Windows Emulators

Windows Emulators

Post by Jabali Pragy » Thu, 26 Dec 2002 04:49:03




> If what you want to run are applications like MS Office, Quickbooks,
> Framemaker and thing like that then Win4Lin is the best choice. I've
> been using it for several years and it works very well for basic Win9x
> applications. Win4Lin gives you full Win9x networking on top of full
> Linux networking (for example you can mount SAMBA and Windoze shares
> directly thorugh network neighborhood, in addition you can mount any
> Linux partition as a drive so NFS mounted directories are accessible to
> Windoze applications). The performance of Windoze applications on
> Win4Lin is as good or better than native performance and Win98 on
> Win4Lin is significantly more stable than native Win98.

Do I have to install windows applications in the win4lin directory or
would it pick them up from a Fat32 partition as Wine does ?

--

jabali

 
 
 

Windows Emulators

Post by B. Joshua Rose » Thu, 26 Dec 2002 13:42:17




>> If what you want to run are applications like MS Office, Quickbooks,
>> Framemaker and thing like that then Win4Lin is the best choice. I've
>> been using it for several years and it works very well for basic Win9x
>> applications. Win4Lin gives you full Win9x networking on top of full
>> Linux networking (for example you can mount SAMBA and Windoze shares
>> directly thorugh network neighborhood, in addition you can mount any
>> Linux partition as a drive so NFS mounted directories are accessible to
>> Windoze applications). The performance of Windoze applications on
>> Win4Lin is as good or better than native performance and Win98 on
>> Win4Lin is significantly more stable than native Win98.

> Do I have to install windows applications in the win4lin directory or
> would it pick them up from a Fat32 partition as Wine does ?

You have to install them in the win directory. Wine can use the same win
directory so it's possible to use win4lin to do the install and then run
the application under wine.

 
 
 

Windows Emulators

Post by Lee Sau Da » Tue, 31 Dec 2002 19:02:15


    >> That wouldn't be a "port" as I understand the term.

    Peter> Well, you have a point. A port usually uses whitebox info,
    Peter> because it's assumed that the target system is like the
    Peter> source system, and so the code is of use.  But it doesn't
    Peter> have to.  

It  does!   The  word  "port"  itself means  to  "bring  something  to
somewhere new".   e.g. If you're  planting your own crops,  you aren't
im"port"ing them.

    Peter> The linux kernel ports, for example, often don't
    Peter> have any way to make use of each others code, but yes, they
    Peter> do benefit from seeing how it's been done on the other
    Peter> system.

As long as  they adapt the original code to  suit the new environment,
that's technically called "porting".

    Peter> Still, a port is quite possible with only a blackbox to
    Peter> look at.  

That's not called "porting" in  any technical sense.  You can say it's
"cloning" or "(cleanroom) reimplementation".

    Peter> And in the case of wine, people run disassemblers, tracers,
    Peter> and so on, to get greybox data.

Are  you sure  they really  use  disassemblers and  tracers?  In  some
countries, these are disallowed by copyright.  But reverse-engineering
is usually OK.

I  think the  WINE team  only developed  the API,  and used  their own
implementation to watch the behaviours of some Windows programs.  They
did reverse-engineering some non-conformant behaviours.  (Yes, even M$
does not  conform to the APIs that  they define, and they  often do it
buggily.)

--


Home page: http://www.informatik.uni-freiburg.de/~danlee

 
 
 

Windows Emulators

Post by Peter T. Breue » Tue, 31 Dec 2002 21:04:35




>     >> That wouldn't be a "port" as I understand the term.
>     Peter> Well, you have a point. A port usually uses whitebox info,
>     Peter> because it's assumed that the target system is like the
>     Peter> source system, and so the code is of use.  But it doesn't
>     Peter> have to.  
> It  does!   The  word  "port"  itself means  to  "bring  something  to
> somewhere new".   e.g. If you're  planting your own crops,  you aren't
> im"port"ing them.

I take your point, but disagree about the absoluteness of your claim -
not looking at the original source at all is merely a trivial case of
looking at it somewhat.  I've often reimplemented C library functions
(strstr, ..) without looking at the C library code, for example, simply
to support platforms which don't have that particular functionality in
their system.  The manpage is all I need in order to do a "port".

Quote:>     Peter> The linux kernel ports, for example, often don't
>     Peter> have any way to make use of each others code, but yes, they
>     Peter> do benefit from seeing how it's been done on the other
>     Peter> system.
> As long as  they adapt the original code to  suit the new environment,
> that's technically called "porting".

