Wikipedia talk:Stub Contest/2014 archive

Some thoughts for next time

There has been some discussion on the usefulness of re-rating stubs that are no longer stubs versus expanding those that are true stubs. My personal opinion is that there is more value to Wikipedia users if they are offered more comprehensive articles, and to them, it probably is of little significance that there is an article class system in the first instance. I guess that in itself shouldn't be a controversial statement, as you earned 10 points for expanding a stub, and 1 point for re-rating one. I would like to suggest, though, that this ratio isn't high enough, and contributors found it easier to re-rate 10 articles to expanding 1 stub. Hence, most of the points scored during the contest was for re-rating rather than expansions. There is, of course, value in having articles rated correctly, but what could be considered is to sum up all the re-rate points, some up all the expansion points, and the ratio of those two numbers could be taken as a guide for how much higher the point value of expanding stubs should have been.

Say, for example, that re-rating has a point total of 8,000, and expansion has a total of 2,000 (i.e. 200 articles). That could be taken to mean that for the next contest, the value for expanding a stub could be 40 points (i.e. 8,000 divided by 200). With such an approach, we may find that people put less effort into re-rating, and more effort into expanding articles. If my above hypothesis is right, that will be of more value to users of Wikipedia.

I'm sure that others will have some thoughts on this. Schwede66 00:45, 5 January 2014 (UTC)

Support. At first I actually thought that the scoring was done perfectly, as the goal of the contest was to reduce the number of stubs on Wikipedia. However, after seeing your point about what Wikipedia readers care about, you changed my opinion. Like you said, the truth is that most Wikipedia readers don't care whether an article is a stub, a start-class, or a featured article... they care about getting quality, correct content. I agree with your method for deciding the value of points. Newyorkadam (talk) 00:53, 5 January 2014 (UTC)Newyorkadam

Comment - my provisional thinking of the correct ration was more like 100:1 - i.e. base 100 points for expanding a stub and 1 for re-rating, based on what I have seen in the scoring. I'd be keen to hear from contestants who did both as to what they thought a fair ratio was. Cas Liber (talk · contribs) 01:01, 5 January 2014 (UTC)

<Edit conflict> Note that the above example was based on numbers that I've made up. I'm not saying that 40:1 is the right ratio, but I'm giving an example of how to work out that ratio, and we could wait till all the judging has been done, and continue the conversation from there based on real numbers. Schwede66 01:14, 5 January 2014 (UTC)

Yeah, I hadn't planned on opening this up till the dust had settled but someone started it now. Cas Liber (talk · contribs) 01:18, 5 January 2014 (UTC)

As much a fan I am of uprating and fixing all the back-end stuff, I'd go a step further on a second contest and have it just be the stub expansions, work on benefiting the reader side of things. Wizardman 01:04, 5 January 2014 (UTC)

Which personally as a judge I would agree with. Mitch32^{(The man who renounces himself, comes to himself.)} 17:49, 5 January 2014 (UTC)

I could definitely see the usefulness of this as well. To go further, they could be separate contests entirely. One focused on cleaning up the assessments (basically everything that doesn't require a formal review, this could also include current stubs that should be deleted / redirected / merged), and one focused solely on expanding existing, correctly labeled stubs. Canada Hky (talk) 17:52, 5 January 2014 (UTC)

Comment: I don't think the bonus were high enough. If you use the 1,500 DYK cut off for prose and start with something that was around 500-1,000 and has some wp:proseline and/or a small or one line lead. You could clean up the wording, add a proper lead, and some minor additions to hit the 1,500 mark. Take the same size page and expand it to a 4,500 article requires more time and research and only gave you an additional 5 points. It would be much easier to to clean-up and slightly expand 3 pages of 1,500 (which is the same readable pros ans the minimum bonus) than to do a full on expansion and you receive double the score. While the encyclopedia undoubtedly needed the clean-ups and expansion the contest didn't encourage greater expansion of stubs. Which is probably why there were only 42 (20 coming from 2 editors) of the 667 (6%) were given bonuses. If the scoring is adjusted as mentioned above the bonus is something that should be looked into to encourage larger expansions IMO. --Mo Rock...Monstrous (leech44) 01:50, 5 January 2014 (UTC)

Ten of those were mine and, selfishly, I agree! I had the idea going in of expanding a bunch of articles to about C class for this contest, but once I saw the re-rates going wild (never expected that at all), I just went back to my normal goal of trying to expand as complete as possible. And of those ten, one is already a GA, and four more are in the queue. I'm quite happy with that. As it is, looks like I am going to finish 10th, and the contest certainly pushed me into a strong editing phase. I'm quite happy with how this turned out overall. Resolute 18:53, 5 January 2014 (UTC)

What about a competitor writing a huge article and getting a huge bones on an article with nearly no page views. Snowman (talk) 11:41, 25 March 2014 (UTC)

• You could run seperate contests or parts of the same contest, with bigger prizes for expanding than rerating. Then you don't have to worry about the ratios of mixing points, just a proportion of the prizes. Easier somehow, though people are likely only to attempt one section. Johnbod (talk) 20:21, 5 January 2014 (UTC)

What I think we'll do once we've finished scoring etc. is start with a yes/no poll on whether we include rerating. I guess I was originally thinking of it as an editor, I find the wikiproject assessment graphs really useful, and also having some idea of how many articles are still rated as stubs. This has been a fascinating exercise for this. The counter-view is that, duh, we know there are loads so let's just keep on expanding them, as rating scales mean little to readers. Cas Liber (talk · contribs) 21:58, 5 January 2014 (UTC)

Running again?

Is the contest going to be run again? If it is, I don't see any reason to wait an entire year... why not within the next couple of months? -Newyorkadam (talk) 15:40, 28 January 2014 (UTC)Newyorkadam

I am thinking May/June-ish. Would like to get a few things straight for next time and was going to archive this page once all prizes had been received. Can everyone else give a nod that they have received them? Cas Liber (talk · contribs) 19:49, 28 January 2014 (UTC)

I have received and activated my prize. (It was in my spam folder for some reason.) Thanks, ~HueSatLum 20:59, 28 January 2014 (UTC)

A Stub-Contest at Wikimania 2014 in London

Please consider a Stub-Contest at "Wikimania 2014 London". in August.--DThomsen8 (talk) 03:24, 9 February 2014 (UTC)

• Can people participate on-line without being in London? Please include a link to the details of the competition. Snowman (talk) 15:18, 11 February 2014 (UTC)

Fine-tuning the rules before we run another Stub Contest

Right - I was looking back over this and feeling quite happy about how it all turned out, as I think it has alot broader appeal than the Core Contest. I am bvreking this up into sections to consider and would be grateful if folks comment underneath. Cas Liber (talk · contribs) 04:18, 25 March 2014 (UTC)

Rerating stubs

From discussions above, and from thinking about it (i.e. stub re-rating is of negligible interest/benfit to readers), I am strongly considering removing the re-rating of stubs from scoring in this contest...unless a bunch of people really really want to keep this aspect in, in which case we could be need to adjust the scoring ratio (probably to 100:1 or something). So speak now or forever hold yer peace! Cas Liber (talk · contribs) 04:18, 25 March 2014 (UTC)

