On every project I've ever worked on, specs were ephemera, falling out of date almost immediately despite occasional herculean efforts to keep them up-to-date.
Personally, the best answer I've found to the spec/code dilemma to be automated tests, as they evolve along with the code and fail if they're out of date. It sounds also like the project you're talking about had a lot of good ol' fashioned technical debt. Code won't be readable if it's badly written - but neither will natural language. Cheers, Chris On 22 May 2013 20:29, Tim Daly <d...@axiom-developer.org> wrote: > Is "the ground truth" your spec or your code? > > Here is an interesting read: > http://shanecelis.github.io/2013/05/20/why-im-trying-literate-programming > > Shane started with a co-worker, working from a spec, to create a program. > He eventually found that only he could make changes because only he > understood the code and the spec was out of date. > > The last big project I worked on had 6 people for 6 years. The central > data structure eventually became complex. It had optimizations and > mountains of code that depended on them. When we tried to write a new > and "better" central algorithm it turned out that nobody knew all the > various substructures embedded in the main data structure so we > couldn't make the improvement. The person who "managed" the main data > structure had left the project, taking with him all the knowledge. > The program died. > > Are you limiting our ability to collaborate because you don't communicate? > Do we have to read (and reverse engineer) your code before we can write? > Does your code "add, change, or extend" the spec with special cases? > > Can you keep up with the whole team using reverse engineering? > Can you identify "the person" who holds it all together? > Is your whole project "dead code" if certain people leave? > > If you want your code to live, communicate. > Write words for people who will maintain your code but you'll never meet. > > Tim Daly > Knuth fanboi > > -- > -- > You received this message because you are subscribed to the Google > Groups "Clojure" group. > To post to this group, send email to clojure@googlegroups.com > Note that posts from new members are moderated - please be patient with > your first post. > To unsubscribe from this group, send email to > clojure+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/clojure?hl=en > --- > You received this message because you are subscribed to the Google Groups > "Clojure" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to clojure+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
