[Rubygems-developers] Rubygems Packaging Enhancements and Designs

Hugh Sasse hgs at dmu.ac.uk
Wed Mar 31 20:41:59 EDT 2010

On Thu, 1 Apr 2010, Luis Lavena wrote:

> On Thu, Apr 1, 2010 at 1:45 AM, Hugh Sasse <hgs at dmu.ac.uk> wrote:
> > On Thu, 1 Apr 2010, James Tucker wrote:
> >
> >>
> >> On 31 Mar 2010, at 21:05, Hugh Sasse wrote:
> >> > That's clear and readily found.  What isn't so clear is what is going on
> >> > with require, gem versioning, and the deprecation of require_gem. Examples
> >> > in some books will need to be changed.
> >>
> >> I have done a first pass through the first book and fixed many of the broken items and loose markup, as well as added updates to reflect recent rubygems.org introduction and the gem push command.
> >
> > I was still talking about examples from print books that are out
> > there.  Some will be made obsolete by more recent editions, we face
> > the same upper bounds imposed by availability of people to do this,
> > but I'm coming from the position that a continual criticism of Ruby
> > has been the limits on its documentation.  Provision of these
> > corrections seems like it could be a good aspiration, even if we
> > only make limited progress.
> >
> Sorry to chime in, but is ilogical to assume that we can control an
> update every book out there.

Agreed, we can't control or force updates, and some are far too old
or obscure to merit much 3rd party effort..

> What you're suggesting is that backward compatibility should be
> maintained because "Programming Ruby", based on Ruby 1.6 is still out
> there in the hands of someone.

No, I'm suggesting that it would be useful to provide some information
such that people can find this in relation to their book.  (I'm certainly
not suggesting we hold back development in order to meet constraints
imposed by old texts.)  I doubt much progress could be made in such an
effort, but some might be possible, and it would help people get started.
Clearly it is a low priority, but I'd only ask that it be kept in mind.

I can state a case for this, but if everyone is so opposed that it is
to be ruled out already, then I'll save the bandwidth.
> It is true that RubyGems documentation site hasn't been updated, but
> if noone volunteers or provide feedback enough, that is not going to
> change.

> Developers are used to peek into the source. To step up as maintainer
> of a project you need to know what is made of and what are the tools
> to make it work.

Yes, agreed.  There are problems with this model, and the open source
movement doesn't seem to address them....

> That is true for any project out there.

...and the problems apply as broadly, in my experience.

More information about the Rubygems-developers mailing list