I am all for scrapping the re-rating aspect. Sven Manguard Wha? 06:32, 25 March 2014 (UTC)

• Doing semi-automated edits efficiently and accurately with contentment as a hobby may be restricted to those interested in computer science. I may have raised the profile of semi-automatic editing; however, I am not interested in promoting semi-automated editing, because I would think that those interested would have taken it up early in their Wiki editing hobby and gradually learned how to use the relevant tools. May I say that I needed a number of scripts and semi-automatic tools to do that amount of re-rating accurately and I think that it is very likely that I used these sort of tools better than any other competitor. If re-rates were awarded only 1/100 of the points of a expanded Stub, then this would effectively scrap the re-rating part of the contest. Surely, there is the value having accurately rated articles in the Wikipedia. Does the contest want to attract competitors who can do re-rates at a high rate with the help of script assisted semi-automatic edits? I have no particular problems about not entering the next Stub competition, if I thought that I would not win under a new scoring system. Snowman (talk) 13:06, 25 March 2014 (UTC)

• Stub tags on non-stub article pages do affect the reader by (a) confusing the reader about what is a stub, (b) visual clutter and (c) few milliseconds of extra download time. All minor, but cleanup of stub tags from article space should still be rewarded. --ELEKHH^T 00:14, 26 March 2014 (UTC)

@Elekhh: - do you have a rough idea on ratio to score them? i.e. how much a rerate should be compared with an expansion? Cas Liber (talk · contribs) 00:29, 26 March 2014 (UTC)

There are not that many of these, so would go with the 1:10 ratio as last time. --ELEKHH^T 01:08, 26 March 2014 (UTC)

I have been musing on running a separate 24 or 48 hour re-rating blitz contest instead for these - as I don't think I could face checking 30,000 of them maybe. Cas Liber (talk · contribs) 00:54, 26 March 2014 (UTC)

What User ELEKHH seems to be suggesting is only removing the Stub tag from an article that is higher than a Stub. The previous Stub contest involved re-rating WP Banners on talk pages as well as removing the Stub tag (if present) from the article. I would think that there are 100,000s of non-Stubs tagged as Stubs. Snowman (talk) 10:50, 26 March 2014 (UTC)

Snowman, can you think of an easier way to mark them? Also, would you be happy to have two separate competitions or do you prefer them to be part of the one contest? Cas Liber (talk · contribs) 11:22, 26 March 2014 (UTC)

I use scanning scripts fairly often, but the short answer is that I cannot write a scanning script to accurately scrutinize re-rated scripts. My simple scripts and AWB can be useful to speed up the process of inspecting re-rated Stubs, but it would also involve a lot of repetitive manual reading and checking as well. Even if I wrote a script to fully automatically count references, record the total size of the article, note the presence of Stub templates in the article, and examine the WP banner on the talk page, then at best it is only a screening tool. Have you ever used AWB? If so then I might have a better idea on where to pitch a reply. Snowman (talk) 14:54, 27 March 2014 (UTC)

I personally feel that rerating is incredibly valuable - it's something that needs human activity and can't be easily automated, that we historically haven't been able to do on a rolling basis, and without it we simply have no idea what our quality is like. However, I do agree that scoring it alongside expansion is problematic, and indeed checking/scoring on the scale that would be needed is very complex. A separate blitz approach (with, eg, automatic worklists and so on) might be the way to go. Andrew Gray (talk) 12:58, 27 March 2014 (UTC)

Defining a stub

One of the aspects I liked when designing the contest is that it might make attractive the reffing and fixing of larger stubs that were too large for an easy 5x expansion at DYK, yet might be easier to fix than starting from a one-sentence stub (which is easier to 5x expand at DYK). However, there were times when there was some difference of opinion over what constitutes the barrier between a stub and a start-class article. If one looks here, we have The article is either a very short article or a rough collection of information that will need much work to become a meaningful article. It is usually very short; but, if the material is irrelevant or incomprehensible, an article of any length falls into this category. yet we have start class as An article that is developing, but which is quite incomplete and, most notably, lacks adequate reliable sources (more detailed) The article has a usable amount of good content but is weak in many areas, usually in referencing. Quality of the prose may be distinctly unencyclopedic, and MoS compliance non-existent; but the article should satisfy fundamental content policies, such as BLP, and provide enough sources to establish verifiability. No Start-Class article should be in any danger of being speedily deleted.

Thus - I think it'd be good to clarify and determine a consensus (for the sake of the competition), how many sentences an article can have and still be a stub, and is that different whether it has zero, one, or two or more inline references. All input welcome. Cas Liber (talk · contribs) 04:18, 25 March 2014 (UTC)

This might be controversial, but to me, the difference between a Stub and a Start is negligible. Ugly Americans: Apocalypsegeddon is a stub. It is measured at 1467 B (237 words) "readable prose size". It covers the basic aspects, but it's not very complete. Death By Cube is a start class article. It is measured at 2078 B (350 words) "readable prose size". It covers the basis aspects, but is not very complete. I don't really see much of a difference between one and the other, but because of a difference of just over 100 bytes (thirteen words), one is a stub and one is a start class article. While I am thoroughly disappointed at the mediocrity on show in each of them (they're my work, I can insult them), they are at least coherent and well sourced.

To me a stub is any article with less than 2000 characters of readable prose. A dire stub is any article with less than 1000 characters of readable prose, or any regular stub that has only one source (or zero, although I would delete sourceless articles on principle). An embarrassingly dire stub is any article with less than 500 characters of readable prose. They do exist, and they are embarrassing. I would create a sliding scale of points based on how bad an article was to start and how much it was expanded. For example, using numbers I crafted from looking mostly at my own work, I came up the table on the right. The vertical axis is start size (in tiers like "less than 500 characters of readable prose), and the horizontal axis is end size (in tiers like "4000 characters of readable prose"). To put my tiers into context, here is what what 2400 characters looks like, what 4250 characters looks like, and what 6850 characters looks like.

- - -	2500	4000	6500
< 500	15	25	35
< 1000	5	15	25
< 2000	-	5	15

What it means is that if you start with a < 500 character article and bring it up to 4000 characters, it's worth 25 points. If you bring a < 2000 up to that size though, it's worth only 5 points. This is meant to reward people who make large expansions to articles in the worst shape, and also reflects the differing amounts of effort needed to bring articles of varying start sizes up to the 6500 character range.

I considered adding bonus points for DYK, but decided that it would be too complicated to do fairly, as 5x expansion is easier to meet with smaller articles, and my system is already giving more points to improvements to those articles. I also considered giving a blanket +25 point (or so) bonus for getting a stub to GA level, but I don't know about the timeline for the contest, and how GAN would fit in to that timeline.

