🚀 go-pugleaf

In article <474nvl$f3@aztec.co.za> panic@atec.co.za (Panic) writes:
>In message <46q6ev$8ii@silver.jba.co.uk> - gbjbaiap@jba.co.uk (Alan R. Parker) 
>writes:
>>
>>"Guy O. Cox, Jr." <gcox@mitre.org> wrote:
>>
>>>There is a comparison Paper on the Microsoft Site. I think that 
>>>I found it on the gopher site.
>>Papers are also available on the IBM sites(www.austin.ibm.com I think)
>>
>>amusingly enough both companies use an engine/car analogy to say why
>>the other is wrong and they are right.
>>
>>I suppose it can be summed up in a few(ish) short lines.
>>
>>COM provides a language-neutral binary interface specification,
>>basically a set of rules that must be implemented in every application
>>implementing 'Windows Objects'.  COM is Object based, and uses an
>>unfamiliar (to OO programmers generally) methodology.  COM is a
>>specification and an implementation via OLE2.  
>>
>>SOM is a model for developing and packaging object-oriented software.
>>SOM supports the definition and use of language neutral o-o systems,
>>independent of programming languages.
>>SOM is object oriented, using a familiar (to OO programmers)
>>methodology.
>>
>>
>>In summation OO gurus may have a problem with the Microsoft
>>interpretation of what an object is, inheritance, version control,
>>object lifecycle, and platform support.  BUT it is first to
>>marketplace.  SOM may be the better technology, indeed it is possible
>>that the proposed OpenDoc technology will prove a winner, but in the
>>mean time OLE2 is outthere.
>
>Well, sort of. SOM is out there and working right now. You can even get SOM 
>for Windows NT - last I heard. However, it is not correct to assume that SOM 
>and OpenDoc are tied together in the same way that you describe the relation 
>between COM and OLE2. OpenDoc is based on SOM, but SOM stands by itself. SOM 
>is by no means a specification that has OpenDoc as an implementation. Both 
>are technologies that have their own specifications and implementations. If 
>you like, SOM is an implementation (and extension of, which may or may not 
>be a Good Thing (tm)) the CORBA specifications with respect to inheritance, 
>version control, lifecycle services etc.
>
>
>
>Bye, have a nice...
>John Anderson
>
>

Hmmm... now how does CORBA compare to DCE? 
I think we can argue MS vs. IBM for a while; but what is really needed
is a good paper on network transparent objects/environments so we can skin
the marketing cats. Anyone know of any good, recent, and technical papers?
A survey of offerings s.a. MS COM, DEC COM, and OpenDoc as well as OO OS's 
s.a. Next, Clouds, Cairo would be really nice as well.

P.S> if( CHAOS && PressedForTime() )  ChooseMostObvious();
-- 
Tim Jowers  AT&T GIS Windows NT Development  GSI  Ph./Fax:1-803-776-1483
"It always was and always will be the power to understand and
the ability to act that turns information into profits."
- Gerald M. Loeb, "The Battle for Investment Survival"