November 27,1999
Dear Toronto Colleagues,

	I want to thank you for the good suggestions you have made, in 
particular the finding of mistakes. This is really useful- in the welter of
many changes mistakes creep in, particularly late in the game when the
authors are completely worn out with responding. 

	However I would like to make a suggestion. Could you take a look 
at item App. B.1 in the CDF guidelines for publication? It explicitly asks
that collaborators NOT rewrite papers. There are many different styles of
writing, and as long as it's clear, and above all, correct, the prose should
be left to the authors and the literary godparent. Why, you might ask, should
this be so, when we obviously like it to be different? It is because of the
effect I referred to above: eventually one does more harm than good by
making changes. I referee many papers from D0, and often find sentences without
verbs, missing articles, etc.; how, you may ask, does this happen in a 
collaboration of 500 people. Are they less literate than we are? What is the
mechanism that obvious errors creep in and are not caught before it's mailed?

	First, the question of whether D0 is less literate than we are. It's
unlikely on a statistical basis. More than that, our papers suffer from the
same disease; our dilepton mass paper went out without a reference to the
D0 paper, to pick an example where I know that the authors are exceptionally
careful and conscientious. How did this happen?

	I have long claimed these obvious errors get through because we do not
focus on the important issues in the review process for our papers. Very few
collaborators read our papers, and those who do spend much time and effort in
rewriting. Every rewrite introduces errors: it's a standard rule of thumb in the
software industry that every change has a 50\% chance of introducing a new
error. For papers, I would guess it's not far off: the integrity and coherence
of having a paper written by authors who take full responsibility gets diluted,
and eventually the damn thing just gets mailed.

	The biggest issue in the prose, I would posit, is not whether we really
want to use a comma there, or replace "contribution from" with "branching ratio
for", or remove the word "rather", but are there any glaring blunders, or
sentences that just aren't clear? Rewriting it your way will most likely
make it worse; take a look at Ben Franklin's essay on writing by committee.
(I really recommend this- take it to heart!).

	Lastly, I could use some help on some serious authorship issues. We
still don't have a top cross-section; the written record of the CDF top
cross section is, to put it politely, a mess. Would you be willing to use
some of your (obvious) excess time and energy to push on getting the top
cross section paper out? It's much more important than many of the nits you
have picked. It's important for young folk to focus on the important issues
in a big collaboration; remember Pauli: 'so young, and already done so little'.

	So, in conclusion, I really appreciate the comments related to content;
finding the errors you've commented on has been really important. The Toronto
group is doing a real service by reading the papers carefully. All I ask is
that you restrain yourselves in rewriting; it generally makes things worse,
and is contrary to CDF policy for that reason. And, if you can identify the
really important papers that we HAVEN'T written, and can use your talents
to writing them, rather than rewriting extant papers, you will make a major
contribution to the collaboration.

				Sincerely, and best wishes,
					Henry