Just some early thoughts. Sven Manguard Wha? 06:30, 25 March 2014 (UTC)

@Sven Manguard: See I'd be thinking Ugly Americans: Apocalypsegeddon was a start article as it had references and covered the subject enough so you had a fair idea about it. 5 Spots Party (909 B (173 words) "readable prose size") is more in stub territory but even this has two references although coverage is really patchy. Now, question is does one distinguish between fully inlined and unreferenced articles and how - maybe 1000 bytes of prose if >50% referenced or 1500 bytes if <50% referenced...? I am not keen on involving other review systems as they have their own timelines and often considerable delay that might outlast the run of this competition (unless we run over more than four weeks.....actually that is worth bringing up too....) Cas Liber (talk · contribs) 09:29, 25 March 2014 (UTC)

During the running of the stub contest I spent virtually my whole Wikipedia time on improving stubs. I wouldn't have wanted to spare time to nominate articles for DYK or GA and deal with issues in these venues. Nor were the resulting articles rounded wholes that I would have been happy to have written from scratch. They had deficiencies and I did not check the original references for inaccuracies or taxoboxes for inconsistencies (for which I was attacked by my chief critic!). I would suggest that for a stub to be expanded and scored in the contest, it needs to end up with at least 1500 B, be at least double the original size and be reasonably well referenced. Cwmhiraeth (talk) 10:53, 25 March 2014 (UTC)

So what do you think should be the maximum starting size? The idea I had was that we'd clean up some of the messier larger stub articles too....Cas Liber (talk · contribs) 10:57, 25 March 2014 (UTC)

How about anything over 1500 B needing approval from a judge as to its original stub status? The competitor could put an asterisk beside the entry to denote this or could list it somewhere for prior approval. Cwmhiraeth (talk) 11:29, 25 March 2014 (UTC)

May I suggest simple methods that will involve the judges in a minimum of work. Snowman (talk) 12:20, 25 March 2014 (UTC)

Actually, I am not a fan of editing contests on the Wiki, because of what seems to be to be an apparent conflict between the wish of competitors to expand Stubs quickly to accumulate points and time needed to enhance the authority of the resulting text. How is the next competition going to deal with this apparent conflict between quality and quantity? In the previous Stub contest, I made some small helpful improvements (including script assisted semi-automatic edits to taxoboxes) to Stubs that were expanded by competitors who were doing Stubs at a high rate. Snowman (talk) 12:20, 25 March 2014 (UTC)

Bonus points for full inline referencing? And spot checking references for paraphrasing or OR? Cas Liber (talk · contribs) 13:13, 25 March 2014 (UTC)

"Bonus points for full inline referencing?" - could be. "Spot checking references for paraphrasing or OR?" - you gotta be joking! Expanding stubs is about adding content and finding references, not snooping. I prefer to assume good faith. Cwmhiraeth (talk) 05:59, 26 March 2014 (UTC)

Sorry, I meant the judges would be spot checking references for paraphrasing or OR - not contestants. That was in response to Snowman's comment about quality. Cas Liber (talk · contribs) 09:02, 26 March 2014 (UTC)

Sorry, I misunderstood what you were suggesting. Cwmhiraeth (talk) 11:12, 26 March 2014 (UTC)

Learning from the previous Stub contest?

• It seems to me that in the last contest the judges were on a learning curve, so I think that the next contest could be designed to minimize anticipated problems and minimize repetitive tedious judging work. If a bonus system is going to be used next time, then I would suggest that it is set up to work with automated scoring. A useful bonus system could be based on the number of hits an article attained on the day prior to the start of the competition, and a script could do the calculations. The bonus need not be in direct proportion with the number of hits; perhaps the square root, cube root, or based a per-determined bonus for certain ranges of hits. I expect that this would need liaison with a person who can write computer programs (also called scripts). Snowman (talk) 11:26, 25 March 2014 (UTC)

I would be keen to see this. Cas Liber (talk · contribs) 13:11, 25 March 2014 (UTC)

I was somewhat disappointed to see that a sizable proportion of the Stubs expanded in the previous Stub contest were not very popular with readers. Perhaps, the next Stub contest could reward expanding "popular" Stubs and not Stubs never likely to have more than a tiny number of views per month. I would think that bonuses based on hits would be a useful driver. The idea of automating the calculation of bonus scores derived from hit counts, is not entirely theoretical, because I have already written a script that finds relevant page hits and used it on Stubs expanded in the previous Stub contest. I am not promising anything. Snowman (talk) 14:14, 25 March 2014 (UTC)

• I think that the judges should be strict to apply the rules including the submitting format and not hesitate to discount particular submissions that are not compliant with the rules. Of course, judges (or anyone else) can prompt competitors to amend problems during the competition, but the listing at the end of the competition should fully compliant with the rules and if there are submissions that are not compliant then each non-compliant submission is discounted. Snowman (talk) 11:35, 25 March 2014 (UTC)
• I think that an extra rule is needed which explicitly says that the submission for each article must be submitted within 1 to 2 days of completion of work on each individual article. This would not prevent submitting a block of submissions providing that all the stubs in the block were completed in the previous two days; however, this rule should be worded to prevent accumulating submissions over a period of time and then submitting in a block near the end of the competition. Also, saving up minor changes to a number of nearly finished articles and then submitting these in a block should also be excluded. The spirit of the time of making submissions would be for clarity to other competitors. Snowman (talk) 11:26, 25 March 2014 (UTC)

I think we can urge people to do so, but I am not sure how obligatory we can make it. Some people might be tied up in Real Life issues for a few days. The wikicup has no obligation like this and works ok. Cas Liber (talk · contribs) 23:20, 25 March 2014 (UTC)

Urging might not be effective. I expect that the judges would use some discretion based on an explanation provided by competitors for making late submissions. WikiCup has a similar dysfunction, and I understand that accumulating unrecorded points and making late submissions has been used as a tactic in the WikiCup to overtake other competitors just before the finishing line. Is this a fair tactic? If people were tied up in real life for a while, then they would not be massing unrecorded points during that time. Snowman (talk) 08:57, 26 March 2014 (UTC)

You refer to a WikiCup dysfunction above but your "understanding" is not accurate. The only comments made by the judges over at the WikiCup when the matter was discussed was that judging was made easier when points were claimed in a timely manner. If you want a time limit here, I suggest 5 days or a week. Cwmhiraeth (talk) 09:37, 26 March 2014 (UTC)

As far as I am aware, my understanding about the possible disruptive effect of late submissions in the WikiCup has a solid foundation. My understanding is based on User:Sasata's, a WikiCup competitor, comment on this talk page in January 2014 (now archived). Snowman (talk) 09:55, 26 March 2014 (UTC)

The matter was discussed here. Cwmhiraeth (talk) 11:07, 26 March 2014 (UTC)

I note various competitors apparent interest in this topic on the WikiCup talk page. I think that it would be beneficial to discuss the issues again here following the events of the last few days of the previous Stub Contest. Snowman (talk) 11:22, 26 March 2014 (UTC)

• A bonus for the oldest stub should not be awarded, because the articles made at the start of the Wiki were in CamelFormat and modified to the new format making date of creation unreliable. Snowman (talk) 11:26, 25 March 2014 (UTC)

Hadn't thought of this, but happy to ditch it. Cas Liber (talk · contribs) 13:11, 25 March 2014 (UTC)

