User talk:SuggestBot/Archive 2

My Feedback

1)You might want to archive this page. 2) If there is any way to have more up-to-date versions of pages, it would be good, because one of the ones I was recommended for was deleted. Thanks, Genius101^Guestbook 19:57, 27 April 2009 (UTC)

Pathetic Bot

Sorry ForteTuba, but the bot is pathetic. I have done a lot of work related to the Percy Jackson articles, and so it suggested me a list of articles containing the word Percy and several Language related articles, which I am not interested in. Please try to improve the bot from this state. Thanks and don't take offense. I just gave feedback as a friendly suggestion. Pmlinediter  ^Talk 15:10, 28 April 2009 (UTC)

Yeah, one of its recommenders was a straight text-based match. You'll be glad to know, then, that the crash that took out the bot for the summer also killed the text-matching recommender at least for the short term. -- ForteTuba (talk) 13:05, 11 August 2009 (UTC)

Still active

Is this bot still active?

dfrankow (talk) 19:08, 30 April 2009 (UTC)

Feedback

Suggestbot is pretty good, but I added my name onto '/Regulars' and I've only got one recommendation. Also, I have to address this point: It's suggesting things I never saw any relation to. Several suggestions I got are just related, but I noticed that several 'stub' suggestions, such as Kowloon-Canton Railway Corporation, aren't stubs. Could you please improce the 'stub' suggestions, and also find a way of doing regulars (such as programming the bot to move all the users in '/Regulars' into in normal list every run, or writing a new bot that might do that).

Okay, I spent a long time on suggestions. But I have to say, it's good, so keep improving it!--Leolisa1997 (talk) 12:54, 26 July 2009 (UTC)

It went down for a long while with a machine crash. It should be coming back up. Stubs (and other work categories) will also have a more recent list; they hadn't been updated for some time. -- ForteTuba (talk) 13:04, 11 August 2009 (UTC)

More feedback

Most of the suggestions I got were fairly relevant, but I somehow managed to get both December and June (as far as I know, I've never edited a month page), and Ontario Australian Football League (probably from an Ontario-related page on my watchlist, but I usually don't go for sports). It's also listing a disambig page (Dolly) as a stub, not sure why. Thanks though, Nikkimaria (talk) 01:46, 13 August 2009 (UTC)

Stubs

I got a few articles in my "stubs" section which weren't stubs... Other than that it's really good. Of course not all of them will interest me. Mainly due to my vandalism reverts. But that's BOUND to happen. Great bot, not-so-great stubs :) Marx01 ^{Tell me about it} 01:54, 14 August 2009 (UTC)

Ditto here, a LOT of stubs and articles that haven't been stubs for years. Bot has a hiccup! Montanabw^(talk) 04:03, 14 August 2009 (UTC)

It was just brought back online. I'm sure there is a lot of dust in the database that will need to be blown out before everything is working again. --EpochFail ^{(talk|contribs)} 19:44, 14 August 2009 (UTC)

Username Change

I changed by username from User:Juthani1 to User:World. The bot has been placing suggesstions on my old talk page which now redirects to my new page. I would like it if this change can be made so that article suggesstions come only to User talk:World. Thanks World (talk • contributions) 16:41, 14 August 2009 (UTC)

 Done --EpochFail ^{(talk|contribs)} 19:42, 14 August 2009 (UTC)

Adding peer review?

I was wondering - we always can do with more reviewers at Peer Review - whether suggestbot can suggest articles currently at Peer Review for people to look at. Peer Review is underpopulated by editors, and this lag in reviewing leads to more work at GAN and FAC - I figured as Peer Review is unstructured, it was a good way of introducing editors to reviewing articles there. Casliber (talk · contribs) 05:37, 17 August 2009 (UTC)

Disambiguation Page = Stub?

The bot listed Pauline under the Stub section for some reason. Aside from the fact that I'm not entirely certain why I have this listed (although I have a hunch), it's linking to a disambiguation page and calling it a stub. BAPACop (converse) 06:02, 14 September 2009 (UTC)

Outdated suggestions

Hi, Just an FYI if you look at the suggestions SuggestBot made on my page, it identified several articles as stubs that have not been stubs for well over a year, maybe two (I know this because I was the one who expanded them! LOL!) Somehow, it isn't reading and recognizing current tags. Don't know how to fix, but FYI. Incidentally, I don't really mind the occasional odd-to-explain-why-it's-there suggestion, sometimes leads in an interesting direction I hadn't thought of. Montanabw^(talk) 02:39, 17 September 2009 (UTC)

Yeah, it has some issues with how often it re-loads the work categories. It's both manual so easy to forget and fragile so sometimes doesn't work when I try. There's some new energy forming around people who want to work on it so I'm hoping I'll be able to freshen it up soon. -- 128.84.98.50 (talk) 20:54, 5 November 2009 (UTC)

New bot or Flagged Purpose

I think I would like a bot (hopefully yours) to suggest things from all categories associated to my User Page (or UserBox page...) that have NOT been edited in a while (especially with any WikiFlags). Could I create some kind of /suggestbot page that contained flags for controlling certain behavior? -- Mjquin_id (talk) 04:42, 21 September 2009 (UTC)

Also, although I found the page an interesting read; this page really, really needs archiving; maybe 180 days User:MiszaBot/Archive_HowTo? Ask and I will help. -- Mjquin_id (talk) 04:53, 21 September 2009 (UTC)

Miscategorization?

SuggestBot seems to have categorized the article, Havana, under "Stub", where it clearly is not. Cheers. --Intelligentsium 23:55, 21 September 2009 (UTC)

B e a u t i f u l

The useful associations are really good (excluded below), but these tell a story I am soon to forget:

Anticipation
Marine worm
Sphincter
Meat grinder
Chameleon
Compersion
Melancholia
Fanaticism
Dissappointment
Humiliation
Jealousy

Perhaps an interest in anxiety (a negative feeling), complementary medicine (at times crazy adherents) and the brain (human anatomy) might explain some of the above
Nevertheless, anticipation is spot on, but I included it for the sake of the storyline.
I'm so happy I have to say it in Japanese: Ureshii desu!

Thank you SuggestBot - Pleasure meets utility

Ostracon (talk) 20:32, 22 September 2009 (UTC)

ANI

Please see Wikipedia:Administrators'_noticeboard/Incidents#SuggestBot_spamming_new_User_talk_pages. 99.166.95.142 (talk) 15:59, 23 October 2009 (UTC)

Feedback from my suggestions

SuggestBot predicts that you will enjoy editing some of these articles. Have fun!

Stubs People N, not a stub Moron N N, not a stub, and the name of a bad word Yeast N N, a very long article ChatZilla ? Archaeology of the Americas N, not a stub Hat N, not a stub Fart N N, not a stub Rap N N, redirect, there must be a bug Whip N, not a stub, there must be a bug Madfox Y Y Thomas N, disambiguation page Zhukov N N, redirect page, please fix this East Siberian Sea Y Sneeze N N, not a stub, please fix this Amy N N, disambiguation page that is longer than a stub Knowledge Adventure ? ? Loren D. Estleman ? June N, not a stub Sarah N N, not a stub

Cleanup Plastic wrap ? Chapultepec Castle N, does not have a cleanup tag Black people N, no cleanup tag Merge Breakdance N N, already is merged Acne vulgaris N, no merge tag Christmas truce Y Add Sources Ben Y Hemicellulose ? Wednesday Y Wikify Jean Piaget ? Hit and run (baseball) N Microsoft Certified Professional ? ? Expand Military communications Y Alucard (Hellsing) N, section with tag already has enough content Southern Cone ?

SuggestBot picks articles in a number of ways based on other articles you've edited, including straight text similarity, following wikilinks, and matching your editing patterns against those of other Wikipedians. It tries to recommend only articles that other Wikipedians have marked as needing work. Your contributions make Wikipedia better -- thanks for helping.

If you have feedback on how to make SuggestBot better, please tell me on SuggestBot's talk page. Thanks from ForteTuba, SuggestBot's caretaker.

P.S. You received these suggestions because your name was listed on the SuggestBot request page. If this was in error, sorry about the confusion. -- SuggestBot (talk) 00:50, 18 December 2009 (UTC)

• Y This is a good choice.
• N This is a bad choice or may not fit the category that it is listed in.
• ? Unfamiliar, but I might edit it or might not edit it.
• Y Y This is a very good choice.
• N N FortaTuba should fix any issue with the bot.
• ? ? Never heard of before, but I'm, indeed, interested in it.

There are a lot of non-stubs listed in the stubs section. Please fix. December21st2012Freak ^{Happy Holidays!} 01:30, 18 December 2009 (UTC)

User:Thejadefalcon

Regarding these suggestions: pretty good. I think I'll sign up again one time. There were a few issues:

• Duras takes me to a disambiguation page, rather than the correct House of Duras.
• Xindi takes me to a flute, rather that the correct mass murderers, Xindi (Star Trek).
• Some things, such as Blossoms in the Dust and Holiday (1930 film) are of zero interest to me and presumably come from my new page patrolling. May I suggest a limit to SuggestBot. Say... it only bases suggestions on articles that you have more than five-ten edits on?
• Two pages, Perfect Dark Zero and Dominion War, are already on my watchlist, though I haven't made any edits to them yet.

As far as I can see, that covers my issues this time. I'm going to go put my name on the list for repeat suggestions. Any idea how long those can take/any idea when the bot will run constantly? --Thejadefalcon^{Sing your song}_{The bird's seeds} 05:24, 18 December 2009 (UTC)

Can you give me a list of articles?

