September 9th, 2003 07:21 PM
The Problem with Programmers (rant)
Computers and computing has enjoyed unprecedented growth for the last three, going on four decades. This growth has lead to both an arrogance and lack of reflection, both of which are readily apparent in virtually all modern operating systems and by leafing through any of the countless open source project forums. No one wants to learn from the past it seems, and so were are left with each generation of programs getting no further than the previous generation. Sure the languages change, way back when it was Modula-2, yesterday it was C++, tomorrow itíll be OCML?
IS security history really should be taught in the CS/IS/IT programs, but is it? No of course not. Neither is historical design. What are we left with? NT has inferior ACLs to MULTICS and FreeBSD now has inferior ACLs to NT. Doesnít anyone learn?? The need for security kernels was realized back in 1973, yet I donít anyone who is reading this runs a system with on one it. Automated intrusion detection, the dangers of insufficient argument validation were all known. People talk about Automated intrusion detection like it is some new thing (in fact such a post on this forum inspired this) and does a week go by without reading about a bof or god knows what on bugtraq? Either all of this software is over 30 years old, or really there is no excuse at all.
Perhaps it isnít the programmersí fault, perhaps it is the education system. I donít recall a single utterance of things like ďfinite-sate system.Ē However, it seems with open source, so very few programmers these days actually went to school for such things.
Letís seeÖ bashed on NT and FreeBSD who is left? Linux developers of the world, listen carefullyÖ no matter how much tweaking you do, no matter how many nifty tools you make, not mater how many women in bikinis you photoshop tattoos of Tux on, isnít going to make Linux a modern OS. It has a flawed core design that can never be verified, why not spend your time elsewhere that might actually be useful? Oh and one more note about FreeBSD, you guys are not as inventive as you think, shared memory segments, network firewall, and typed files all existed well before BSD4.2. Oh and Microsoft, Palladium? What the hell are you thinking?
Anyhow programmers, before your next project, please do a little research and see if it has already been done, if so why not work on something else. I know itíll take some extra effort to be creative, but Iím sure you can manage.
(watch this earn me heaps of negative points heh)
September 9th, 2003 07:57 PM
The basic problems that I have noticed from working with newly qualified IT people is that :
1. they don't seem to have been given any sort of exposure to the history of the Industry.
2. There seeems to be an emphasis on the latest technology/marketable skillset, rather than underlying principles.
The end result is the system seems to churn out five year wonders, rather than true professionals. All very good for the people running certification courses etc. but I wonder if it is good for the Industry long term?
just my £0.02
If you cannot do someone any good: don't do them any harm....
As long as you did this to one of these, the least of my little ones............you did it unto Me.
What profiteth a man if he gains the entire World at the expense of his immortal soul?
September 9th, 2003 09:33 PM
Part of the problem is a lot of the programers out there went to tech schools insted of universities. But some programmers are quick to forget what they learn. A lot of MS guys come from U of I, and I know they learn about the evils of bof's and how to think in terms of finite state machines there.
Who is more trustworthy then all of the gurus or Buddhaís?