As far as I understand the early Wiki, I think that a number of data transfers were made which resulted in the loss of a batch of very early editing history and so some of the very old apparent creation dates are wrong. Having said that I think that very few of the converted CamelCase articles are still Stubs. With a scanning script, I came across one that was tagged as a Stub, it was largely unreferenced, but it was slightly longer than a Stub. Finding old Stubs is an interesting task, but I wonder if competitors' time used in finding old Stubs, could be directed into something more beneficial to the Wiki. I recall that finding the oldest Stub was a problem for the judges to do manually. If the interest value in finding old Stubs is overwhelming, then I might be able to run a script weekly during the competition over the submitted expanded Stubs to find candidates for the oldest Stub. I am not promising anything. Snowman (talk) 09:37, 26 March 2014 (UTC)

• A bonus for the most humorous stub should not be awarded, because it is too subjective. Snowman (talk) 11:26, 25 March 2014 (UTC)

Agreed. Cas Liber (talk · contribs) 13:11, 25 March 2014 (UTC)

Stub Challenge

I enjoyed the previous competition, and I am saddened that some of the other competitors seem to have been stressed. I am also saddened to hear that an editor appears to be dissatisfied with rushed editing and that another editor seems to have struggled to gain complete command of AWB. For me, the most stressful part was after the competition when the competitors had to sort out the results by themselves at a time when the judges were not advancing the discussion, or that is the way it seemed to me. This is why I think that there are lessons to be learned from the previous Stub contest mainly to reduce the stress of competitors. Stress seems worse when things are not fair and so I think that all aspects of fairness should be examined. The issue of the exact number of points awarded for a re-rate and for an expanded Stub appears to be controversial and may harbor an unspoken feeling of unfairness, so it may be better to separate the two sides of the competition. On the re-rating side, I note that some editors have said that the high scores of re-rates is easy. In-the-round, on the Stub expansion side, I would think that the Stub contest could have done better than produce so many expanded Stubs that were not popular and viewed only a few times by readers. Another model could be a challenge. The challenge is that anyone who expands more than 50 Stubs (or a certain amount) all share a set prize equally. This will make a lot more people winners and everyone could be a winner. The Stub contest would then be more about completing a challenge than producing one winner and some runners up. The Stub contest could be renamed the "Stub challenge" - to expand 50 Stubs, or perhaps 60 Stubs (about two per day). Specifications for the Stubs can be predetermined - ie should have at least xyz hits, and so on. Snowman (talk) 14:54, 27 March 2014 (UTC)

(belated reply) Hmmm, humans are naturally competitive, plus there is a limited pool of funds (250 pounds) to get prizes from, which makes giving a prize for everyone passing 50 stubs tricky. I will reopen debate very soon. I am thinking that a "rerate blitz" over the course of 48 hours might be a good contest - and not too exhausting to judge! Cas Liber (talk · contribs) 13:18, 29 April 2014 (UTC)

Rules for 2014 rerun

Right folks, thinking of rerunning it and figuring on how best to improve wikipedia. When I've been looking for existing stubs to expand for DYK, I've been put off by articles that have significant amounts of unreferenced text, many of which are broader more straightforward subjects (such as household tools for instance) - so am thinking of how best to maximise what we do here - all thoughts appreciated. Please indicate which rules folks like or dislike, or offer new options below. Options that help improving broader or more popular content would be good.Cas Liber (talk · contribs) 06:24, 21 June 2014 (UTC)

Note that I think re-rating will be done as a separate competition - a rerate blitz - see above entry. Cas Liber (talk · contribs) 06:36, 21 June 2014 (UTC)

Right YOOHOOO EVERYBODY! .....Sven Manguard Bobnorwal GoingBatty Leech44 Dana boomer Jamesx12345 Canada Hky Matt294069 Acather96 Coin945 Chris857 Samwalton9 Guerillero HueSatLum Cerebellum Tom Morris Cwmhiraeth Gobonobo Iselilja Resolute MasterOfHisOwnDomain Wizardman Ruby2010 Seattle czar Miyagawa Fortdj33 Andrew Gray Muboshgu Secret Ealdgyth Snowmanradio Schwede66 ChrisGualtieri Newyorkadam Go Phightins! Elekhh Sturmvogel_66 Carcharoth Mitchazenia Sbharris Circeus

Defining a stub

With above in mind, maybe a broader definition, particularly for unreferenced material is good. So maybe we define a stub as:

• If lacking any inline references at all, then an article of up to 1500 chars or 300 words.

• If has one or more inline references, then capped at 1000 chars or 200 words.

(Please discuss or offer alternative definitions of stubs - eg no distinguishing between number of refs - or divide into 0, 1-2, 3 or more....) Cas Liber (talk · contribs) 06:24, 21 June 2014 (UTC)

Pointscoring

• 10 base points for expanding to 1500 chars or 300 words, with >50% sourced with inline citations (double points if all material fully sourced) (or do we insist all material is referenced from the get-go?)

• +10 points for a stub with a score over 1200 on Wikipedia Release Version Tool (existing rule)

• +5 points if rated as top or high importance by any WikiProject. (existing rule)

• +5 points if reaches 4500 b "readable prose size". (existing rule)

• +5 points if article created before 1 January 2009 (??? - alternate date - ???)

• +20 points if article created before 1 January 2004 (??? - alternate date - ???)

Comments

• I would lower the threshold for the +10 points to 1,000 points (than ca. 10K stubs would qualify out of 2.5 million stubs we have). The 1,200 threshold last time was surely too high, with very few articles within any given topic (only 500 articles qualified overall) and very few participants scoring on that criteria. Regarding the new rule proposals, I don't see how old articles would be more important just because they are older. --ELEKHH^T 08:32, 21 June 2014 (UTC)

The rationale was that really old stubs were more likely to be more basic and less specialised material...the question is then, is removing some of the most longstanding stubs a worthwhile aim? Cas Liber (talk · contribs) 08:51, 21 June 2014 (UTC)

If we have evidence that such tendency exists than ok, but would need to be well calibrated. I'm sure there were bot runs before 2009 that added thousands of stubs, such as the 17,000 Category:Main Belt asteroid stubs. --ELEKHH^T 09:04, 21 June 2014 (UTC)

And asteroid stubs have been created in the early years already (2005). By the end of 2008 we had over 1 million stubs. --ELEKHH^T 09:22, 21 June 2014 (UTC)

Maybe really early then? 2005? Cas Liber (talk · contribs) 09:23, 21 June 2014 (UTC)

How exactly do you measure sourcing as a percentage? By sentence? I am not sure that's the best solution. Consider something like "Main points in the article, and anything that might be disputed, are cited. No entirely unsourced paragraphs." This would better accommodate the range of citation styles, and gets at what is really needed. Yes, judges will have to check to see if the major points are sourced (and ask for them to be sourced if they're not), but we'd have to count sentences and citations in the original plan, so my proposal is no more work. Sven Manguard Wha? 15:13, 21 June 2014 (UTC)

I am open to suggestions and yours sounds reasonable I think. Cas Liber (talk · contribs) 16:45, 21 June 2014 (UTC)

Apologies for not replying sooner. My interest is mainly in the assessment side rather than the expansion side (call it a deep and long-standing curiosity as to just how many bad articles we have...) but a couple of thoughts:

• One issue that I noticed with the old contest was people claiming (in good faith) an expanded stub where the original had a stub tag but was way over stub level by any reasonable definition. A clear metric for "stub" seems good, but it does pose the interesting question of whether we're implicitly defining everything bigger than that as definitely-not-a-stub ;-)
• 1500 chars = assuming this is 1500 bytes article size then I think it's reasonable. Using 1500 readable prose as the threshold is too high for a stub, I think, at least based on my gut feeling of what is and isn't one.
• We previously required people to rerate to claim the points, and this seems good to encourage/require this time around.

