Talk:GOV.UK Verify/GA1

GA Review

Article (edit | visual edit | history) · Article talk (edit | history) · Watch

Reviewer: Kingsif (talk · contribs) 02:12, 5 January 2020 (UTC)

starting review Kingsif (talk) 02:12, 5 January 2020 (UTC)

• This article is very list-y, with lots of short sections. This isn't good prose style.
• The prose in the article is good grammar, but uses a lot of short and/or simple sentences. This could be improved by adding sentence variety.
• Many sentences or pairs of sentences stand alone as very short paragraphs. Some should be combined and given connectors.
• The see also section has far too many tangential links
• Style criteria - needs work

• Very light on coverage
• It was accepted from AfC as start-class in 2017, and has slowly gained: 1 history paragraph (June 2018), 2 certified companies paragraphs (October 2018), additions to the list of connected services, 1 private sector paragraph (October 2018), the 1-sentence Use by foreign citizens section (September 2018), 2 Criticism paragraphs (July and October 2018), and the Future developments list (December 2018) - none of these additions have been more than a few lines, not really enough to bump its coverage up.
• Last update was in September 2019, I worry how in-date the information is.
• Are there no articles on the fact the system doesn't work making it hard for people? I imagine the criticism section could have a lot more in it.
• Coverage criteria - fails

• Only illustration is a table-formatted list. The other lists are not table-formatted, so it sticks out.
• Some illustration of the system, companies, etc - gov.uk web page, perhaps - would be useful
• Illustration criteria - needs work

• Neutrality seems generally fine
• Why is the cost of the program listed under criticism? There is no suggestion that this was criticism, so this is NPOV - move it history or the development section that should exist.
• Neutrality criteria - needs work

• The table appears to be copied word-for-word and with matching layout from this source. Now I see why it's the only list in a table.
• This part The Infrastructure and Projects Authority (IPA) conducted a review of Verify in July 2018, and found Whitehall departments were reluctant to continue funding the project. A subsequent report by the IPA recommended that the GOV.UK Verify identity assurance programme should be terminated. is also copyvio
• So is The system is intended to provide a single trusted login across all UK government digital services, verifying the user’s identity in 15 minutes.
• Close paraphrasing from other sources
• Copyright criteria - fails

• History is clean
• Stability criteria - passes

• Heavily relies on primary sources

• Which will be given more weight as these are government sources

• Other sources seem reliable
• Verifiability criteria - fails

• Overall Would need a major expansion^A and for (something like) half the sources to be replaced. It is currently assessed as start class and I see nothing that would let me disagree with that. Kingsif (talk) 02:38, 5 January 2020 (UTC

^A This is not a fail on length - short articles are just as able to be good articles as any other - but this is low on coverage, which is where it needs expansion. Kingsif (talk) 02:40, 5 January 2020 (UTC)

@Kingsif: Thank you for your review. I am surprised though, that numerous of your comments do not map to the six Wikipedia:Good article criteria? For example, the length of sentences, tangential links or the nature of edits. There's nothing in the criteria saying that an article can't be GA as soon as it is created, so the number of edits since creation don't influence GA surely? Is or isn't "the prose is clear, concise, and understandable to an appropriately broad audience; spelling and grammar are correct"?

Re illustration, "the presence of media is not, in itself, a requirement. However, if media with acceptable copyright status is appropriate and readily available, then such media should be provided." so how can it be a fail if there's no illustration?

Re it being failed on coverage, "The 'broad in its coverage' criterion is significantly weaker than the 'comprehensiveness' required of featured articles. It allows shorter articles, articles that do not cover every major fact or detail, and overviews of large topics." So are you saying it is is short on facts? How so?

WP:SEEALSO says "one purpose of 'See also' links is to enable readers to explore tangentially related topic." It is up to editorial judgment how many links there should be. Every one of those links is relevant. If the list is lengthy (which it isn't) WP:SEEALSO says it can be formatted in multiple columns...

The list, which you say is a copyvio, is clearly sourced to this page which is licenced under CC0 compatible scheme so is not a copyvio.

