1 mloc

What's your point? I've made my living designing and building software products this size and larger. No, they weren't bloated, they just tackled large problems. And most of the numbers postulated are crap, BTW.

Do I have to have a point?

John

Well, you don't normally post random crap, so you must have thought that someone somewhere would derive some benefit from reading that article. My question is who, and what benefit did you envisage?

Actually, I was sort of hoping to be nominated for the Nobel Prize.

John

Are you claiming you're pointless?

Robert

--
Posted via a free Usenet account from http://www.teranews.com

Aren't we all?

John

Now you've got me flashing on Calvin and Hobbs (and a Tiger sleeping with his points up)

Robert

--
Posted via a free Usenet account from http://www.teranews.com

As a friend remarked about cats, you've got to be careful around any critter that's sharp on 5 out of 6 of its ends.

John

At the DreamWorld theme park on the Queensland Gold Coast, they used to walk a tiger around the park. One keeper walking in physical contact with each side of the tiger's neck, and one more 20 feet in front saying "Stay away from the sharp end, go around the back if you want to pat the tiger" continuously in a loud voice. This with a ten foot long white tiger which all three together would have had no chance of restraining. Exciting... but they don't do that any more. Wonder why :-)

BTW, I'm still genuinely mystified why you posted the OP. It wasn't frivolous, so you must have had something in mind?

You haven't lived until you've had a cat panic and run to you for safety, but climb right up your body like you were a tree... didn't stop until he was perched on the top of my head... man, did that hurt ;-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
         America: Land of the Free, Because of the Brave

I just thought it was interesting. I've dealt with a lot of bad embedded code lately, on otherwise good electronics.

John

You didn't write the million lines of code. And you didn't have to live with the results (customers did). Now, if the vast majority of your code was dearly-paid-for and heavily- sweated-over IP, that may be another matter. But if not, I'd say you may well be part of the problem, not part of the solution.

Every item of consumer electronics I've ever used more than a couple of times has had software defects, even where it's doubtful that they contained more than 200 lines of code. Here I mean microwaves, VCRs, watering timers, etc. Every single item. Small defects mostly, or usability ones, but defects nonetheless.

Software is hard, and the hard part is to accurately define correctness. Once you can do that, you can test for it. The definition is easier for hardware, though implementing and testing for it is much harder. I have huge respect for the engineering standards exemplified by the EE community, but though their ragging the SW community isn't without justification, there are genuine differences in the disciplines.

Your idle and ignorant speculations are noted.

A visit to

is quite eye-opening with regards to the quality of code that ends up in some products. They seem to focus more on "Internet code," but I suspect that the quality level is little different for embedded software. I've personally witnessed code like this:

if (num == 2) ch = '2'; if (num == 4) ch = '4'; if (num == 6) ch = '6'; ...etc... being used in a Real Product! (Odd numbers were "not supposed to be used" in calling this function... although there was zero effort to detect or enforce that requirement...)

Granted, that code isn't actually buggy, but it definitely sets off my sense of, "oh oh, I just know the rest of the code is going to have problems..."

One difference I notice a lot is that with hardware you spend most of your time making the widget do what it's *supposed* to do, and going back and adding, e.g., over voltage/current/power protection is usually a fast and easy thing. With software, it often seems as though there's almost as much "stuff" (code) that validates inputs and tries to do something "reasonable" with bogus inputs as there is stuff that does the real work.

---Joel

I dunno what John had in mind, but the post evokes the thought of how software has evolved - from lines of code, to k lines of code, to m lines of code to ?? I think there was a magazine maybe 30 (?) years ago that had contests to write something useful in 64 bytes. Maybe someone who knows will chime in here.

Anyway, John's post was interesting to me - I can't speak for others.

Ed