07:21:22 <Palmik> Hi guys. I have add IxSet to the benchmark https://github.com/Palmik/data-store/blob/v0.2/benchmarks/src/01.hs I have added version of the benchmark that works on smaller data sets so that IxSet and Tables do not cause stack overflow. The results can  be seen in bench.out (https://github.com/Palmik/data-store/blob/v0.2/bench.out) bench.html and bench_cmp.out
07:26:53 <Palmik> I also encourage you to take a look at how the benchmark is implemented for the store (ixset/tables) you are familiar with so that you can point out possible inefficiencies in the benchmark code itself.
13:58:52 <Taneb> How does digestive-functors mix with Happstack?
14:00:20 <Taneb> Hmm
14:01:06 <Taneb> I think the answer is "pretty good"
14:01:21 <Taneb> Next question is, do I want digestive-functors-happstack or digestive-functors-heist or both
14:01:26 <Taneb> (I use Heist)
14:37:35 <Palmik> Hmm, I just probably realized that the benchmarks are completely useless since I use nf... it reevaluates the whole store, always... but if I do not use it, it will probably be useless too because of the varying level of lazyness.
14:38:16 <Palmik> (of the implementations)
14:40:40 <Palmik> Well, not exactly... the lookup benchmarks should actually be OK, since we want to evaluate the result anyway.
14:40:53 <Palmik> But not the benchmakrs that measure how long it takes to insert one element.
14:41:00 <Palmik> Those are flawed.
16:48:09 <stepkut> Taneb: both probably
16:48:49 <stepkut> Taneb: the -happstack part is the glue that allows df to extract the values from the submitted form data, the -heist part is the glue that allows df to generate the form html
16:49:08 <Taneb> Right