Otherwise, hope it runs again! Splitting the two aspects seems good and hopefully will streamline things. Andrew Gray (talk) 21:53, 3 July 2014 (UTC)

Andrew Gray, now my turn to apologise for belated reply. Some folks would class a larger unreferenced article up to this size a stub. It got me thinking that this might fulfill a role as some of these articles have too much prose to make for an easy 5x expansion for (say) DYK. Examples might be Apple corer and Chocolate syrup - actually may be worth a mention on DYK.....Cas Liber (talk · contribs) 21:10, 6 July 2014 (UTC)

Right, in case anyone is wondering - if w can fix old stubs then that is great, hence the bonus introduced. Regarding text, I have allowed for a generous classification of unreferencd articles as this is a bit of a counterfoil to DYK, and we can get some of these cleaned up. Cas Liber (talk · contribs) 14:25, 9 August 2014 (UTC)

Rerating taken to separate contest

Ok folks see User:Casliber/Template blitz and discuss on talk page. Hatching an idea on how to do this. Cas Liber (talk · contribs) 22:26, 7 July 2014 (UTC)

Is this another way of saying; "lets not bother with re-rating. This is the main talk page and the other talk page (actually a subpage's talk page) about re-rating will be forgotten"? Snowman (talk) 21:32, 13 August 2014 (UTC)

No, this is a way of saying we'll have a separate contest for that. I think the things are separate enough to do so. Also, the rate at which people can tag could be exhausting so I don't want a month-long contest - I'd find it exhausting. The idea is that after some discussion I will apply for another microgrant for that one separately. Cas Liber (talk · contribs) 13:59, 15 August 2014 (UTC)

BLP stub bonus

I think it's a good idea to add a bonus five points for every completely unreferenced BLP de-stubbed. The threshold would be the same for completely unreferenced stubs: 300 words and/or 8 sentences. Thoughts? Seattle (talk) 14:37, 9 August 2014 (UTC)

Hmmm, sounds like a good idea. will see what other folks think. Cas Liber (talk · contribs) 14:49, 9 August 2014 (UTC)

I agree. It's just that I don't know if there is any unsourced BLPS out there. Looks like they are all either deleted, tagged with PROD or in draft space. TheQ Editor (Talk) 20:41, 9 August 2014 (UTC)

There are about 1700 BLPs tagged as unreferenced, but you have to be careful: some consider a BLP with any relevant external link, not necessarily in a reference, to be sourced, so many articles in that category are not actually unsourced by that definition. We'd have to set down specific rules as to what constitutes an unsourced BLP. /~huesatlum/ 21:45, 9 August 2014 (UTC)

I would consider all BLPs without an inline citation as unsourced, because any material without an inline citation, particularly in BLPs, can be reasonably challenged. That opens the most possibilities for entrants to improve BLP stubs too, which is probably a subsection of that 1700. Seattle (talk) 23:01, 9 August 2014 (UTC)

@Mitchazenia: Thoughts? Seattle (talk) 21:45, 12 August 2014 (UTC)

Ok - IMHO, "referenced" means inline citations - a link at the bottom doesn't cut it. period. However if a good link clearly does give someone a headstart in reffing an article :) Cas Liber (talk · contribs) 22:46, 12 August 2014 (UTC)

Points

Lots of stubs have above 1500 b "readable prose size". How will their expansion be evaluated. Example -ASSOCHAM has 1600+characters at present, if a user expands it to 5000 characters how many points will be awarded?--Skr15081997 (talk) 16:30, 9 August 2014 (UTC)

There are many like that, alas, and of course those are not really stubs. You're welcome to remove the tag and update the talk page assessment, but that won't be a part of this contest. Wizardman 19:49, 9 August 2014 (UTC)

Advertisement

I was wondering if there is a wiki ad for this contest. Thanks, TheQ Editor (Talk) 20:47, 9 August 2014 (UTC)

I have posted it in a few places. And will get a note in the signpost. Other than that, what had you in mind? Cas Liber (talk · contribs) 21:31, 9 August 2014 (UTC)

If we could get a global notice, it would be great. Mitch32^{(Protection is not a principle, but an expedient.)} 01:45, 11 August 2014 (UTC)

Global notices are usually for pretty major things, but I think a watchlist announcement might be okay - presumably starting on Day 1 of the contest? Don't want to do it too early... Andrew Gray (talk) 10:19, 11 August 2014 (UTC)

I meant the watchlist global notice, but I usually call that one the global notice. Mitch32^{(Protection is not a principle, but an expedient.)} 00:38, 12 August 2014 (UTC)

Fair enough - happy to do this if you remind me closer to the time. Andrew Gray (talk) 22:05, 13 August 2014 (UTC)

@Mitchazenia: - still want a notice? If you give me suggested wording I'll put it up ASAP. Andrew Gray (talk) 21:25, 2 September 2014 (UTC)

Stub creation cut-off?

So could I spend the next three weeks creating stubs, and then de-stub them for contest entries?

On a serious note, is there a cut-off for stub creation? Would stubs created during the contest (or since the announcement of the contest in general) be eligible? Would self-created stubs be eligible (and if so, again, would there be some sort of cut-off in terms of creation dates)? Guettarda (talk) 01:54, 11 August 2014 (UTC)

Keep in mind the idea of the contest is that spirit of the contest is to can stubs that already exist. If we create more stubs, it seriously defies the idea that we're getting rid of them. (This is of course to why I have over 200 sandboxes). Mitch32^{(Protection is not a principle, but an expedient.)} 02:08, 11 August 2014 (UTC)

Hmm...good question. The whole idea is the improvement of existing stubs not adding more (DYK is (presumably) the incentive for new articles). So in the spirit of this I will say that only stubs created before 1st August 2014 are eligible. Cas Liber (talk · contribs) 02:10, 11 August 2014 (UTC)

Yes, this sounds eminently sensible. Andrew Gray (talk) 10:17, 11 August 2014 (UTC)

1500b readable prose

Is there a tool on Wikipedia to where you can check the amount of readable prose a given article has - minus code like for inline citations? LuciferMorgan (talk) 21:15, 12 August 2014 (UTC)

Yes - here is the page --> User:Dr pda/prosesize.js. Let us know if you have any problems. Cas Liber (talk · contribs) 21:20, 12 August 2014 (UTC)