..that I would like to edit? Just some sugesstions. Fuzzyhair2 (talk) 02:11, 14 January 2010 (UTC)

Title links

Please provide a link to a page in each of category. Ie, what is wikify versus cleanup, expand versus stub, etc? They seem similar. I am not sure when I go to a page what exactly the objection is. Was it just that some person tagged the article or has text processing been done and some criteria meant (lack of links, etc.). A link to a page on how to find this information or some wiki key policy would a least be a step forward. Bpringlemeir (talk) 20:32, 21 January 2010 (UTC)

Bot standing?

I've noticed the bot hasn't handled any "I want to try" requests lately. Is the bot out of use? Pitke (talk) 10:58, 25 February 2010 (UTC)

Quick Comment

When I got my new list, one of the stubs was the NES Cleaning Kit. As you can see, its deleted. Can you remove it or something? GamerPro64 (talk) 20:34, 15 March 2010 (UTC)

How long it takes...

How long does it take to get the suggestions back and how many will you recieve back

The delay depends on what the status of the bot is (we could perhaps add some info abut that). Currently the recommendation template lists 34 articles. (I'm mostly replying to this so MiszaBot can archive stuff) Nettrom (talk) 03:48, 16 April 2010 (UTC)

Comments

I cant complain about the post i received,most of them did not interest me, but that is my fault, most of them i could link to something i had edited, i am certainally going to sign up for regular ones, if i edited more articles that interested me, i would have got different results. i have marked a lot of articles as 'Speedy Deleting' or 'Stubs' Recently so have gone through a range of articles. below are what i got, please excuse my spelling, im full of typos.

Stubs Elliot Wilhelm-Dead Link- i must have done a william page Tage William-Olsson- Never Heard of him, thing i did an inter wikilink with it. another william one. Fitzpatrick Nuclear Generating Station- Did something on a nuclear thing Native American hip hop-Where did that come from? Cupidon (1875)- related to a guy called william Otto-Henry, Duke of Burgundydid some bio thing about austria, could that relate? Beaver Valley Nuclear Generating Stationnuclear. Brasero-Disambig, one links was to a page i had already done. Millstone Nuclear Power Plantnuclear again. William-Henry Gauvinanother william one, although i did edit this :) The Nut Gatherersanother william one Attack (song)-where did this come from? La Dansesame william Hope Creek Nuclear Generating Stationnuclear NGC 2997some thing about a galaxy China Institute of Industrial Relationswhere did this comefrom. Alone in the Worldwilliam guy again Robert J. Chasselli had edited a software article The Knitting Girlanother william one

Cleanup SimMars- i edited the sims 2, i dint find this one interesting. Mario Party 5see above Pörkölthungary again Merge Personalization List of U.S. state insects Rubbia reactorredirected, nuclear Add Sources List of U.S. states' largest cities- must have done an arirican linked page Hooton Park? List of the most popular names in the 1890s in the United States Wikify Suggestion Sundance Vacations Celine Dion awards and accomplishments? Expand Demographic history of the United Statesamerica Count KVDA?

I'd like to add that we've done some trials with users on Norwegian Wikipedia and found out that patrolling and list-related activities skewed the recommendations in a non-positive way. The bot might therefore filter minor edits and reverts, which we found was quite a success, provided the user uses those mechanisms in a predictable fashion. Nettrom (talk) 03:52, 16 April 2010 (UTC)

Real life Privacy violation?

Please stop!!! Why your bot keeps suggesting me to edit the article Italian Communist Party? I never edited anything related to that. Never, and at this time I'm not going to add words that I would later regret. best regards - --Doktor Who (talk) 03:05, 8 April 2010 (UTC)

Thanks for the feedback, I'm sorry to hear that the bot recommended an article that's upsetting you. The algorithms SuggestBot uses will unfortunately lead to some of these kind of situations as there is not a single straightforward way that leads to its recommendations. I've made a note of looking at this when we later work to improve the bot, for example by allowing users to exclude articles or categories, but I currently do not know when we will have time to implement any changes. Nettrom (talk) 14:37, 11 April 2010 (UTC)

I don't like to be negative...

But most of the suggestions were irrelevant:

Stubs Bruno Langley he left coronation street ages ago, and i don't know anything about him Al Attles never heard of him Wendi Peters i don't know anything about her Sfumato never heard of MSN Music i hate music Shortland Street not related to coronation street Asiatic Warriors the only warriors i have ever edited is the book series, not this Boston Market no interest Kimnyole never heard of Northwestern United States no interest; i am english Skeleton warrior i never played a game with one in Katherine Tudor never heard of Timberline High School (Weippe, Idaho) never heard of The Warriors EP, Volume 2 the only warriors i have ever edited is the book series, not this Warriors Three the only warriors i have ever edited is the book series, not this The Warriors EP the only warriors i have ever edited is the book series, not this Coca-Cola Cherry i hate coca cola Zardoz never heard of Crème fraîche never heard of

Cleanup Contrary Warrior the only warriors i have ever edited is the book series, not this Second sight maybe their because of the book the sight Noel Edmonds i despise him and wouldn't touch his article with a 10 foot pole Merge Tallarn never heard of Equilibrioception no Warriors, Inc. the only warriors i have ever edited is the book series, not this Add Sources Henry, Duke of Cornwall no idea Regional lockout i hate regional lockouts David Walliams i trust what wikipedia says (mostly), i have no need for sources Wikify Khumalo what is wikify Sphynx (cat) " Amy Barlow " Expand Cell microprocessor they are cool but i know next to nothing about how they work Azad (rapper) i hate music Verizon Wireless never heard of

as i say, i don't like to be negative...

Sorry to hear that. I know ForteTuba made a lot of improvements to the recommendations, and we'll also be trying to improve things in the future. Nettrom (talk) 23:25, 25 April 2010 (UTC)

Slightly odd Suggestions

I have made almost all of my edits on novels and literature, none of these suggestions relate to either of those, I got transportation in america or something like that, when I havent even ever edited a page dealing with america in general

Even though we're aiming to recommend articles that are closely related to an editor's recent edits, there is always a possibility of getting what appears as unrelated articles. For some editors, those recommendations will still be useful, while for others they're not. Nettrom (talk) 23:28, 25 April 2010 (UTC)

Feedback from YixilTesiphon

Mostly pretty good, though I'm a little confused why it gave me Beleriand. Definitely a useful tool. —Preceding unsigned comment added by YixilTesiphon (talk • contribs)

Because of the way the recommender works, it can find what appears to be unrelated articles and recommend those. As long as they're articles you are able to improve on, my thought is that it's a successful recommendation. If it's not at all useful, and the confusion baffles you, it's of course not successful. It might be that we'll have better tools available for explanation of recommendations in the future, as that's on our list of ideas. Nettrom (talk) 23:33, 25 April 2010 (UTC)

Feedback from X!

Stubs John-Paul Clarkin Flag of Suriname Pope Victor I Norman St John-Stevas, Baron St John of Fawsley Bundesversammlung (Germany) Piancastelli Continuous obsolescence Pope Innocent V Arirang Festival Pope Adeodatus II Pope Cornelius Military awards of World War II Richard Parker (economist) Ottone Kijiji Pope Caius Hatfield Swamp Karl Koch (botanist) Thirteen Days (book)

Cleanup Pope Paul II Pope Leo I Containment Merge List of United States Presidents by date of birth Digital download Electrical double layer Add Sources List of the most popular names in the 1900s in the United States Pope Eusebius List of the most popular names in the 1890s in the United States Wikify Identity (social science) Norm (sociology) Karl-Erik Norman Expand Pope Benedict II Children of the Prime Ministers of Canada Direct manipulation interface

Just responding to this so the bot can archive it later. Cheers! Nettrom (talk) 23:34, 25 April 2010 (UTC)

Feedback 2 from User:Mark Chung

Stubs Penilaian Menengah RendahY Y Interlocking directorate? ? Mukah DivisionY East MalaysiaY Y JulauN Kuching DivisionY Y Melaka Maju JayaN Samarahan DivisionY Y Bintulu DivisionY HalogenN Miri DivisionY Kapit DivisionY Fu-un ShinsengumiN N MeradongN N Hartig DrugN N TatauN Intermediate Unit? Sarikei DivisionY Borneo PostY

Cleanup Asian valuesY Intranet? EgyptAir Flight 648Y Merge List of Intermediate UnitsN Software cracking? Tui naY Add Sources St. Francis Institution? ? Universiti Malaysia SarawakN Abang Haji OpengN N Wikify MetallographyN N BitboardN N Newsprint? Expand German Red CrossY The StandN Tobacco advertisingY Y

• Y Good. It's somehow related to what I typically search.
• N No! Bad choice! There must be a glitch in the bot...
• ? Unfamiliar, but no strong objection.
• Y Y Perfect, Magnificent, Awesome choice!
• N N Stupid counterfeit bot...
• ? ? Never heard of before, but I'm, indeed, interested in it.

Also, just responding to this so the bot can archive it later. Nettrom (talk) 23:35, 25 April 2010 (UTC)

Juve10

This first list I recieved wasn't too bad at all, and actually it was a pretty desent list given to me, but possibly we could make it better by adding articles that are also not yet created that you suggest is an article that should be created.

