Allison's Reflections

Just another Weblogs at Harvard site

Blog #0010: IPRs and Open Source

Filed under: Where Wizards Stay Up Late — allee at 1:46 pm on Thursday, September 15, 2016

For me, one of the most frequently mentioned topics in high school was intellectual property. Lexington High School’s Honor Code addressed the necessity to respect individuals’ intellectual property; my economics class discussed the implications of IPRs on a country’s economic productivity; my US history class would often have debates around IPR policy as part of our current events section. Thus, when IPRs made an appearance in Where Wizards Stay Up Late, I was intrigued.

BBN’s initial refusal to release the IMP code was a blatant attempt to control every part of the current network— essentially, monopolize control of a unique resource. While the source code for the IMPs is not exactly a product being sold on the market, I find that many economic ideas are still relevant. If I may be visual for a second, here is a graph of a perfectly competitive firm:

screen-shot-2016-09-14-at-1-52-55-pm

We can see that economic welfare (basically the benefits society reaps due to the sale of the product) is quite plentiful. On the graph, the total economic welfare is represented by the sum of consumer and producer surplus.

Now, here is the graph for a monopoly:

screen-shot-2016-09-14-at-1-53-26-pm

(Source for graphs: Essential Foundations of Economics by Robin Bade and Michael Parkin, 7th Edition)

Because a monopoly will choose to produce at a level below the demand for the commodity to raise prices, the total economic welfare is reduced. In our reading, BBN would be the monopolistic firm in question. The text mentioned “deadweight loss” (harm to society) such as the Network Measurement Center at UCLA being unable to function efficiently.

In this specific case, it was quite clear what the benefits and detriments of BBN keeping the source code private were. However, this caused me to wonder about the converse scenario— open sourcing. I remember hearing about Google choosing to open source TensorFlow and chose to read about it (here is the link, if anyone is interested). The basic idea of this particular article is that Jeff Dean believed that open-sourcing would make collaboration between Google’s researchers and other scientific communities easier and faster. In addition, individuals could improve the source code with few barriers.

Of course, as Satell includes as a caveat in his piece, total openness would harm a firm (hence Google keeping its search engine’s workings a secret). But generally, I see open source code as a great thing. Much like the RFCs had been at the beginning, I feel like they’re an invitation to join a larger community. They share a spirit with the ARPANET’s first users, who tinkered with the network on their own and contributed ideas freely. That’s how electronic mail came to be, and while I have a love-hate relationship with my inbox, it’s certainly connected the world in a new way. It’s evidence of how much this kind of innovative environment can cause great improvements in society.

However, I’m sure there are even more subtleties to IPRs and the choice between privacy and open-sourcing. I’d love to examine TensorFlow or another case study next week in our seminar. Until next time, then!

1 Comment »

11

Comment by Mike Smith

September 18, 2016 @ 10:16 pm

Love your numbering system.

Great topic! There are lots of complexities in the intellectual property space, and I look forward to reading how your understanding grows and changes. Here’s one thought to add to your thinking: Intellectual property is thought of differently in the software industry than in, say, the biotech industry. Why might that be? In what ways are they different?

I took a look at the Satell story. I think he got some pieces right, but doesn’t probably have the entire picture. Yes, building an ecosystem of value off your platform is important when a space is about to explode in interest, as everyone thinks the space of applications based on machine learning is. But other market pressures also drive some companies to make some of their platform open source. You get a hint of this when Satell says “Facebook announced it would open source its own library of AI tools.” Competition. Your secret sauce isn’t that valuable if others are building basically the same thing. Companies have learned that the one that captures the “free labor” out in the world is often the one that wins. Can you think of past examples where this occurred? Hint: There’s a 25th anniversary for this open-source operating system this year.

RSS feed for comments on this post. TrackBack URI

Leave a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>