|
I used to work on mainframe operating systems (GCOS-8) and have seen code dating back to the 70's, IIRC! GCOS was around when they were first writing UNIX - I believe you can still find vestigal remanents of hacks made in UNIX for GCOS support
"Sucks less" isn't progress - Kent Beck [^]
Awasu 1.1 [^]: A free RSS reader with support for Code Project.
|
|
|
|
|
Today everything is .Net and Microsoft have made new programming language for it: C#
Do you have any code in C# older then two or three years? I don't think so...
Microsoft is very good in propaganda and in the near future we'll make Windows programs only in C# (or Visual Basic.Net)
Should we move to Linux, which is backward compatible with old 386's?
|
|
|
|
|
So what?
I have tons of old codepieces, all C++, that are no longer suitable for a) current hardware or b) my current views of"minimum good programming style".
I can't accept a FFT kernel that uses 8MB static arrays in a multi-threaded environment. I have to write an adaptor class that hides all the C-style tricks and feats in the interface.
So our tools change, and so does the code.
"Der Geist des Kriegers ist erwacht / Ich hab die Macht" StS
sighist | Agile Programming | doxygen
|
|
|
|
|
You are right. I don't have any code in a .NET language older than 3 years. However I started commercial software development about 7 years ago and I know that large portions of C++ code that I wrote way back then are still in use today in a current application (which was the question). I no longer work on these applications - but since I keep in touch with the people who do work on it I know that my code is still in use.
My God! I dread to think of all that code - written by me, fresh out of University - still being used. I was so naïve about maintainability and performance then!
|
|
|
|
|
Colin Angus Mackay wrote:
My God! I dread to think of all that code - written by me, fresh out of University - still being used.
I really know that feeling !!
Regardz
Colin J Davies
* WARNING * This could be addictive The minion's version of "Catch "
It's a real shame that people as stupid as you can work out how to use a computer. said by Christian Graus in the Soapbox
|
|
|
|
|
I know this feeling too
C-Code of mine is running for 16 years now.
Parts of this C-code, general purpose
libraries written in the past, are still
in use in modern C and C++ Projekts on
Windows maschines.
This stragtegie with c# and the
accompaning rules i´m quite unhappy
about. Because this means nonportability
of code to other platforms.
That is the reason that we stick to
the old 6.0 Compiler
|
|
|
|
|
Now that we've established that code lives longer than 3 years, what can we as a programming community do to help prevent this type of legislation from taking place? Has anything actually been started to bring about publicizing code - or is it just in the talk phase?
|
|
|
|
|
|
I m using a code that was written in 1992.
Its almost 11 years that i m using the same C++ algorithms.
rgds,
Rana
By the last night i m counting upon the transistors in P IV chip.
|
|
|
|
|
Dear all,
I agree with the argument "Old code works", but one thing is for sure that we should always look forward to find better ways of implementation. Futre / Requirements are so demanding that we can't sit back idle with old code. Computer science is so much evolving that we can always find new options for most of the times.
-- Nikhil Alulkar.:
|
|
|
|
|
hi, i think it is a constantly changing world, so is the IT field, old code, take my word is of no use after some time since new and better approaches/concepts just keep on coming, and in order to keep up with them, we have to keep on changing/optimizing our code.
|
|
|
|
|
In plain words you throw new developments as bone in front of buyer. Apart from to be in competttion it is necessary to have new technology.
- nikhil alulkar.
|
|
|
|
|
I still use the original rounding routines that I wrote back in 1988 using Microsoft C 5.0. Works great, less filling.
I think that if a function or class is still relavant and still compiles properly (and you can guarantee no insulating foam will hit it's wing on launch), there is no need to trash it just because of it's age. See Nish's comments about strcpy.
onwards and upwards...
|
|
|
|
|
|
I wrote a screen management package back in the late 1970s on a custom UNIX desktop machine and then ported the code to MS DOS in the early 1980's. The package allowed interaction with the local screen as well as (optionally) providing an ANSI data stream thru the COM ports for remote access. That code was incorporated (with very little modifications) in several Motorola CPU based RTX imbedded applications about 5 years ago and uses the COM stream to interact with the equipment. The code is in active use at dozens of major US airports.
Steve
|
|
|
|
|
I wrote an application for some healthcare consultants in FoxBase in 1987. Installed it and did some minor debugging. I thought that it had been retired a long time ago. Then last March, I ran into the owner of the company on an airliner. Boy was I surprised to find that they were still using that program! He said that it had just gotten better as the PC's they run it on got faster.
Still in Shock
|
|
|
|
|
If it aint broke, don't mess with it!
onwards and upwards...
|
|
|
|
|
Since my my current product[^] hasn't been around for 2 years yet, I went with under 2.
However, when I last worked on NAV (~3 years ago) there was code that was 10 years old. Well, I should say the comments were 10 years old, not sure if the code itself had been touched. That was the result of the code going through several companies when mergers/takeovers happened, and after that happened a few times, no one knew how to properly maintain it. But it mostly worked, so it was left alone.
--Mike--
"Big handwavy generalizations made from a position of deep ignorance is one of the biggest wastes of time on the net today.
-- Joel Spolsky
Ericahist | Homepage | RightClick-Encrypt | 1ClickPicGrabber
|
|
|
|
|
Michael Dunn wrote:
But it mostly worked, so it was left alone.
Mostly is the key word here.
-Nick Parker
|
|
|
|
|
So how old does code have to be before it's considered "legacy"? I suppose the definition for a product is easier... once a newer version is out then the older version can be called "legacy", but what about code?
"When a man sits with a pretty girl for an hour, it seems like a minute. But let him sit on a hot stove for a minute and it's longer than any hour. That's relativity." - Albert Einstein
|
|
|
|
|
As long as the code can be used without change in new projects.
Legacey: When the code is no longer usefull without change.
INTP
|
|
|
|
|
.... is based on the fact the question asked about codes in a current application, which does not have to be mine.
My company is maintaining codes that are more than 10 years old, written by my CEO, when they used to be programmers. They won't allow us to rewrite or modify it cause they believe it's "perfect"
Oh, Incidently, I'm working in a software house.
Nick Seng (the programmer formerly known as Notorious SMC)
God, I pity me! - Phoncible P. Bone
|
|
|
|
|
...(i.e. >20 years ago) I worked on a DOS/FORTRAN engineering application that's still in daily use today. (The program generates AutoCAD DXF drawing files and bills of materials from data entered into forms.)
I was recently commissioned to perform a face-lift on this app and was surprised how quickly I could find my way around the old source code. It took about 2 days to get back into the swing of writing FORTRAN code after a >15 year break.
It's kind of gratifying to think that code written so long ago is still ticking along in daily use. There are plans in place to use it on another project next year too!
Yours in geezerdom,
Steve
|
|
|
|
|
Exactly! And should that code be forced to be public domain and lose it's copyright? That's what some non-programmers are suggesting publicly because they think source code is obsolete after 3 years.
------------
|
|
|
|
|
Just out of curiosity, are you going to use Lahey/Fujitsu Fortran for .NET to update the app.
The last time I worked with Fortran was back in 1985 and the fact that there is a .NET version is, for lack of a better term, cool.
Konstantine
|
|
|
|