- engage relevant people with what I am doing
- get constructive feedback on the paper and the science
- get the paper published as smoothly and quickly as possible.
Here are my usual steps:
1. get a local colleague to read the paper for feedback
2. put the paper on the arXiv [and write a post about it on this blog].
3. send the preprint to a few people who might be referees or be able to provide useful feedback
4. revise the paper in response to feedback, including another proof read
5. submit the paper to a journal, sometimes suggesting people who have provided positive feedback already as possible referees.
A few comments and rationale.
My target journals are mostly Physical Review B and Journal of Chemical Physics. About once a year I send something to PRL. For idealistic reasons, I abstain from the luxury journals and want to minimise publishing in journals from commercial publishers such as Elsevier. Going down the journal "food chain" involves continually reformatting a paper, wasting a lot of time.
If you publish in American Chemical Society journals it is not clear you can post on the arXiv.
Why not exchange steps 2 and 3? i.e. send it out for private comment before putting on the arXiv. I used to sometimes do that. However, I had a few instances where people would say things like, "we have some similar results, can we submit at the same time" or "we have some related experimental results can we publish together" or "lets combine your paper with ours" or "you really should also calculate XYZ ....". This can lead to significant time delays and (particularly in hindsight) debatable benefits. Ignoring their advice or requests can be awkward. If the paper is already on the arXiv it pre-empts some of this.
I welcome comments. What is your preferred strategy?