But they don't. It's usually a "write from scratch". They didn't start
with the original code and go through line by line swapping one thing
for another, which is what I think you are thinking of. They had to
implement an interface and they did it. They might have had to look at
the original code to get the semantics of the interface they were supposed
to be implementing, but that's because they don't have a man page.

Quote:>     Peter> Still, a port is quite possible with only a blackbox to
>     Peter> look at.  
> That's not called "porting" in  any technical sense.  You can say it's
> "cloning" or "(cleanroom) reimplementation".

Well, I'd call it a port.

Quote:>     Peter> And in the case of wine, people run disassemblers, tracers,
>     Peter> and so on, to get greybox data.
> Are  you sure  they really  use  disassemblers and  tracers?  In  some

Yes.

Quote:> countries, these are disallowed by copyright.  But reverse-engineering

They're specifically allowed in all countries I know!  Copyright
cannot control such a thing. You can do what you like with your OWN
copy. It would be a licensing issue if anything, not copyright, and
license cannot abrogate your right to reverse engineer (in europe
at least).

Quote:> is usually OK.
> I  think the  WINE team  only developed  the API,  and used  their own
> implementation to watch the behaviours of some Windows programs.  They
> did reverse-engineering some non-conformant behaviours.  (Yes, even M$
> does not  conform to the APIs that  they define, and they  often do it
> buggily.)

That is correct.

Peter

 
 
 

Windows Emulators

Post by Lee Sau Da » Fri, 03 Jan 2003 16:56:31


    >> It does!  The word "port" itself means to "bring something to
    >> somewhere new".  e.g. If you're planting your own crops, you
    >> aren't im"port"ing them.

    Peter> I take your point, but disagree about the absoluteness of
    Peter> your claim - not looking at the original source at all is
    Peter> merely a trivial case of looking at it somewhat.  

That's not  "porting".  In the software field,  "porting" means taking
an existing program and adapting  it to a new environment.  That means
patching the original code so that it works in the new environment.

    Peter> I've often reimplemented C library functions (strstr, ..)
    Peter> without looking at the C library code, for example, simply
    Peter> to support platforms which don't have that particular
    Peter> functionality in their system.  

That's not porting.  That's a re-implementation of those functions.

    Peter> The manpage is all I need in order to do a "port".

That's  not  a  port.   What  you're doing  is  not  porting.   That's
reimplementing.

    Peter> The linux kernel ports, for example, often don't have any
    Peter> way to make use of each others code, but yes, they do
    Peter> benefit from seeing how it's been done on the other system.

    >> As long as they adapt the original code to suit the new
    >> environment, that's technically called "porting".

    Peter> But they don't. It's usually a "write from scratch". They
    Peter> didn't start with the original code and go through line by
    Peter> line swapping one thing for another, which is what I think
    Peter> you are thinking of.

That's still porting.  You have to take the whole kernel as a whole in
this  case.   Every  porting  task  involves  some  parts  that  needs
rewriting.  The task on the whole is still called "porting".

If everything is rewritten from  scratch, that's a reimplementing of a
given *interface*, not the porting of old code.

    >> Are you sure they really use disassemblers and tracers?  In
    >> some

    Peter> Yes.

    >> countries, these are disallowed by copyright.  But
    >> reverse-engineering

    Peter> They're specifically allowed in all countries I know!
    Peter> Copyright cannot control such a thing. You can do what you
    Peter> like with your OWN copy.

No.  You  aren't allowed to make  copies other than  backups.  This is
what  copyright protects  against.  Neither  are you  allowed to  do a
translation (if  that's a  book, or porting  --- if  that's software).
Modifications are  not allowed, either.   Translation and modification
are  called "adaptation"  in copyright  laws, and  this  is prohibited
together with "copying".

--


Home page: http://www.informatik.uni-freiburg.de/~danlee

 
 
 

Windows Emulators

Post by Peter T. Breue » Fri, 03 Jan 2003 21:21:21



Quote:>     >> countries, these are disallowed by copyright.  But
>     >> reverse-engineering
>     Peter> They're specifically allowed in all countries I know!
>     Peter> Copyright cannot control such a thing. You can do what you
>     Peter> like with your OWN copy.
> No.  You  aren't allowed to make  copies other than  backups.  This is

You can do what you like for your own use under /copyright/.  It's
simply not in the domain of copyright law.

Quote:> what  copyright protects  against.  Neither  are you  allowed to  do a

You mistake "not specifically allowed" for "specifically not allowed".
I'm afraid this is not the domain of copyright law.

Peter