@agile_geek you know I was jesting…. 🙂

morning

morning

morning

looking forward to seeing a bunch of you on Thursday at XT16

Oops, I have slides to write! Maybe tmo

hmmmmm

get on it Trojer

we have about 2-3 tickets left if anyone on here wants to come - ping me directly

Morn'

Much as I'd love to come to XT16 as I'm currently out of work I better not burn anymore company funds on expenses!

probably a wise thing to do @agile_geek but we’ll drop your name a few times 😉

@thomas cheers 😉

Be good to see folks at XT16 - the goto conference is just a bit too big, looking forwards to a conference that “fits in my head” !

@korny everytime I go to a conference the previous one falls out of my head!

jonpither: Hope the conference goes well... Would love to be going, but the timings pretty inconvenient for us. Hopefully next year! 🙂

I want to know what matching engines 'do', what kind of algorithms are they based on, hmm. @seancorfield ? 😉

Can anyone point me in the direction of any papers to read, tech blogs, etc. ?

I recently discovered https://www.semanticscholar.org/ which is pretty good for finding papers on topics (as a general solution to anyone with an interest in papers)

@malcolmsparks https://github.com/papers-we-love/papers-we-love/tree/master/pattern_matching has ML and racket pattern matching. http://stackoverflow.com/a/7883554 might help too

I wish Slack would improve the way links are displayed. Less pictures, more horizontal.

@malcolmsparks As in dating site matching engines? That’s usually each company’s "secret sauce"...

oh haha

@seancorfield yeah, I realised after I sent that I'm probably asking a bit too much! I'm just starting to read up

At WS we use a fuzzy search engine (heavily used by realtors for matching houses against online searches) and we’ve developed a custom set of criteria extracted from member profiles. It was a pretty complex slab of CFML code for years, then we converted it to a pretty complex slab of Clojure code 🙂

i’m impressed that it got migrated

that sort of thing often ends up a ball of do-not-touch

apparently pretty much all car rental pricing goes back to one company’s pile of COBOL

Well, our business requirements change a lot over time so some areas of the code see a lot of churn — and rewriting certain areas from scratch is often the better solution. That means we have a lot of support for modernizing code on a pretty much continual basis.

Which is very nice.