What happens when all the software is written that you would write for
money, 
what happens to programmers, do they end up configuring software, do they
become 
system administrators? Isn't that the same plight of those taking on
Windows 
platforms? I would rather stear everyone into open source object oriented 
systems, so at least we have ways to interface with proprietary designs
without 
comrpomising integrity of a working process for software. The problem with
proprietary 
software is it tends, for business purposes, to leverage in bad designs or
designs that 
are not logical in the interest of making money in some unreasonable way,
and usually 
down the road when 100% of your trust is in the software.. With objects, at
least 
the software can be rewritten, but there is no way to leverage in a new
design, unless 
we use XML which unlike CORBA, doesn't uphold a consistency of interfacing
or a 
consistent upgrade path (how many interface schemes in XML allow for
inheritance of 
existing interface standards?). Where Windows rules the world is on the
subsystems (the operating 
system) and integration with the hardware, our file formats (container
formats), and where we interface 
with others (communications).. That's how its not possible to move to linux
and remain 
independent of Windows.. Also all our file formats are data based, chunks
and headers, 
isn't it about time the files could maintain their own contents like
objects? Microsoft relishes the 
adoption of pure text based and binary based file formats, they don't want
you to adopt 
standardized open source objects, that would put them on a level playing
field 
with everyone..  Just to think that interfaces to communication, files,
operating systems, 
API, etc.. could be openly described and set in stone, that would be the
end of the world for 
them, and that is precisely how to are able to keep people on Windows.. 

Does Linux have its own container format for movies, documents, sounds,
presentations, spreadsheets, etc.. 
Does Linux's files maintain their own integrity, or is this in the
operating system (the data is made an object and 
the applications interface with it as an object, not as a data file), or
does each application use the 
files in different ways, inconsistently?  If its the last case, I can
guarantee you 100%  Windows will continue, 
no matter what you say about linux.. But Linux was not designed as a user's
system.. To make it as such, 
we should outline what makes Linux a user's system and make those the goals
for open source development. 

But what works against open source is also, its being made for little or no
money at all, at best 
the only software that comes out of that is software people like to do, and
software they 
can do themselves, but generally without guidelines, planning, agreement..
You have many 
programmers making the exact same thing, and changing it a little.. So
people have a choice, 
but of what, 3000 versions of some skin, or 5 versions of the same source
code with 
various modifications? Where is the consistency? There is none with linux,
its rather inconsistent, 
so you are asking Windows users to accept inconsistency in design with the
freedom to choose..
Its this not being able to decide what is the correct thing to do for Linux
that is acting against it, 
too much choices, not enough focus, not enough perfection, not enough
decisions on what to do.. 

So how is linux going to save us from system administration tasks? Its the
same 
problem with Windows, its just open source.. Determine what a operating
system must do absolutely, 
what file formats it should have, design program to convert Windows and Mac
formats to the Linux 
platform, design it so that there is one well thought up interface to do
Internet communications, 
one well thought up one for file formats, one well thought up for
interfacing with the API, and so on.. 
Eventually you will come up with something that can't be disputed and will
be used by everyone 
because it simply does everything that needs to be done, simply.. It is
possible, just we are 
working against programmers who are watching their future dwindle away to
developing 
database applications and writing legacy software.. I'm sick of humpteen
gillion libraries 
that all do the same thing, with each trying to make a buck somehow, and no
way to 
take the application in my head and make it real because of all the
dependence on low-level 
libraries and bloated standards, and so few programmers with a faith in
objects.. 

Java failed because it was designed to leverage SUN's hardware.. Smalltalk
failed because it 
didn't allow for binaries to be sold apart from the operating system, to
distribute your application 
you have to distribute modifications of the environment or the system image
(that's also why Java 
was created).. CORBA failed because they left the interface standards open
just enough to 
allow vendors to leverage proprietary partially non-compliant ORB's, and
the OMG favors 
vendors (where does the money come from?), not users, also how do you
arrive at a standard 
without a consensus of approval of how to do things (OMG does this by
committees, but 
really shouldn't it be determined by the users/programmers?)..  XML will
fail eventually because 
its open enough to be leveraged proprietarily, unless we quickly determine
a working 
information process that can be standardized, and its a question, is XML
extensible really, 
is there a format for inheritance or multiple inheritance, could a XML file
reformat itself 
to be properly interpretted by a older applications that had used a minimal 
version of the standard, do we have to replicate the same data in each
format 
under the same tree, do request information in a particular format , is
there standards for 
asking of this information, how do we denote object oriented relationships,
can 
the objects in XML format maintain their own contents or do we need
libraries to 
determine the data format and maint them appropriately for the objects in
XML?

I see a future of objects, basically every movie file, document,
spreadsheet, what have you, 
is an object, and if you want those objects in any other format, you just
ask the objects: 
of a movie -> getFrame(123), storeFrame(123), getName, getHeader, etc.. 
of spreadsheet -> getCell(123,345), getName, getHeader
...

Rather than asking a library, the library executes the object, in a
sandbox, and asks this 
of the object.. Maybe the object calls services in the library to handle 
interpretation, but in essence its possible and it would be much harder to
leverage 
commercially in the interest of moving people to a proprietary platform.
Because the 
objects (the files) can interpret themselves independent of the operating
system 
using them and of the libraries interfacing with them (try to get rid of
the notion that 
you read and write to a file, think of it that you are interfacing with the
file!). 

And that's where Linux should go.. But I have lost faith in Open Source
already.. 
I think mostly due to what I'm seeing in Linux users and this muted pride
in 
having something alternative but not being able to really point out what is
so 
significant about it (okay I'm challenging you, but if you ask yourslef,
what trully 
is linux doing that makes logical [in terms of computer science] sense and
is 
simple, axiomatic and will be impossible to leverage proprietarily, and 
you will find something that hasn't been found yet). Its just that nagging
feeling, 
how is this going to help me get a job in the future, that is the evil..
You should think, 
how is what we have now not answering the problems that could be answered 
more simply, and what other challenges await that we don't know about yet, 
all because we choose to play int he muddy details of semantics and
proprietary 
interfaces.. 


[EMAIL PROTECTED]
http://www.bl3nder.com/
_______________________________________________
VNC-List mailing list
[EMAIL PROTECTED]
To remove yourself from the list visit:
http://www.realvnc.com/mailman/listinfo/vnc-list

Reply via email to