Ben, you are not alone. Have worked in IT for more years than I may care to
admit. Windows-based IDE's are for wimps . . . REAL programmers write
assembler, and use a line editor on a VT100 or Tektronix 4014! (OK, so
later we wrote Cobol/Fortran with vi or EMACS, connected to a "thicknet"
cable . . . and I do mean THICK-net, as that cable looked like a big orange
snake!)
Any DoD realtime developers left out there? Cut my software teeth writing
8-bit octal code for DEC PDP-8's (paper punch tape and all) and 16-bit
assembler/mnemonic interrupt handlers for UYK/SEL/Gould guidance and flight
data collection systems. Thought we were living large when the then-massive
Winchester 80MB platter drives became available. Wow. Later graduated to
assembler and Fortran utilities for Univac 1100's -- in "elements", of
course. And you are correct . . . 70K lines is nothing . . . try debugging
a few MILLION lines of code.
Our younger members might also be surprised to hear the SEI CMM/CMMi
standards grew out of earlier work by Mitre (defense contractor) and the
infamous DoD-STD-2167A. Many aspects of IT have a MUCH longer history than
you may imagine. Take a long look back if you want to appreciate how easy
we have it today.
-- Bing
Bradford Bingel ("Bing")
ITM3 California
http://www.itm3.com/
[EMAIL PROTECTED] (email)
925-260-6394 (mobile)
-----Original Message-----
From: Action Request System discussion list(ARSList)
[mailto:[EMAIL PROTECTED] On Behalf Of Ben Chernys
Sent: Thursday, July 10, 2008 5:27 AM
To: [email protected]
Subject: Re: "Lines of code" in Remedy?
Hi Dan,
You are correct, cc 72 was left blank in all languages unless you wanted to
continue the line. If non-blank (any character) the line was continued.
73-80 was ignored by the compilers and generally used to sequence the cards.
So, perhaps I have dated myself a tad now. I still use the same editor that
came out just after the punch card era. And, if you know "pig iron", you
can still get a job supporting, developing, etc
And, in those days we had a proper change management application!
As for lines of code: Meta-Update is currently 70K lines. I've worked on
systems with over 5M. I prefer to ask what ratio of full line comments do
you have in your code? I average around 25%. Anything above 1% is above
average!
Someone writing Java here told me converting a date (using the Remedy
function) was a single line. But when I looked (not believing him) I
counted about 10 lines.
Of course, in most languages, you could simply get rid of the lines and end
up with one very very long line.
As for "decompiling", you end up with machine code, NOT source code, no
comments, no headers, etc, so you cannot determine the number of lines that
way.
Happy (early) Friday.
Ben Chernys
>----- ------- Original Message ------- -----
>From: Daniel Bloom
><[EMAIL PROTECTED]>
>To: [email protected]
>Sent: Thu, 10 Jul 2008 07:55:49
>
>Okay, I will reset the clock on the list server so we are now Friday.
>>From the original question, any Friday in the
>1970s or 80s.
>
>Don't forget to use 71 character lines (if I remember correctly and
>probably don't, column 72 was for an X to say this line is extended for
>at least Fortran and the rest were for sequencing your card deck in
>case they fell off where you put them and spread themselves over the
>floor).
>
>Anybody who knows the correct answer has *really* dated themselves.
>For the rest of you, I am going back 32 years, the first and last year
>I used a card punch.
>
>So David, bundle up all the responses from the arslist, bind them, Pick
>a number(as recommended by your peers, either random number or An
>inaccurate calculated one), attach a printout of the .def file and All
>supporting code from mid-tier, integrations etc. and hand it in :-)
>
>... Dan
>p.s. has everyone requested funding for the BMC UserWorld in Miami?
>-----Original Message-----
>From: Action Request System discussion
>list(ARSList)
>[mailto:[EMAIL PROTECTED] On Behalf Of Bradford Bingel
>Sent: April 29, 2003 6:13 PM
>To: [email protected]
>Subject: Re: "Lines of code" in Remedy?
>
>Geez . . . no one has used the "lines of source code" (SLOC)
>measurement since the 1980's! It was a poor metric then with
>monolithic languages (Cobol, Fortran, etc.), and it's an even poorer
>metric today using object-oriented software and N-tier architectures.
>
>But you may still need to provide a valid number.
>Can anyone from Remedy
>provide a ballpark SLOC metric by application?
>
><history snip>
>
>___________________________________________________
>____________________________
>UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Platinum
>Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are"
____________________________________________________________________________
___
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Platinum Sponsor:
www.rmsportal.com ARSlist: "Where the Answers Are"
_______________________________________________________________________________
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are"
