Re: [DotGNU]Implement RDF in a Universal Data Structure

[Seth Johnson]

Norbert Bollow wrote:
> 
> Seth Johnson <address@hidden> wrote:
> 
> > But the truth is, the whole database industry is unaware that this
> > is a possibility.
> 
> How can you be sure that they're not busy filing patent applications
> while we speak?


I am less concerned about this, because what I'm doing is intended to be
"pure."  I explicitly leave out anything that is esoteric.  What I mean is,
the formulation of the universal data structure is the pursuit of science,
knowledge that should not be owned, because as such it can't be owned.

Sure someone might have a patent on some part of it.  But if you get a
universal out into everybody's use, it's the sort of thing that they will
all say is just not right to be patented.  This is one of the things I'm
speaking of when I say the universal data structure handles a lot of
information freedom political fights all at once.  Yeah, yeah, they can
pursue patents regardless -- but GNU wins with this.

Then again, this is also why I only speak of this thing at a surfacey level.


> > I think for a couple of reasons some people are going to have to see
> > it to believe in it first
> 
> Count me in among the unbelievers :-)


I gathered as much. ;-)


> > and it is not the sort of thing I think should be done on the
> > "release early, release often" principle.  The first reason is that
> > DotGNU's basic inspiration is implementing the CLR .NET kind of
> > thing, so something that purports to be more of a universal
> > application than a universal runtime necessarily brings about
> > cognitive dissonance.
> 
> I don't agree with what you say here about "implementing the CLR .NET
> kind of thing" being "DotGNU's basic inspiration".   I think that it's
> much closer to the truth to say that "DotGNU's basic inspiration is
> that in every matter, there should be multiple good options that do
> not require any sacrifice of freedom".


Very good.  A universal application server should intrigue you.


> So for example, besides IL
> bytecode, cscc is also able to generate JVM bytecode and support for
> further bytecode systems may be added at a later stage.  Multiple VM
> support is among the features provided by the new version of DGEE,
> (DotGNU Execution Environment, a basic webservice server) which has
> been released today.


Yes, I think the runtime engines are essential.  Bear in mind that the
implications of a universal application server include that it may as well
serve at the OS and file system level.


> In my mind, the cognitive dissonance starts already when something is
> called "universal". :-)


Yep.  I know.  Yet this is what science aspires to.  It's certainly better
to postulate theorems than to refuse to imagine any truths might exist.


> > I showed RMS an early server about a year or so ago, and his main
> > thing was that I ought to show it to database type folks who could
> > comment on whether it's already been or being done.
> 
> I think RMS has a very good point... it seems plausible to me that
> the kind of thing that you're talking about will be very useful in
> the context of databases, namely to create a database engine that
> can automatically optimize the way that data is stored depending
> on how the data will typically be used.


It sure is.  Yet, for instance, the entire industry believes that there are
two domains of applications: operational, transactional "front line" systems
geared toward capturing information rigorously in full detail (using complex
entity-relation models to capture the complexity of reality), and strategic,
data warehouse, dimensional systems geared toward rapid queries and data
mining (using denormalized models).  There's an entire market in "extract,
transfer and load" systems, based on the supposed need to transfer data from
the operational side to the strategic side.  The universal data structure,
on the other hand, handles both.  Plus, it's easier to model your business
processes with a universal application server, than to try to build
interfaces that allow you to continue to use legacy systems for operational
systems purposes, while you continually extract to the data warehouse as you
go.

They don't think a universal data structure is possible.  Somebody might be
working on it, migth even have something pretty similar.  The key though is
to demonstrate its universality and get it into general use.  Then it
becomes a major factor in contending with the idea of software (and now,
standards) patents.  Paradoxically, once you've worked it out, the model is
extremely intuitive, not the sort of thing anybody would believe ought to be
"property."  Once you see it, it's really hard not to automatically use it
in your ordinary thinking processes.


Seth Johnson


> --
> Founder & Steering Committee member of http://gnu.org/projects/dotgnu/
> Free Software Business Strategy Guide   --->  http://FreeStrategy.info
> Norbert Bollow, Weidlistr.18, CH-8624 Gruet (near Zurich, Switzerland)
> Tel +41 1 972 20 59        Fax +41 1 972 20 69       http://norbert.ch

-- 

DRM is Theft!  We are the Stakeholders!

New Yorkers for Fair Use
http://www.nyfairuse.org

[CC] Counter-copyright: http://cyber.law.harvard.edu/cc/cc.html

I reserve no rights restricting copying, modification or distribution of
this incidentally recorded communication.  Original authorship should be
attributed reasonably, but only so far as such an expectation might hold for
usual practice in ordinary social discourse to which one holds no claim of
exclusive rights.