Curb Safe Charmer (talk) 15:31, 5 January 2020 (UTC)

• To respond in order:

• I've done a lot of GA reviews now, so I don't necessarily add the specific criteria to comments.
• Length of sentences and tangential links are covered by style. Is it easy to read? Not with so many staccato sentences, because a reader gets bored and tunes out.
• Appropriate wikilinks are in the criteria description.
• Number of edits is not in the criteria, but if you fully read that comment you'll see it said that the article has changed little since it was first made, when it was rated by an AfC reviewer as 'start class'. If it hasn't changed much (i.e. not many edits), then neither has the class.
• Illustration includes tables, etc., and since the style here is inconsistent (which I mentioned), then it fails.
• For your note on coverage, I very explicitly said this wasn't a fail on length. But comprehensiveness does not mean leave out facts, and there is barely enough coverage here to establish notability (i.e. just enough detail to allow this article to exist). It lacks in facts, then, yes - in the manner that a reader can only just get an idea of what the subject is.
• Re. see also - on many articles there are long debates about what is closely connected enough to warrant inclusion. The general assumption (and I apologize if all those links are truly very relevant) is that if it gets too long, editors are just adding more tenuous ones.
• Finally, quite importantly: being sourced to the page that it duplicates things from doesn't prevent it being copyvio. I looked at that page and, minus a few rows that duplicate text from later sources, the entire table could have been copy-pasted. You're not a new editor, that last one should be obvious. As said, I looked on that page - I didn't see any license information. I've been burned before assuming everything government-produced is CC0. (I also read that comment while editing, so skimmed, thinking it was another source link). Fair, strike that comment, but it still fails sourcing as it's mostly gov.uk sources. Kingsif (talk) 16:28, 5 January 2020 (UTC)

Kingsif (talk) 16:25, 5 January 2020 (UTC)

Forgive me for not following your assessment clearly. If you'll humour me, I've reproduced the list here and annotated accordingly. Does this represent the situation?

1. Well written ✗ Fail

• clear
• consise
• understandable to a broad audience
• spelling and grammar correct ✓ Pass
• complies with MOS for

• lead sections
• layout
• words to watch
• fiction
• list incorporation

2. Verifiable ✗ Fail

• contains list of refs?
• inline citations from RS?

please be specific as to which sources are unreliable. There are 8 government sources out of 35 (22%)

• no OR
• no CV

are you happy that this is now a pass?

3. Broad in its coverage ✗ Fail

• Addresses main aspects of the topic?
• Stays focused - no unnecessary detail?

4. Neutral ✗ Fail

The one piece of neutrality feedback is "Why is the cost of the program listed under criticism? There is no suggestion that this was criticism, so this is NPOV - move it history or the development section that should exist." I can't see why the statement about what it cost is not neutral? If you look at the source, the statement was clearly made in the context of criticism.

5. Stable ✓ Pass

6. Illustrated ✗ Fail

I cannot see how the table data should be better presented than in a table?

For clarity, sure, why not.

• Style:
  • Clear ✗ Fail
  • Concise ✓ Pass
  • Understandable to a broad audience ?
  • Spelling and grammar correct ✓ Pass
    • complies with MOS for
      • lead sections ? - accurately summarises the article, but could also define a standard level of assurance instead of using this phrase, which doesn't mean much
      • layout ? - lots of small sections that could be structured better
      • words to watch ? - elements of jargon present (particularly political jargon, which is very close to weasel words)
      • fiction ✓ Pass
      • list incorporation ✗ Fail - it's largely short lists. Prose greatly preferred where possible, which it is here
• Verifiable:
  • has refs ✓ Pass
  • inline citations ✓ Pass
  • RS ? - a lot of the valuable information is from those 8 primary sources. As I said, being government sources, these could carry more POV than other primary sources, state sources can do that. Especially with this system, which is a massive failure, but the government don't want to admit that.
  • no OR ✓ Pass
  • no CV ✗ Fail - there were other instances, and close paraphrasing throughout, as mentioned above
• Coverage
  • Addresses main aspects of the topic ✗ Fail - gives not enough detail to really inform the reader (me, here) of what this is and what it does. I know, but not because of this article.
  • Stays focused - no unnecessary detail ✓ Pass - If you're saying all those see also links are needed, then this is fine
• Neutral ✗ Fail
  • If the comment on cost was made in the context of criticism, it should be written that way. Otherwise it's a plain statement and it looks like Wikipedia has decided that it's something to be criticised. Say "X person criticised the government for spending Y amount", not "X said it cost Y".
• Stable ✓ Pass
• Illustration ✗ Fail
  • Table could be prose or a plain list, because the other abundant lists are.

This is also taking into account the spirit of the criteria - after a review if I think I've been harsh, I go back and imagine I've just come to the article and it has the GA icon, then read it and see if I'd question the status. This one, I would, so there may be things I've missed or can't put a name to that can be improved. Kingsif (talk) 18:29, 5 January 2020 (UTC)