Recommending redlinks is on the list of ideas we have, and it might be possible to do something about it. We'll probably explore those ideas some time in the near future when things are running smoothly. Nettrom (talk) 23:36, 25 April 2010 (UTC)

Feedback 1 from User:Senthryl

Stubs Electropositivity ? Scott Adams (game designer) Y Jalgaon Rural (Assembly constituency) N Papi Hills N Pelican Accessories Y Rachyal N Herman Chin Loy N Union of Democratic Forces for Unity N John Swett High School ? Ducati Moto Y Marinesco–Sjögren syndrome ? 6th Virginia Regiment N Interstate Athletic Conference N N Rhymers' Club N N Robin of Sherlock Y Gardner High School ? Red 5 Studios Y Steamboat Pilot & Today N Annie Hayden N

Cleanup Moody House N N Mothering (magazine) N N N Costello syndrome ? Merge Jimmy Carter N 2000s in science and technology Y Harassment by computer Y Add Sources Popeye (arcade game) Y Toby Keith N N Mao Yisheng N Wikify Avi Wigderson Y Y Heerlen photo gallery N Park Place Productions Y Expand Smart Bomb Interactive Y Rush Limbaugh N N N Mortal Kombat Y

• N N N So unrelated to my interests that it is contrary or even funny.
• N N Completely unrelated to my interests. Definitely won't edit.
• N Unrelated to my interests - probably won't edit.
• Y Related to my interests - might edit.
• Y Y Great find or already on watchlist. Will definitely edit.
• Y Y Y Dropping everything and editing it immediately!
• ? May be related to my interests, but I lack the expertise to edit.

---

Relevance Score: 30.2%

— Senthryl ^(talk) 21:20, 3 June 2010 (UTC)

Great feedback, thank you for taking the time to report it! I'll be going through these in detail and compare it to what our recommender did, and see how we can improve the performance. Nettrom (talk) 16:05, 5 June 2010 (UTC)

Feedback from User:Thibbs

Stubs Benjamin William Mkapa Special Economic Zone N - I rarely edit political articles. Straight Life (Freddie Hubbard album) ? - I like jazz but rarely edit album articles. Heme-regulated inhibitor kinase ? - Basic interest in biology but why so many enzyme articles? Scot Eaton Y Alpha,alpha-phosphotrehalase ? - Another enzyme. Gianluca Freddi N Alpha,alpha-trehalase ? - And another. Dextrin dextranase ? - And again. SuggestBot must think I'm mainly here to edit on enzymes. SERPINB6 ? - And now a gene. Kind of straying far from my normal macro-bio interests. Anthology of Tom Waits N - Tom Waits is cool but I rarely edit album articles. List of fictional lions Y ? - I see why but this is not a stub. Frederick Ransome ? - Totally neutral on this. Jessica Priest N - I rarely touch comic book character articles. Excitebike (series) Y ? - Perfect, but not a stub. 8 March 2009 Baghdad police recruitment centre bombing N - I rarely edit political articles. Skitz N - I rarely touch comic book character articles. Nidovirales Y - Organism biology is closer to my interests. Marshmallow Alpha-Bits ? - Neutral and mystified. List of best-selling video game franchises Y ? - Perfect but not a stub.

Cleanup Abby Klein ? Homer Price ? Nintendo Y ? - Perfect but in no need of cleanup Merge List of Mario games by genre Y Gameplay of Pokémon Y Ric Lansing N - I never touch soap opera star articles. Add Sources Homophone Word Game ? Epic hero ? Princes of the Universe N - I rarely edit album articles. Wikify Holly Sutton-Scorpio N - I never touch soap opera star articles. Dental pellicle ? - A protein article. Atypical for me. Sharon Newman N - I never touch soap opera star articles. Expand Echoing Green N New Play Control! Y Tori Scott N ? - I rarely touch actor articles. Better placed as a stub.

• Y Good. I either edit the page or can see a rational nexus. (Total: 27% with 12% of these having problems. So 15% perfect ideal suggestions.)
• N Bad. This is not what I am interested in nor can I see any rational nexus. (Total: 35% with 3% of these having problems.)
• ? Huh? I might be interested but I have no idea where you got this from (Total: 38%)
• ? ? Problem. There is a problem here that should be looked at. (Total: 15% problematic suggestions - these have been overlapped into Good and Bad categories depending on whether they met my interests even though there was a misidentification of the problem.)

Cheers, -Thibbs (talk) 22:34, 3 June 2010 (UTC)

Excellent feedback, thanks for taking the time! I see there are some obvious shortcomings like lists and articles that aren't in the right categories, so we'll have to figure out what's been going on and why these show up. I'll also go through the other recommendations to see how we can improve performance. Nettrom (talk) 16:09, 5 June 2010 (UTC)

Feedback:Isabella and Lego Liker

My thought on the recommended articles with a rating:

1=not familiar with article 2=may not edit 3=may/may not edit 4=will place on watchlist 5=will find info when possible

Stubs Olivia Olson=4 The Great Jazz Piano of Phineas Newborn, Jr=1 Leonard Bernstein=1 Disney Channel Greece=2 Phineas Priesthood=1 Samantha Droke=1 Tony Oller=1 Barack Obama speech to joint session of Congress, February 2009=1 Phineas Bogg=1 Jake T. Austin=1 Noah Munck=1 Phineas Maimela=1 Fred Meijer=1 ElgooG=1 Meat pie=1 List of dog breeds=2 Kent Osborne=1 Maria Canals Barrera=4 Jim Bernstein=1

Cleanup Disney Channel (Central and Eastern Europe)=3 TFOU TV=1 SpongeBob's Atlantis SquarePantis=3 Merge Family of Barack Obama=1 Clare (Power Rangers)=1 Period 8 element=1 Add Sources The Replacements (TV series)=2 Tiya Sircar=1 Phineas Pett=1 Wikify 1950s=2 Martha Quintana=1 West Springfield Highlanders=1 Expand Gerard Way=1 Hannah Montana=1 Phineas Poe=1

Results on a one-off try: No 5's, 2 4's, 2 3's, 4 2's, and 26 1's. Final score: F(but nice to try)

Isabella and Lego Liker _{Whatcha doin'?} 02:03, 28 June 2010 (UTC)

Thanks for the feedback! Too bad we couldn't get you any 5's, and there's obviously too few good recommendations. I'll go through them in detail and see what we can do about this. Thanks again! Cheers, Nettrom (talk) 16:20, 30 June 2010 (UTC)

Feedback 1 from Forty two

Stubs Handa BhondaYYY KharijY Devdas (1979 film)Y Peacock Bay? GNUnifyN Symbiosis International SchoolN S. B. MujumdarN BillooY GhulamiY Sopan BaugN Devdas (1928 film)Y ParindaY Pavana RiverN Vidya YeravdekarN Center for Social Evolution and Symbiosis (Copenhagen University)NNN John Hackett (musician)N Symbiosis (film)N Pleiadians (band)N LohegaonN

Cleanup Bachelor of Business ManagementNN Culture of PuneN The Colored MuseumNNN Merge NationalismY SymbotNNN Australian and New Zealand meat pieNNN Add Sources Superboy (TV series)N Tony Hawk's Underground 2N Alex RodriguezNNN Wikify Pune Municipal CorporationN Association of Asia-Pacific Business SchoolsNNN Department of Management Studies, IIT DelhiNNN Expand European School of EconomicsNNN Georgia Institute of Technology College of ComputingNNN WarjeN

Legend--Forty two^{You talkin' to me?} 09:23, 30 June 2010 (UTC)

Thanks for the feedback! Nice to see we were able to get some good stub recommendations in, but it's apparent we've still got ways to go. As with the others, I'll go through these in details and see what we can do about that. Thanks again! Nettrom (talk) 16:23, 30 June 2010 (UTC)

Feedback from Arfn24

Stubs Agent (video game) Y DGamer Y Butte, Anaconda and Pacific Railway N Sega Racing Studio Y Follow N Rock Springs – Sweetwater County Airport Maybe - ? The Outsider (video game) Y Mickey's Pirate and Princess Party N Clap Hanz Y Celtuce N Gibran Museum N Syrian Television Um... N Coosawattee River N World Forge ? Saddam (name) N Romaine lettuce N Pauli Levokari N Baradostian culture N Disney.com ?

Cleanup Red Dead Redemption Y Bakugan (video game) ? When Do We Eat? (2006 film) N Merge Nintendo DSi Y Bedouin Soundclash N Drag racing ? Add Sources Grand Theft Auto (series) Y Cloudy with a Chance of Meatballs (video game) Y Amy Rose Y Wikify Garry's Mod Y House Lannister N Lettuce big vein virus ? Expand Max Payne 3 Y Rick Shapiro N Lo Que Son Las Cosas N

Here you go. Hope this can be useful for you :) I used the ? for ones that I was unsure of or were weird. Arfn24 (talk) 21:25, 1 July 2010 (UTC)

It's definitely useful, thanks for taking the time! Will check it out and see where we can improve things. Cheers, Nettrom (talk) 19:20, 9 July 2010 (UTC)

Repeated suggestions?

I thought you might want to take a look at this.

