langreiter.com plain, simple: comments-2004-11-19

LANGREITER.COM plain, simple

START

INDEX

Freitag, 19. November 2004

Die Zeit vergeht.

2004-11-19-t

What a wonderful invention: The IS NOT OPERATOR. Congratulations, Paul Vick, Corneliu Barsan and Amanda Silver, congratulations, Microsoft.

I hereby retract all criticism of software patents ever expressed; such devices of beauty, such machineries of unquestionable elegance like the IS NOT OPERATOR must be protected by every means possible. Probably only select members of the Visual Basic team can fathom enough creative power—as demonstrated quite forcefully by this patent application—to imagine the impeding evils, should such stunning inventions fall into the wrong hands (like the dirty ones of those source-sharing code communists).

I hope we'll see many more inventions like that streaming forth from that fount of imaginativeness at Redmond—just imagine how much easier life would be with the awesome power of, say, the IS OPERATOR!

2004-11-19-gpdj

Adam Bosworth: "And it is my considered and thoughtful judgement that while the concept underlying RDF is simple, even brillliantly simple, it isn't how most of us think about data. In short, while it is simple, it isn't intuitive."

I guess I have to agree (based on quite some time of dealing with data in pure triple form). Just because it's possible to express any computation using Turing's machine or lambda calculus, it doesn't mean that those are effective ways for people to think and express computations in. I think what we really want in terms of data models is something fairly simple: uniquely identified dictionaries of simple values (strings, numbers, dates &c.) plus typed links to other such dictionaries (or resources or whatever we're going to call them); in short, u-forms, code-free object graphs. Even if a store is implemented using triple structures, that really shouldn't be what surfaces to the user (developer). It's fairly easy for people to think in terms of linked entities with properties (object graphs, relational tables, even web pages), but I've yet to meet someone who is really productive manipulating raw triples. The pivoting we can have cheaper.

Hint: Gobble got a lot right.

And now some excerpts from a quite wonderful talk Adam Bosworth recently held:

"Consider user interface. When HTML first came out it was unbelievably sloppy and forgiving, permissive and ambiguous. I remember listening many years ago to the head, then and now, of Microsoft Office, saying contemptuously in 1995 that HTML would never succeed because it was so primitive and that Word would win because Word documents were so rich and controlled in their layout."

"The value is in the information and its ability to be effortlessly aggregated, evaluated, filtered, and enhanced."

"You want to see the future. Don't look at Longhorn. Look at Slashdot. 500,000 nerds coming together everyday just to manage information overload."

"Learning Avalon or Swing isn't going to matter. Machine learning and inference and data mining will. For the first time since computers came along, AI is the mainstream."

"I encourage all of you to remember, that in the long run, we are all human and, as you add value, add it in ways that are simple, flexible, sloppy, and, in the end, everything that the Platonists in you abhor."

Heck yes.

2004-11-19-tripleTrip

man-uel 7337 days AGO
Ich finde man muss Tripel analog zu Buchstaben in der Schrift und Conses in Lisp sehen. Keines der drei ist an sich effektiv, erst durch smarte Notation und Verbindungen von vielen solchen einfachen Objekten in Komplexe wird die Power von Worten, Sätzen und S-expressions möglich. Ich meine, diese ganze Seite besteht nur aus Buchstaben, aber trotzdem sind Buchstaben kein effektiver Weg für Leute mit Konzepten umzugehen.

Lisp definiert (+ 1 2) als Notation für die drei Conse (+ . (1 . (2 . nil))). Plötzlich hat man die Grundlage für eine der mächtigsten Programmiersprachen.

Eine u-form sehe ich deshalb analog zu einem Satz oder Wort: Sie ist Element des Diskurses, aber als Medium wäre sie zu wenig granular, zu grob. Als Medium braucht es etwas formbares und simples wie Buchstaben, Tripel, Atome. Die interessanten Objekte sind aber die Moleküle.

chris 7337 days AGO
Well put! Und ganz recht hast, wobei ich mittlerweile nicht mehr glaube, dass Triples notwendigerweise den atomaren Level repräsentieren müssen ...

Please log in (you may want to register first) to post comments!