Add links to all issues and add missing word

Author: timholy

blog/_posts/2016-03-27-arrays-iteration.md

@@ -156,7 +156,7 @@ set the latter problem on fire.
 
 Some possible new `AbstractArray` types pose novel challenges.
 
-## `ReshapedArrays` (#15449)
+## `ReshapedArrays` ([#15449](https://github.com/JuliaLang/julia/pull/15449))
 
 These are the front-and-center motivation for this post. These are
 motivated by a desire to ensure that `reshape(A, dims)` always returns
@@ -174,7 +174,7 @@ problem is that computing the correct `k,l,m` might result in a call
 to `div`. So ReshapedArrays violate a crutch of our current ecosystem,
 in that indexing with `N` integers might not be the fastest way to
 access elements of `B`. From a performance perspective, this problem
-is substantial (see #15449, about five- to ten-fold).
+is substantial (see [#15449](https://github.com/JuliaLang/julia/pull/15449), about five- to ten-fold).
 
 In simple cases, there's an easy way to circumvent this performance
 problem: define a new iterator type that (internally) iterates over
@@ -185,8 +185,8 @@ has "ideal" performance.
 Unfortunately, this strategy runs into a lot of trouble when you need
 to keep two arrays in sync: if you want to adopt this strategy, you
 simply can't write `B[i,j]*v[j]` for matrix-vector multiplication
-anymore.  A potential way around *this* problem is define a new class
-of iterators that operate on specific dimensions of an array (#15459),
+anymore.  A potential way around *this* problem is to define a new class
+of iterators that operate on specific dimensions of an array ([#15459](https://github.com/JuliaLang/julia/pull/15459)),
 writing `B[ii,jj]*v[j]`.  `jj` (whatever that is) and `j` need to be
 in-sync, but they don't necessarily need to both be integers. Using
 this kind of strategy, matrix-vector multiplication
@@ -237,8 +237,8 @@ bases---it might be convenient to support row-major arrays, where
 instead `A[i,j]` and `A[i,j+1]` are in adjacent memory locations. More
 fundamentally, this is partially related to one of the most
 commented-on issues in all of julia's development history, known as
-"taking transposes seriously" aka #4774.  There have been at least two
-attempts at implementation, #6837 and the `mb/transpose` branch, and
+"taking transposes seriously" aka [#4774](https://github.com/JuliaLang/julia/issues/4774).  There have been at least two
+attempts at implementation, [#6837](https://github.com/JuliaLang/julia/pull/6837) and the `mb/transpose` branch, and
 for the latter a summary of benefits and challenges was [posted](https://github.com/JuliaLang/julia/issues/4774#issuecomment-149349751).
 
 One of the biggest challenges mentioned was the huge explosion of