Template talk:Academic-written review

This template was considered for deletion on 2018 April 10. The result of the discussion was "keep".

Why not just use the Academic peer reviewed template?

After a cursory inspection of the articles which use the academic peer reviewed template, that template appears to apply to Wikipedia articles where the corresponding externally reviewed article contains verbatim or near-verbatim text. The articles using this template were updated by an expert in the field, but are not textually identical to the corresponding externally peer-reviewed article. In other words, these articles were updated with content written by an academic expert while he/she was writing a peer-reviewed, subject-specific review (secondary source).

What journals have used the dual publication model?

RNA Biology
Gene

Better names for this template?

Suggestions welcome.

Articles to update with this template

There are about 90-95 gene-specific articles which can be updated with this template based on the Gene Wiki Review series (Gene). I've only updated 3-4 so far to give a chance for the community to comment on this template. I know there are Wikipedia articles that were updated via a similar initiative from RNA Biology, but I don't have a list of those articles. Gtsulab (talk) 17:35, 20 September 2017 (UTC)[reply]

Will update about 10-20 of these gene specific articles which already have assigned PMCIDs in the next few days. Followed up with the journal to find more about the articles missing PMCIDs. Gtsulab (talk) 18:03, 28 September 2017 (UTC)[reply]

Consensus

Is there any consensus to be using templates like this that pull data from wikidata, the general consensus appears to be to have all of the data in the wikitext of the article rather than in an off wiki repository. Keith D (talk) 19:44, 20 September 2017 (UTC)[reply]

Great question. Unfortunately, I don't know about the consensus on this right now, but there seems to be a trend in towards pulling from Wikidata for infoboxes. This seems to be happening as more data is added to Wikidata. Since the Wikidata community seems to have systematically added pubmed citations, I figured it would be a matter of time before references work similarly. I was hoping to preempt the need to do the conversion at a later point by doing it now. Gtsulab (talk) 17:24, 22 September 2017 (UTC)[reply]

Looking at the discussion on the Cite Q template, it doesn't look like there is consensus. Gtsulab (talk) 17:24, 26 September 2017 (UTC)[reply]

|vauthors= errors, and other problems

At Category:CS1 errors: Vancouver style there are several pages the use this template. That caused me to look into the code of the template.

This template uses {{cite journal}}, reformatted here for clarity:

{{cite journal
|vauthors={{{authors|}}}
|date={{{date|}}}
|title={{{title|{{PAGENAME}}}}}
|url={{trim|{{#if:{{{doi|}}}|http://dx.doi.org/{{{doi}}}|{{#if:{{{pmid|}}}|http://www.ncbi.nlm.nih.gov/pubmed/{{{pmid}}}|{{{link|{{{url|http://www.example.com}}}}}}}}}}}}
|journal={{{1|{{{journal|}}}}}}
|volume={{{volume|}}}
|issue={{{issue|}}}
|pages={{{pages|}}}
|doi={{{doi|}}}
|pmc={{{pmc|}}}
|pmid={{{pmid|}}}}}

|vauthors= expects the assigned value ({{{authors|}}}) to be correctly formatted else it shows an error message. Perhaps this template should drop {{{authors|}}} and use {{{vauthors|}}} with appropriate documentation so that editors can know what is expected of them.

But, that doesn't help when fetching author names from wikidata which is problematic. My experience with author name string (P2093) – admittedly limited – suggests that the return from that property is never properly formatted for |vauthors=. The template could use |authors= but there is a down-side: the author data will not be included in the citation's metadata (even though you might not notice that it's missing, those who consume cs1|2 citations via the metadata will notice that this important information is missing).

It is relatively easy to support a limited number (say, 10) |authorn= parameters for manual (non-wikidata) use.

Yeah, authors are problematic.

Why handle |url= that way? If set, |pmc= will automatically link |title= so that part is not needed. |pmid= doesn't link directly to an article so using it to link |title= is misleading. Links applied to |title= are assumed to be free-to-read. The identifiers, like |doi=, are not because sources reached through identifiers typically lie behind paywalls. Instead of misleading readers into assuming that the source in |title= is free-to-read, perhaps it would be better to use |doi-access=free to add the access icon to the rendered citation when |doi= links to a free-to-read source.

—Trappist the monk (talk) 12:08, 21 September 2017 (UTC)[reply]

Thank you so much for helping to look further into the source of the vauthors error, offering ideas and drawbacks for potential fixes. I'll hold off on attempting to fix this until more ppl have had a chance to chime in.

--After reading previous discussions on the vauthors, authors, and corporate authors issue, I've decided to use authors for now, and will note the limitations of this approach in the documentation. Thank you for all your help. Gtsulab (talk) 20:57, 26 September 2017 (UTC)[reply]

With regards to the handling the url, I was trying to be consistent with a template that was similar, but didn't cover this use case. Although the pmid link will not link directly to the article, it will usually take you to the pubmed abstract which will have the PMC link (once available), which can be used to reach the free article. This series does require that the journal deposit the paper into PMC (it's part of the Gene Wiki initiative), so the articles will be free to read, but it may take a year after the initial publication for that to happen (so the PMC ids aren't assigned immediately).

