I really do not understad  what the suggestion, practically should be! Actually a paper is in a continuous provision for deletion/amendment. If some error is detected immediately there is a correction publication. I know that the Clay Organization has two years waiting before money is payed:

"Before consideration, a proposed solution must be published in a refereed mathematics publication of worldwide repute (or such other form as the SAB shall determine qualifies), and it must also have general acceptance in the mathematics community two years after. Following this two-year waiting period,"

http://www.claymath.org/millennium-problems/rules-millennium-prizes  

It seems the question needs to be elaborated.

OK. Let us see this through the following example:

Article: Some physical properties of wheat and paddy and their relationships.

AMA (1987), vol 18, No. 1, pp 45 -50.

In this study, an empirical equation was developed to predict the volume of a single wheat grain as a function of the grain dimensions and a comparison was made with Mohsenin’s equation. It was reported that Mohsenin’s prediction involved an error of 69.3% whereas this one showed an error of 10.22%. This was alarming as Mohsenin, regarded as an authority in properties of biological materials, should not have recommended a prediction equation with 69.3% deviation.

Mohsenin, in fact, hadn’t!!!

A gross misinterpretation exercised by the authors was that they considered the wheat grain as an oblate spheroid and applied the corresponding volume equation leading to the reported error, (69.3%), which was absolutely inappropriate. Actually a wheat grain resembles a prolate spheroid and the corresponding volume equation would yield the predicted volume values with errors around 3% (upholding Mohsenin's reputation).

The article, published in 1987, has been in circulation for about 27 years, during which period it must have been (actually, has been) cited and applied (wrongly) by many researchers in their subsequent research articles further propagating misleading erroneous results.

Hope this example drives home the impact of an erroneous presentation. My question may be viewed in this background.

Thanks Dr Kamal.

If a mistake happens one time after submission, without provision and modification, this mistake appears every day as if it is made every day, and will propagate exponentially, eventually it is taken by young researchers as true science. 

I think it was difficult to modify a published paper in the past, because the article is printed out on thousands of copies all over the world. however with the advances in IT it is easy to do such amendments.

For more information about what and how those mistakes are made please refer to the work of Fausto Galetto and his famous cases at: https://www.researchgate.net/profile/Fausto_Galetto/contributions?ev=brs_act

Most of the items in the checklist are common to all scientific reporting, but virtually all of them have been modified to reflect the unique nature of improvement work. The impact of discrepancy should be mitigated by the inclusion of more information about aspects of the providers practice settings that could affect the study outcomes.

http://squire-statement.org/guidelines/explanation_and_elaboration/16_limitations/

Very fine suggestion to me, dear Shanker!

What is the definition of an error? Is old knowledge that has been rejected by new findings an 'error' and should it therefore be deleted from history?

I really do not understad  what the suggestion, practically should be! Actually a paper is in a continuous provision for deletion/amendment. If some error is detected immediately there is a correction publication. I know that the Clay Organization has two years waiting before money is payed:

"Before consideration, a proposed solution must be published in a refereed mathematics publication of worldwide repute (or such other form as the SAB shall determine qualifies), and it must also have general acceptance in the mathematics community two years after. Following this two-year waiting period,"

http://www.claymath.org/millennium-problems/rules-millennium-prizes  

Dear Dr Marcel,

I have explained the nature of error by citing (please see the example cited by me above) a published article with a gross mistake in  adopting an established mathematical formula and wrongly claiming an inferior prediction to be better than the established literature showing it as an absurd prediction. Adopting the correct formula yields a far better prediction. Thus the erroneous article (cited by followers) is  based on a wrong foundation leading to an irrelevant outcome.

Progressive developments in accepted theories are quite natural and do not fall under the cited circumstances. The question relates to the cases where established theories are wrongly contradicted through irrelevant analysis.

Dear Shanker,

thank you very much for the precision.

Dear Shankar,

In school and engineering we still read about Newton (and we know most of the theory don't stand now) and what is the guarantee that whatever you are taking as truth as of now  is also right.

Deletion is not the solution but to review each paper thoroughly is.

Regards,

Bhushan Poojary

Dear Bhushan,

Please read the question and the subsequent discussion carefully and then answer.

Your comment indicates that you have not read the cited example and the discussion that followed, carefully.

