Giving Away Software For Free Costs More Than You Would Think - Discussion

[MaximB]

I read the series of articles about the "total" price of Ubuntu is $7,033,290,160
Here is a link to the third article : http://www.fsckin.com/2007/10/05/giving-away-software-for-free-costs-more-than-you-would-think-part-3/

I use and love Ubuntu but this article is just wrong.
Wayne the author of this article "calculated" every program that available in the Ubuntu repros.
This is very misleading as the Ubuntu developers didn't actually develop any of those programs, they might changed some lines to make those programs work better in Ubuntu but they certainly didn't develop Open Office or Firefox.

What I would like to propose is :

1. calculate only the GNU and Linux (kernel) sources to present the actual OS without any of the programs.
Because if you want to calculate every program in the repros you should also understand that many of the programs have a Windows clients also AND then you must calculate all the windows programs - and believe me, it's almost impossible to do.

2. If you really want to calculate the codes made by the Ubuntu developers you must make a "diff" between the source code reritten by the Debian developers and the rewritten by the Ubuntu developers source code, then you can calculate the actual changes and additions.

So, this article could be great if it was done right.
But it's just doesn't represent the truth.

[tehet]

OTOH it wouldn't supprise me that for Debian with some ~1000(?) devs spending a lot of time on stabilizing it $ x bn. would be an accurate figure. Then Ubuntu takes that and adds some more value to it.
While the Debian and Ubuntu devs may not have actually written all the code, it is still valuable code and it would require a huge amount of cash and time if you'd hire someone to make it for you.
So it's kind of true. No wonder why it's so much better than Windows ;)

[az]

You're missing the point.

It's not to say that so-and-so wrote the code. The point is that in some instances, it's worthwhile to estimate how much the code would have cost to write from scratch.

For example, in this report from last year (http://ec.europa.eu/enterprise/ict/policy/doc/2006-11-20-flossimpact.pdf) it was necessary to take into account the value of the code.
Quote:
"FLOSS potentially saves industry over 36% in software R&D investment that can result
in increased profits or be more usefully spent in further innovation.
• ICT infrastructure has a 10% share of European GDP, providing a basis for a further
2.5% share of GDP in the form of the non-ICT information content industry. However, a
large and increasing share of user-generated content is not accounted for and needs to be
addressed by policy makers; FLOSS increases the value of the ICT infrastructure,
supporting this wider information ecosystem.
• Increased FLOSS use may provide a way for Europe to compensate for a low GDP share
of ICT investment relative to the US. A growth and innovation simulation model shows
that increasing the FLOSS share of software investment from 20% to 40% would lead to
a 0.1% increase in annual EU GDP growth excluding benefits within the ICT industry
itself – i.e. over Euro 10 billion annually."

Although the code in of itself does not offer any value - that's the business model for FLOSS - the value is in making your computer do something. In the proprietary world, your software has (for a very limited time) a monetary value. That's artificial value since the fact that the software is sitting on your shelf adds to your assets on paper, but that's not the end-goal for having bought it. The end goal is to get some task done. In free software, you go straight to the end-goal - the value - you pay someone to make a computer do a certain thing.

The code that was written to get the job done gets passed on and reused, which is why FLOSS is such a powerful model.

So, to get back to the point, who cares who wrote it. The point is that if you want to relate it to proprietary code which has a sale price, you can.

[Sporkman]

Per wikipedia:


Using the Constructive Cost Model, the study estimated that this distribution required about eight thousand man-years of development time. According to the study, if all this software had been developed by conventional proprietary means, it would have cost about 1.08 billion dollars (year 2000 U.S. dollars) to develop in the United States.

& gives the following as the reference:

http://www.dwheeler.com/sloc/redhat71-v1/redhat71sloc.html