> The original paper is no longer accessible. (An alternative is to add the equivalent of a big red stamp on it saying "RETRACTED" while leaving the content acc

> The original paper is no longer accessible. (An alternative is to add the equivalent of a big red stamp on it saying "RETRACTED" while leaving the content accessible.) I think you are slightly confused - the "alternative" that you describe is actually what happened in this case (and in all the cases of retracted papers that I've encountered). A Retraction Notice has been published at a separate publication [1] with its own DOI, and it cites the original paper [2]. The original paper [2] has been updated to indicate that it has been retracted. In the HTML version, there is an orange banner saying "This article has been retracted". In the PDF version [3] there is a red header, and the retraction notice has been appended as an extra page. But you are correct that it is important to be able to update PDFs if a significant mistake (or deliberate fraud) is identified. [1]: https://www.science.org/doi/10.1126/scisignal.abn0168 https://www.science.org/doi/10.1126/scisignal.abn0168 [2]: https://www.science.org/doi/10.1126/scisignal.aag1064 https://www.science.org/doi/10.1126/scisignal.aag1064 [3]: https://www.science.org/doi/epdf/10.1126/scisignal.aag1064 https://www.science.org/doi/epdf/10.1126/scisignal.aag1064