How deeply embedded are those chips?

greenspun.com : LUSENET : Electric Utilities and Y2K : One Thread

This article is by a mainframe programmer who worked in the field since 1967. It is not a hoax. She worked at Leeds Northrup R&D from 1967 -1971, then got a degree in Computer Science from the Moore School of Electrical Engineering (where the computer was invented) at the University of Pennsylvania. From 1975 - 1996 she was a manager of software development projects. In 1996 she left the software field permanently.

http://www.drivezero.com/herbal/framec/lampoilc.html

-- Anonymous, July 08, 1999

Answers

I have seen this letter posted before on other sites. It was challenged as to accuracy of the technical statements, specifically there were some posters who felt it was not correct enough to be the report of someone who had been in the IT field for so many years. So, while I personally take the embedded chip issue very seriously, I'm not sure this letter reflects actual experience in that area. Beware.

-- Anonymous, July 08, 1999

Anyone who knows anything about how the grid works, or generators, or chips knows that this is complete B.S. I think whoever wrote this was confused about the Mhz of chip (it's internal clock so to speak) with an external clock.

This goes back to the story that was posted around about chips having different embedded clocks, depending on where they were manufactured around the world. Complete Hooey.

-- Anonymous, July 08, 1999


Complete horse-hockey.

There is plenty of mistaken and exagerated content in that article but this is completely bogus:

"And two lines of High Level code would replace two hundred lines of Assembler code, which replaced two thousand lines of Machine Language code."

This works out to: 2 lines of high level code equals 2000 lines of machine language. This is utter nonsense.

Two lines of high level code might replace ten lines of assembler code. Quite often assembler code maps one-to-one with machine code. In complex machine language instruction set architectures such as DEC VAX or Intel x86 one assembly language instruction may sometimes equate to 5 bytes of machine code (but it's still one line).

In reality 2 lines of high level code would equal between 2 and 50 lines of assembly code. Usually closer to the lower number.

-- Anonymous, July 08, 1999


Sorry, you can't discredit the paper based on 2 lines of high level code = 2000 "lines" of machine language. First machine language has no lines. It is measured in words that are sized to the processor, e.g. 32bit or 64bit processors. Second, one line of high level code could easily translate into 2000+ machine instructions. Have you ever traced a FILEWRITE type of command? How about a CREATEWINDOW? or maybe a CREATEPROCESS? Some of these commands execute 10 of thousands of machine instructions. Be very careful of your facts!

-- Anonymous, July 09, 1999

I saw this posted one other place. The misc.survivalism forum. From what I read, it wasn't so much refuted as it was sniped at.

I saw no valid refutation of this other than some minor pettiness.

-- Anonymous, July 09, 1999



"And two lines of High Level code would replace two hundred lines of Assembler code, which replaced two thousand lines of Machine Language code."

Anybody writing about coffee in terms of programming is obviously trying to write for a computer illiterate audience. So, I suspect she was just trying to convey in concrete terms the idea that higher- level languages simplify coding by an order of magnitude, more or less. Nothing wrong with that, except she should have made it quite clear that she was doing so; otherwise, as we see, one invites irrelevant objections.

-- Anonymous, July 09, 1999


i agree with lane. i believe that she was targeting her paper towards jane q. public. i could be wrong but i think the first time that i saw that paper was at a site that also referenced y2k recipes, etc., and i believe it was her website.

one of the most difficult aspects of conveying the seriousness of the y2k issue to the general public is that it is a technological problem and is largely incomprehensible to those with little or no exposure to the it industry. now, having said that i know that i am going to get slammed from somewhere, by someone who is going to argue that it is also a management problem, etc.

what i am referring to here is understanding the problem itself... not the repercussions to society nor who is responsible for said problem.

also, i am not referring to the logical extrapolations that may be drawn due to the inherent weaknesses of the fragile interconnected infrastructure upon which our modern society resides.

i have received email from people that lurk at this forum and a commonly repeated theme is the difficulty in following, not only the it issues but the it issues as they impact the electrical industry.

long, laborious, tedious, and detailed technological explanations do not go a long way in reaching the psyche of the american public or... any public for that matter. those types of explanations are offputting and intimidating to the average person.

as a very wise man once said... keep it simple stupid.

-- Anonymous, July 09, 1999


Computers are driven in their guts by an Array Table. Array Tables are driven by clocks. .... computers are driven by their clock.

The computers needed a means to calculate position. They needed a measure of linear progress. We were young. We were brilliant. We knew the DATE would make a fine progress array table.

Is this the part that's bullcrap? If so, in what way is it false?

-- Anonymous, July 09, 1999


Moderation questions? read the FAQ