Stubs Time's Champion * Matinee Mouse * P.R.O.B.E. N Fred Quimby * The Little Whirlwind * Peter Bennett (producer) N Tyke (Tom and Jerry) Y Wartime (Doctor Who) * Adam Hargreaves Y Tom Cat Y Marky Ramone and the Intruders * Jerry Mouse Y Buddies Thicker Than Water * Ryan Butler * Donald's Quack Attack N Slap Happy Lion * Beth Willis (producer) ? Cat and Mouse in Partnership * Mouse Soup *

Cleanup Push-Button Kitty * Tom and Jerry Y Doctor Who merchandise Y Merge Reel N Zyurangers N Pointer trails * Add Sources Two Little Indians * Part Time Pal Y Mouse for Sale Y Wikify Lucky Star (novel) * Antony Root * The Toy Castle N Expand The Mad Doctor N Droopy, Master Detective N A Bug's Life Y

Stubs Pots and Pans (cartoon) N Slap Happy Lion * Ryan Butler * Advance and Be Mechanized Y Mouse Soup * Cat and Mouse in Partnership * Time's Champion * Ed Barge ? The Voidoids N Fred Quimby * Marky Ramone and the Intruders * Apple Wireless Mouse N Plane Dumb N Matinee Mouse * Cambio (band) N Wartime (Doctor Who) * The Little Whirlwind * Barbara Clegg N Buddies Thicker Than Water *

Cleanup Push-Button Kitty * Apple USB Mouse N IntelliPoint N Merge Montauk Highway N Pointer trails * Sea Devil (Doctor Who) Y Add Sources Tom's Photo Finish Y Safety Second Y Two Little Indians * Wikify Antony Root * Keith Chapman ? Lucky Star (novel) * Expand Tenth Doctor Y List of Torchwood creatures and aliens Y Bessie Cursons ?

I've redone my lists for this, and as you can see nearly half of my second lot of suggestions were repeated ones from my first (marked with a *), just to let you know. Oh, and to point out, Push-Button Kitty managed to be relisted in the exact same spot it was before. --Andromedabluesphere440 (talk) 17:15, 9 July 2010 (UTC)

You're absolutely right, I do want to have a look at this! Thanks so much, both for spotting these, and taking the time to figure out the matches! The problem of repeat suggestions is on our list of issues to look into. I've seen some users who see a couple of the same articles show up, but when it becomes half of them it's obviously something that should be taken care of. It's now on my todo list, we'll work on fixing that problem. Thanks again! Cheers, Nettrom (talk) 19:31, 9 July 2010 (UTC)

Feedback

Hi Nettrom,

One thing sticks out for me -- it would be nice if the suggestions reflected what articles I've worked on in the past. It's a little odd to be suggested articles that I've already worked on; but if that fact was reflected somehow in the suggestion, that would make more sense.

Might be worth looking into the "bacon" tool I mentioned -- I'm going to get you some more info about that shortly. (It's the tool that determines what articles you've worked on in common with another user.) -Pete (talk) 17:27, 19 July 2010 (UTC)

Thanks for your comments! We filter out articles that you've already edited, but as we currently do not gather a complete list of your contributions, there might be suggestions to articles that you edited in the past. Unless you do a large amount of patrolling they should not be recently edited articles though. Is this something you've noticed occur frequently?

With regards to your other comment, I understand it to mean you'd like to see something similar to Amazon.com's recommendations where they explain what products you bought or looked at that resulted in you being recommended Product X, is that right? I think that is on our list of ideas, but I'll make sure it gets there. Transparency in the recommendations is a useful option to have available, so we might make it possible for you to see why if upon request, but not necessarily explain all of them in the suggestions we post.

Thanks again! Cheers, Nettrom (talk) 18:35, 21 July 2010 (UTC)

Feedback from Pitke

Stubs MareYY Buddhist musicNN Derby (horse race)Y WarajiY TinkerNN Procedural lawNNN EnduranceNN AsaY Ultra JumpY Demographic profileNNN Highland cattleYYY RodeoY Bay Horse ! Horse-drawn boatY Shinto musicN Jive (dance)N Group SoundsNN StallionYY Tea ceremonyY

Cleanup Gypsy Vanner horseYYY Horse breakingYY CerebellumNN Merge Bunraku Barefoot horsesY CremelloYYY Add Sources YukataYYY Show hunterYY Callahan's Crosstime Saloon??? Wikify 1990s? SEX (boutique)??? Dinah Cancer??? Expand Law of the United KingdomNNN EquestrianismYY Connemara ponyYYY

!: I can see how the bot was confused... but this is a hamlet, nothing to do with horses. Pitke (talk) 19:01, 3 August 2010 (UTC)

Thanks for the feedback! I was happy to see that it gave you a significant number of pretty good recommendations. It's also obvious that we've still got some ways to go, for instance since we missed that Bay Horse has nothing to do with horses. Will keep that in mind as we start improving things, thanks again! Nettrom (talk) 20:34, 11 August 2010 (UTC)

Great job; I'm curious what the next list of suggestions will be.

You did such a great job of mind-reading with your first list of suggestions in early July that I thought now was the time to ask for a second list of suggestions, as I have branched out into editing new kinds of articles. SuggestBot seems to be a very useful tool, and I look forward to seeing what it suggests next time. Keep up the good work. -- WeijiBaikeBianji (talk) 23:41, 10 August 2010 (UTC)

Thanks for the positive feedback! The bot's working on getting your new list of suggestions as I'm typing this. Hope we're still doing a good job, and if not, feel free to let us know where we've missed. Cheers, Nettrom (talk) 20:36, 11 August 2010 (UTC)

Feedback from bobrayner

Hi,
Thanks - suggestbot gave me some very good suggestions!
My only qualm is that the suggestions seem to be very much based on topic rather than the type of work. For instance, if ArbitraryUser spends all day copyediting articles about films, I expect suggestbot would still give ArbitraryUser a recommendation of 19 stubs of films (maybe a few directors or actresses or whatever), 3 films that need cleanup, &c - rather than suggesting a dozen unrelated articles which need copyediting. However, I understand that it would be practically impossible to write a bot that understands what type of work people do, rather than which articles they edit :-)
Thanks again. bobrayner (talk) 09:48, 18 August 2010 (UTC)

Just wondering...

...when I might get another lot of suggestions? The last one was in July, and I miss it... Cheers, C628 (talk) 01:58, 7 September 2010 (UTC)

I welcome your wondering! It's been a little busy and confusing over the summer while we've been working on research and some improvements to the bot, so the regular recommendations ended up being forgotten. We're currently updating our data to make sure we don't recommend too much old stuff, and I'll make sure the suggestions get posted once it's done. Might take a day or two before we get going. Thanks again for the reminder! Nettrom (talk) 15:13, 7 September 2010 (UTC)

Yep, got it. Many thanks! C628 (talk) 11:35, 10 September 2010 (UTC)

Hello

Hi, I would like to discuss about the deletion of Retro Force with no evidence provided, I don't know how to wikify the page and I don't know the basics, but I clearly need your help to prevent this page from being deleted, because in the next six days, It's going to be deleted, can you please help me on preventing this page from being deleted? ~SC3nutter (talk) 18:42, 10 September 2010 (UTC)

I'll post a response to this on your user talk page as it's a discussion that doesn't really concern SuggestBot all that much. Nettrom (talk) 19:55, 10 September 2010 (UTC)

Another maintenance category

Is there a possibility of adding articles from Category:Orphaned articles to the mix. These generally are backwater and don't get a lot of eyes and we have a huge backlog.... They often have other problems as well and fixing the orphan issue would be rather easy for most topics. Any thoughts? Sadads (talk) 21:47, 6 October 2010 (UTC)

I'm always open to ideas regarding what categories we should grab articles from, and also how we show them to the user. From your description it sounds like orphaned articles would make a good addition (or swap for some other category). We're currently working on some enhancements to the bot, and I'll see if this can make it in together with those. Thanks for the idea! Cheers, Nettrom (talk) 16:06, 7 October 2010 (UTC)

I don't know why

SuggestBot added Speeedfins to my list of suggested articles. AFAIK I haven't edited any surfing-related articles. --Chemicalinterest (talk) 19:42, 12 October 2010 (UTC)

Hi, I had a look at SuggestBot's logs, and that particular recommendation comes from comparing your edits to those of other Wikipedia editors and then suggesting articles they have edited but you have not. It's quite similar to how "People who bought X, Y, Z also bought..."-recommendations on some on-line shopping sites work. As you see it can lead to suggestions that appear unrelated to what you have edited. In some cases they are articles you still can edit because the type of work the article needs is something you can do, while in other cases it might not work at all. Hope the explanation helps, let me know if you have any other questions. Cheers, Nettrom (talk) 20:55, 12 October 2010 (UTC)

Other than that, it works fine! Good work. --Chemicalinterest (talk) 23:05, 12 October 2010 (UTC)

By the way, if you would like to see how I used SuggestBot, look at my talk page. --Chemicalinterest (talk) 14:48, 16 October 2010 (UTC)

That's awesome, good work, and thanks for letting me know! Should be useful for us as we work to improve how SuggestBot works. Cheers, Nettrom (talk) 19:37, 18 October 2010 (UTC)

Thank you

Thank you for the suggestions, all interesting. A ntv (talk) 15:34, 16 October 2010 (UTC)

Great to hear that! I noticed you got a suggestion for an article that's been deleted, which shouldn't normally be happening, so we'll look into that. Cheers, Nettrom (talk) 19:39, 18 October 2010 (UTC)

A bit innacurate, but great

Well, I certainly didn't need the "Percy" suggestions, but great overall! Περσεύς|Talk to me 18:44, 18 November 2010 (UTC)

