I've never worked at anything but a company with "closed-source" code, but all engineers within the company had access to all product code via SVN. From reading other comments, it seems the quants are paid relative to their code's performance in the market, and this culture of not sharing code is probably understandable under those circumstances.
In other words, Shaw doesn't seem to be recognizing the competitive nature of the HFT business, even within a given organization, and is more accustomed to a team effort.
I don't know, to cultivate a culture of absolutely not sharing sounds like bad business, to me. Sure it's good for each individual, but aren't you severely limiting the organization's product (let's call HFT software a product for lack of a better word) quality and competitiveness?
I'm not talking about whether or not I'd mind working in such a culture. I'm talking about how ridiculously stupid it is to encourage such a culture.
Isn't there some team of highly motivated, cooperative good engineers who are beating all the solo pilots, right now, somewhere?
I'm sure the same's true in any industry where code contains tradable trade secrets, keeping code on a need-to-know basis makes sense. For example I imagine not everyone at Google has access to the page ranking score algorithm, because a rogue employee could easily sell that information for millions of dollars.
As opposed to no-one having access, a restricted access is a much better idea.
I was not aware that HFT algorithms are that 'personal', now that article makes even less sense! What seems extremely blasphemous is that this person outrightly denies the importance of creativity and innovation that goes into Product designing at (pardon my ignorance, and bias for real tech firms), say, Google.
What I do get from the port is this: HFT people are extremely smart - extremely closed circle. They know algorithms and optimization. Speed is of utmost importance. They know more ways to optimize than I will ever know (Actually I don't even need to. I can very well work on something more significant). And they earn a lot (Well I will earn enough for my satisfaction, so I don't give a flying SegFault).
In other words, Shaw doesn't seem to be recognizing the competitive nature of the HFT business, even within a given organization, and is more accustomed to a team effort.