There are
benefits and drawbacks to either approach. Some simply can
maintain a better narrative flow if they wait until they are finished to
add the sourcing. They may have a stack of sources sitting beside their
computer, but not be sure which will be relevant until they are well
advanced intheir writing. Presumably they are making private notes as
they go along so that they will be able to add the references when they
get to the bottom of the article. If an article appears to be a work
that is actively in progress it is good etiquette to let the editor
finish what he is doing before complaining about missing sources.
Oh certainly, you don't need to add the sources to the code as you go
along, but you need to know what the sources are. The sources should
be added in the same editing session as the article is written, the
exact order within that session is up to the writer. Someone else
coming along after you've written the article and trying to find
sources for your statements is completely wrong - it's not what
"source" means. They are effectively rewriting the article.
If you want different people to do the sourcing and the writing, then
the sourcer should find the information and add it as bullet points to
the article, with correct references, and then the writer comes along
and makes it read eloquently. The other way round is completely wrong.
I'm not at all saying that an editor should depend on others to find
sources. I am saying that there is no urgency to add them in each
editing session. When you look at the article's history, and see a
regular and continuing series of diffs that keep adding material to the
article you need to be ready to cut that person some slack.
Ec