1. 28
  1.  

  2. 9

    Hey, mods? For the last five years or so, I’ve titled these articles “Jepsen: [Database Name] [Version]”. I’ve been informed that that’s no longer an acceptable title for Jepsen reports, because “Jepsen” is also the name of the site. Can I get a policy clarification here? Do I need to make up a lobste.rs-friendly title that doesn’t include the word “Jepsen” for each submission from now on?

    1. 7

      The story submission guidelines state, in part: “Please remove extraneous components from titles such as the name of the site, blog, section, and author.” The story was originally submitted with the story title “Jepsen: YugaByte DB 1.3.1” from the domain jepsen.io. Given that the site name appears in the story title, and consistent with the story submission guidelines, I replaced “Jepsen: “ with “Testing “ as the best operational description of the article I could find.

      It has since been brought to my attention that Jepsen is also the name of a Clojure library. As a site name, the story submission guidelines are clear. As a library or tool name, they are mute. As Jepsen is both, but given the existence of an interpretation of the article title that does not violate the site guidelines, I have restored the story title as originally submitted.

      1. 4

        It has since been brought to my attention that Jepsen is also the name of a Clojure library. As a site name, the story submission guidelines are clear. As a library or tool name, they are mute. As Jepsen is both, but given the existence of an interpretation of the article title that does not violate the site guidelines, I have restored the story title as originally submitted.

        This is such a wacky justification! Somewhere, a lawyer smiles without knowing why.

        But I like the “Jepsen: WebscaleDB 0.1.1-alpha2” titles, so I’ll take it.

        1. 3

          Absent a Damas-Hindley-Milner typechecker for story titles, it’s just us manually splitting hairs to try to keep up their signal:noise ratio. I was also surprised by Jepsen being the name of the tool used for the analysis as well as the project and consulting company I knew about. I can only ask that @aphyr disambiguate with Hungarian Notation - perhaps he’d like to adopt Jepsen_clj, Jepsen_blog, and Jepsen_LLC so we can more efficiently mangle his titles in the future.

          1. 5

            I like to think the usage in the submission title here is a fourth meaning of “Jepsen”, being a proper name not of the blog itself but of the kind of article the blog contains, i.e. “a Jepsen of YugaByte DB”

        2. 3

          Thank you. For what it’s worth, I prefer to see Jepsen in the title, because in my mind, it adds a ton of credibility for the content.

      2. 5

        This is awesome. I am far from an expert in databases but being able to push this hard and break database consistency requirements (and find bugs that are the combo of really high impact and hard to debug and time sensitive!) so often is really impressive.

        @aphyr btw I saw your talk at Deconstruct this year and it was one of my favorite talks I’d ever seen

        Jepsen is an international treasure, and I’m glad I can read these. If only I could sit around and do these kinds of tests all day….