kambodza wrote: |
On every Open Source forum I see many posts
about weaknesses of Windows and how Linux is much better (where Linux =
good example for Open Source).Let's talk little in oposite direction...
First of all, it is impossible to compare different operating
systems architectures: Windows, Linux, MacOS, etc. Why Windows "is more
vulnerable"? Simply, because there are much more people who use Windows
(more than 90%), so it is natural that they discover more bugs. Linux
has also many defects (for example, why common user can have temporary
root privilegies if he launch some of the system processes), as it have
some advantages. One of the greatest antivirus experts in
world, Kaspersky, said that Linux is more vulnerable. |
|
Well, I don't agree at all with this:
--A normal Windows user founds a bug? Are you sure? I have always
thought that the bugs where found by hackers/crackers in Windows (yeah,
Sasser was made by a common-windows user, who found "a bug")...and
where should a common-Windows-user post his bug (I don't complete trust
the automatic version of Windows XP...I prefer a big forum, with
experts, where I woulb be able to write all the process which
determined a bug...
--Well, common users somehow can have root privilegies. Wow. Don't you
think that happens also in Windows somehow, but you cannot see it?
Let's talk a reboot without root privilegies, or accessing System
API's...
kambodza wrote: |
What is bad in Linux security? Exactly the fact that you have open
source of program code, so you can scroll free throw it and see some
mistakes if you are expert. In closed source, you have to use reverse
engineering techniques to get source from binary file, if you wan't to
find some vulnerability. But the catch is, that when you
decompile/dissasemble binary file to source, that is not same code of
author, it is only approximately same, but in open source it is 100%
original code. So, it is much easier to find a bug... Take a look at
phpBB2 forums, they are exploited every second week....
|
|
But with Open Source you can fix more easily a problem: more people
know the code, and there are the chances that more people can find the
bug in lesser time...with closed source? If it happens to be no
documentation? If the experts aren't at the moment available?...not an
happy situation :))
kambodza wrote: |
Also, there is bad hardware compatibility for Linux. You need loose lot
of precious time to setup some simple things like scanner, web camera,
modem, printer. It is bad support for new technologies ( when usb 2.0
standard was invented, he was available in Windows about half year or a
year before than in most of Linux distributions). What for I
need operating systems if I can't quickly make it work efficiently with
my hardware? (I'm talking from the view of common user)
|
|
...never had any problem with ALL, *ALL* of my hardware :), and I
never spent more than 10 minutes in searching with google/Linux sites
in order to find a solution...and I'm talking about Gentoo, and not a
Mandriva, that's got a guided installation like windows ;))
kambodza wrote: |
Also there is bad project management in develloping small open source
applications (this is not case with Linux kernel, KDE, Appache, etc.
don't missunderstand me). Simple, you have very, very large
number of people which are working, and some of them are not much
experienced (what can't happen in Microsoft for
example), so you can't coordinate that efficiently and
there must be many bad lines of code, it's obvious.
|
|
Are you completely sure about this? Don't you think that the testing with Open Source is more common and easy to do?
Think about it: Sourceforge.net, you can access all the Open Source
projects, choose one and download the latest version that you can find
on the CVS. You can try it, and see if it works. You can just write a
post in the forum telling "Hey, program X fails to load module Y. My
machine is Z, and..."...do you think that happens to Windows? Do you
think there are more people testing Windows Vista or the latest Kernel of
Linux? I made only an example, but this happens for ALL the programs...OpenSource programs are just better tested...
Another thing is documentation: closed source programs usually have
_VERY_BAD_ documentation...have you ever tried developing a plugin for
Outlook?
I did...and the documentation is totally a mess.
Have you ever tried to develop a program to set a Label to an AppointmentItem in Outlook?
I've to translate this horrible code into C#, in order to make it work with my plugin:
http://www.outlookcode.com/codedetail.aspx?id=139
(I'm thinking that Outlook's developer had been drinking to much beer ;)))
BTW, do you know what's my plugin,
http://sourceforge.net/projects/remotecalendars ? Do you know what is
an iCalendar? Did you know that iCalendar (RFC 2445, you can find it on
the IETF site) was wrote by two person: a Lotus and a *****WINDOWS*****
developer...
So Windows sets a standard...and it doesn't use it? (Try to Goooooogle Outlook and iCalendar, and you'll see what I mean).
Instead, you can find a program, version 0.2, OpenSource, that use this
standard...Sunbird...oh, I forgot KOrganizer...maybe there are other...
I hust wrote this to tell you the _biggest_ problem of the closed
source, like windows...how are they doing with standards? Open Source
programs usually respect much more standards...
BTW...I think you're asking yourself why I did an OpenSource plugin for
Outlook...well, I had to do it to get my degree :)...but after I
graduated, I decided to publish it as Open Source project...and you can
see (by simply going to that site) how much that program has grown with
the free contribute of people: it's getting better & better day by
day...
Please don't kill OpenSource if you haven't tried to develop an
OpenSource project...I wasn't so fanatic, and I didn't understand my
friends who kept talking about
"OpenSource!!"OpenSource!!"OpenSource!!..."
But now I understand them :)
LastHope