Re: New blog is posted

D and C - you are both forgetting that in this country there is no justice,
only law.
BobJ
"dawn" <dawnwolthuis@xxxxxxxxx> wrote in message
news:1137079315.402288.97330@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

murthi wrote:
> Thanks much, my OP was somewhat in the nature of a snide joke, I guess
> I'll
> have to get with the emoticons sometime. In reality, of course I read the
> whole thing and thought it great. Keep it up.

And I, of course, knew you did ;-) But I like to be precise and I
researched your statement and decided it was probably fine if I removed
them.

>
> I don't now much about the mathematics involved in NF1, but the thought
> occurred to me that maybe Codd evolved the mathematics first and then
> applied it to practice,

Yes, that is pretty much the case. He wasn't heavily involved in the
development of SQL, however, so it is others who have worked to apply
his theory.

> akin to a scientific survey that selects the results
> first and generates data that substantiates the results.

I see it somewhat that way too, but the reason that others do not is
that once you make the giant leap from reality into the mathematical
model, you can prove some significant things. For example, you can
show that you can prepare a query language that uses only 1st order
logic if you use normalized relations. That has obvious practical
implications for those writing a query language. You cannot prove that
someone will enjoy using your product, however.

> Obviously, the math
> for NF1 must be simpler than for our MV (or NF2 as you call it). In fact,
> the other question is whether there is any math modeling of MV at all?

Yes, indeed, there is. Just like RDBMS's, the MV isn't an exact match
to a complete mathematical model, but there are a variety of
mathematical models that can be used, typically working with directed
graphs (which is what the web is). The RM folks will state it to be in
their favor that RDBMS's are based on theory rather than backing into
the theory from the intuitively-and-pragmatically-developed solutions.
I have a healthy regard for theory, but I would much rather use a
time-tested solution that provides what customers need than one that
should work in theory.

> If
> not, and I suspect not, would this be a useful topic to explore? Is it
> even
> possible?

This is one of the things I've been researching, but without doing the
math myself, just the modeling. I have read through a bunch of papers
of folks putting the mathematics behind such models. For example, in
some future blog I'll mention that it requires a second order logic for
a query language, but that this is not problematic (that is, it is
still provable, np-complete, or whatever cs types say) because it is a
monadic second order logic. (If you haven't read my blog yet, don't
assume it will be as dull as that last sentence might have sounded ;-).

> On a somewhat OT note, I did find it revelatory that you actually checked
> with vendors on the tm/c issue. It would never have occurred to me to do
> so;
> as I said, blogs and emails and other such material are not commercial
> speech and do not have to follow those standards other than, of course,
> libel etc. (free speech lives!)

I had researched the first with the poster and flash cards and had not
researched the matter of blogs not being commercial speech. I found
that interesting.

> I see tm/c marks as being for the owners to
> establish ownership claims when competitors use them in ads etc., to show
> clear provenance. In this case, the vendor in question should have been
> glad
> of the exposure; my reaction would've been, no doubt unproductively, to
> tell
> 'em to take a hike.

So you are not always pragmatic then, eh?

> But this, alone with Tony's previous comments about libel, shows how much
> we are cowed by lawyers.

I try not to be. Since I know that I intend no harm, if someone thinks
I have wronged them, I would think that we could discuss it and make
things right. But that isn't how the stories go.

> The probability of being sued for activities such
> as these is basically zero, but worry trumps common sense.

I decided that was what I was doing, so I changed it. Cheers! --dawn

> Chandru Murthi
>
> "dawn" <dawnwolthuis@xxxxxxxxx> wrote in message
> news:1137038907.918444.126380@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
>
> murthi wrote:
> > Finally got round to looking at. Great reading until...
> >
> > If there's one thing that gets me going more than the misplaced
> > apostrophe
> > (see Eats, Shoots and Leaves by that curmudgeonly Brit woman), it's (not
> > its!) the unnecessary use of those stupid tm and c symbols. Get it
> > straight
> > folks--*you don't have to put them in anything but commercial speech
> > (ads)*
> > and even then it is debatable! You will note that it's only in the less
> > successful magazines that you'll see them, probably because they're
> > using
> > unedited handouts from companies. I *couldn't go beyond this para, it's
> > like
> > having speedbumps on a freeway*:
> > Products employing an MV or NF2 data model include the IBM® U2 products,
> > Temenos jBASE®, Revelation OpenInsightT, Raining DataT D3®, Northgate
> > Reality, EDP Plc UniVisionT, Ladybridge Systems OpenQMT, and
> > InterSystems
> > CachéT. etc.
> >
> > Get rid of them and I might continue reading...
>
>
> OK, Chandru. To show you how much respect I have for your opinion, they
> are gone, brother. You may now continue your reading at
>
> http://www.tincat-group.com/mewsings
>
> I thought I was picky with an aversion to split infinitives, but with
> some very educated (and some delightfully anal) readers, I have learned
> a lot about what people care about (if I put a parenthetical comment
> here, then I will not be ending this sentence with a preposition,
> right?). I'm sure I will offend someone with each blog entry, but I'll
> see if I can avoid flipping your bit again.
>
> smiles. --dawn
>
> >
> > Btw, looks fine on IE 5.5 and its (not it's!) siblings.
> >
> > C.handru Murthi
> >
> > "dawn" <dawnwolthuis@xxxxxxxxx> wrote in message
> > news:1136912747.098987.261850@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> > > I'm going to gradually announce this blog over the course of the day.
> > > If you are in multiple forums, apologies in advance that you will see
> > > this from me several times today.
> > >
> > > For what I hope is a good read, please visit this first posting of the
> > > tincat mewsings blog at
> > >
> > > http://www.tincat-group.com/mewsings
> > >
> > > I'm hoping some of you will feel free to get it started with some
> > > comments too. If you see anything amiss in form or content, please
> > > let
> > > me know. Thanks. --dawn
> > >


.

Relevant Pages

  • Re: New blog is posted
    ... > I don't now much about the mathematics involved in NF1, ... (If you haven't read my blog yet, ... > Chandru Murthi ...
    (comp.databases.pick)
  • Re: New blog is posted
    ... In reality, ... Chandru Murthi ... Great reading until... ... I'm sure I will offend someone with each blog entry, ...
    (comp.databases.pick)
  • Re: Need help in understanding a python code
    ... Mathematics doesn't have an "else", so you write hybrid functions by ... Which is also not valid terminology for hybrid functions. ... dismissing an entire blog. ... He's meant to be writing Python code, ...
    (comp.lang.python)
  • My blog about (learning) mathematics (and physics)
    ... I'm a mathematics student. ... I started to write a blog: ... someday, hopefully, contemporary :-) things in mathematics and physics. ... I'll write about some physics I've ...
    (sci.physics)
  • Re: My blog about (learning) mathematics (and physics)
    ... | I'm a mathematics student. ... I started to write a blog: ... | someday, hopefully, contemporary :-) things in mathematics and physics. ... I'll write about some physics I've ...
    (sci.physics)
Loading