With regards to the DOI's, thank you for clarifying that distinction. I don't think the DOI's for this series will pull a non-paywalled page so I'll remove the DOI from the conditional for the url. Gtsulab (talk) 18:12, 22 September 2017 (UTC)[reply]

If this series ... [requires] that the journal deposit the paper into PMC, then all you need is |pmc= and perhaps |embargo= for those papers that are embargoed – when the embargo period expires, cs1|2 will automatically link |title= to the copy at PMC. But, it cannot if you have put anything else into |url=. Use |doi= and |pmid= so that readers may, if they choose, use those identifiers to get to the source article – even if it costs them a few bob. |url= should be reserved for free-to-read and should go directly to the article without an intermediate stop along the way.

—Trappist the monk (talk) 21:45, 22 September 2017 (UTC)[reply]

Thank you for clarifying that! I will make the recommended changes for the URL and title to the template Gtsulab (talk) 15:31, 25 September 2017 (UTC)[reply]

Issues addressed (formerly "To Fix")

Reference formatting issues. The use of Wikidata items for the template has some formatting issues that need to be addressed for the citation template to work properly. For generality of the template, these might be best handled at the use of the template and not hard-coded into the template.

This needs fixing before any further use of this template. Also date need to match rest of those used in the article, Month first, day first or ISO. Keith D (talk) 19:44, 20 September 2017 (UTC)[reply]

100% Agreed, hence the comment in the next section about just using this template on a few articles to serve as examples and waiting for comments before moving forward. Regarding the date. This template uses the cite journal template which does the formatting for all the references. In comparing the date in the example with the table on Help:CS1_errors#bad_date, the date pulled from Wikidata seems to provide the date in an acceptable format, so I'm not really sure how to fix this--any suggestions? Gtsulab (talk) 18:27, 22 September 2017 (UTC)[reply]

It turns out the date pulled from Wikidata has extra spans. The example has been fixed to address this issue. The only issue left with regards to this template is the authors, which is described in better detail in the last section Gtsulab (talk) 20:43, 25 September 2017 (UTC)[reply]

I don't know why {{#statements:P577|from=Q38916407}} has extraneous ... tags:

30 July 2016

Is that a bug?

Instead, you might want to use:

{{#property:P577|from=Q38916407}} → 30 July 2016

or this instead of nested {{replace}} templates (if you must use {{#statements}}):

{{#invoke:String|replace|source={{#statements:P577|from=Q38916407}} |pattern=%b<>|replace= |plain=false}}

30 July 2016

{{trim}} seems unnecessary because mediawiki trims leading and trailing white space from named template parameter values before they are handed off to Module:Citation/CS1

—Trappist the monk (talk) 21:50, 25 September 2017 (UTC)[reply]

@Trappist the monk: Thanks! I wonder if it would be better to transclude the Cite Q template instead of the cite journal template. On the one hand it looks like that would fix my author problem. On the other hand, I'll have to re-do this if the Cite Q template gets deleted. Gtsulab (talk) 17:16, 26 September 2017 (UTC)[reply]

Author vs Authors

Scholarly articles in Wikidata may be inconsistently expressed using property P50 (author) and/or property P2093(authors). This is the template to exclude authors described by property P50 in the template. Not sure whether to update the template or edit the Wikidata entries to be more consistent. Suggestions? Gtsulab (talk) 16:43, 19 July 2019 (UTC)[reply]

In the long-run, I think that updating wikidata items to P50 is the more valuable and versatile solution (e.g. better interaction with scholia). However it should be possible o make an interim fix on this template. I've added in the ability to use the |first1=, |last1= syntax for now. I'll look into whether it'll be possible to just pull this info straight from wikidata via Module:WikidataIB. Currently I think you'd need to enter the wikidata QID, but there might be a way to do it knowing only the doi (discussion). T.Shafee(Evo&Evo)^talk 04:04, 20 July 2019 (UTC)[reply]

You might want to read Template talk:Academic-written review#|vauthors= errors, and other problems and then undo part of your most recent change.

See Category:CS1 errors: Vancouver style for the reason why.

—Trappist the monk (talk) ~~09:34, 20 July 2019 (UTC)~~ 09:45, 20 July 2019 (UTC)[reply]

Ah, thanks Trappist - I'd not caught up on that. I've also added in the option to use {{cite_Q}} if a QID is provided to avoid needing to use repeated use of {{#property:P1476|from=Q39151066}} etc in the parameters (example)? T.Shafee(Evo&Evo)^talk 11:56, 20 July 2019 (UTC)[reply]

Thank you, Trappist the monk and T.Shafee for the feedback and fast responses. The {{cite_Q}} seems to work for pages where P50 is used, so I'll switch the the template on pages where authors are missing due to the use of property P50 until a better solution becomes available.Gtsulab (talk) 20:01, 22 July 2019 (UTC)[reply]

use Module:template wrapper

I have hacked the sandbox to use Module:template wrapper. Doing this allows this template to receive and send on to {{cite journal}} any parameter supported by {{cite journal}} without the needed to provide passthrough assignments. Doing this allows the reference rendered by this template to look and act just like a {{cite journal}} template in the accompanying article. I have added another test cases to ~/testcases.

Without objection, I shall update the live module to use the version currently in the sandbox.

—Trappist the monk (talk) 18:47, 26 November 2023 (UTC)[reply]

There having been no objection, updated.

—Trappist the monk (talk) 17:43, 3 December 2023 (UTC)[reply]