RE: [xml-dev] XML-DEV list

["Len Bullard" <cbullard@hiwaay.net>]

For anyone coming in late, this is the "kill all the lawyers" episode.  In
this series, I am the bad guy and Andrew and Ben are the brave heroes who
have come to overthrow Wolfram and Hart.

From: andrew welch [mailto:andrew.j.welch@gmail.com] 

On 9/28/06, Len Bullard <cbullard@hiwaay.net> wrote:
> My problem with that description is how many web programmers or designers
> are willing to suggest or document solutions without doing the research
into
> the topic to discover what solutions have been tried and what the results
> were.

>Does that really matter?  If two people do similar work with similar
>results, but aren't aware of each other, is that a problem?

It doesn't have to be.  If two people are doing the same work at the same
time, that is normal.  They are solving obvious or at least evident
problems.  If a person comes along and solves a problem others have already
solved, that person may be wasting time, wasting resources, or may be
improving on the solution.  Style sheet based linking has been solved
multiple times.  As controversies go, it has a long lineage and because it
is an obvious approach, it will go on and on.  I've no problem with that.
I'm asking questions and citing previous attempts to determine if the thread
I started (Xlink Dead?) is revealing fresh information or thinking.  So far,
no.  It outs yet another 'style sheets should encode the semantics'
initiative.  And I've no problem with that other than I like to know if such
an effort is necessary or simply convenient.

>> Two very serious problems emerge out of that:  claiming credit where
>> there is prior art making the IP situation difficult for everyone, and
>> claims that lead to false eigen-index locking (the Google/Wikipedia
>>effect).

>Is IP there an issue here?

I don't know.  If he cites the prior art, there won't be one unless he
stumbles into a patent.   No one knew that Eolas was lurking until the
lawsuits started.  There are patents on XML.  There is also prodigious prior
art.  Most of it goes unnoticed by Google because Google has a shallow and
recent memory (the world of URIs).  Fortunately, the humans have a longer
one and so some people make a healthy living as consultants on patent cases.
That is a side issue but I bring it up because the practice of 'we'll code
now and ask for forgiveness later' is ok until one proposes a change to a
standard that stumbles into the patent.  Take your own risks at your own
will, but don't ask others to do it without being willing to answer
questions and withstand critique.

>Also, googling for "false eigen-index locking" returns 3 results.  So
>even google doesn't know what you're talking about!

Oh good.  A fresh idea?  Well, not really.  

http://pagerank.suchmaschinen-doktor.de/index/math.html
http://www.cs.ubc.ca/~greif/Papers/gg_BIT.pdf

Force is the feedback effect over time of the naïve indices in use.  You may
have missed earlier discussions on this list of the ease with which Google
can be gamed.  All inbound-linking systems that scale out of some boundary
share that characteristic without filtering controls.  One of those is
vetting assertions against other assertions with time-variant properties or
restricting the domain of the citation (eg, inverted indices restrict the
domain to the book; library cards restrict it to the book title, author,
date, etc; cross-domain indices make no assertions beyond location, and so
on).  Google knows exactly what I am talking about.  They don't like to talk
about it.  The idea that a Google search returns an authoritative assertion
is not an idea they endorse; naïve users do that for them.

><rant>
>Sorry to be blunt Len, but my problem with some of your posts is the
>sheer amount of noise.  

See above.  I ask questions because I am familiar with the history of
attempts to use style sheet declarations to associate functions to markup.
Linking is a function but because of the noisy definitions, it has, as
others have noted, typically become parameter-heavy; the case for any
ill-defined problem.

>Maybe it's just me (in which case I apologise
>and I'll go back to not reading them) but if you are going to be
>critical of anyone else's work or posts, take a good look at your own
>first.

Certainly.

1.  Style based linking is an old idea.
2.  Adding more semantics to CSS bloats the browser.
3.  Bloating the browser may be a good tradeoff if authors are more  
    productive or interoperability improves.  The first is likely but the   
    second is not.
4.  Rational argument compares options first than proceeds.  Otherwise hack
    as you will and increase the overall entropy while enriching the local
    language.  It's a selfish behavior but (See Ayn Rand) a lot of progress
    can be made that way and a lot of progress is undone that way.

len