my sincere apologies.
Imagine my chagrin. I answer a seemingly innocent question on how to implement nested loops in R and get lectured on what should and shouldn't be done in R and how to do it properly with the entire process culminating in the post below.
Back to the markets,
-----Original Message----- From: tech-spec-bounce@xxxxxxxxxxxxx [mailto:tech-spec-bounce@xxxxxxxxxxxxx]On Behalf Of Tom McCubbin Sent: Tuesday, November 16, 2004 5:42 PM To: tech-spec@xxxxxxxxxxxxx Subject: [tech-spec] Re: Nested Loops
After rigorous battle w/ my oldest son at bedtime, I was ready to return
to do battle w/ the infamous BBands...then, like a wise father, S. Wisdom stepped in to smooth
the waters. I felt reduced to a sniveling school yard child, having realized the point moot, and
my religious zealousness silly.
But how can i lay, after a 'paul graham' shellacking?
First, I am not a hacker. I am a professional developer. The libraries and tools i have
lovingly, and sometimes in obscure low level, ugly, but well commented assembly, C, and C++,
written underly some of the largest historic time-series data warehouses, perf attr, risk mgmt, and
modelling / back-testing systems in place today.
Keane has pitted me against several of the established leaders in both the historic data and
relational model space, and each time i have come out on top. Top in terse memory requirements,
top for scalability, top for runtime performance by orders of magnitudes. And of course tops
for hardware requirements. If you want i can send pretty reports saying so.
DMG, JPMorgan, BoA, Goldman, American Century, Fidelity, Putnam, State Street, Allianz, Dresdner,
the ECB, the Beaureau of Economic Analysis, Haver Analytic, the Bank of International Settlements
and many others all rely on my unreadable code. And soon the FED will be migrating their
historic dw's onto my platform as well. Why? Because, i have a clue what i am doing in the
little space which is my world.
I don't need Graham's rhetoric. I can find plenty of opinions on both sides of the fence.
So please don't lecture me on being a programmer. I'd rather learn from your experience and
wisdom in the financial world. Everyone knows who you are. I am no one. Can you please leave me the paltry dignity of knowing my own profession? Or must you claim that as well?
R is a language i have only recently begun to use. I do however have extensive experience with
similar proprietary languages, and understand the concept space / knowledge domain just fine.
In fact, i have spoken and taught on several occasions at central banks, i-banks, and the like.
I defer to Dirk for my R lessons. Clearly, he is the resident guru.
I defer to you for the applied finance CS lessons. And steve. And others who i hope will excuse
my not mentioning each of their names.
So please, lets stop the insanity, and return to you guys teaching me something...for my own
thank you, my apologies if i have exceeded the flame limit, all due respect, and it hurts
when you step on my toes,
ps - how could perl possibly still exist?
people to rememberCode is not meant to be understood...that's what comments are for.
( you do comment your code, right ? )
I'll let Paul Graham reply for me:
"Source code, too, should explain itself. If I could get
just one quote about programming, it would be the one at thebeginning of
incidentally forStructure and Interpretation of Computer Programs.
Programs should be written for people to read, and only
machines to execute.your readers. It's
You need to have empathy not just for your users, but for
in your interest, because you'll be one of them. Many ahacker has written a
program only to find on returning to it six months laterthat he has no idea
how it works. I know several people who've sworn off Perl after such experiences."