Most published scientific research papers are wrong, according to a new analysis. Assuming that the new paper is itself correct, problems with experimental and statistical methods mean that there is less than a 50% chance that the results of any randomly chosen scientific paper are true.

Reasons combine to make most research findings false.

http://www.newscientist.com/article/dn7915-most-scientific-papers-are-probably-wrong.html#.VDSyKVckyX0

Thanks Dr Krishnan. Ascertaining the points mentioned as well as those in the link suggested by you is the very responsibility of the reviewers and editors. The job of  reviewing, therefore, should not be assigned at random. Moreover, before accepting or rejecting an article, the Editors must evaluate the article as well as reviewers' comments very critically themselves instead of relying solely upon the reviewers. In fact, in one of my other questions [What is the recognition accorded to a manuscript reviewer?], I have stressed upon the accountability of the reviewers and also the need for according them a due recognition in order to make the system transparent and reliable.

A case of an article deleted after its publication in a journal with a high IF was experienced by myself. Some authors from other Italian towns and me raised an ethical problem about the use of an antibiotic in H. pylori eradication regimen. Our assessments induced a strong disagreement by an American group of researchers. The disagreement threatened to result in a "legal battle", and so the Editor decided to cancel  our article and the response of the other group to avoid of involving the issue in a "scientific.brawl". A real experienece, indeed.

What a unpleasant experience dear @Lerardi! I do think that editor should not make this action, but to publish the addendum to your article explaining the results of American group of researchers!!!

Withdrawal of a published article is certainly a disappointment to the authors. Hence it is essential that such a withdrawal must be effected only after ascertaining/ establishing  beyond doubt that the errors noticed/pointed out are real and must not be propagated so as to  uphold the standards of research. Once this is done, there should be no  misgivings/ ego clash among all those concerned (whether in favour or otherwise).

The question that I raised, w.r.t. a specific cited example and the like, pertains to situations where the error involved is crystal clear, as clear as somebody claiming that 2 x 2 = 17. I hope nobody, including the author, will have any issues in accepting such an error which certainly ( / most probably) must have occurred because of an oversight.

Indulging in a legal battle to establish a scientific fact is highly unprofessional.

@Shanker, I agree with your consideration about the unprofessional behaviour. However, also in this case the power has the upper hand. In my case, the group, which has threatened legal retaliation, because it was caught out on an ethical issue, is a very powerful group in the aspect editorial. So the topic was closed because it could also have implications for our subsequent scientific production. I do not want, but I prefer to forget, in fact I have already forgotten. 

@Ierardi 

The instance you have cited is really unfortunate and the very purpose of my question was intended to create an awareness that irrelevant literature is not propagated.  I hope with the introduction of the proposal to publish all the articles on RG and making them available for public opinion will provide a transparent mechanism for a free and fair evaluation of each article, be it right or wrong. Wrong articles will then be obviously / naturally / procedurally / democratically / scientifically sidelined. 

My instance regars a a single, strange episode which may be of e teaching for some researchers. Nowadays, in almost all the journals is possible with the use of letters to the editor disputing or questioning the results of work already published. Obviously democratically and with the right of reply by the author. I do not know if the hunt for the error that you propose is applicable in all disciplines. On RG is already possible to do a firther review of an article already published. The novelty is that you propose the deletion of the article in case of serious errors in it. From an ideal point of view, it is is a sacred thing, but I am not able to see examples to consider. Your help in this regard would be usefull to all the discussants of this interesting forum.

@Ierardi

Please read the example cited by me. 

@Shanker, I red your example, but on the best of my knowledge no similar examples have been found in my field. So, I'm sorry for my inappropriate anwers and unable to suggest you something of constructive.

Dear Dr Ierardi, your own example may serve as an ideal case if you have a firm conviction about its relevance. If you post your referred article(the one removed from a high IF Journal) on RG, it might draw an appreciation from a large number of readers from your field, thereby justifying and upholding the relevance of your creation. It will also prove that the act of removing it from the Journal was an injustice. Best wishes.

Dear all,

Thanks for the constructive/critical responses. I have myself come across this relevant  article which I wish to share here:

http://publicationethics.org/files/retraction%20guidelines.pdf

[RETRACTION GUIDELINES]

See this link:

http://iai.asm.org/content/79/10/3855.full