[Rubygems-developers] is this thing on?

James Tucker jftucker at gmail.com
Sun Oct 17 15:14:22 EDT 2010


On 17 Oct 2010, at 13:39, Jon wrote:

>>>> So I've been wondering: what is going on? -- this goes specially to
>>>> Eric, who is lead and maintainer of RubyGems -- are you too busy?
>>> 
>>> This is just how Eric is. Frustrating as it may be at times, Eric works on what he wants to work on when he wants to work on it. It keeps him sane. Currently his focus is on rdoc. He'll cycle back around to rubygems before long. I through all my years of working with him have not found an effective means of escalating an issue to him when he doesn't want to pay attention to it.
>>> 
> 
> This response, while informative, is both unhelpful and may fuel existing perceptions that needlessly undermines both the project and Eric-as-Lead.  I'm not saying this to start a flame, but the response misses the root issue and does nothing to better perceptions.

Ok.

> As a RubyGems user all I really want to see is that the project is managed so that key fixes and features don't linger any longer than needed to create robust solutions.  How this is implemented isn't (and shouldn't be) my business.

IMO rushing in features is not a good idea for a project of this scope. Not sure if you've looked, but a lot of the patches we receive have platform or design issues, or lack of tests.

As for key issues, well the key issues no one seems to supply patches for, and they aren't always trivial to approach cross-platform (that is both platform wise, and interpreter wise, etc).

> IMO, the root issue appears simply to be that a trusted backup lead/maintainer hasn't been identified who works closely with the lead and who's primary role is to keep things from stagnating when the primary lead/maintainer is unavailable.  This has nothing specifically to do with Eric.  It's simply a practical acknowledgement that all RubyGems contributors are (a) working for free, (b) have lives/passions outside of Rubygems, and (c) need "me time" to keep sane and passionate.

People who supply good patches with tests get on the team if they ask. That's why Eric gave me a commit bit, hell, one of my patches even had some re-arrangement details for win32, so, there's not even an expectation of perfection at all, just that you supply committable patches often enough for the devs to want to "get out of your way". Honestly, we don't get many people fall into that category.

> Bluntly, I think the "fix" is for Eric and the contributors to take some time and go offline to quickly figure out a more flexible and realistic way to manage the project.

What exactly is it that you think is so wrong? Maybe if you can explain what the actual specific problem is, I can address it.


More information about the Rubygems-developers mailing list