Serious problem with the addition of multiple citations from ACM
I just used the tool to add multiple articles at once to Zotero (the yellow directory icon), in order to add some results of a search for future analysis. It usually only download the title and abstracts of the articles, which suits me fine.
This time however, while browsing conferences in the ACM digital library, Zotero went crazy and decided to download massive amounts of random PDF. This caused Firefox to crash, but not before it downloaded no less than 250 articles I did not request, all named "ACM Full Text PDF" and all shoved in two unrelated citations. I restarted Firefox and retried the same operation, which worked perfectly the second time around. Unfortunately, the damage was already done.
The sudden download of 250 full articles in a few seconds did not went unnoticed by the ACM library: I AM NOW BANNED FROM USING IT. As much as I love Zotero... I cannot use it anymore. The risk of being banned from other databases is simply too great. I have deadlines looming, and I cannot afford to be blocked from more services.
Whether the problem comes from ill-formatted citations from the ACM or from a Zotero bug is irrelevant. Article databases have the big end of the stick here ; If I am banned from accessing them, I loose my job.
I do not know what the solution is here. Maybe add a confirmation when the volume of addition is over a certain threshold? Maybe perform a throttle on addition of new content? At any rate, this should be addressed quickly, before more researchers are affected.
This time however, while browsing conferences in the ACM digital library, Zotero went crazy and decided to download massive amounts of random PDF. This caused Firefox to crash, but not before it downloaded no less than 250 articles I did not request, all named "ACM Full Text PDF" and all shoved in two unrelated citations. I restarted Firefox and retried the same operation, which worked perfectly the second time around. Unfortunately, the damage was already done.
The sudden download of 250 full articles in a few seconds did not went unnoticed by the ACM library: I AM NOW BANNED FROM USING IT. As much as I love Zotero... I cannot use it anymore. The risk of being banned from other databases is simply too great. I have deadlines looming, and I cannot afford to be blocked from more services.
Whether the problem comes from ill-formatted citations from the ACM or from a Zotero bug is irrelevant. Article databases have the big end of the stick here ; If I am banned from accessing them, I loose my job.
I do not know what the solution is here. Maybe add a confirmation when the volume of addition is over a certain threshold? Maybe perform a throttle on addition of new content? At any rate, this should be addressed quickly, before more researchers are affected.
As for your access to ACM, have you contacted their technical support to explain the situation? It would make sense if this glitch triggered an automatic rate throttling or a temporary (several hours) ban, but anything more serious is unreasonable. Please advise them to contact the Zotero developers if they are concerned with the software's usage of their systems. And we'll work to make sure this doesn't happen any more.
For good measure, I just made a change to the ACM translator to automatically cut off at 20 attachments; that version will come out soon.
I cannot access the ACM website anymore, so I cannot give you the exact URL. The two offending articles were "Bringing the Systems Analysis and design Course into the 21^st Century: A Case Study in Implementing Modern Software Engineering Principles" by Christopher G Jones (2007) which had 96 attached PDFs, and "Producing more reliable software: mature software engineering process vs. state-of-the-art technology?" by James C Widmaier (2000) which had 155 attached PDFs. I must mention that the PDFs attached have no link with the citation. I have no idea why they were included. The exact number of attachments could also be higher, as I forcefully exited Firefox before it could finish. I must also mention that doing the exact same thing again did not produce aberrant results.
I contacted the ACM's technical support, but no answer so far. I do not know how long the ban will last. I know the school has already been in trouble for this kind of behavior, hence my concern over this matter. I can thankfully skip the ACM for now and focus on other databases. I am currently working manually; downloading RIS files and adding them one by one to Zotero.
Great idea on the limit to 20 attachments. It is much simpler than throttles or whatnot, and should have the same effect.
http://portal.acm.org/citation.cfm?id=337192