Thursday, June 20, 2019

NarraScope 2019: complete

NarraScope 2019 is over. Here's what I wrote about it.
You've probably already read this blog post. But NarraScope was such a large part of my past year that I need to at least mention it on my personal blog.
Yowza, that was great.
See you in 2020.


  1. ...And, apparently Graham Nelson announced at NarraScope that Inform7 isn't dead after all, but in fact is freshly rewritten and soon to be open-source (but with Graham still at the helm). This is wonderful news--the biggest and best IF news I've heard in a long time. (A bit surprised not to have read about it here yet?) Looking forward to your take on this: seems like a major shift in the overall text IF situation.

  2. I'm not sure I have a take beyond "this will be good, let's see where it goes".

    I7 not being open-source was a problem, but the problems it caused were second-order. It was hard to keep Linux builds up to date on all Linux variants. You couldn't be sure of being able to build old projects because the compiler binaries were out of date. Some annoying bugs lingered for years and people kept tripping over them.

    I don't think these were serious blockers to the use of I7. They were friction. With the upcoming release (which will include source repositories of the last several I7 releases), the friction will be reduced, but I'm not sure it will amount to a phase change.

    The use of Inter as a new intermediate format is fascinating, but we're still several steps away from the implementation of a new target platform.

    What I'm hoping for in the short term is a more cohesive volunteer ecosystem *around* I7: extensions management, bug management, etc. Some of us (including Graham) had a bit of a conversation about this at the con. IFTF may have a role in this; nothing is decided though.

    1. Fair enough. For me personally, I felt like the "bug lingered for years" and "build old projects" part - with I7 news kind of drying up as well - just seemed more and more like a general warning not to rely too heavily on the future of an ambitious, closed, small team project. Not that I don't trust Graham (the project's history had been a pretty smooth upward slope until the last few years, with exhaustive bugfix rollups and lovely change logs) but the bus factor was starting to seem like a real long-term liability. That's sorted now, in what looks to me like the best possible way. I agree the specific bugs themselves are just friction; hopefully this means faster fixes for any problems, but to me the big shift is that now, where there's friction, there will be oiling ports and 'user serviceable parts' going forward, not just the one genius janitor who knows how to talk to the plumbing.

      Of course this is an outside view rooted in generalities, and I doubt I'd personally understand the code well enough to fix bugs, but I do like the idea that more people might eventually be available to look at my bug _reports_, leaving Graham more time for the bigger ideas.

      I'm also thrilled about the progress towards multiple targets, even if that's a long way off. Inform7 is so nice for modeling "situations", I'd love to be using it (someday) as something like a game engine scripting language.