Thanks for that, but unfortunately I had difficulty following the instructions. LuciferMorgan (talk) 05:19, 13 August 2014 (UTC)

I can see something on your monobook page. I am not hugely skilled at these thing either.....Cas Liber (talk · contribs) 06:04, 13 August 2014 (UTC)

You know what a monobook page is, which makes you more skilled than me... LuciferMorgan (talk) 06:21, 13 August 2014 (UTC)

is here at User:LuciferMorgan/monobook.js. Looks like a bit has been added in 2007.....10:42, 13 August 2014 (UTC)

I tried plugging it in myself. Let me know if it works; the button should show up under page information on the left side. Wizardman 12:06, 13 August 2014 (UTC)

Wizardman I am surprised you hadn't had it before now. A very useful tool that one. Cas Liber (talk · contribs) 12:16, 13 August 2014 (UTC)

Thanks guys. I guess the information I need is "Page length (in bytes)"? I only ask, just incase that includes inline citations and so on as well. LuciferMorgan (talk) 17:43, 13 August 2014 (UTC)

Once you have it, you have 'page size' appear in the tools section of the far left-hand column of things under the wikipedia ball on your page. Is it working for you now? Cas Liber (talk · contribs) 20:25, 13 August 2014 (UTC)

Unfortunately not. LuciferMorgan (talk) 20:43, 13 August 2014 (UTC)

Tried purging? Cas Liber (talk · contribs) 20:48, 13 August 2014 (UTC)

Tried that. LuciferMorgan (talk) 20:53, 13 August 2014 (UTC)

Can anyone help?LuciferMorgan (talk) 12:11, 15 August 2014 (UTC)

I tried modifying your monobook again; purge and try. If that doesn't work I'm out of ideas and you'll have to ask pda about it. Wizardman 12:13, 15 August 2014 (UTC)

Sadly didn't work, but thanks for trying. Will get in touch with him. LuciferMorgan (talk) 12:19, 15 August 2014 (UTC)

You might want to change the units for bonuses next time round ... "4500 b" reads "4500 bits". See Bit#Unit and symbol. — Brianhe (talk) 19:40, 30 September 2014 (UTC)

Looking for 3rd judge

Howdy. I am announcing that Casliber and I have agreed that we are looking for a third judge this time around. If people are interested in becoming a third judge, I would ask that you email Casliber and me your interest so we can discuss it and approve someone. Mitch32^{(Protection is not a principle, but an expedient.)} 02:42, 11 August 2014 (UTC)

 Done Sorted now. Cas Liber (talk · contribs) 23:44, 26 August 2014 (UTC)

And/Or

Maybe I am dull, but how shall I interprete "maximum of 150 words and/or 4 sentences"? Is it a stub if it has fewer than 150 words but 5 sentences? Iselilja (talk) 20:35, 26 August 2014 (UTC)

@Iselilja: happy to cut folks some slack on this - the and/or means you can take either to give the definition of a stub - i..e this example would count as a stub. Cas Liber (talk · contribs) 23:46, 26 August 2014 (UTC)

I agree that the use of "maximum of X and/or X" here makes no sense, in logical terms. A definition can either contain the phrase "shorter than 150 words or 4 sentences" or "shorter than 150 words and 4 sentences", the former being what Casliber had the kindness to inform us that is meant here. I hope noone minds me changing it. Desiderius82 (talk) 06:57, 21 September 2014 (UTC)

(And congratulations for the contest)Desiderius82 (talk) 07:07, 21 September 2014 (UTC)

Cool! thx for tweaking it...Cas Liber (talk · contribs) 08:06, 21 September 2014 (UTC)

On-the-fly scoring begins...

Just to say I've started marking off some of the submissions (but will wait unil I've run through everyone before updating the table) - starting from the bottom for no particular reason. A few have been marked as pending for a second opinion - @Casliber and Mitchazenia:, if you could have a look and let me know your thoughts. All are "was it a stub to begin with?" cases... Andrew Gray (talk) 22:24, 4 September 2014 (UTC)

Half-way done, will get the rest tomorrow. Andrew Gray (talk) 23:26, 4 September 2014 (UTC)

Sorry, just noticed the funny marks. Will take a look. Cas Liber (talk · contribs) 21:21, 6 September 2014 (UTC)

Ok, sadly Josh_Dueck does not qualify as at its pre-expansion version it is 177 words and seven sentences and has inline refs. Cas Liber (talk · contribs) 21:38, 6 September 2014 (UTC)

Perry Berezan had one one inline ref pre-expansion, which meant that it had to be under 150 not 300 words. Sorry, Resolute Cas Liber (talk · contribs) 21:49, 6 September 2014 (UTC)

S'alright. Suppose I should have read up on the rules rather than relying on what we did last year! Resolute 13:51, 8 September 2014 (UTC)

Resolute nevermind, your next nomination is all good. cheers, Cas Liber (talk · contribs) 09:21, 10 September 2014 (UTC)

Eugene Koonin is interesting - the prosesize tool does not count bulleted lists, and hence the preexisting stub had 127 words, so hence passes. Worth discussing at some time maybe. Cas Liber (talk · contribs) 23:18, 6 September 2014 (UTC)

Could I get a second look at Bioinformatics (journal)? Maybe I'm missing something but that's clearly under 1500 bytes right now despite it being checked (prosesize has 1280). Wizardman 03:30, 7 September 2014 (UTC)

Whoops - well-spotted. placed back on hold till expanded. Cas Liber (talk · contribs) 05:15, 7 September 2014 (UTC)

Thanks Wizardman for spotting this, part of my edit was reverted. Just keep on hold just now and I'll see if I can add something else. Thanks, Amkilpatrick (talk) 07:12, 8 September 2014 (UTC)

Hi Casliber, I've added a bit more to the Bioinformatics (journal) article and updated the diff on the entries page, hope this is ok! Thanks both! Amkilpatrick (talk) 08:50, 10 September 2014 (UTC)

Amkilpatrick Yep, that looks fine now. Thanks for expanding! cheers, Cas Liber (talk · contribs) 09:14, 10 September 2014 (UTC)

• Just gone through and updated all scores as of Wednesday night here. 92 stubs exapnded in the first 9 days. Interestingly, #2-4 are all neck-and-neck, and not far behind #1... Andrew Gray (talk) 22:48, 10 September 2014 (UTC)

291 stubs marked expanded as of tonight, of which 31 (!) have been stubs for over a decade. Andrew Gray (talk) 19:41, 27 September 2014 (UTC)

Diffs

Help needed! I have forgotten, or never knew, how to record a diff where several edits have been made during the expansion of the stub. (Or can I just provide a link to the article's history?) Cwmhiraeth (talk) 05:37, 5 September 2014 (UTC)

Go to the history page, select the edit before your first one with the first radio button and your final edit with the second, click "compare selected revisions". Andrew Gray (talk) 16:29, 5 September 2014 (UTC)

Ah! I don't think I have come across that before. Thanks. Cwmhiraeth (talk) 18:05, 5 September 2014 (UTC)

Prizes in currency where the winner lives?

