7 Apr
2005
7 Apr
'05
9:25 a.m.
On Thu, Apr 07, 2005 at 11:20:16AM +0000, Jules Richardson wrote:
I wonder if it's actually "whitespace
return" - i.e. a tab and a return
would do it too. I'm guessing someone was trying to optimise storage for
the last line and screwed it up ;)
More likely, "space-return" as listed in the original bug was used
internally as an end-of-file marker or something. Stupid design
decision.
I agree with Sellam to be honest - I think
software's probably worse
these days. Not if you compare it against features maybe, but certainly
if you compare it against the core functionality that you actually use.
Too many code muppets (I hate monkeys) diverted onto piling in extra
junk (and introducing strange side-effects) at the expense of getting
core functionality right IMHO.
The problem with much of today's software is that we are collapsing
under our own weight. It's a fairly obvious cycle: The longer a
product is around, the more a company has to add to it every few
years to keep getting new licenses and generate cash flow. The
upshot is that there is an expectation for all software, existing
and new, to have an incredibly massive feature set to compete in the
marketplace. Having such a massive feature set means more
opportunity for error when creating it, especially when you've got
multiple people creating different chunks of it. (To be fair, I
think the industry has always had the featuritis bug -- it's only in
recent times that personal computers have grown so much that they
are finally at the level where they can physically house such
massive applications -- kind-of like how our collection grows to
fill the space provided for it :-)
Another thing facilitating this is the mindset: Thanks to
widespread adoption of the WWW, cheap broadband internet access, and
Microsoft Windows 3.x and 95, it is now "accepted" in the minds of
most consumers and developers that:
- *Everyone* writes buggy software since it's so complex nowadays
- I can always download a patch to fix something later
- I reboot Windows once a day, why should applications be any different
- It's so easy to contact tech support that it's not a big deal to do so
...etc. Back in the 1970s/1980s, there simply couldn't be bugs in
the software because there was no way to get fixes to consumers.
Luckily, most software back then was of the one-man-wonder variety
and a competent progammer usually didn't make mind-blowing terrible
mistakes. (I agree with Jules in that many bugs of the "collosal"
variety I've encountered with large-team projects are the result of
two people making a minor bug in their own components that,
combined, cause a real problem.)
It's important to remember that there were bugs and upgrades in the
1980s too (although mostly limited to the PC and Mac platforms). I
can remember connecting to Wordperfect and Sierra BBSes to download
patches that were uploaded by tech support to fix bugs in their
products... Were there less of them? Yes. But I just wanted to
point out they *did* exist because too many times I have talked to
people who INSIST that early software was BUG FREE.
--
Jim Leonard http://www.oldskool.org/ Email: trixter at oldskool.org
Like PC games? Help support the MobyGames database: http://www.mobygames.com/
Or taste a slice of the demoscene at http://www.mindcandydvd.com/