diff --git a/.travis.yml b/.travis.yml
index b568f87f..22fddd50 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -3,7 +3,6 @@ language: julia
 os:
   - linux
 julia:
-  - 0.4
   - 0.5
   - 0.6
   - nightly
diff --git a/REQUIRE b/REQUIRE
index 427b0e0f..d6ad0822 100644
--- a/REQUIRE
+++ b/REQUIRE
@@ -1,4 +1,4 @@
-julia 0.4
+julia 0.5
 BenchmarkTools 0.0.7
-Compat 0.19.0
+Compat 0.30.0
 JLD
diff --git a/appveyor.yml b/appveyor.yml
index 0515ca7e..9e24c5c1 100644
--- a/appveyor.yml
+++ b/appveyor.yml
@@ -1,7 +1,5 @@
 environment:
   matrix:
-  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x86/0.4/julia-0.4-latest-win32.exe"
-  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x64/0.4/julia-0.4-latest-win64.exe"
   - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x86/0.5/julia-0.5-latest-win32.exe"
   - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x64/0.5/julia-0.5-latest-win64.exe"
   - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x86/0.6/julia-0.6-latest-win32.exe"
diff --git a/src/io/IOBenchmarks.jl b/src/io/IOBenchmarks.jl
index 84f8ed6c..b65629a4 100644
--- a/src/io/IOBenchmarks.jl
+++ b/src/io/IOBenchmarks.jl
@@ -28,7 +28,7 @@ g = addgroup!(SUITE, "read", ["buffer", "stream", "string"])
 testbuf = IOBuffer(randstring(RandUtils.SEED, 10^4))
 
 g["read"]       = @benchmarkable perf_read!($testbuf)
-g["readstring"] = @benchmarkable readstring($testbuf)
+g["readstring"] = @benchmarkable read($testbuf, String)
 
 #################################
 # serialization (#18633, #7893) #
diff --git a/src/problem/ProblemBenchmarks.jl b/src/problem/ProblemBenchmarks.jl
index b23b6f84..06e1c95f 100644
--- a/src/problem/ProblemBenchmarks.jl
+++ b/src/problem/ProblemBenchmarks.jl
@@ -81,7 +81,7 @@ include("JSONParse.jl")
 
 g = addgroup!(SUITE, "json", ["parse", "closure"])
 
-jstr = readstring(joinpath(PROBLEM_DATA_DIR, "test.json"))
+jstr = read(joinpath(PROBLEM_DATA_DIR, "test.json"), String)
 
 g["parse_json"] = @benchmarkable JSONParse.perf_parse_json($(jstr))
 
diff --git a/src/problem/SpellCheck.jl b/src/problem/SpellCheck.jl
index 106b6aee..417c0528 100644
--- a/src/problem/SpellCheck.jl
+++ b/src/problem/SpellCheck.jl
@@ -24,7 +24,7 @@ function train(features)
     return model
 end
 
-const NWORDS = train(words(readstring(joinpath(PROBLEM_DATA_DIR, "norvig_spellcheck.txt"))))
+const NWORDS = train(words(read(joinpath(PROBLEM_DATA_DIR, "norvig_spellcheck.txt"), String)))
 
 function edits(word::AbstractString)
     splits     = [(word[1:i], word[i+1:end]) for i=0:length(word) ]
diff --git a/src/shootout/k_nucleotide.jl b/src/shootout/k_nucleotide.jl
index e68440be..9dcd5fd3 100644
--- a/src/shootout/k_nucleotide.jl
+++ b/src/shootout/k_nucleotide.jl
@@ -68,7 +68,7 @@ function perf_k_nucleotide()
             break
         end
     end
-    data = collect(readstring(input))
+    data = collect(read(input, String))
     # delete the newlines and convert to upper case
     i, j = 1, 1
     while i <= length(data)
diff --git a/src/shootout/regex_dna.jl b/src/shootout/regex_dna.jl
index e87cd8e7..eade4ac0 100644
--- a/src/shootout/regex_dna.jl
+++ b/src/shootout/regex_dna.jl
@@ -32,7 +32,7 @@ const subs = [
 
 function perf_regex_dna()
     infile = joinpath(SHOOTOUT_DATA_PATH, "regexdna-input.txt")
-    seq = readstring(infile)
+    seq = read(infile, String)
     l1 = length(seq)
 
     seq = replace(seq, r">.*\n|\n", "")