I live in the US and use dollars. Many editors live outside the €uro zone. Could prizes be in the currency that a winner chooses?

(Yes, a winner could have a bank change the €uros into dollar$, but they take a large percentage as a "service" fee.)

Thanks either way -len

It'd be Amazon and done digitally by them as they are international. Should be pretty minimal loss I expect. Cas Liber (talk · contribs) 21:15, 6 September 2014 (UTC)

Thanks. — Lentower (talk) 02:44, 7 September 2014 (UTC)

Are you sure Cas? They used not to be able to do that. Johnbod (talk) 15:32, 7 September 2014 (UTC)

We should I guess ask some of the recipients what has happened so far. I haven't had anyone complain since we've been doing this with core and stub contests....Cas Liber (talk · contribs) 01:48, 8 September 2014 (UTC)

Untagged stubs

I'm seriously considering jumping in, and I'm noticing that a lot of those tagged in my areas of interest are too long for this contest, while a significant number of those that are short enough aren't actually tagged as stubs (even though they are). Do these count? The Drover's Wife (talk) 15:29, 7 September 2014 (UTC)

Ones that are tagged as stubs but too long (ie, >150wds content) aren't eligible for expansion in the contest (but please do take the tags off if you spot them!); ones that are "really" stubs but don't have a stub tag on them are eligible. Andrew Gray (talk) 21:04, 7 September 2014 (UTC)

Yep. what he said. :) Cas Liber (talk · contribs) 01:47, 8 September 2014 (UTC)

Thanks heaps! The Drover's Wife (talk) 01:49, 8 September 2014 (UTC)

Second question - I'm unclear about the 4-sentences thing. Does that mean if it's under 150 words, but over 4 sentences I can't use it? The Drover's Wife (talk) 03:37, 8 September 2014 (UTC)

No it's nice and flexible for the editors' benefit - hence if it is either 150 words or 4 sentences then it can qualify as a stub Cas Liber (talk · contribs) 03:47, 8 September 2014 (UTC)

Why was my entry rejected?

My expansion on the Cape Town International Convention Centre was rejected even though I expanded from less than 300 word and no reliable sources (it linked to a coi blog)? Thanks. Nathan121212 (talk) 12:28, 10 September 2014 (UTC)

Good point - I didn't notice the source was not reliable and in effect a non-source. Corrected now. Cas Liber (talk · contribs) 13:20, 10 September 2014 (UTC)

Minimum requirements clarification

Just a quick note that I've reworded the rules a little and added a sentence - the rules haven't changed but it should hopefully be a little clearer that only stubs which were below the minimum size and have now been expanded past it will count for points. Other judges please feel free to revert if you think this isn't needed :-) Andrew Gray (talk) 21:54, 16 September 2014 (UTC)

Looks fine Cas Liber (talk · contribs) 02:54, 17 September 2014 (UTC)

Province of Cáceres

I'm rather concerned that entries like this are being approved when it's obviously of poor quality and almost entirely reliant on one source. Some form of quality should also be considered here and if editors are going to translate or write text from languages they're not half proficient in then they need to get somebody who is to proof it. "It was formed a province in 1833", "Except the northern and southern parts the whole of the province is plain and fertile area ", "The water of river Tagus", "Gabriel y Galán dam was made on the Alagón River" and "Las Hurdes was one of the poorest region". Las Hurdes is not a region anyway, it's a comarca, it's misleading. Even Encyclopedia Brittanica isn't spelled correctly.♦ Dr. Blofeld 12:06, 18 September 2014 (UTC)

All of these will be fixed shortly. Thanks for pointing them out.--Skr15081997 (talk) 14:03, 18 September 2014 (UTC)

My apologies, the user name didn't ring a bell. Just try to take care with the prose that's all! Some of your other ones look better in fairness.♦ Dr. Blofeld 15:49, 18 September 2014 (UTC)

Thanks for the last comment. I will follow the advice. Thanks again.--Skr15081997 (talk) 16:09, 18 September 2014 (UTC)

Finding articles

If you are having trouble finding articles under 150 words, Wikipedia:CatScan is useful - just set a file size of 2000 or 2500 and you'll find plenty. James^x12345 13:47, 18 September 2014 (UTC)

Shouldn't one check to see if one actually added info or just changed it to a start instead?

Like lets say user a put in tons of stuff but forgot to change the tag so then someone else took the credit. That could happen. Wgolf (talk) 22:04, 19 September 2014 (UTC)

People are submitting their diffs. Easy for me to see history. But it is all supposed to be collaborative, so not a hanging offence neither. Cas Liber (talk · contribs) 00:21, 20 September 2014 (UTC)

Stub contest entries

Can folks please take a look at the Entries page and - if any entries have an C symbol, it means they need to be expanded further to qualify - can folks please update the entry list if/once they've expanded any? Also don't forget to change the template on the article talk page to "start-" or "C-class" or whatever. Finally, in the interests of data gathering, once this contest is concluded I'd really like to get an idea of the ages of the stubs expanded. Easiest is if everybody places a year of creation in parentheses after their entry. If everyone could do their own that would be great...otherwise it'll be alotta work for me afterwards. Cas Liber (talk · contribs) 23:31, 24 September 2014 (UTC)

@Casliber: - I've checked most of the recent entries but left query marks next to half a dozen of Blofeld's, because I was having trouble with the release tool earlier. They're all checked other than this (and not yet added to the scoreboard) - if you have a chance today could you do them? If not I'll try and finish it tomorrow. Andrew Gray (talk) 17:00, 25 September 2014 (UTC)

Ok - should do. Cas Liber (talk · contribs) 20:16, 25 September 2014 (UTC)

Be patient when clicking the tool server site, what I do it have it open on a different tab and get on with something else while it loads. Most of the recent expansions are from page 2 or 3 I think of the tool so it takes even longer to find! It does load eventually. just takes a few minutes a time.♦ Dr. Blofeld 22:27, 26 September 2014 (UTC)

I was doing the initial scoring on the train, and the connection was sufficiently flaky I gave up trying after a while :-). Will do the updates just now. Andrew Gray (talk) 18:17, 27 September 2014 (UTC)

@Casliber: and @Andrew Gray: Can you check my latest entries, for some reason for Provinces of Brindisi and Lecco I can't find the entries on +1200 which I could have sworn were there before. It seems odd that the other Italian provinces were rated over 1200 and not those.♦ Dr. Blofeld 15:41, 29 September 2014 (UTC)

I get 1200+ for Brindisi (and most others), but only 1155 for Lecco (no idea why). Andrew Gray (talk) 19:25, 29 September 2014 (UTC)

Regarding the province...maybe there was some confusion with Province of Lecce? Cas Liber (talk · contribs) 08:43, 1 October 2014 (UTC)

Redirects eligible?

