How I reduced my Haskell CI time by 84%


Happy to hear any feedback!

Quick personal feedback: based on the title I was expecting to discover some amazing, breakthrough, never-seen-before technique that adds a new dimension to “doing CI fast” with Haskell.

Then 2 minutes into the article, I was a bit underwhelmed, thinking to myself (sarcastically) “ah yeah, well you implemented caching, good job”.

Apart from that, I can feel an overuse of “super-fast”, “lightning-speed” and similar qualifiers that are more an annoyance than anything else.

All in all, article tone and title are overselling what is essentially a basic 101 practice. It’s a good and useful thing to explain how to implement CI caching specifically in Haskell, but maybe the article would gain being more honest and to-the-point.

Other than that it’s very clear and easy to read, and informative.

1 Like

I’m not the only person to over-sell a post in the title… I don’t think it’s that sneaky, but I apologise if it was. I didn’t say ‘super-fast’ - I gave an empirical metric based on my data.

Yes, everything there is completely straightforward to anyone with any CI experience. But many people are not familiar with CI, even intimidated by it (I was.) They were my target audience - I wanted to demonstrate that a large and complex config file is unneccesary, with only 14 lines of the unfamiliar technology you can go beyond making CI work - you can make it work for you.

Maybe I wasn’t sufficiently clear that the post wasn’t very Haskell-specific.

Thank you for your feedback!


Nice post! Clear and to the point!