06:05 <ddfreyne> I'd like to split up nanoc in three parts:

06:06 <ddfreyne> * core - containing only interfaces to external stores (data sources, but also the item rep writer)

06:06 <ddfreyne> * base - containing basic things necessary to work with core. filesystem data source, standard item rep writer, erb filter. Maybe the CLI as well.

06:07 <ddfreyne> * extra - containing all the rest, i.e. the plugin

06:07 <ddfreyne> s

06:07 <ddfreyne> What do you think?

08:22 <darix> ddfreyne: "why?"

08:28 <guardian> o/

08:29 <guardian> musicmatze> Fuckfuckfuck. Just did a git reset --hard on changes I wanted to keep!!! AAAAAH FUUUUCK!

08:29 <guardian> that's why you want to commit work in progress commits often

08:30 <guardian> then git rebase -i to concatenate them in a single commit eventually

08:30 <guardian> so that git reset —hard can be saved by looking at git reflog

08:30 <guardian> of course you still lose everything that's not committed with git reset —hard

08:31 <musicmatze> guardian: That's what I did, but I resetted too much!

08:32 <musicmatze> I don't even know anymore if I have lost anything.

08:32 <guardian> if you committed you can look at git reflog to see if you find a SHA1 which content matches what you believe you were at

08:33 <cDlm> ddfreyne: why the core/base separation ?

08:33 <musicmatze> jo, I did that. I don't think there was something lost...

08:34 <musicmatze> anyway, the kernel will be almost ready in two or three weeks. Then, I have to check the compiler error output and check the whole thing for bugs and so on.

08:34 <musicmatze> But that's not the topic of this channel...

08:56 <ariejan> I have a site with 600+ blog posts. Doing design development is pretty tedious as recompiles take 10-20 seconds. Besides moving posts to a draft folder during development, are there any other solutions to this?