This’ll have to be quick as I am writing this on an iPhone and the tiny keyboard and auto-misscorrection is killing me. I’m away from home and should be relaxing sipping beer, but instead I’m micro-analysing our ludum dare jam entry.

Our game, the small world of professor strange ( http://www.ludumdare.com/compo/ludum-dare-23/?action=preview&uid=10313 ) was quite daring in the sense that very few text adventures are entered. We wanted to be different, plus we had our own ideas how text adventures could be made more up to date. Text adventures are vintage, but I remember spending hours playing them in the 80s. Everyone had a friend who’d already completed them all so you knew it was possible, you just couldn’t find the right key or tool or wherever.

Want went right:
The artwork (my part) went much better than expected. We chose to be different and hand draw everything in ms paint and rendered in GIMP. The real reason was because it’s the fastest way I could create masses of artwork. We drew a sketchmap first, then I rendered it (4-6 hrs) then we started characters, sketches and interiors.

I also think the coding and text content went great considering the time limits. The whole project was smoothly coordinated, except calling James in at short notice to help with some text.

What we could do better:
For sure we underestimated the problem of not having good text parsing for commands. We knew we couldn’t write a decent natural language parser in 72hrs so we didn’t even try. You have to enter valid commands as laid out in the JavaScript data file. This is an obvious barrier to entry, many players get frustrated very quickly. The most consistent criticism we’ve had is that text parsing is missing and help not obvious enough (we improved that in a post entry update).

Another pain point was text saturation. We placed text descriptions in the relevant location on the map, but these soon get overcrowded and overlapping. Again we improved this after entry but it’s definitely something that could do with more attention. The fundemental idea of putting the text in the location is a good one I think.

Amusingly we also got someone pointing out spelling mistakes and grammar (mostly bad cut and pasts actually). Oh well you cannot please everyone.

Overall I’m quite pleased. Looking forward to LD#24, I think well keep the team and do something quite different.