|
If you aren't a kdb+ user, you can probably stop reading here. I also apologize in advance for the open, anonymous solicitation, but I'm not sure I know a better way to go about this. I am in fact a long-time lurker, and I've opted to post here instead of in the Software phorum so as to be less spammy.
I'm looking to unload a set of software packages that I developed; they're written in some C, some Q, and mostly K4 (the latest version of the low-level language that sits underneath Q and kdb+). I wrote it all for fun/research (with a mind towards maybe one day trading with it). I never wound up working in finance for anyone else, and I've since moved away from any interest in finance at all; I work happily in infrastructure engineering at a non-finance tech company. In other words, I really have no use for this software now except in terms of its value to someone else, hence I'd like to sell it. I'd be selling the code and docs as-is along with ownership / full rights to do with it whatever you please (i.e., without ongoing support).
The packages are tested, filled with comments and thorough documentation, feature-rich, and in most cases micro-optimized (e.g., for execution time and internal kdb+ bytecode length). Altogether, there are two major units (a feed-handling unit and a market execution simulator unit) and a set of miscellaneous utility scripts (on-disk OHLC database creation from extant higher-precision databases, detection of missing/unexpected dates in databases, SEC/EDGAR data management, numerous feed roll-up subscribers, C plug-in implementing common English word stemming algorithms, etc.).
At this stage, due to time constraints, I'm just gauging interest, but if you email me, I will follow-up with you. Serious inquiries from reputable individuals and companies only, please. I'll be happy to provide more information, reasonable code samples, and eventually benchmarks and/or in-person demos (within New York City) as necessary.
|
|
|
|
|
If it's an alpha generating machine that's one thing, but if it's hobby code -- plumbing and data-munging utilities -- I can't think of any reason anyone would buy it without ongoing support. The feed handling part might interest someone, depending on what feeds it handles out of the box.
Have you considered going open source with it? That way you could keep abreast of your project while meeting other K4 users. |
|
|
|
| Well, I think the value is in the feature set and optimizations that have been made. The feed-handling solution, for example, is equal in scope to kdb+tick/taq/tow and is fast and more general (can easily handle text or newswire feeds that lack a ticker symbol column, for example, plus utilities for gracefully handling feeds for which fitting a day in memory is unrealistic). I do agree that support adds significant value, and I'm not entirely closed off to negotiation on that front. If there's no interest, I suppose open source is not outside the realm of possibility. |
|
|
|
quantie
|
|
| Total Posts: 861 |
| Joined: Jun 2004 |
| |
|
yeah agree with athletico. I have been tinkering with K recently and once you glue it to R you can do a lot of data munging in R. Not that I recommend it as it is probably infinitely better to learn K properly.
Anyway if you want this code to live and grow open-source(under appropriate license) it. Perhaps at a future date someone will coming knocking asking you for help (while using your code). |
|
|
