One example - I've heard OpenOffice (maybe StarOffice) is going to or
already supports MS Office Macros which has been a vulnerable area.
Depends how they execute them. If they merely interpret them and not
try to recreate all the capabilities in terms of system access, it
might not be a problem. Also the file system permissions will still
be in effect which will limit a rogue macro's impact.
Also, there have been accounts of people become susceptible to Windows
based viruses by running Windows apps under WINE.
Here's a comment about the Sir Cam virus running under WINE I just
found via Google via a Vnunet article:
=========================================================
Reports emerging over the weekend have confirmed that the Sir Cam
virus, which spread across the internet throughout July, runs under
Wine.
However, it has been noted that because of the way Wine is
constructed, Sir Cam is unable to create the relevant registry entries
to make itself relaunch at boot. And for users of Wine without a
Windows mail client installed, the virus is unable to mail itself out
to names in the address book.
One Wine user commented: "The effect was that Sir Cam was exposed but
not functional, and I was able to explore its code without fear. There
were no registries to infect, no exchange list to exploit, and the
'hidden' Trojans were easily seen and removed. Sir Cam is totally
harmless on Linux under Wine."
The discovery caused more than a few chuckles amongst the die-hard
Linux community, with many joking that the only way Linux can run
viruses is under emulation, "otherwise it wouldn't have any."
Another user commented: "Wine supporters have finally ported the
single most popular Windows application to Wine. It took a lot of work
and years of research and determined effort, but it can finally be put
to rest. Yes, thanks to the efforts of hackers worldwide, Linux is now
capable of running Virus programs designed for Windows."
========================================================
I think it's a losing battle to try to take
market share from an OS by providing emulators to allow users to do
this. As good as some emulators are, they typically are noticeably
less stable and more sluggish than the real thing. I think a
brilliant example of how this approach fails is OS/2.
WINE - as the acronym title states - is not an emulator. It is a
reverse-engineering of the Windows APIs to work on Linux. I don't
think the point is to take market share - although the obvious
complaint most non-techie users have about Linux is it won't run
Windows programs. Well, it will run a lot of them. The entire
Microsoft Office suite through the XP version runs using the (not
freeware) Codeweavers CrossOver Office utility which is an enhancement
of WINE. The point is to allow Linux to run the scores of thousands
of Windows software that is out there. There are also emulators for
Linux for virtually every other OS that has ever existed - from CP/M
to MacIntosh and videogame OS's.
Also you can't compare OS/2 to Linux in this regard for two reasons:
first, although OS/2 had its own applications, it was pretty much all
closed source. Secondly, IBM's horrendously bad marketing was the
cause of OS/2's "demise" (it's still around somewhere, isn't it?)
Remember, too, that originally OS/2 was a combined IBM-Microsoft
project. OS/2, when it became an IBM-only product, was technically
superior to Windows - but IBM couldn't match Microsoft's marketing or
head-start with developers. Linux doesn't have either problem - the
"marketing" is community-driven (and supported by the big boys) and
while a lot of commercial developers aren't porting Windows stuff to
Linux, the open-source community has developed enough to generate tons
of Linux apps on their own. Finally, the attitude toward Microsoft by
a lot of developers and consumers has changed since the early 90's. A
lot of people are switching to Linux more because they are pissed off
at Microsoft than for purely technical reasons. So the lack of as
many apps as on the Windows platform does not necessarily deter them.
That's true in my case, for example. I know I don't need all the apps
on Windows - I just need the ones I need and I really hate Windows 98
for the unreliable crap it is. If I was using Windows 2000 or XP,
which are more stable, I'd still hate Microsoft and want to get out
from under. Not to mention the money issue.
And I acknowledge it as well. I just happen to feel that the
magnitude of this difference is not a large as commonly "advertised".
Well, what is "magnitude" and who is advertising it? That's a
subjective perception. We just agreed on that. Hundreds versus ten
is still at least an order of magnitude difference - objectively.
Not all of it. In general, I would without a doubt say that Windows
as far as ease of use for the typical user is still significantly
ahead. There are possible two types of users that this would not be
true of. First, the super geek who would prefer the CLI over the GUI
and second, the user who never, ever diverges from clicking on two,
maybe three icons - email and web browsing. But that's really a
different debate.
I still have to differ. For most of the functions running on the
USUAL user's system, Linux can run them all from a GUI - just like
Windows - not the command line. Every time a new release of a distro
comes out, something that used to be command line has a new GUI
front-end for it. There isn't much left that needs to be run from the
command-line. While it occasionally helps to know that in fact a
command-line utility is being run by the GUI, or the details of that
command-line utility's options, most of the time you can run it just
like any Windows GUI app. The average user doesn't know or care what
hoops Windows jumps through to do a task - and they won't care on
Linux either if nobody tells them that their "program" is just a Perl
script or a GUI front-end calling a command-line tool like cdrecord.
There are some problems with consistency with cut-and-past apparently
in some apps, I can't remember what they are - but in general the same
GUI concepts apply in both OS's. And a lot of people in businesses
who have bit the bullet and switched their users to a Linux desktop
have found that their users pick up the differences quickly - the
retraining is not nearly as extreme as some people think it is.
The only problems that arise occur when an end user has to futz with
hardware that doesn't work. But even in Windows, there are times you
have to run msconfig or ipconfig or whatever to troubleshoot a problem
with hardware or software. That's not significantly different from
having to edit a config file under Linux. If it needs to be done, the
user has to find someone who can tell him what to do and how to do it.
On the other hand, although I can't prove it and don't have enough
experience using Linux apps versus Windows apps personally to prove
it, I suspect there is a lot less conflicts between programs under
Linux than Windows. Getting a Linux program to run in the first place
can be a problem due to dependencies on specific program libraries - a
result of the continually changing Linux environment - a problem which
is similar to Windows "DLL hell" - but once a Linux program is
running, I don't think they tend to conflict with other programs, or
at least less so than on Windows. This might be less of a problem on
the later Windows than on Windows 95 and 98, but it's a major source
of problems for many Windows users. And on those versions of Windows,
again, the problem of an app taking down the OS is a major irritation.
On Linux, while this can happen, most of the time you just kill the
offending task - and unlike the task explorer on Windows, a Linux
process gets killed FAST and permanently - none of this "wait twenty
seconds then complain again about it" crap - which drives me nuts in
Windows 98.
So all in all I suspect that once one gets used to the very real
differences between Linux and Windows - less great than between, say,
DOS and Windows, something we older users had to go through - I think
end users can use Linux just as easily as Windows. And this will get
better as inconsistencies between desktops like KDE and GNOME are
ironed out over time.
Recent articles in the last couple weeks indicate the "Big Boys" are
planning a major push to put Linux on corporate desktops over the next
couple years. Added to the groundswell against Microsoft in many
foreign countries, this could eventually lead to proving that Linux is
as good a desktop as Windows is. I think the next five years will be
pivotal for Linux in that regard. And the news is going to be all bad
for Microsoft.
