Re: Bug#652432: Acknowledgement (ITP: v3c-dcom -- Baby steps to DCOM)
On 18/12/11 20:50, John D. Hendrickson and Sara Darnell wrote:
DCOM's package description. DCOM's danger.
I studied Microsoft's DCOM. It's a lesser hack of Sun Java technology
(which Microsoft patently attempted to steal, hide, and destroy).
Object interfacing. (ie, apple's corba) It came out predictably much
later than Java.
You forgot to mention DCE.
While I think it's great to provide support or alternates for PATENDED
material like DCOM. Surely it was allot of work I appreciate that.
It's being used in Samba on Linux now. Not forgetting Wine.
I think it's misleading to sweepingly say "virtually unlimited
configuration and customization. What isn't? "Users and client
programs can even create sandboxes on the fly" "for use with linux
Makefiles." (how is dcom related to unix Makefiles again ??)
The package provides a plug-in system. Each plug-in file contains two
dictionaries,
one mapping ProdIDs to GUIDS and another mapping GUIDS to plug-in filenames.
I don't recall saying "for use with linux Makefiles.", please include
the relevant text.
Who knows a DCOM copy cat would probably bring yet another Microsoft
lawsuit toward linux. Microsoft has often stole the X of Xerox Windows
(ie, X-box which did not use any X technology, while sony ps3 does or
did). That DOESN'T mean microsoft won't try to sue if it's the other
way around. Doesn't anyone remember "lindows"? Wishing to make
computing ubiquitous? That linux team got sued and LOST in court.
Remember anyone?
Qt provides a plug-in system but it's system-wide and shared between
programs and users.
v3c-dcom provides the same system but inside a file, and programs and
users can choose
to share them by specifying their path in an environment variable.
There's not much to v3c-dcom, and it really could become part of a boot
loader.
What I mean is: "Baby steps toward DCOM?" Yea. But is this baby a
500lb Gorilla baby?
It's SURELY against Debian Rules to write incorrect package
descriptions. DCOM doesn't provide sandboxes.
v3c-dcom provides a library to allow anyone to create a sandbox - it's a
file.
Description: see Microsoft for copyright material on DCOM's purpose,
function, form, and compatibility. repeating it out of band could be
infringement.
Again, Samba, Wine.
I'm sorry. Microsoft "paid to make it" (or said they did) and they
don't wish to share it, am I not completely correct?
Then there's ATL - the C++ wrapper. I deliberately limited myself to the
contents of
"Inside ATL" Copyright© 1999 by George Shepherd, Brad King, ISBN
1-57231-858-9.
Yes, it states that "No part of the contents of this book may be
reproduced or transmitted in any form or by any means without the
written permission of the publisher." and that's
Microsoft Press.
If you think about it, I broke copyright right there, by telling you how
I broke copyright!
There is an aspect of copyright law that allows people to discuss and
communicate their
opinions on published works, by making reference to them, and their
contents.
Oh and, by the way, you can't reference anything you read in this email.
Hell, you can't even read it.
We're all free to voice legal-sounding mumbo-jumbo that will never see a
court room.
Here in Ireland there's a company called UPC and they state in some
legaleze document
I read somewhere, something along the lines of
"non-UPC equipment cannot be connected to the UPC TV outlet".
How about a TV then?
Believe it or not there are libraries for sale that you have to sign a
non-disclosure agreement
for before you can even read the documentation. And they don't have
snippet galleries,
discussion forums, help groups learning centres that are two clicks away
from a web search.
I think that's called due diligence, but I'm not at liberty to discuss it.
There comes a point when discussions reference other discussions and the
subject matter
becomes so widely discussed on the web that it enters the public domain,
and someone
who hasn't bought or read about the subject from a privileged source can
become quite
knowledgeable about it.
And so to the crux of the matter,
I wrote v3c-dcom firstly because I think it's a really neat plug-in system.
I used Microsoft's naming scheme as it may be familiar with some
software developers.
I could change all the names by adding an "idily" at the end -
CoCreateInstanceExIdily() -
would that be OK? Then someone would publish a header file to #define
them back so
they could compile their code on Windows and Linux.
If you like, I can change the projects name to "v3c-dcomidily".
I hope you're getting the point by now.
That's life I'm not saying I like it or not. Nothing to like or not
like about object interfaces after all (security lapses aside).
v3c-dcom provides a plug-in system as an alternative COM
implementation.
Unlike COM, v3c-dcom encourages the use of "sandboxes" of registered
plug-ins,
Regards,
Philip Ashmore
Reply to: