Attribution

[sindresorhus]

I noticed you copied a lot of code from https://github.com/yeoman/insight

Definitely ok, but you forgot attribution ;)

From the MIT license:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

https://github.com/twokul/leek/blob/master/lib/send.js
https://github.com/yeoman/insight/blob/master/lib/push.js

[sindresorhus]

Also just curious, why create another one instead of improving Insight?

[twokul]

@sindresorhus Initial implementation was based on insight and I didn't mention it in the README and I should have. I apologize for that.

I actually didn't mean it to be just a copy. I wanted to rewrite it and make the implementation as simple as possible, I just didn't get have time to get to it. Today I refactored insight's original implementation (via 33ab2fd).

As for motivations:

• I don't understand why inquierer is a part of insight. It seems like it has to be a part of the app that uses analytics library, not the analytics library itself.
• I also wanted support for tracking errors, events and timings and insight didn't provide the functionality (I will gladly submit a PR to insight to add it).
• Each request is processed in a separate process. I possibly miss some context, but why?

[twokul]

I updated the README.

[sindresorhus]

wanted to rewrite it and make the implementation as simple as possible

❤️

I don't understand why inquierer is a part of insight. It seems like it has to be a part of the app that uses analytics library, not the analytics library itself.

It's used when asking for permission to track, though in hindsight that should have been a separate component.

I also wanted support for tracking errors, events and timings and insight didn't provide the functionality (I will gladly submit a PR to insight to add it).

Error tracking weren't available when Insight was created, but it's a great idea!

Each request is processed in a separate process. I possibly miss some context, but why?

Insight was mainly created for CLI tools which are usually short-lived. It's also a lot simpler than having to manage a long-lived forked process that might act up, freeze, go crazy and all that. But yeah, reusing the forked process would be better.

It seems like we want the same thing, an awesome GA tracking lib, and I honestly don't see the point in duplicating efforts.

Would you be open to either:

1. merge the two projects and you could become the owner of Insight
2. insight could be rewritten to depend on Leek, becoming just a small sugar layer
3. ?

?

// @addyosmani @passy @stephenplusplus @SBoudrias

[SBoudrias]

I'd definitely give writing rights on Insight to @twokul. Let him improve and rewrite it.

Insight need more love and a maintainer. It'd be great to have someone better taking care of it.

[passy]

Indeed! It's used in Bower as well and we're currently limited by the features exposed through insight. It would be fantastic if either insight or leek would be actively maintained.

[twokul]

@sindresorhus

It seems like we want the same thing, an awesome GA tracking lib, and I honestly don't see the point in duplicating efforts.

agreed.

insight could be rewritten to depend on Leek, becoming just a small sugar layer

that could be an option.

let's chat via gtalk or irc. I'm alex.navasardyan@gmail.com on gtalk and twokul on irc

[stefanpenner]

@twokul likely forking in the future rather would be good. Authors don't like to see there code end up in other repo's missing historical context.

Also, when users want to see "why" something exists, the historical divergence leaves them without context.

[knownasilya]

@twokul what's the status on this?

[twokul]

@knownasilya in progress.I kinda dropped the ball on this one. Will pick it up again.