Bumping in here a little late; great to hear the suggestions were good overall! Sorry about all the Percy suggestions, I'll have a closer look at that issue and see what improvements we can make. Cheers, Nettrom (talk) 15:34, 12 January 2011 (UTC)

Very Interesting, Thank you

Thank you for your suggestions, I am interesting in editing the majority, if not all, of them. --George2001^hi 13:51, 30 November 2010 (UTC)

Great to hear that, thanks! Nettrom (talk) 15:38, 12 January 2011 (UTC)

Works good

Works pretty good! Did come up with a few completely random pages though. Sumsum2010·T·C 04:47, 18 December 2010 (UTC)

Happy to hear it's doing a fair job! Some suggestions tend to appear fairly random, or are by our measures related to articles you've edited but are outside your area of interest. It's of course something we want to see if we can improve as we work on the bot, although we'll probably not remove all randomness as serendipity might make sure our users find something interesting. Cheers, Nettrom (talk) 15:40, 12 January 2011 (UTC)

Interesting picks

Promising set of possibilities - good fun hopefully ahead! Jacobisq (talk) 04:25, 10 January 2011 (UTC)

Great to hear that, hope you've found some articles well worth editing by now! Cheers, Nettrom (talk) 15:45, 12 January 2011 (UTC)

Hello

According to these active bots SuggestBot seems to be the ONLY bot doing reccomendations. Great Going! I am experimenting with it will let you know feedback.Other than Norwegian version do you know any other reccomendation bots? GreenEdu (talk) 10:20, 20 January 2011 (UTC)

Hi, unfortunately I don't know of any other bots that do article recommendations in a way similar to what SuggestBot does. My knowledge of what goes on outside of English and Norwegian Wikipedias is very limited, so it might be that there's one running in another language, just that I've never heard about it.

And thanks for trying out the bot, I hope you found some interesting articles to edit! Please do let me know if you have any comments, as they're always helpful to us. Cheers, Nettrom (talk) 15:16, 20 January 2011 (UTC)

Is there?

Is there an option for "once a week" suggestion, instead of the "once fortnightly" or "monthly" option? -- Sp33dyphil ^{(Talk) (Contributions)(I love Wikipedia!)} 22:23, 28 January 2011 (UTC)

It's currently an undocumented feature. Off the top of my head there's a couple of reasons why we haven't advertised it yet: there's a high possibility of seeing many of the same articles suggested every week, and we're slightly concerned about system performance if many people sign up. We'll be working on the first problem shortly as that's a major concern regardless of how often you receive suggestions. Because we filter out articles people have edited they'll likely see the same uninteresting suggestions again and again. We're also planning on looking into the second issue. Feel free to sign up for weekly recommendations if you want to, just be aware of these problems until we've got them taken care of. Cheers, Nettrom (talk) 16:05, 1 February 2011 (UTC)

Thanks. Sp33dyphil ^{(Talk) (Contributions)(I love Wikipedia!)} 21:42, 1 February 2011 (UTC)

Sourced article

Hi. I suggest expansion & improving article Ola-abaza. It doe sin fact have sources which are independent and there are more. A notability criteria thing is not quite correct but has not been taken out --kellyrussell34 (talk) 09:52, 13 February 2011 (UTC)

SuggestBot and the Case of the Reel

Ignore uncatchy title. Anyhow, it's me again, I've found a very odd problem with SuggestBot. Take a look at this...

Merge

Reel

Zyurangers

Pointer trails

Merge

Montauk Highway

Pointer trails

Sea Devil (Doctor Who)

Merge

Reel

Pointer trails

Captain Becker

Merge

Cameron Ministry

Reel

Manned mission to Mars

Merge

Manned mission to Mars

Reel

Fly Me to the Moon

Merge

Reel

Merlin the Magic Mouse

Little Miss Naughty and the Good Fairy

Merge

Reel

Little Miss Naughty and the Good Fairy

Manned mission to Mars

Merge

Reel

Manned mission to Mars

Ice Age (film series)

These are taken from here, here and here. As you can see, Reel has appeared on 7 out of 8 of the suggestions I've been given from SuggestBot. Other articles also appear too, but not as much as Reel. I honestly don't know why Reel has kept appearing, can't think of any edits at all relating to it. So, yeah. --Andromedabluesphere440 (talk) 11:40, 24 February 2011 (UTC)

First of all, thanks for bringing this up! I dug through our logs and Reel is an article that is suggested because it is linked from one (or more) of the articles you've edited. As you see it always shows up in the "Merge" category, so there's also the issue of finding articles that can be suggested to you in that category, it might be that there aren't any really good ones (I noticed that Reel is fairly far down the list of potential suggestions).

The good news is that this shouldn't happen again. Previously we would only stop suggesting an article if you actually edited it, while now we also make sure that you aren't suggested the same articles over and over again (later we might further improve on this feature). So the next time SuggestBot stops by, it shouldn't recommend you Reel. If it does, please do let me know, as that would be a serious bug.

I hope this explains what's been going on, let me know if I've been unclear. Cheers, Nettrom (talk) 15:47, 24 February 2011 (UTC)

Thank you Nettrom for explaining this. I look forward to hopefully, not seeing Reel on my next list. --Andromedabluesphere440 (talk) 16:41, 24 February 2011 (UTC)

Thanks for your help, Nettrom. Reel didn't appear on my list this time. Hopefully, Reel won't appear next time either. Again thanks for your help Nettrom. --Andromedabluesphere440 (talk) 11:02, 7 March 2011 (UTC)

I'm happy to hear it's working the way it should, thanks for letting me know about it! Cheers, Nettrom (talk) 15:42, 8 March 2011 (UTC)

what about a menu where readers can set the algorithm for themselves?

For instance some people like looking at stubs, others not. Some might like random suggestions (many not). Let people select what they want. Also, I think it's more important to give people what they want, then to push some sort of agendas (wikifiy and all that). — Preceding unsigned comment added by TCO (talk • contribs) 18:38, March 7, 2011 (UTC)

Thanks for the feedback! I agree that more control over the recommendations would be preferable, and it's on our list of things that we'll be working on in the future. While I cannot say for certain, I think we will probably aim to balance user control with what the Wikipedia community thinks is important. For instance, there's currently more than 1.5 million stub articles in our database, so based on raw numbers it seems clear that the community thinks stubs need attention. Instead of pointing our users to random articles we (mostly) point them towards articles that are relevant based on the contributions the users have already done, so we're not asking them to look at random stubs, for instance.

With regards to random suggestions, our research has previously shown that they are less successful than our recommendation algorithms, and therefore we only fall back on random suggestions if our algorithms fail. Thanks again for your comments! Cheers, Nettrom (talk) 16:47, 8 March 2011 (UTC)

That they are looking to help at all should be good enough. After that, I would do everything you can to make the results pleasant and ("pull", NOT "push"). Only if users choose an option that "helps Wiki" would I slant it that way. This whole thing is volunteer based. Keep the sugar to medicine ratio very high.TCO (talk) 17:55, 8 March 2011 (UTC)

Research on software and other communities has shown that most users do not take the time to change options, so we'll work on making sure that SuggestBot has sensible default choices while we also try to allow those who wish to modify how the bot works. Cheers, Nettrom (talk) 15:58, 10 March 2011 (UTC)

Delete old suggestions?

I'm not sure if this has been brought up before, but is it possible for SuggestBot to replace old suggestions with the new ones? Right now, suggestions pile up on the talk page and we have to manually delete the old suggestions. Thanks. --RDavi404 (talk) 13:54, 10 March 2011 (UTC)

Don't think I've seen that idea suggested, thanks for bringing it up! I am not sure how easy it is to do, but will make sure we look into it. In the meantime I think there are a couple of options available:

1. Create a subpage in your userspace (e.g. User:Rdavi404/SuggestBot or something like that), add/move the config template there, and add that page to your watchlist. Then SuggestBot will start posting its suggestions to that page instead of your user talk page. You won't get the "You have new messages" notification, but your watchlist will show when you've received new suggestions.
2. Set MiszaBot up to archive your user talk page. I think it'll remove the suggestions automatically after a period of time, but I'm not completely sure as I've so far only used MiszaBot to auto-archive SuggestBot's talk page.

Thanks again for the idea! Cheers, Nettrom (talk) 16:08, 10 March 2011 (UTC)

How do you do this?

