An apology

November 11, 2010 in General

I wrote a post yesterday referencing Forbes' coverage of R, which I was psyched about. But as I was reading, one quote in the post stuck in my head. I couldn't get past how bad I thought it made statisticians look, and ultimately that became the focus of my post. I was pretty off base with the whole thing. Nonetheless, Steve McNally, the author of the post, wrote me a comment in which he apologized for "giving quants a bad name", as I had put it. I began to reply in line but decided to write this apology as a separate post instead.

Steve, there's no need to be apologetic. Instead, let me apologize to you. You set out to shine the spotlight on a relatively obscure corner of the world and somehow I ended up biting the hand that feeds - that was really stupid of me. Furthermore, I should have noted more clearly that the offending line was not your words at all, but lifted directly from a summary of a presentation (also not in any way at fault, as we will see).

Finally, it wasn't my intent to criticize at all, but to lament how difficult it is to write about statistics in a manner that "normal people" can appreciate. Obviously, I failed to convey that message in every way -- and in doing so, I also failed the mission of TGR. Somewhere around here it says (or used to say) that TGR is about democratizing information -- taking complicated topics and breaking them down to make them more understandable. That objective really began to take shape after I saw the overwhelming response to a series on the Gaussian copula, and it has been my hope since to take ideas and present them in as clear and informative a manner as possible.

Anyway, Steve wrote a piece about R. He did a really great job. He provided examples, explanations and justifications. He linked to everything. And I jumped on it for a line that wasn't even his.

I argued that the line misrepresented statisticians by not providing a context for the analysis. I still think that's true in an absolute sense, but certainly not in the context of the article and even more certainly not in any way that actually matters. Steve was just trying to give an example of a large company like Facebook that used R. He was not writing about the analysis itself. In fact, I assume he found the result interesting or he wouldn't have chosen to write about it - fully undermining my argument in the first place. The original author of the quote, Michael Driscoll of the Bay Area R Users Group, did present the context (albeit briefly, given that his audience was the aforementioned Users Group) because he actually was writing about the result, and therefore the context mattered.

No, the only person who screwed up here is me -- I should have supplied the context that I thought the line was missing AND recognized that the context didn't actually belong in that post (and this is the super important part) in a manner that could not be construed as an attack on Steve's work.

In the meantime, I have committed a cardinal sin of blogging and removed the post. I made that decision because in retrospect I can't really stand by what I wrote, and it benefits no one to have it up.

I hope Steve - and my readers -- will accept my apology. I missed an opportunity to provide supplemental information that some may have found beneficial. I will try my best to make up for it.


Leave a Comment

{ 1 trackback }

Previous post:

Next post: