Weblog


Weblog

FAST TCP

06.06.2003, 10:57 · Martin · Permalink

some papers etc. From the source at Caltech

Hack the Planet: on WASTE

05.06.2003, 22:15 · Martin · Permalink

I looked at the WASTE design document and as I suspected the protocol is a piece of junk. The gnutellaism of broadcasting all requests just makes no sense. Since I've already gotten some pushback, let me give a specific example: when you use WASTE to send an IM to another person, under the hood that IM goes to everyone in the group. This is not acceptable behavior for IM software.

BTW, here's a theory I haven't seen in any of the WASTE coverage: WASTE is a clone of Microsoft's threedegrees.

small scheme

05.06.2003, 09:43 · Martin · Permalink

BIT: A Very Compact Scheme System for Embedded Applications
"We present an implementation of Scheme for microcontroller that is very compact and includes a real-time garbage collector. The compiler, which runs on a normal workstation, produces byte-code from the source program and the bytecode is linked with a runtime module. With this system, we demonstrate that it's clearly possible to run Scheme programs on a microcontroller with 64 KB of memory such as the Motorola 68HC11. Executables that include the whole library can be as little as 22 KB."

via Gavin
The paper can be foung here

Lessing Copyrights should be limited "to Authors"

04.06.2003, 22:55 · Martin · Permalink

We believe it is right for the law to protect "Authors." But corporations are not "Authors." We believe it is right for the law to ban knock-off pirates. But transformative work is not "piracy." And we believe it is right to grant a copyright for "limited times." But a term lasting more than a century is not a "limited time[]."

If you'd like some insight into this 'copyright for authors' you might read this entry on Neil Gaiman's Journal... About the stuff the 'authors' not being in a possition to grant permission to use thier own material... That's what's the problem.

EZ Bayesy

04.06.2003, 18:58 · Martin · Permalink

Naive Bayesian classification is like a good magic trick, perplexing until explained and
continuing to provide entertainment. The idea is to use observations from the past to classify a new obeservation. The classic Bayesian example made famous by Paul Graham is the Bayesian spam classifier, which is first trained on some messages at explicitly told which messages are "spam" and which are "clean". After training, it is presented with each new email and "guesses" whether it is spam or clean, based on the "features" of the message.

by The author of Reverend, "a general purpose Bayesian classifier" written in Python
[via Keith]

I need more air!

04.06.2003, 15:00 · Martin · Permalink

I'm currently positively choking. It's simply too hot for my comfort. And that's me who will be spending the next week in grill-hot Hungary for vacation... Oh well, down-under-water I guess

freshmeat.net: Quantian 0.2-Knoppix_2003-05-20

04.06.2003, 11:42 · Martin · Permalink

Quantian is a directly bootable and self-configuring Linux system on a single CD-ROM. It is based on Knoppix and adds about 500MB of software with a quantitative, numerical, or scientific focus such as R, Octave, GiNaC, GSL, Maxima, OpenDX, Pari, PSPP, QuantLib, Scilab, XLisp-Stat, and Yorick.

wow look at these flames

04.06.2003, 11:41 · Martin · Permalink

generated with GPL code

Wikis look bad...

04.06.2003, 09:47 · Martin · Permalink

says Brent Simmons and I do mostly agree with what he says on the surface, but not in whole.

I disagree his statement that text-left-to-right is a problem because it forces you to resize the browser window... This is only true if you come from 'normal' pages to a WiKi like page... if you browse a WiKi a lot, and adjust your browser window once, the benefit in screen real-esate is a deal.

I guess WiKis only look ghood if you set sensible browser defaults for fonts (sit´ze/family) and that's one of the virtues of a WiKI. It values my settings, and not those of the site-designer.

To return to Brent's page, I simpüly can't read his text until I up the font-size several notches, and then the line-spacing becomes a problem...

I think WiKis are in the same league as RSS feeds on this. They care about content, and not about presentation. Presentation is left for the reader.

Only problem here is the expectation of the normal webuser. Who even know he can change the default font on hie/her browser?

So maybe WiKIs should be split consquently into a data-layer and a presentation layer. Some CSS magic should do nicely here. (comapre this original WiKi page lua-users.org/wiki/LuaCheia to the mirrored and slightly altered version with some CSS)

on how

03.06.2003, 22:20 · Martin · Permalink

searching should be more about context than about words... This appearently goes back quite some time (2002-03-24) for me, I'd actually forgotten about that post.

PapaScott is my hub

03.06.2003, 21:59 · Martin · Permalink

to the world. Really. Acording to Downhill. Almost any random blog I tried, the last step to mine was PapaScott (ok, Leerlaufprozess and weblogs.com too)

Lua 5 for EPOC

03.06.2003, 20:53 · Martin · Permalink

- port of Lua for Symbian OS v1-5.

Simon on bookmarklets

03.06.2003, 15:30 · Martin · Permalink

Very usefull stuff to learn CSS and (X)HTML etc.

ongoing · Subscribers Redux

02.06.2003, 10:55 · Martin · Permalink

I'm just trying to see if, and how, we can fit this into an economic framework that learns some lessons from the current publishing business. The basic bargain I'm proposing - I'll write stuff for you, and in exchange, you let me count you - feels pretty fair to me.

The Code Project - PugXML - A Small, Pugnacious XML Parser - SOAP and XML

02.06.2003, 09:55 · Martin · Permalink

Presented is a small, fast, non-validating DOM XML parser, contained in a single header, having no dependencies other than the standard C libraries, and <iostream> (KERNEL32.DLL with WIN32). This XML parser segments a given string in situ (like strtok), performing scanning/tokenization, and parsing in a single pass.

Don't let yorself be fooled by the 'login required to download' stuff... the parser itself is 'in the public domain'

HTML Techniques for Web Content Accessibility Guidelines 1.0

02.06.2003, 09:44 · Martin · Permalink

Converting HTML to CSS

02.06.2003, 09:43 · Martin · Permalink

The Third Eye: Further on context sensitive RSS aggies

30.05.2003, 18:04 · Martin · Permalink

via Jim O'Halloran's Weblog

They re talking keywords yet again ;) When will they ever get that keywords go only so far? You can perfectly use feedster for 'items with these words in it'...

QuickColor

30.05.2003, 15:07 · Martin · Permalink

colormatch.dk in Flash

A Personal, Customizable Web Crawler

30.05.2003, 14:31 · Martin · Permalink

WebSPHINX ( Website-Specific Processors for HTML INformation eXtraction) is a Java class library and interactive development environment for web crawlers. A web crawler (also called a robot or spider) is a program that browses and processes Web pages automatically.

WebSPHINX consists of two parts: the Crawler Workbench and the WebSPHINX class library.

Anybody interested in crawling and info-retrieval should download and play with this a little. It's like the RegexCoach for Crawlers


alles Bild, Text und Tonmaterial ist © Martin Spernau, Verwendung und Reproduktion erfordert die Zustimmung des Authors

Martin Spernau
© 1994-2026


amazon.de Wunschliste

Facebook me!
Google+

powered by Traumtank