It's amazing. I was just studying Ancient India in Social Studies class at school, and it's amazing how much I know about that now. I have no idea how you do it SuggestBot, but you do it right.Mountain Girl 77 (talk) 00:34, 13 March 2011 (UTC)

Oops, sorry for not responding to this earlier! Happy to hear we're getting things right, appreciate that. As for how SuggestBot does it... parts of it is looking at what you've edited and finding similar articles. Apart from that I'm not totally sure, maybe he spends his evening chatting with Watson over a WikiPint? ;) Cheers, Nettrom (talk) 14:43, 28 March 2011 (UTC)

Innapropriate

Hi, I got an inappropriate page in my suggest bot message, surely these should be blocked? -- Thomas888b (Say Hi) 21:00, 15 March 2011 (UTC)

Unless you can define what "inappropriate" stands for in a manner that can be implemented in software, it is unlikely that the bot could filter out results other than through a blacklist. Looking at the bot's delivery on your talk page, I cannot even spot which entry you find inappropriate? —  HELLKNOWZ  ▎TALK 10:42, 16 March 2011 (UTC)

I'm sorry to hear you received suggestions you found inappropriate. Currently we do not filter the results in any way. As Hellknowz mentions, it would require a definition of "inappropriate" that is so crisp and clear that we can implement it in software. This kind of problem has been brought to our attention earlier by another user, but we have not yet found a good solution. So, it's on our list of ideas but I cannot say when we'll be able to fix it, nor exactly how a fix will be done. Again, sorry about the problem. Cheers, Nettrom (talk) 15:54, 16 March 2011 (UTC)

An idea: leveraging WikiProjects

Hi, when I was a new editor, I liked what SuggestBot suggested; by now I have enough ideas of where to help. Some of these ideas come from the quality-importance assessments of the particular project/projects that overlap most of my editing. So perhaps this might be an idea for the SuggestBot to use: If you don't find anything via your regular method, perhaps look at which WikiProjects the user's contributions have involved. Then find suggestions from their importance/quality assessments (perhaps emphasising low-quality with high-importance). Or perhaps you already take this into account. -- Nczempin (talk) 14:13, 25 March 2011 (UTC)

Thanks for the suggestion! We've had ideas about how to use WikiProjects for suggestions, but not yet made it into anything concrete. Your idea sounds interesting, and since we've already been thinking about WikiProjects I'll make sure we bring it with us. In some way or another I am sure SuggestBot will start using that kind of information in the not-too-distant future. Thanks again! Cheers, Nettrom (talk) 14:49, 28 March 2011 (UTC)

Kommissar Hjuler shall become deleted?

Hello, I am Kommissar Hjuler, I have an account here for sometime, but only now started to work by myself on the article someone started on me. Also made some articles on my wife Mama Baer, Dietmar Kirves, a colleague at the NO!art Movement and Barbara Rapp, another colleague.

I try to learn by doing, but anytime I open the artice, I am told, that I violated some rules.

I hope I did not, I am not sure, how I violated any rules, and now I am told, that the article can become deleted.

I need any help I can get,

Kommissar Hjuler [...@versanet.de +Germany+[...] +Germany+[...} —Preceding unsigned comment added by 141.91.129.7 (talk) 06:29, 8 April 2011 (UTC)

I don't think the SuggestBot development team can help you with that. -- Nczempin (talk) 07:16, 8 April 2011 (UTC)

Blank Bot Approval Page

Your bot approval page is blank... feeling neglected? Or disapproved? twilsonb (talk) 14:26, 12 April 2011 (UTC)

From what I've been able to piece together after the fact, SuggestBot first got approval to run when there was no formal bot approval process. After that there's been three other approvals. I've been concerned about the link not working, but I failed to notice that there's a BRFA field in the bot template. It has been fixed now, pointing to the latest one which was approved in late February. Cheers, Nettrom (talk) 14:59, 12 April 2011 (UTC)

Suggestions Request to my user page

Bear (talk) 07:25, 11 April 2011 (UTC)'Bear620 ^{Talk• Contributions} 03:25, 11 April 2011 (UTC)Bear (talk) 03:30, 11 April 2011 (UTC)

Suggestions are now posted to your user talk page :) Cheers, Nettrom (talk) 14:43, 14 April 2011 (UTC)

SuggestBot in mobile apps?

Hi, is anybody interested in thinking how SuggestBot could help Wikipedia mobile users? Of course the same suggestions for pages to edit could be given based on previous edits, but in addition to this there are other use cases relevant in mobile where SuggestBot could be a good base technology:

• Suggest pages to view based on your edited/viewed pages.
• Add your current location to the equation.
• Look for local cases like geolocation missing in this page, a picture is welcome to this page...

Please check the rationale of the fresh Wikipedia mobile app we are after. Hopefully SuggestBot can help there making happier mobile users and turning some of them into active participants.--Qgil (talk) 22:12, 14 April 2011 (UTC)

Hi, sorry for not responding to this sooner, mainly I've spent the time thinking about how SuggestBot can help in this kind of scenario. I think the ideas of adding geolocation information and pictures to articles are really good, particularly because they can be done in a way that's user friendly. Likewise, I'm not sure how useful general suggestions for articles to edit would be, as a mobile users probably does not want to type in a lot of Wiki markup from their phone (maybe if it's a tablet). I'll have another look at the app rationale on Meta and have a discussion with my colleagues. Thanks for bringing this to my attention, hopefully we can find some way of incorporation SuggestBot that'll be beneficial for everyone. Cheers, Nettrom (talk) 16:21, 25 April 2011 (UTC)

Thanks for your time and brain on this. :) You have a good point, in fact we have two scenarios: mobile users getting suggestions for "mobile contributions" (things you can do on the go) and mobile users just getting generic suggestions. Let's not dismiss this latter use case: many mobile users are also PC/laptop users and will welcome suggestions even if not all of them are perfectly suitable for mobile devices. One good thing about the mobile app we are defining here is that users will be implicitly encouraged to be all the time logged-in, even if it's only to browser pages. This could make SuggestBot a good fit even for those mobile users that haven't contributed much or at all before.--Qgil (talk) 18:08, 25 April 2011 (UTC)

a problem

I have set SuggestBot to leave suggestions on my talk page twice a month. However, I have gotten nothing from this bot for weeks. I got one post from this bot in March and none in April. Help, please? Crazymonkey1123 (Jacob) ^T/_S 00:20, 21 April 2011 (UTC)

As far as I can see from our logs and your talk page, you received a set of suggestions on March 25 (currently section 22 on your user talk page), and another set on April 10 (currently section 40).

If you don't want the suggestions to get lost in the rest of the traffic on your user talk page, you can move the config template to a subpage (e.g User:Crazymonkey1123/SuggestBot), and the bot will post the suggestions there instead (just add it to your watchlist, and you'll notice when the suggestions are posted).

Hope that helps, let me know if there's anything else. Cheers, Nettrom (talk) 01:52, 21 April 2011 (UTC)

Looks like the bot is posting on my talk page again. Thanks for the help. Crazymonkey1123 (Jacob) ^T/_S 15:20, 24 April 2011 (UTC)

I have the code:

{{bots|allow=SuggestBot}} {{User:SuggestBot/config |frequency = twice a month }}

on the subpage User talk:Crazymonkey1123/Suggestions from SuggestBot of my main talk page. Will I receive suggestions there? Am I doing anything wrong? Crazymonkey1123 (Jacob) ^T/_S 21:07, 24 April 2011 (UTC)

Your setup is correct, and SuggestBot will start posting the recommendations there. The next one is due May 8 or thereabouts, so you won't see it until then. Let me know if there's anything else. Cheers, Nettrom (talk) 16:14, 25 April 2011 (UTC)

I hate you

QUIT SUGGESTING THINGS TO ME!

Does not compute, I never suggested anything to you. SuggestBot (talk) 16:57, 27 April 2011 (UTC)

User talk: Lihaas

i havent received by suggestion in a while..Lihaas (talk) 18:56, 12 May 2011 (UTC)

As far as I can see from your user talk page, the last set of suggestions (section 80) were posted May 2, ten days ago. I also found three other sets, posted approximately a month apart (February, March, and April), which your configuration has set the bot to do, so you'll receive the next post in early June.

I see there's quite a lot of traffic on your user talk page, so you might want to consider asking SuggestBot to post to a sub-page in your userspace (we discussed that on my user talk page in February).

Hope that helps solve the problem, let me know if you still have any questions. Cheers, Nettrom (talk) 19:41, 12 May 2011 (UTC)

okay sorry, i soaw someone else had ti recently and i didnt so i queried it. no probs ;)Lihaas (talk) 00:15, 13 May 2011 (UTC)

Can I have more suggestions?

Hi, I'm on the list (i think), and I got suggestions, but could I get more? Like once a month? That would be great, and if I could get the suggestions on a range of subjects, that would be AWESOME. Thank you! Cheers, Mountain Girl 77 (talk) 19:32, 12 May 2011 (UTC)

SuggestBot has stopped by and posted some new suggestions to your user talk page. Couldn't find you in our list of regular users, so if you want to get suggestions montly, just add our template or userbox. Let me know if you have any questions, I'd be happy to help you out! Cheers, Nettrom (talk) 21:27, 12 May 2011 (UTC)

Feedback from Philosopher

Not sure if you're still collecting these, but...

