatible.  
> Oh, wait, no, it succeeded like nothing before.

PCI _was_ backwards compatible - you could have an ISA bus alongside 
it.  That was a design requirement.  Not so with MCA.  MCA was about 
control.

>   MCA failed because IBM wanted to charge you bucks to be able to use 
> it. So the only company using MCA was... IBM.

And Tandy, NCR, Olvetti, some others... (notably not Compaq).

>   Meanwhile, the rest of the industry got together and came up with an 
> alternative that, while not as elegant, was free.  So everybody 
> started making EISA/VLB cards.  Now MCA was not only not helping IBM, 
> but a liability.

In my alternate reality, if MCA could sit alongside an ISA bus and IBM 
had implemented that, EISA/VLB would not have been undertaken.  People 
were _tweaked_ that IBM wanted them to repurchase all their peripheral 
cards, whether they were available or not.  Folks were pumped for the 
big "PS/2" announcement and then read about the MCA bus and said, 
"cool! - but where are the ISA slots?"  Then there was lots of "WTF"?  
So a bunch of EE's sat down and did EISA, then when that was way too 
expensive they tapped the 486 memory controller to do VLB.

> -- Ben "chmod a-x ~/bin/tact" Scott

Penguins have thick skin! [insert blubber joke]

-Bill
-----
Bill McGonigle, Owner           Work: 603.448.4440
BFC Computing, LLC              Home: 603.448.1668
bill at bfccomputing.com           Mobile: 603.252.2606
http://www.bfccomputing.com/    Pager: 603.442.1833
Jabber: flowerpt at gmail.com      Text: bill+text at bfccomputing.com
Blog: http://blog.bfccomputing.com/




More information about the gnhlug-discuss mailing list