Wednesday, October 15, 2003

Mark Hughes shows how creative, targeted marketing breaks through the noise - here's how he got the attention of a Harvey Weinstein, the Miramax boss.

(Auf Deutsch!)

Werbe-Revoluzzer Hughes: Der König der Glückskekse - 2 - Wirtschaft - SPIEGEL ONLINE: "Zuerst schaute er in einem Kinolexikon nach. 1950, fand er heraus, wurde tatsächlich ein Hollywood-Film namens 'Harvey' gedreht, James Stewart spielte die Hauptrolle. Hughes suchte bei eBay nach Memorabilien. Für gut 300 Dollar ersteigerte er ein Original-Kinoposter. Das schickte er zu Weinstein nach Hollywood - zusammen mit der Offerte und ein paar Päckchen Keksen. "

First page link.
There's an interesting post on VentureBlog about venture valuations:

VentureBlog: Beauty Contests and Venture Valuations: "As an entrepreneur, does it often seem that venture valuations are arbitrary? If so, don't worry, they seem that way to investors too, but there's not much we can do about it. "

Certainly interesting to get this perspective from a VC insider.

It mentions a very good presentation from Ion Equity, Fundraising and Valuation in the Current Environment - be sure to read it, too. It shows an article saying that the average company makes 70 investor pitches to get to funded.

Until now I thought it was hard work - and with this metric in mind, my previous experiences with VCs turn out to have been the easiests, smoothest walk-in-the-park experiences in comparison.

Isn't it fascinating how a single number can change your perspective on things?

Gregor Kiczales has written another noteworthy piece on aspect oriented programming - this time an easy-to-read and convincing introduction to AOP and why you should start using it, too.

Software Development Online: Testing the Waters: "Testing the Waters
With new tools emerging and the buzz increasing, it’s time to start working with aspect-oriented programming. Here’s a plan for doing so, in four simple steps.
By Gregor Kiczales "

Tuesday, October 14, 2003

The past couple of days a couple of very odd articles have popped up. Odd in the sense that I hope you do not need to read them for any other reason than curiousity or entertainment value:

Localization in COBOL .NET Do you create applications for use in multiple locations and languages?

In COBOL? No, sorry, not at all, ever! Please! (But I am guilty of talking to a company about building a .NET GUI for their 1980'es Cobol ERP system last year).

An the other one:

IBM Enterprise PL/I for z/OS V3.3 integrates PL/I and Web-oriented business processes

Enterprise PL/I V3.3 enables developers to leverage more than 30 years' worth of applications in new endeavors. This compiler gives you the needed PL/I function to begin to integrate both PL/I and Web-oriented business processes.

Notice the clever use of language - "leverage more than 30 years worh of applications"...

Well - I agree with Joel Spolsky that you should never rewrite code from scratch (great article, BTW!) ... but 30 year old code? It seems that IBM is certainly drinking that same cool-aid...

As people move to .NET I really hope they drink a sip from the refactoring tap, too. In fact, you could argue that one of the major USPs for .NET is that it provides a great way to extend legacy code using modern tools, languages and class-libraries. Just recompiling for the CLR is a major refactoring step in itself.

Sunday, October 12, 2003

Read this fascinating quantitative study of the dynamics of escape panic: Panic: A Quantitative Analysis.

It comes with a number of visual simulations of escape panic - some interesting lessons apply. For example, widening a corridor in a section reduces the overall flow compared to a straight corridor.

I believe there are some interesting parallels to here to road network design. For example, the article concludes that under certain conditions "faster is slower" - meaning that the faster people want to go in a crowd, the lower the average speed.

Maybe lowering the speed-limits during rush-our could increase the flow?

...of course, mountain-bikers should be exempt from this - our strategy for increasing flow is to expand the search space by relaxing constraints on where cyclists are supposed to go :-)

This page is powered by Blogger. Isn't yours?