Instructions for Reviewers

You may review pretty much anything -- paper, book, software, demo, etc. -- as long it is likely to be of technical interest to people in the database field. You will need to be able to provide a URL of the object you are reviewing, or at least a URL to a complete citation. You will also need to make up a unique ID for the object that you are reviewing. All of this book-keeping is taken care of for you, if you choose to review "standard" material, as described next.

The vast majority of reviews are of papers referenced in the Uni. Trier DBLP Bibliography. Use the special search form we have to find the paper you wish to review in this bibliography. Once you find the paper, our tools will automatically create the URL and paperID.

If you read a tech report, or other unpublished work, and would like to review it, your first step should be to contact the author, declare your intentions, and see if the author is willing to place the report in CoRR so that you can obtain a valid ID for an archived document. Once more, if you choose a paper from CoRR using our interface, the paperID and URL will be populated automatically.

The easiest way to to identify a document for review is to find an existing review and click on the "review same" link. You will immediately be able to enter your views on the paper that has previously been reviewed by someone else for Digital Review.

All reviews are expected to restrict themselves to technical observations germane to critiquing the paper at hand. Ad hominem attacks, personal insults, and flames, are not acceptable.

How much you write in a review is up to you. However, keep in mind that the review will be public with you identified as the reviewer. If all you have to say is "great paper", you may develop a reputation as a reviewer that you do not really want.

Conceivably, you could modify your review if you change your opinion with time. The way to do this is to re-enter your review, with the needed modifications. Reviews are keyed by paperID and reviewerID.. The new review will replace the old review by you for the specified paper. If you assigned the paper ID yourself the first time, make sure that you use the same ID when you revise. The easiest way to do so is to use the "review same" link after finding your old review.

If you wish to change the paper reference, for instance what was an unpublished tech. report gets listed in DBLP, you will need to contact us. We will put in the change by hand.

Use HTML to enter your reviews. If you use plain ASCII, any formatting in your reviews (such as paragraph boundaries, and lists) will be lost. Use of HTML tags is critical. If you wish to reference some other work, use an HREF to an appropriate URL where possible.

You will be asked to sign electronically a "permission to reproduce" form. If you have included third party material in your review, it is your responsibility to get the requisite permissions.