Owner for the generator system #193

Closed
addyosmani opened this Issue Mar 20, 2013 · 13 comments

Projects

None yet

4 participants

Owner

Something that would help the project out tremendously is having a new owner for the generator system - someone that has studied the implementation, understands the insides and would be willing to help us review issues and pull requests for this part of the project.

To give you some insight into how important it is for us to find an owner, one of the main things blocking out 1.0 release is actually good docs on our generator system. This system was previously maintained by @mklabs but as he's understandably strapped for time these days, a new owner would be great.

I am happy to help whoever is able to assist with addressing issues / docs as time allows.

@kevva we were thinking this might be something you might consider getting involved with - interested? :). It's pretty central to the project but is also all kinds of amazing.

cc @sindresorhus @passy

The single most important tool missing from the tooling ecosystem right now is a proper, long-term generator package.

To my understanding, Grunt has given way to yeoman by intentionally not focusing on grunt-init.

What is the reason that the generator project isn't the one thing that yeoman does perfectly? Meaning, why isn't the team focused on creating a robust generator system?

Owner

A long-term generator package is definitely needed right now - we all agree
with that.

It was however only recently that scaffolding became a core focus for
this project (we previously tried doing a lot more) and the system will
take time to improve.

Why is it taking so long? Not enough team members are familiar with our
current implementation and we've been lately trying to better understand
what is missing and what can be proved.

Give us a little time and I have no doubt we will be able to deliver a rock
solid solution the community can rely on.

On Wednesday, March 20, 2013, Thanasis Polychronakis wrote:

The single most important tool missing from the tooling ecosystem right
now is a proper, long-term generator package.

To my understanding, Grunt has given way to yeoman by intentionally not
focusing on grunt-init.

What is the reason that the generator project isn't the one thing that
yeoman does perfectly? Meaning, why isn't the team focused on creating a
robust generator system?


Reply to this email directly or view it on GitHubhttps://github.com/yeoman/generator/issues/193#issuecomment-15200669
.

Addy Osmani

Developer Programs Engineer at Google
Blogger at: http://addyosmani.com
Phone: +44 7771 334170

Owner

To my understanding, Grunt has given way to yeoman by intentionally not focusing on grunt-init.

Not really. Init didn't really fit into grunt at all, so Ben decided to deprecate it. Anyone is free to create a replacement. Yo is just one of them.

@addyosmani i came to the conclusion that yeoman should mainly focus on scaffolding by realizing that there is absolutely no way that i will not lay a hand on Gruntfile.js. It is actually impossible to not edit the file.

Which has as an effect that web development using only the yeoman cli is impossible too.

What this means to me, is that less effort should be given in trying to abstract Gruntfile as Gruntfile is enough of an abstraction by itself. I am talking about the rigorous wiring / templating i've seen in yeoman generated Gruntfiles.

Creating a powerful scaffolding engine and providing thin glue for Grunt and Bower is my projection of what yeoman should be... But that's just me. I think that the most important issue is to define who is the end-user of yeoman. You can't serve them all. Not now, with so few resources at least.

@sindresorhus what you say is true... But i am confident i saw in Grunt docs somewhere that the init task gets deprecated and mentioning yeoman explicitly as the project to look for that functionality.

Owner

It is actually impossible to not edit the file.

You're thinking about 0.9.6, in 1.0 you're free to edit the Gruntfile as much as you like.

Creating a powerful scaffolding engine and providing thin glue for Grunt and Bower is my projection of what yeoman should be...

I don't agree. For me scaffolding is only a small part (though still important) of Yeoman. It's a means to an end for us to get customized workflows out to the user. But the main mission of Yeoman is still to optimize developer happiness by creating a great workflow for users, whatever tools and guides that entails.

Grunt docs somewhere that the init task gets deprecated and mentioning yeoman explicitly as the project to look for that functionality.

That's correct. If Ben likes the generator system he will recommend it as a replacement.

But the main mission of Yeoman is still to optimize developer happiness by creating a great workflow for users

Yes. What is missing from a great workflow atm is scaffolding. The rest is taken care of by Grunt and Bower, are we talking curating plugins?

@kevva kevva referenced this issue in yeoman/yeoman Mar 28, 2013
Closed

Meta-bug for task ownership #997

0 of 15 tasks complete
Owner

We're still looking for an owner of the generator system. Comment if you're interested ;)

I am interested. Can make it work mid to long term

Contributor

I still have some studying to do on the internals of the generator system, but looking over the meta ticket, I'd be willing to help with the "face of Yeoman" tasks, ie docs, tutorials, help on SO & here. At the very least, if anyone else takes on the "ownership" role, I'd be equally happy to work with them to find a fluid design/documentation approach, helping wherever possible.

Owner

Over the past month the team (in particular @passy @kevva etc) have really stepped up to help us with the generator system. @stephenplusplus I would love some help with the generator docs if you have time to spare. The reception to the recent updates has been warm, but we could definitely add more succinct examples. Happy to point you at some specific feedback if you'd like to help :)

see #206

Contributor

I'm all over it, good sirs!

After four minutes of searching for the right emoticon, I give up. Please don't judge my writing skills on lack of emoticon usage. :)

Contributor

Hey @addyosmani, I'm still mulling over some doc ideas, and kicking around some generator ideas. Just wanted to let you know I haven't forgotten about the task and if all goes well, will have something to show soon!

Contributor

Anyone who might be looking here, but not following #206, please go check it out! I've posted a link to a walkthrough I've drafted, and would appreciate any of your available time and feedback.

@addyosmani addyosmani closed this Jul 5, 2013
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment