Timeline for How are scientific papers uniquely identified?
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 26, 2018 at 17:33 | vote | accept | Agelos | ||
| Oct 25, 2018 at 0:15 | comment | added | user71659 | @Trilarion The problem is, by definition, hashes completely change in cases where the actual content has not. When PDF becomes obsolete and papers get converted into whatever new format, you'll get a change... Some publishers put a cover page on papers with an ad, or they watermark the paper with the downloading IP, also causing changes. DOIs don't have this issue by not assuming binary representations = content. | |
| Oct 24, 2018 at 8:29 | comment | added | NoDataDumpNoContribution | @BryanKrause "...the unique IDs aren't for "us".." If they aren't for us then the obvious solution are just hashes of the electronic files. Makes everything universally (almost always) uniquely identifiable. It probably depends on what the current digital underlings and future overlords intent to do with these scientific publications. | |
| Oct 23, 2018 at 15:55 | comment | added | Bryan Krause♦ | I think the key piece of this answer is the parenthetical at the end: the unique IDs aren't for "us", they are for our digital underlings and future overlords to prevent them from making mistakes. All of the other citation stuff is for "us" and is sufficiently unique going back a long time. | |
| Oct 23, 2018 at 14:52 | comment | added | henning no longer feeds AI | @Anyon, Konrad Rudolph, you're both correct, but the need for unique IDs presumably is a result of digitization in the sense of computerization, since computers don't deal well with the ambiguity of traditional citations. | |
| Oct 23, 2018 at 14:11 | comment | added | Konrad Rudolph | @Anyon This overloading was inadvertent: You’re right that analog ISBNs were already sequence of digits (i.e. “digital”) but I was actually referring to the fact that they are nowadays digitised. | |
| Oct 23, 2018 at 14:04 | comment | added | Anyon | @KonradRudolph First, I think there's an overloading of the word "digital" here. ISBNs were introduced as an ISO standard in 1970, and while they (and other serial numbers) consist of a series of digits (one meaning of 'digital'), this was certainly before the "digital age". Second, I agree with you. I also don't see any particular reason why a DOI-like number for research papers couldn't have been introduced way before 2000, say in the early 1900s. There probably wasn't much need for it, but it could, in principle, have been done. | |
| Oct 23, 2018 at 13:38 | comment | added | Konrad Rudolph | I’m not sure I agree with the reasoning: books predate the digital age, yet even those books (up to a point) all have digital identifiers (ISBN for most of them, and other — albeit not uniform — identifiers for older works). Likewise, older publications can be (and are) digitally archived and identified in retrospect. | |
| Oct 23, 2018 at 9:33 | history | edited | henning no longer feeds AI | CC BY-SA 4.0 |
added 137 characters in body
|
| Oct 23, 2018 at 8:01 | history | edited | henning no longer feeds AI | CC BY-SA 4.0 |
deleted 2 characters in body
|
| Oct 23, 2018 at 7:40 | history | edited | henning no longer feeds AI | CC BY-SA 4.0 |
added 57 characters in body
|
| Oct 23, 2018 at 7:29 | history | answered | henning no longer feeds AI | CC BY-SA 4.0 |