From the last suggestions you gave me (and I love the feature, thanks!):

Y = Good suggestion;   = I already edit this article occasionally or already intended to begin editing this article occasionally; N = Not interested; ? = Seems random (I also mark these with Y or N)

The  s and Ns seem to be largely state legislators (I'm already editing the ones from Iowa and not interested in the others). Several of the ?s are on articles related to homosexuality, either pro- or anti-; I have no idea where they came from. --Philosopher ^{Let us reason together.} 00:10, 13 May 2011 (UTC)

Stubs YIowa Democratic Party  Pat Grassley NHinton Mitchem NDiane B. Snelling N?Scott Lively NSam Granato  Kevin McCarthy (Iowa politician) NWilliam E. Peterson NWisconsin's 7th congressional district  Jack Drake (Iowa politician)  Scott Raecker NEdd Houck  Tyler Olson N?Transsexual News Telegraph NBruce Patterson (Michigan politician) NTom Graves (politician) Y?Senate Taiwan Caucus  Jo Oldson  Betty De Boef

Cleanup NJeff Kruse NVincent Fort N?The Cleveland Show Merge NKosher foods N?Bari people (Sudan) N?Soulforce (organization) Add Sources YDavid Loebsack YBourke B. Hickenlooper NUnited States Senate election in Kentucky, 2010 Wikify N?Chipmunk (rapper) N?BYU Creamery YA Preface to Paradise Lost Expand NKeota, Iowa NNathaniel J. McFadden NOmdurman Islamic University

This was great feedback, thanks for taking the time to do this! I do find some interesting trends in these, but I'll need to also have a look at some of the other feedback we've gotten to be more confident about where to work on improving the recommendations. As you mention, some of the articles are ones you already edit occasionally (e.g., Kevin McCarthy (Iowa politician)), which I think we fail to remove because we don't look that far back in your contribution history. Over the summer I plan to spend some time trying to improve the recommendations both by doing a better job of picking out the relevant contributions, removing already edited articles, and make sure our recommenders work well. I'm happy to see we were able to send you some good suggestions, and of course also that you find SuggestBot useful. Thanks again for the feedback! Cheers, Nettrom (talk) 18:34, 13 May 2011 (UTC)

Suggestions for WikiProjects

Hey,

A random idea (not necessarily very well thought through)… Could SuggestBot be used to find articles that are likely to be within the scope of a particular WikiProject?
As an example, WikiProject Shakespeare covers everything from articles on William Shakespeare and his family (e.g. Judith Quiney), his plays (e.g. Hamlet, Romeo and Juliet, etc.), his biographers and important scholars (e.g. Edmond Malone), adaptations of his plays (e.g. Romeo + Juliet), related concepts (e.g. Shakespeare garden, Shakespeare (programming language)), important manuscript collections (e.g. MS Egerton 1994), and so forth. At the same time there are just an insane number of articles that link to or simply name Shakespeare or one of the plays, so just searching for those key words and going through them is not practical. So the idea is something like going through all articles tagged by a WikiProject and finding articles that are similar but which do not have the WikiProject tag. Possible? Practical? --Xover (talk) 11:35, 8 May 2011 (UTC)

That would be really interesting, because it might help many projects ferret out articles that haven't been properly categorized or labeled, Sadads (talk) 11:46, 8 May 2011 (UTC)

If you're suggesting that this could be added to the ways SuggestBot recommends articles, I see a few possible pitfalls. It could create additional false-positives for people who edit articles in broad WikiProjects, such as geographically-based Projects, or really large ones, such as WikiProject Biography. In the first case, there are many disparate topics in the Project, and in the second case, virtually everyone on Wikipedia has edited articles belonging to that WikiProject at some point. Just using it for WikiProject Shakespeare may work, but most projects' subjects aren't that focused. --Philosopher ^{Let us reason together.} 12:51, 11 May 2011 (UTC)

Well of course the massive projects like Biography or Military History wouldn't benefit, but their task forces might and any project that sits in the 3000-4000 article or less range would certainly benefit, and are really the projects that would want to find more content, Sadads (talk) 13:04, 11 May 2011 (UTC)

Oh? Maybe I misunderstood you. I thought you were suggesting that the individual recommendations SuggestBot already gives be influenced by the banners on the pages they edit. Which is where you would run into serious problems with false positives. If you're suggesting that SuggestBot give group recommendations, that's a whole different story, though as a non-programmer, I'd be interested to know whether that task is even possible for a bot that uses an individual's contributions as the primary basis for recommendations (as there are no individual contributions for a WikiProject). --Philosopher ^{Let us reason together.} 17:45, 11 May 2011 (UTC)

Let me see if I can respond to everyone at the same time without creating confusion. There are several great ideas here, and they echo some of the things we have in our notebook of ideas already. Philosopher is correct that group suggestions would require us to rework our system, as individual contributions do not make sense. Instead we would probably try to suggest articles that appear to be within scope of a given WikiProject, but haven't been claimed by them, and we already have some ideas on how to find those (one is used in The Effects of Diversity on Group Productivity and Member Withdrawal in Online Volunteer Groups). We could perhaps also attempt to draw attention to specific articles within scope of a WikiProject, but I am not sure if we are able to improve on the project's own importance measure.

When suggesting articles to an individual user we could possibly suggest they join a specific WikiProject based on their interest (and then strive to handle the really large projects like Biography intelligently, it might be more useful for a user to join a smaller project), or suggesting articles within scope of one project (e.g., "You seem to edit articles within WikiProject Shakespeare, here are some articles important to this project that needs work but have not been edited recently…").

Since the topic of WikiProjects has been mentioned a couple of times already this year, and there are a few different ideas floating around for what we can do, I'll bring it up in our research group meeting and see if it gains some traction. Thanks again for bringing up good ideas, let me know if there are any questions! Cheers, Nettrom (talk) 19:09, 13 May 2011 (UTC)

Removal

Remove me please as currently inactive. I don't use the template method and seem no instructions on how to remove yourself if you signed up long ago. Regards, SunCreator ^(talk) 12:34, 15 May 2011 (UTC)

I've removed the template from your user talk page (it was added by me on February 18 (here's the diff), we added everyone who had signed up earlier), and you should no longer receive any updates. Let me know if there's anything else. Cheers, Nettrom (talk) 18:24, 15 May 2011 (UTC)

My bad. I couldn't find it anywhere on my pages. I though it would of had a category, but seems not. Anyway, thanks for the removal, it's an excellent tool and I will add it again if I become active. Regards, SunCreator ^(talk) 19:30, 15 May 2011 (UTC)

No problem, happy I could help you out. Also great to hear you found SuggestBot useful, hope to see you again some time in the future! Cheers, Nettrom (talk) 20:43, 16 May 2011 (UTC)

Feedback 5.2011

Hi. I've had two sets of recommendations now from SuggestBot, both stimulating. Only improvement I could suggest (which may not be technically possible) is to screen out repeat suggestions from one list to later ones, on the grounds that those articles left untackled the first time will probably be left on the side of the plate second time round as well, if you see what I mean?! Anyway, many thanks for much interesting stimulation, LOL, Jacobisq (talk) 11:10, 17 May 2011 (UTC)

Hi, thanks for the positive comments, I'm happy to hear you like the suggestions! We did change SuggestBot a few months ago to do exactly what you ask for, so the next time you get suggestions there should be no repeats. Cheers, Nettrom (talk) 14:30, 17 May 2011 (UTC)

Getting SuggestBot on local mediawiki

Hi, I have recently installed a local intranet based mediawiki instance for my team (at my workplace). Can I get suggestbot working for that wiki ?? Is this piece of code open source ? in public domain ?? --Raghav 15:15, 29 May 2011 (UTC) — Preceding unsigned comment added by S raghu20 (talk • contribs) 15:15, 29 May 2011 (UTC)

Looks like source code is "currently not" available: Wikipedia:Bots/Requests for approval/SuggestBot 4. I was also hoping for source code, so I could see the specifics of how it's working and possibly help improve it. —Mu Mind (talk) 15:07, 5 June 2011 (UTC)

First of all, sorry for not responding to Raghav's question earlier; had a busy week and simply forgot about it, partly because I wanted to discuss this with a couple of my colleagues and never got around to doing so. Currently we haven't published the source code anywhere. It's fairly specific to Wikipedia, so I am unsure whether it'll easily work with a local MediaWiki instance. Also, it's rather installation-unfriendly because it's only used on a couple of Wikipedias. Once I've spoken to my colleagues I'll get back with a clearer response. Cheers, Nettrom (talk) 15:56, 5 June 2011 (UTC)

Feedback from EdwardLane

First column - can I see why it was suggested Y or N or ?
Second column - am I interested Y or N or ?

So

YY I can see why it was suggested and I have some interest in editting these articles
YN I can see why it was suggested but I'm not interested
NY I can't see why it was suggested but I am interested !!
NN I can't see why it was suggested at all and I'm not interested but nice try

Some places I have ? meaning I might be able to guess why it was suggested but it's a bit tenous, or I might have vague interest in editing the article.

I think the section title expand is making too much of an assumption of knowledge. I know I've only done a few hundred edits so the history is probably not enough to make a better judgement. But yes very good all in all - especially finding Eunicycle which was an article I'd tried looking for but not been able to find due to not knowing what it was called, just knowing what it was. Thanks very much :)

Stubs Y?Water basketball ?NSwansea East (UK Parliament constituency) Y?Beach basketball YYHand walkingY NNÍF Fuglafjørður Y?Hybrid sportY YNIntercrosse YNUnderwater ice hockey YNGuadalajara Cartel Y?Cardiff International Sports Stadium YNGonzaga, Minas Gerais YNSilvio Valenti Gonzaga YYEunicycleY NNSaterland ?NLink Top ?NNational Nature Reserves in Somerset ??Footbag net YNHouse of Gonzaga ??Ki-o-rahiY

Cleanup Y?Deepwater Horizon oil spill Y?Evangelicalism Y?List of breakdance moves Merge Y?Religious behaviour YYComputus YNSperm motility Add Sources NYKho khoY Y?1990 (breakdance move) YNFerdinando Gonzaga, Duke of Mantua Wikify Y?Freedom of religion in Panama YNRidda wars YNEnglish-language editions of The Hobbit Expand ?NHeron Carvic ??Golden Crescent ??Red Dead Redemption

EdwardLane (talk) 11:26, 17 May 2011 (UTC)

Thanks for your feedback, I greatly appreciate it! I'm of course also happy to see that we were able to give you some great suggestions, and I love that we found Eunicycle for you. Perusing our log files I do see some interesting things that I'll see if we can use to improve the quality of the suggestions.

