Merge pull request #258 from dataopt/patch-1

Fix typo in probprog.md

docs/docs/probprog.md

@@ -185,7 +185,7 @@ All inference methods in monad-bayes work with all distributions. The only excep
 
 The challenge of inference is that most distributions that are of interest are not as simple as `sprinkler`. They could have continuous random variables, a huge number of them, or even a number of them that is itself random. They could involve a series of observations, interspersed with other sources of randomness.
 
-Designing a language in which you can specify arbitrarily complex (computable) distributions as probabilistic programs turns out to be a largely solved problem. The tools given about are sufficient for that. 
+Designing a language in which you can specify arbitrarily complex (computable) distributions as probabilistic programs turns out to be a largely solved problem. The tools given above are sufficient for that. 
 
 The hard part is designing a language where you can specify how you want to do inference, because sophisticated, often approximate, inference methods are almost always necessary for the models involved in solving real world problems.