Re: Report?


Subject: Re: Report?
From: Arthur Corliss (arthur@corlissfamily.org)
Date: Wed Jul 03 2002 - 07:07:51 AKDT


> I got into an argument today about Linux vs Micro$oft with my
> boss. He wants me to write him a report on 'Why Linux is Better'
> Some ideas would be appericated. (Besides the common 'More
> Stable' 'Free' etc.)

<G> The fact that the upcoming SP3 for Win2K adds almost *800* bug fixes
isn't enough (and since the patch set has been frozen for awhile now,
it's already out of date)?

How about the fact that MS' programmers continually demonstrate their
incompetence? I've been pretty low profile on the list because of time
constraints, but I'm ripe for a rant, so here goes:

Anyone here work with automation objects in WSH? The network object, as
an example, includes methods that allow you to programmatically map
network drives, etc. Those boobs decided that an return value on the
method was just too damned hard (read: two lines of code, max), so you
get no immediate feedback as to whether or not the mapping failed. And
to add injury to insult, the only way to determine this is to grab their
net drive collections object and iterate over each one to see if it's
present. But then the idiots use misleading methods like 'Count', which
isn't the number of mapped drives, it's the number of elements in an
ordinal array that the entire collection consists of, with each drive
being two elements (local mapping, remote location). So what should have
only taken them two lines of code now costs me about ten (depending on
the language used) in *every* script that needs that safety check.

Or lets talk about their brain-dead attempt to be POSIX compliance.
NTFS, which is .1 compliant, supports a lot of features that the OS knows
nothing about. Hard links, for instance, are supported on the
filesystem, but the apps themselves have no idea what to do with it. The
filesystem supports case-sensitive names, but if you tell Windows to
delete "Foo", it deletes "foo", "FOO", and any other combination of case
along with it.

Or what about ACLs on shares? They had to complicate things by allowing
you to define specific file ops on the *share* in addition to the
*filesystem*. ACEs are cummulative in one context or the other, but
combined, only the intersection of rights apply. WTF? Why on earth do
we need two sets of ACEs? Why not just control who can make a remote
connection and let the filesystem ACL do its job?

And what about these geniuses that write languages incapable of testing
for Null directly in a conditional statement (you actually have to use a
separate *function* to do the test)?

I could go on, but I think you get the point. . .

--

--Arthur Corliss Bolverk's Lair -- http://arthur.corlissfamily.org/ Digital Mages -- http://www.digitalmages.com/ "Live Free or Die, the Only Way to Live" -- NH State Motto

--------- To unsubscribe, send email to <aklug-request@aklug.org> with 'unsubscribe' in the message body.



This archive was generated by hypermail 2a23 : Wed Jul 03 2002 - 07:56:34 AKDT