The intent of this is to clarify the bar for novelty requirements for the main proceedings (full papers and notes) at UbiComp, while also enabling submission to the adjunct tracks (demos, posters, videos, etc) without concern over blocking one's own subsequent main proceeding submission.
Anything in the ACM Digital Library or equivalent formal repository is technically a prior publication for the purposes of review. This includes (but is not limited to) proceedings of workshops, extended abstracts, conferences or journals – anything published in a formal repository. It excludes technical reports and papers that are posted on websites. The PC chairs should be consulted if it is not clear whether a paper is regarded as being part of a formal repository.
However, if a paper makes significant amplifying contributions to prior publication(s) by the same authors, then the reviewers should consider the total contribution (i.e., as if the prior publication(s) did not exist), rather than just the added contribution, when deciding whether to accept the paper. The level of significance is up to the reviewers and program committee to decide. The following are examples of criteria that may be taken into account:
The above is meant to codify existing common reviewing practise, in which early versions of work published in venues such as workshops or as extended abstracts (e.g., accompanying a demo) are not regarded as affecting the contribution level (e.g., the novelty of the core idea) of subsequent mature submissions. It deliberately does not rely on the definitions of “archival” used in the past (since digital publishing has blurred this line), nor on a distinction between workshops and conferences/journals (since the quality bars can overlap).
UbiComp DOES NOT permit simultaneous submission of work with identical contributions to any other publication that is part of a formal repository.
The above is consistent with the relevant ACM policies on Prior Publication and Simultaneous Submissions and Pre-Publication Evaluation.
The most relevant background should be always cited by a paper. This includes the author’s own prior publications or similar simultaneously submitted work.
Relevant self-references must never be omitted or cited as "blank for blind review" for the purposes of anonymization or any other reason.
This includes citations to works that are not yet available to the general public – if something with overlapping contribution (even a minor overlap) is under simultaneous submission elsewhere, then it must be cited (as “In submission to…” or “To appear in…”) and a copy provided to the reviewers (using the supplemental item fields on the submission form, or via the PC chairs). This enables reviewers to consider the novelty of the contributions with respect to one another.
UbiComp follows the ACM policy and procedures on plagiarism (which includes self-plagiarism). As set out by that policy, it is in an author’s best interest to always cite prior work by themselves or others, as this will avoid misperceptions of plagiarism or self-plagiarism. Reviewers should flag any incidence of suspected plagiarism or self-plagiarism to the PC chairs. If the PC chairs reject a paper for plagiarism or self-plagiarism, they should inform the SC chair who shall be responsible for deciding whether to take the case further (e.g. reporting it to the ACM).