I thought that "good" affiliate systems always use cookies to validate the affiliate - am I missing something? Surely cookies would solve this problem?

You're right Frank, but not for the reasons you've mentioned. The great thing about HTML UIs is that they are very easy to use, quick to understand, and well-understood by most users. We don't need rich UIs at all! We think we do, but look at all the stuff you can do online with plain old web forms. DHTML and Ajax are "good enough".

Hey! I did all that already! Where's my six figures? love it :)

It's amazing how naming something almost completely defines it. I never agreed with the name "cost center" - it's just way too negative. Now "profit center" I like very much (shoulda thought of it myself!). Might be a bit of a stretch for some departments, but you never know - just calling them "profit centers" will focus minds the right way.

See, now I wonder. If you add up the time you spent on all of that, you might as well have hired an accountant. That would be my choice anyway :)

Ah - the dreaded negative return from more hardware limit. Ouch.

I think it would be better to let the user try the software for a while first. Then they clearly understand the benefits that it brings and are more likely to submit their details. And people who don't like your software, or more importantly, for whom it does not solve a problem, won't bother. So you get more qualified leads.

And to think I managed to write solver in only 100 or so line of Perl. And it wasn't even complete! Bah.

I'll give the Nod to Balaji and say "how on earth do I apply this to my API documentation writing?" I'd love to make my API docs "fun". They just don't seem to lend themselves to it though...

I guess I'm going to have to sit down and review the Spring source now as a learning exercise! I'll add that to my to-do list... thanks Chris! :)

My number one tip: get an overdraft for when you mess up! (cos you will...)

Wow. That's a bit of a shock. I love to invoke Metcalfe's Law. Bummer! You can't really argue with the math in the article either. I'm convinced. I wonder does it explain all that dark fiber that's supposedly lying around from the dot com expansion...

I'm trying to learn how to touch type at the moment. Thanks for the link that's a handy little site for getting an idea of where I'm at!

I could care less about "could care less"! Glad to know I'm not the only one that has sleepless nights over that particular one.

I concur that Peopleware is an excellent book - a must read for anyone considering this question. On managing team size, when you have a star on the team, the best thing to do is create insulation for them. For example, you can RESTify their interface to the others in the team. This creates a clear boundary that they can build test cases on. The rest of the team can then work against this interface, but the star can continue beyond what has been actually implemented. This works even better when you have more than one star on a team - they'll still argue to death over the interface, but at least they'll leave each other's code alone!

That is the most astonishing thing I have read in a long while. That you can actually survive this type of operation is pretty amazing. From family experience I know that the human body can compensate for the loss of vital organs, but that this also applies to the loss of a brain hemisphere is still surprising for me. I guess this is how people can survive gun shot wounds to the head, which you read about sometimes.

How exactly did the ThreadLocals affect the service. Just curious to know how to avoid this.

So are you up for doing anything like this with Secantus? :)

You're pretty much right. In fact most people respect it when you confirm their requests - it helps them track what they want! :)