The debug text with some of the exercises make levelmeup run * confusing.

[eugeneware]

Eg. I get this code for the batch exercise. When you're trying to print your own debug code out too, it can throw you off:

 Σ levelmeuptest$ levelmeup run batch.js
batchable12 = Install with very commonly did twelve hours there will be slower than a specific LevelDB Logo]  phone : chunk[3] : First does splitting and the checksum is for example used for war propaganda buf buf return a native buffer to another log file are stored directly by user isn't root.
batchable13 = While keeping you sure there's a capturing groups.
batchable35 = check for that wage-earners would be written at the way they can be obtained by consuming them ------------------------------------------------- : output formatter by default fill the in-memory store rather than providing tastes which would have open at a paradise in which they are reserved for future wars or preparation for future wars still waiting for you: * the match filenames for which this formatter or writable but not enough for everybody and filters ------------------- *New BSD Licence].
batchable36 = createValueStream.
batchable42 = cc and highly paid.
batchable78 = forum/node-levelup/decompression library!
batchable86 = Good nature is now /tmp/important.
batchable87 = } }X{ /SiblingPage * lt;Your name aliases and vice-versa.

[rvagg]

hm, perhaps I was trying to be too clever with my gibberish, I just needed a data set to use for text! perhaps you can suggest better data sets for these initial exercises that require blobs of text? I'd prefer either something random or a random selection of a larger data set so that it's a bit more dynamic and there's no way you could hardwire stuff.