I'm not sure if the rules allow expansion from a redirect? Brianhe (talk) 14:09, 26 September 2014 (UTC)

If it is a stubby article that was created and then converted into a redirect I have no problem in including it, as it shows that a stubby article did exist at some point in time. Cas Liber (talk · contribs) 14:42, 26 September 2014 (UTC)

Easily-expandable video game articles

Articles not eligible.--Coin945 (talk) 23:54, 3 October 2014 (UTC)]]
The following discussion has been closed. Please do not modify it.
@Nathan121212, Vin09, Taylor Trescott, Acather96, HueSatLum, Jakec, and TheQ Editor: @Ruby2010, Leech44, SL93, 97198, Wizardman, LuciferMorgan, and Resolute: @Ben Moore, MasterOfHisOwnDomain, Amkilpatrick, Cwmhiraeth, Matt294069, Secret, and Sturmvogel 66: @Jamesx12345, Bobnorwal, Guettarda, Husain007, Anna reg, Worldbruce, and ProudSaffa: @KeepCoolDon'tFreeze, WAVY 10 Fan, Parsecboy, GMTEgirl, ZappaOMati, Brunei, and Miyagawa: @Amandaaa99, Awerage, Skr15081997, Bruin2, Gongojeet, Rsrikanth05, and Teemu08: @Tnophelia, Dr. Blofeld, Mike Searson, Babita Arora, SmileBlueJay97, Brianhe, Lewis Hulbert, and Muboshgu: If you're a contestant in the Stub Contest, and are having trouble finding articles to use for the competition, here's a list of many 2 line stubs that could use some destubbifying: --Coin945 (talk) 17:39, 29 September 2014 (UTC) Electric City: The Revolt Lad (video game) Hot Wheels: World's Best Driver 9 Elefants Monty Python's Cow Tossing Scarface (iOS game) The Relic (video game) Ice Breaker (video game) Bubble Trouble (Atari) Frisbee Forever Sword & Poker Pizza Vs. Skeletons Trainyard (video game) Drive on Moscow Windosill Little Things (video game) Wooords Qvoid Ant Raid Tilt to Live Piyo Blocks PewPew Bug Heroes First Touch Soccer Amazing Breaker Space Miner: Space Ore Bust Dark Nebula (video game) Beat Hazard The Great Jitters: Pudding Panic Reckless Racing Reckless Racing 2 Ragdoll Blaster 2 Ragdoll Blaster 3 Beat Sneak Bandit Glyder 2 Bag It! Punch Quest Papa Sangre II IBlast Moki IBlast Moki 2 Super QuickHook Traveler IQ Challenge Mikey Hooks Mikey Shorts Mikey Boots Table Tennis Touch Traps n’ Gemstones The Ensign (video game) Great idea! Hopefully some will see some expansion....Cas Liber (talk · contribs) 20:29, 29 September 2014 (UTC) These all seem to be created in the past couple of days - unfortunately, this means they're not eligible! Andrew Gray (talk) 20:35, 29 September 2014 (UTC) d'oh! Cas Liber (talk · contribs) 21:56, 29 September 2014 (UTC) d'oh indeed... :/--Coin945 (talk) 16:37, 30 September 2014 (UTC)

Scoring and rulings

A couple of edge cases, for @Casliber and Mitchazenia: to rule on, both to do with defining something as an initial stub:

• Canongate Tolbooth - was too large prior to expansion, but most of this was a textdump that was removed
• Ed Hughes (anchor) - similar, some material removed prior to expansion

There were two (Ingalls, Oklahoma, and Port Talbot, Ontario) which were very close to the initial size cutoff, 151 and 159 words respectively. I've struck them but happy to count them if you'd prefer.

One other oddity:

• French cruiser Ernest Renan - claimed by two users jointly.

I'm happy to defer to you on whether or not the first two are eligible, and I'd suggest splitting the points for the second one. (At the moment, they've both been given full points, I think)

Overall... we asked for submissions by midnight tonight, so assuming nothing more comes in tonight, it looks like neck and next for first place, third & fourth pretty clear, all unaffected by the decisions above. Probably best to go over the totals very carefully, so I'll aim to do this tomorrow night. Andrew Gray (talk) 22:05, 5 October 2014 (UTC)

I just gave 'em full points as we should be encouraging collaborative editing - had no impact on results but on second thoughts could be gamed, so was musing on whether maybe 3/4 cost would be an inducement....? Cas Liber (talk · contribs) 23:01, 5 October 2014 (UTC)

Fair enough, happy to leave it at full points (but perhaps something to consider for a rerun). I've just recounted and we seem to have missed one of Blofeld's multipliers - so scores are now at 1105 Blofeld and 1090 Skr15081997, with Cwmhiraeth in third and Wizardman in fourth. I haven't looked for any points not explicitly claimed by the contestants, but both of the leaders were pretty comprehensive at this so I don't think we're likely to have missed any. Looks like we have our winners, if you're happy to sign off on it :-)

Oldest stub will be Blofeld with Double-hulled tanker, and I haven't yet looked for the most heavily trafficed (is it possible to do this with a script, I wonder? Time-consuming otherwise...) Andrew Gray (talk) 22:33, 6 October 2014 (UTC)

All looks good to me - I'd be very surprised if any stub had a higher view count than September 1 Cas Liber (talk · contribs) 00:38, 7 October 2014 (UTC)

Hmm - I think you're right but I'm reluctant to give it based on a reasonable guess! Do you want to notify the winners for 1st-4th places, and the oldest stub bonus, for now, and I'll see if I can get a way to check for the pageviews? Andrew Gray (talk) 22:21, 7 October 2014 (UTC)

Ok. Will announce shortly....Cas Liber (talk · contribs) 22:41, 7 October 2014 (UTC)

done now Cas Liber (talk · contribs) 12:56, 8 October 2014 (UTC)

@Andrew Gray: any update on an easy way to check the pageviews? Cas Liber (talk · contribs) 14:37, 24 October 2014 (UTC)

@Casliber: - not just yet but I think I may have cracked it (running a script manually over the day's log files). Andrew Gray (talk) 19:06, 27 October 2014 (UTC)

@Andrew Gray: cool! keep us posted....Cas Liber (talk · contribs) 19:53, 27 October 2014 (UTC)

Provisional result is September 1, followed (an order of magnitude later) by Divyanka Tripathi and T. J. House. I'll run the script overnight (just downloading the last of the pageview files now) and confirm in the morning. And then save the script in case we have to do this again :-) Andrew Gray (talk) 21:42, 27 October 2014 (UTC)

Confirmed - September 1, more than sixteen times the traffic of the the nearest competitor. Andrew Gray (talk) 08:07, 28 October 2014 (UTC)

Right, I've emailed Katie Chan (WMUK) and Chase me ladies, I'm the Cavalry, so they should be in contact with prizewinners soon. cheers, Cas Liber (talk · contribs) 09:28, 2 November 2014 (UTC)

Hi Cas Liber. Fabian Tompsett is filling in for Katie at the moment. I've let him know that the stub contest is over and WMUK needs to contact the prize winners. Thanks for running the competition! Richard Nevell (WMUK) (talk) 11:00, 3 November 2014 (UTC)

No problem/it's a pleasure :) Cas Liber (talk · contribs) 12:02, 3 November 2014 (UTC)