The future of pelican

future.rst

Pelican 4.0 - Keep It Simple, Stupid.

I don't know if you agree with me on that, but I think that pelican does more than it should be doing currently. It was at the very beginning a simple tool, without too much corner cases, and now we have something that starts to be too big (at least to me): we cover the creation of a blog from scratch, with the quickstart thing, which generate a makefile for you, for instance, with the pelican dev-server, etc. All these things could be other projects, I think, I would go this direction if that works for you.

This way, we could keep the pelican core easy to contribute to, and we could identify who is the maintainer of which part quite easily.

Here is a proposal to make this happen:

Phase 1 - Cut some stuff out

• focus a bit more on the core value of pelican: what do we want to provide with this tool?
• Move stuff into plugins when it makes sense
• don't accept pull requests if they add features that could be provided otherwise
• Split-off the plugins into another repository (pelican-plugins)
• Remove the assets things, I don't know if that's worth it
• Review all the issues and the pull requests. When they are too old, we should either mark them as "lost in the interwebs" or take the implementation over if we think that makes sense.

Phase 2 - Add niceties

• A new default theme. I quite like the Octopress one for instance, I also love some of the pelican themes, we should maybe contact the authors of these themes about that
• Rework the way the settings work. I'm not sure having everything at the top-level is a good idea

Phase 3 - Communication

• Setup a real mailing list for pelican, authors@getpelican.com should not be used I think, for other things than "thank you" ;)
• Ask for feedback on the pelican blog. We probably should be chattier there.
• what about creating a planet?

And of course, Phase 4 is to release all this and spread the world!

Thoughts? I propose you to discuss this either on #pelican (freenode) or here in the comments.

+1 to everything :)

This also means that a bunch of features may disapear from pelican core, and that the migration from old pelican blogs to the new version may or may not work. I think we're talking about power users so that's not really a problem, but that's only my POV.

As I mentioned on twitter, just from following this project occasionally and using it for my own blog I was starting to come to the same conclusion. Not that it isn't fast, just that it really encompasses a lot. And for future planning, I would certain agree that there should be a mailing list for pelican set up where the discussion can take place. On the other hand, I think what would certainly aid the conversation is to talk about what must be kept and not what must be thrown away. The former keeps the attitude more positive and will hopefully alienate fewer people who've come to rely on some of these extra features.

Cool features :) , as far as migration is concerned, I think for most part people using pelican can handle, though mailing list might be helpful in this case