Closed wg030 closed 2 years ago
Thanks! I've also added your example to the main bib file.
Thanks! I've also added your example to the main bib file.
Great! And if we make sure that the current existing examples are consistent with this pull request, then there will be no risk of people getting confused. Let me quickly update that and create a new pull request for that!
This is quite huge enhancement PR, so thank you for that!
I think we can go with the switch author
→ holder
, and editor
(meaning inventor) → author
, even though the holder
field is only an optional field in pure biblatex. Then we should also deal with a case the entry is without holder and the inventor will be a primary author, shouldn't we?
I think there is a clash with origdate
and date
and their usage. The standard says about "Date of publication.." at the first place, so I think the date
field would be more appropriate for citing and printing as the primary date. I don't know much about patents, but generally only after something is published, it is ready to use. Moreover, in the bib reference it looks quite counterintuitive to have application date repetitively (of course one is more precise), but the publication date is there only once and without any explanation what this date is all about.
I don't really like removing the dot at the very end of each entry in this PR (it's rather hidden here and not relevant to this context at all). We can discuss it in a separate issue #101
Added the localization key application
also to the Translation guidelines @wiki.
Yes, holder
is only optional in the standard.bbx
biblatex style. However I do think that the biblatex manual encourages
developers of new styles to make use of that field for @patents as it explicitlly says
holder list (name) The holder(s) of a @patent, if different from the author.
I also think that from the ISO 690 point of view, it is pretty clear that the holder of the patent (patentee) is more supposed to be the primary creator than the author of the patent (inventor) because in the hierarchy given in section 5.1 in the ISO norm, patentte belongs to group a) and inventor belongs to group b).
Besides, if we happen to have a patent where the patentee is the same as the inventor and a user of this style only uses the author
field, then the resulting entry would be an entry where the name of the author would appears twice: right at the beginning and then again behind the title with the addition of "inventor:".
This fits quite well to the ISO 690 norm in my opinion, which says in section 15.9
The name of the proprietor or applicant creator (see 5.1) should precede the title of the patent resource (contribution). [...] Other information such as the name of the inventor, any classification symbols and the date of application preceded by “Appl:” may be given as additional information between the title and the official designation.
Concerning the dates, it is a little tricky and I wanna point out that I am not a patent expert either.
So at first I thought about implementing it the usual way trying to avoid what you just described. But then I thought about it again and I came to the conclusion that the application date might be rather the primary date. Let me explain that form the example of the biblatex-iso690-examples.bib (see https://patentimages.storage.googleapis.com/e5/ba/57/7c6782a5c9a908/EP1094937B1.pdf)
In date
we put 2008-07-30
and origdate
we or rather I put 1999-05-04
. Now let us check how these dates are described in the patent itself.
2008-07-30: "Date of publication and mention of the grant of the patent"
1999-05-04: "Date of filing"
Moreover the patent shows yet another third date as
2001-05-02: "Date of publication of application"
Well, I tried to do some quick research on how an application of a patent works and as far as I undersood the ""Date of publication and mention of the grant of the patent" is not the first time one can theroteically have access to the patent. There seem to be databases where you can see patents which are in the prosess of application, but have not been granted yet (as you can see in our example the application took 9 years to get granted). Consequently the "date of publication of application" seems to be the best choice for the primary date because this is the date where anyone could theretically have access to the paten application (at least I believe so). But since we don't mention that date in our reference because the ISO does not suggest us to do so, the second best choice for the primary date seems to be the date of filling, i.e. the of application, rather than the date where the patent is offcially mentioned to be granted in my opinion.
You are absolutely right about the authorship of a patent. My point is this line \DeclareNameAlias[patent]{author}{given-family}
, which breaks the consistency rule if there is no holder for the patent. At the moment, I don't know how to differentiate the name format to have family-given
when author
(inventor) is at the beginning of a reference, and given-family
when there is also holder
in a bib file.
Fair enough about the dates, I would question why don't we use a prefix for date
as well, e.g. Publication: YYYY or Publ.: YYYY, so it will get more clear, I think. And it would also support the unambiguity rule (clear identification of a resource).
BTW, related to the dates:
A patent number or sometimes referred to as a publication number is a number identifier that is assigned to a patent application when it is published by the issuing patent office. [1]
which sounds like we don't have a patent number until it is published :open_mouth:
EDIT: we have also patentreq
for patent request, available as a localization key in biblatex, so instead of patent number, user can fill in some application number or so.
I think we can go with the switch
author
→holder
, andeditor
(meaning inventor) →author
, even though theholder
field is only an optional field in pure biblatex. Then we should also deal with a case the entry is without holder and the inventor will be a primary author, shouldn't we?
Damn, I haven't seen that so far. I will try to consider a solution for that, but right now I do not have an idea either.
Fair enough about the dates, I would question why don't we use a prefix for
date
as well, e.g. Publication: YYYY or Publ.: YYYY, so it will get more clear, I think. And it would also support the unambiguity rule (clear identification of a resource).
Very good idea. I strongly agree here.
we have also
patentreq
for patent request, available as a localization key in biblatex, so instead of patent number, user can fill in some application number or so.
Yes, that seems to be the way how to handle the difference between an application that has not become a granted patent and an already granted patent
Added the localization key publication
also to the Translation guidelines @wiki.
OK, I think we can consider this to be done and dusted for now. Once again thank you @wg030 for the hard work :partying_face:
Hi! I improved the @patent entry type according to the description in the "DIN ISO 690:2013-10" (the German version of the norm) by having a closer look at the few examples. Here is an example taken from the iso document itself to illustrate how we can create patent entries from now on.
Note that the
holder
field becomes the primary author here, whereas theauthor
field becomes a subsidiary author representing the inventor(s) of the patent. Moreover theorigdate
field is used to show the date of the application and thedate
field is the publication date of the patent. Though I had to go quite deep into all the biblatex stuff, I think I managed this without messing things up. The only thing that is still missing here are the translations for the new bibliography string "application".Moreover I removed the dot at the very end of each entry because I figured out that this is not done in the examples of the iso document either.