Subject: RE: The Pledge model -- K5 generates 6 mos income in three days
From: "Morhous, John" <morhous@teri.org>
Date: Wed, 26 Jun 2002 21:26:02 -0400

>Historically, I've found that most people won't give you money unless
>you ask them for it, even if they actually owe you the money.  Very
>organized people are exceptions, although businesses larger than 10
>people often hire such a person to handle things like paying
>suppliers.  I don't find it disturbing in the least that K5's users
>share this attribute.

Thats more or less the point I'm trying to make. Disturbing was a poor choice 
of words--what I was trying to say was that the situation was more or less 
glaring in how it highlights (in my eyes at least) some of the core problems I 
see with the open source business model.

Someone today made an analogy to the how open source software was akin to the 
GD allowing people to share tapes of its shows. From a distribution 
standpoint, yes, they are very similar because they both make it much easier 
and cheaper for people to try and enjoy the content, and then in turn become 
"fans" of it. However, where they strongly differ is in the purpose. The GD do 
it because "fans" will (a) buy more CD's, and (b) go to live GD shows. Both 
(a) and (b) make money for the GD, so they obviously found out that benefits 
of allowing tape distribution is greater than the drawbacks of forbidding it.

However, on the open source software side, I'm not seeing the same type of 
correlation. Some people (like Stallman), will go off on this free software = 
free speech idealism tangent (FSF side). Some will say that the purpose is to 
recruit new members (or businesses) who are interested enough to contribute 
code and improve the software (majority of people). Still others say that the 
free software is a "teaser" that lead people to buy more lucrative versions or 
services that generate real revenue (Redhat, Mandrake, etc). But from a 
BUSINESS perspective, I have yet to see a good open source model that 
justifies a solid reason for giving away software for nothing.

Now tying this back with the K5 situation, we've just thrown another kink in 
the chain. GD says that if I give you the tapes for free, you'll eventually 
contribute revenue to me in other methods. But K5 just demonstrated that even 
though I'm giving you the content for free, your STILL not going to give 
anything back to me unless I plea and beg for it. If this type of model were 
the case for GD, they would have ceased allowing people to make tapes a LONG 
time ago.

To me, the problems with open source software have nothing to do with 
Microsoft, but everything to do with laissez-faire licenses like the GPL. 
Granted you don't really see the failures in the small projects, because the 
software is more or less "scratching the itch" of some software problem a 
programmer is having, but in the large projects, from a BUSINESS perspective, 
its killer. I'd like to see licenses that actually require you to support the 
continued development of the software through either cash contributions or in 
"kind" (bug reporting, user testing, mirroring of source, etc). Now there is a 
sustainable model; software gets developed, people can eat, all is well.

Again, keep in mind, I view sites like K5 and Slashdot and Sourceforge as more 
or less open source software projects, except they product web content instead 
of source code. I'm also looking at this from a business perspective (hence 
the list), so flames about how "free software isn't developed to be sold 
commercially" don't really apply. I'm just trying to think of a way for a real 
company to survive, but keep the ideals of open source alive and kicking...

Perhaps I'm just ranting, who knows. Thats my 2 cents...

>I'd like to defer to your experience on this matter, but could you
>elaborate?  I think a million ten-kilobyte pageviews consume on the
>order of[0] $10 worth of bandwidth, wholesale; cheap-junk computers
>you can buy for $250 on craigslist can serve 400 static page hits per
>second, or 34 million hits per day, so web-serving hardware cost
>should be a non-issue if you can afford for it to be unreliable.
>
>I can understand that floorspace in a hosting center can be expensive,
>but the aforementioned million ten-kilobyte pageviews a day is only on
>the order of a megabit per second.  How much does floorspace where you
>can get a megabit cost?  Does it dwarf the cost of the actual megabit?

This is kind of off-topic, so I'll send you a note directly on it. But trust 
me, its expensive.

-JTM