tech-userlevel archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: a 4-part deletion test (was Re: tn3270, mset and map3270)

On Sat, Jan 16, 2010 at 08:17:51PM -0600, David Young wrote:
 > 1) Nobody uses the function, and nobody even wants to.
 > 2) The burden of maintaining the function inhibits the system's
 >    maintenance in some *extraordinary* way.  Consider prioritizing
 >    functions in a few broad categories (kernel & userland, for example)
 >    for deletion.
 > 3) No one offers to maintain the function.
 > 4) The function is not less valuable to its maintainers outside
 >    of the base system than inside it---e.g., there's not any use
 >    cross-building it.
 > Finally, food for thought: if a function that is a candidate for
 > deletion meets criteria 2 through 4, but not criterion 1, perhaps it
 > should be deleted?

This is not the right set of criteria (especially the last proposition
there) as this would allow removing ugly things that people use widely
(like e.g. nullfs) but not allow removing obsolete things that almost
nobody in their right mind cares about (like e.g. rlogind).

That said, anything that's causing a major maintenance burden ought to
be reviewed, not necessarily for removal but more for figuring out why
it's causing that burden and how it can be fixed.

In any case, I don't think the knee-jerk reaction of removing things
when a problem arises is healthy for the project.

David A. Holland

Home | Main Index | Thread Index | Old Index