Re: [syndication] Sneak Preview: my.info

[Aaron Swartz <aswartz@swartzfam.com>]

Per Kreipke <per@onclave.com> wrote:

> I'm not sure I'd be quite that cavalier about it. I don't consider The NY
> Times or the Washington Post the same as most other news sources. Not all
> news is equal in quality or trustworthiness. While the Drudge Report is
> intesting reading, if he ever wrote about foreign policy, I'd doubt its
> expertise and possibly it's veracity. In fact, it's only over time that he
> has developed a measure of trust with his readers.

No, indeed, sources are certainly valuable and that's why they are included
with the information on my.info. However, I care about the fact that my
information is coming from the New York Times, not that it's coming from the
New York Times website. The New York Times is just an entity that places a
"stamp" on the work of its writers (yes, they do do editing and such, but
let's leave that aside for now). That stamp has particular value to you, but
what's important is the stamp -- not that you're receiving your news from a
specific newspaper or website.

>> This is the direction that the web is moving to with XML and
>> RDF -- it's key to the Semantic Web.
> 
> Why do you think it's the key? I don't think one of the goals of the
> Semantic Web is to hide the sources of things. I think it's goal is to make
> data more expressive. That will probably include RDF statements about not
> only *what* things are as well as *where* the came from (and possibly
> qualitative statements about the trustworthiness of that claim (*why*) and
> the (*who*) sources, etc.).

Yes, certainly Digital Signatures are a big part of the Semantic Web as a
means to deal with trustworthy-ness. However, what's special about the
Semantic Web is not that I can make more expressive statements, since that
is certainly not true. (I sincerely doubt that there is something that can
be expressed in XML and RDF and not in English, which is what's currently
being used on the Web.) But instead, that those statements are made in a
uniform, machine-readable way. That is, I can make statements and you can
make statements and so can anyone else, and all of them are capable of being
interpreted and understood by a program. The key is that we agree on a
format -- we all speak the same language -- then, the source doesn't matter.

> It seems to me that RDF is more a way of encoding for machines many of the
> implicit and qualitative measures/statements that humans make or 'know'
> about things (another stab at knowledge representation).

Indeed, but the difference is that it's distributed over the Internet.
Knowledge representation has been attempted at for a long time, the
difference with this is that anyone can make these statements, anywhere
across an anarchic distributed system and we can still make sense out of it.
That's the key. We're continuing the long tradition of anarchic, open
systems for the Internet. The source (and thus proprietary languages) don't
matter -- we can all talk together.

-- 
        Aaron Swartz         |"This information is top security.
<http://swartzfam.com/aaron/>|     When you have read it, destroy yourself."
  <http://www.theinfo.org/>  |             - Marshall McLuhan