User:Happy-melon/WP1

As someone entirely unconnected to the WPIndia project, responding to this request for information, allow me to give my perspective on this issue. The 1.0 assessment scale is and always has been something of an oxymoron: on one hand, it depends utterly on individual editors performing individual assessments, the vast majority of which are not conducted as part of some grand tag-and-assess scheme; on the other hand, its dependants are WikiProjects (and ultimately the 1.0 project) who rely on the scheme having cohesion as a unified whole. It seems almost implausible that such a project could actually work, but it seems that it does, and any modification to that scheme must be careful to ensure that it remains workable for all users.

To that end, consider the various users of the system. The C-Class addition was proposed and implemented at a high level because the 1.0 project stands to gain the most from its implementation. In its mission to maximise both the quality and quantity of static releases of Wikipedia, it needs to be able to reliably identify 'acceptable' articles from a larger set than just FA/A/GA. The 'bar' for an article being considered 'acceptable' for a static release lies somewhere inside the old B-Class category, the version of B-Class that WPIndia still uses. This, it is important to note, is not the same B-Class that is used by WikiProject Military History, so it is disengenuous to compare the quality scales of the two projects. While there are a large number of WPIndia B-Class articles that would be 'acceptable' for a static release, and it is often easy to identify such articles upon reading them, systematically reviewing all 49,000 B-Class articles was not a feasible way of selecting content. The C-Class addition has redefined the lower end of the assessment scale to put that ephemeral borderline of 'acceptability' at the C/B boundary; loosely speaking a B-Class atrticle has the potential to make a static release while a C-Class article does not. Rather than simply shifting the Start/B boundary higher (which was considered and rejected as unlikely to be adhered to), the new class merely subdivides the increased territory between Start and B. The situation with MILHIST is more that they have kept this large area as one "Start-Class"; the criteria for B-Class in the MILHIST system are essentially identical to, and indeed were the basis for, the new 1.0 B-Class criteria. So while previous selections of static release articles have paid only minimal heed to quality assessments, this is mainly because the quality rating was not previously a useful indicator of actual quality! Future assessments, I suspect, will give more weighting to the quality of an article; of course if WPIndia continues to not use C-Class, that fact will not be simply ignored by the 1.0 project; but it will cause unwanted extra work to develop an alternative method of adjudicating the quality of potential release candidates.

From the other end of the scale, the effect of adding a C-Class rating is minimal. Editors reviewing individual articles will have a new option to resolve the situations where an article is not inequivocally either a Start-Class nor a B-Class. The new B-Class criteria form an easy checklist for evaluating whether an article meets that ephemeral 'acceptable' status; it should be both a trivial and an objective exercise for an editor to review an article against these criteria. I think the situation is described beautifully in this quote from Roger Davies. A typical article develops from a 'formative' stage to a level at which it begins to attract interest as some of our better content. It proceeds in a 'ping pong' fashion around successively more thorough review processes until it is validated as some of our best content. It is entirely appropriate that the first step on that ladder is an analysis by one editor who says "yes, this article is worthy of consideration".

In the middle, WikiProjects such as WPIndia are essentially the guiding hands that lead articles (and editors) along that road towards quality content and quality content production. That road has always been divided into two stages, the 'formation' stage where the foundations of the article are laid, and the 'review' stage where the article is brought up to the explicit standards of the various review processes. Most projects contain members interested mainly in one or other area. For those interested mainly in the development and perfection of content, it is invaluable to have a clear 'starting point'(a well-defined B-Class) as a pool of articles from which to select front-runners for the review process. From the other side, the modest yet objective requirements of the new B-Class represent a worthy goal for the development of content from scratch. The whole road from Stub-Class to FA-Class is long and arduous enough that few editors are capable of taking an article all the way along it, and it is the full-time occupation of WikiProjects to provide relay teams to ease the journey. Having clearly defined waypoints along the way provide tangible evidence of the improvement of articles, good for both morale and organisation. Editors who know where their talents lie can more easily find articles where their expertise are needed, and more articles can benefit as a result. In summary, the redefinition of the assessment scale (which is a better way of thinking about it than just "adding C-Class") makes explicit a distinction which has always existed, between articles still in embryonic form and articles worthy of consideration as good content, and thus makes it easier to administrate and facilitate the progression of articles from one stage to the other, and to increase our throughput of quality articles across the board.

In response to your request for information at WP:1.0/A, I wrote a fairly full answer, although it didn't say everything I wanted it to say. Seeing on preview how outrageously long it was, I realised that no one would want to read it, so I made an essay instead :D. So the answers to the questions below are given largely without proof; for detailed argument, you'll have to wade into the full epistle.

Has the implementation of C-class has been effective in general?

It's very difficult to answer this question without an explicit definition of "effective". We now have almost 13,000 C-Class articles, so users have definitely taken to the new class with great enthusiasm. There are certain to be many thousands more that are still being miscategorised due to technical issues (poor formatting of templates, essentially), and so do not appear in the statistics. We will not see the full value of C-Class until the majority of articles have been recategorised appropriately, a process that will inevitably take many months if not years. I estimate that at the present time we actually have about 70,000 C-Class articles and 20,000 B-Class articles, while we record 50,000 B-Class and 13,000 C-Class. I am actually fairly impressed by the rate of takeup of C-Class.

Does C-class articles involve more red tape?

On an individual level, absolutely not. The new assessment scale gives editors the opportunity to identify a relatively narrow group of articles with a higher precision than before. The new B-Class assessment is intended to be a quick, easy and objective way of evaluating articles against a set of clear criteria. Naturally the new criteria take some getting used to, but the process for individual editors shouldn't change, and the new scale should make it easier for editors to assess articles, not harder. On a project-wide scale, it's less obvious. On the surface, the project is responsible for ensuring that its assessments on talk pages are up to date and accurate; it is easy to conclude that adopting the new scale should precipitate a full reassessment of all project articles. In fact, this is entirely unnecessary, because all articles are and should be being reassessed on a continuous basis. Every time an editor, whether from WPIndia or not, encounters an article where the assessment does not reflect the actual state of play, there is a small chance that they will update the assessment accordingly. Over time, reassessment becomes inevitable. Adopting the new 1.0 assessment scheme should merely be a continuation of a project's natural reassessment initiatives and shouldn't be made into a huge operation. Once established, and remember that the new scheme is still establishing itself over en.wiki as a whole, maintaining it is a division-of-labour operation that is greatly facilitated by having such a huge base of users and projects participating in the scheme.

Are the lines of distinction between Start and C, and C and B classes blurred?

The distinction between Start-Class and B-Class in the old scale was extremely fuzzy, with a huge grey area between the two. The new scale tightly defines the C/B boundary, allowing an objective decision to be made about where an article lies in that region of the scale. The Start/C boundary is not so well defined, although the 1.0 project have tried to make it as intuitive as possible.

Does C-class complicate the assessment process?

As I said above, every time an editor, whether from WPIndia or not, encounters an article where the assessment does not reflect the actual state of play, there is a small chance that they will update the assessment accordingly. Issues arise, however, when the reassessing editor is unaware that a particular project such as WPIndia is using a different assessment scale to the majority of WikiProjects. If an editor concludes that an article is C-Class on the 1.0 scale, they are very likely to try and mark it as C-Class on the WPIndia banner as well, creating confusion and wasting time while the reason for the discrepancy is explained. For editors who are not explicitly aware that a particular project does not use the 'standard' scale,