Regarding the Expand category, I'm wondering if the category as a whole is problematic, or is it these specific suggestions that don't work? Would be great if you could explain more as that could give us ideas on how to improve it.

Thanks again for the feedback! Cheers, Nettrom (talk) 15:39, 17 May 2011 (UTC)

Hi, I just spotted (and fixed) one typo in my feedback - the ?NGonzaga, Minas Gerais should have been YNGonzaga, Minas Gerais - obviously due to my acting on the gonzaga boardgame and disambiguation page.

I think the existance of a category called 'expand' is probably problematic in and of itself, searching for the 'this section needs expansion' in articles 'extremely' well matched articles seems fine, bBut if the 'wellMatchedness' is any less good then displaying something else (perhaps another stub?) would seem more useful. I don't know how your code calculates the 'wellMatchedness' so it's difficult to feedback further on that. But articles that are not stubs are going to be difficult to expand for most editors.

Other sections that could perhaps be added and might be possible/interesting?

1. Potluck needs attention. (find an article that doesn't match any of your edits)
2. articles on 'Blah' suggested based on userboxes - something like "if exist User:*/Userbox*/Blah"
3. articles needing attention that are 'similar' to the first article(s) you editted
4. articles needing attention that are 'similar' to the most recent article(s?) you editted
5. articles to create - would require finding any matches in here Wikipedia:Requested_articles

or perhaps if there are lots of potential stubs that could be shown - weighting the ones that are shown toward the matches for articles you most recently editted ?

think that's about all on feedback for the moment, hope it's helpfull EdwardLane (talk) 16:40, 17 May 2011 (UTC)

Thanks for fixing the typo, and thanks for the comments! I find your thoughts about the Expand category compared to Stub very interesting. On first thought I'd think that expanding for instance a section of an article might not be very difficult, as the article is then fairly developed and finding useful sources of information could be easy. While I don't have much data to go on, expanding a stub seems like somewhat difficult work as information on the subject might not be easily retrieved. Perhaps you have a different experience?

Nettrom (talk) 16:12, 19 May 2011 (UTC)

I think my different experience on the expand section is probably because I'm a jack of all trades master of none. Also I've grown quite quick at finding articles via google searches - so if some content exists but needs expansion it's usually difficult for me to find a lot of extra content, while if there is almost no content and a very basic amount of info needs to be added and then left for someone else to expand I've found thats easier for me :)

I've also added a few extra ticks at the end of my feedback bits above - articles I've done edits on or near because of suggestbot - not sure why but near those articles were more things I could contribute to, so perhaps 'density of cleanup templates/stubs nearby' or 'articles containing redlinks' might be good to factor in when looking to find areas that need improvement. EdwardLane (talk) 11:12, 28 May 2011 (UTC)

With regards to your suggestions of sections: the whole list of suggestions is based on similarity to the most recent articles you've edited, but we use article tags to identify those who are marked as in need of attention. From our research we know that the pot luck method (random suggestions) are less likely to be interesting, so we only resort to that if nothing else works. I'll make a note of the userbox idea, together with other indicators of interests on the user page, it might make us able to improve suggestions. We're also working on suggesting articles to create, but I currently do not know when we'll be able to deliver that. :) Cheers, Nettrom (talk) 16:12, 19 May 2011 (UTC)

I had another idea - perhaps suggesting Orphan articles that have some similarity to the previous 500 edits - there is a list here Category:Orphaned_articles_from_May_2011 and some of the articles look to have a reasonable number of information bits about them - but just don't have articles pointing to them. Someone who has been editing related articles might be able to point a few of them at the 'SuggestBot suggests you might be able to find this orphan page a home' page? EdwardLane (talk) 11:12, 28 May 2011 (UTC)

another section that could be suggested [[Category:Articles_without_infoboxes]] EdwardLane (talk) 11:00, 19 June 2011 (UTC)

Thanks for the suggestions! I think both of the categories you mention are useful suggestions, and we've been asked about orphan articles before. Those, as well as perhaps looking at some newer articles, are things that we have on our list. Right now I do not know when changes will be made, but at some point we'll probably give the whole thing a bit of an overhaul, perhaps with some additional control for our users, so that people can find more work that is interesting to them, rather than simply using the general categories we have now. Thanks again! Cheers, Nettrom (talk) 18:38, 30 June 2011 (UTC)

Maybe the bot should make predictions for articles, not users

I didn't expect the bot to know how I think. I recognize some of the content as simply related to a few of the articles I've edited. I was fairly surprised at the lack of molecular biology-related content, since I've edited a few of those, but it's probably difficult to run the links from those. And where on Earth did the Korean calligraphy come from?

Anyway, my feeling at looking at the list is that for me, it's a mishmash of the results I'd expect for individual articles. So why aren't the individual articles what the bot makes predictions for? I think a small box in the article talk page, "See related content..." might be quite useful. It doesn't require the same kind of signup, and humans could be free to dump in their own suggestions. Wnt (talk) 23:30, 15 June 2011 (UTC)

You're right that the bot (currently) predicts based on the articles you've edited, using various methods of calculating similarity and filtering out articles that aren't tagged as needing work. With regards to the Korean calligraphy article, it's one of the articles that other editors who are similar to you have edited :)

I think your idea about "See related content..." is cool, it could perhaps be done as a gadget that adds to the dropdown menu (where "Move" currently is)? Since most of the related content is easily found through the links from a given article to others, I'd have to do some careful thinking on how to improve that (e.g. "Find articles similar to this one that are in need of improvements"). It would enable users to more easily get suggestions for articles to work on, which you also mention, and that's one of the concerns we have about SuggestBot, that it would be great to make it easier for people to use it. Thanks again! Cheers, Nettrom (talk) 18:51, 30 June 2011 (UTC)

Feedback

You make six types of suggestions, and stubs far outnumber the other five. I understand that stubs may require the most work, but some may prefer more suggestions in the other five. Amy Z 21:10, 29 June 2011 (UTC) — Preceding unsigned comment added by Amyxz (talk • contribs)

Hi, and thanks for your feedback! You bring up a good point. I think our main rationale behind suggesting so many stubs is not that they require more work, but that they far outnumber any other category of work needed on Wikipedia (by a factor of 3-100). That doesn't mean that it cannot change, for instance we recently launched SuggestBot on Swedish Wikipedia and altered the categories and number of articles after we received some feedback from users on how to improve them. I have some ideas on how to look into this, and will see what I can do. We do plan to make some cool changes in the not-too-distant future, and that might also turn into changes to the categories we suggest articles from, and how we suggest them. Thanks again for getting in touch! Cheers, Nettrom (talk) 18:59, 30 June 2011 (UTC)

Suggestion - User saves a template that SuggestBot reads

I don't know if it is possible for SuggestBot to read the template saved on the user talk page, but the template or some embedded substitution would permit the user to say how many entries in each category they would like.

For feedback on my suggestions, I think it got pretty close overall. Some of the articles I already have modified. I was surprised to see Access time since I edited it yesterday (and got the suggestion this morning). Hardcard I have edited in the past. Cool application. Nice work. § Music Sorter § (talk) 02:45, 2 July 2011 (UTC)

Thanks for taking the time to comment, and thanks also for the positive feedback! With regards to reading templates, your suggestion is along the lines of the ideas we have been thinking. It's fairly straightforward for those users who sign up to get suggestions regularly as that requires adding a template, we can have some template options that influence what categories they get work from, and the number of articles in those categories. We're currently busy working on some other improvements, so I can't say for certain when we'll work on this, but we plan to get there, somehow. Thanks again! Cheers, Nettrom (talk) 16:39, 6 July 2011 (UTC)

My suggestions

• "Stubs" part is way to big
• Is it possible to check edits on home wiki? On enwiki I mainly add interwiki etc
• Bulwersator (talk) 10:21, 10 July 2011 (UTC)

Thanks for the comments! With regards to the stub section, that might change in the not-too-distant future as we implement various changes. I understand the interwiki-related issue. Right now we have no feature to check other wikis, but that might change too. It's all a question about how we go about gathering data about a user's interests, and looking at their contributions on other wikis might become an important part of that. Thanks again for the comments! Cheers, Nettrom (talk) 14:51, 11 July 2011 (UTC)

Research

Hi Suggestbot,

I am part of a research team at Carnegie Mellon University that is looking to conduct an experiment on Wikipedia. We are interested in a technique to efficiently give one or two suggestions to a user. We would greatly appreciate any advice you have.

Our research is going to examine the effects of different kinds of messages on Wikipedia, one being the directive message that gives a user a task to perform.

We will be sending our proposal to Wikipedia's Research Board shortly, and when it is up, I can absolutely send you a link to it.

If you have any questions, I would be happy to answer them. You can also ask Dr. Robert Kraut.

Thank you for your time, Amy Amy Z 19:43, 12 July 2011 (UTC) — Preceding unsigned comment added by Amyxz (talk • contribs)

Hi Amy! I'd be happy to help you out. Maybe it's easier if we take this conversation off-wiki? Please drop me a line on morten at cs dot umn dot edu. Cheers, Nettrom (talk) 21:15, 13 July 2011 (UTC)

TV Programs

I edit a number of UK related TV serials articles, suggestbot suggests loads of US serials for me yo edit, a subject matter I know very little about, and have no interest in.

I suggest that suggest bot (if its even possible) looks at the geographic location of the prepodonderance of an articles editors and vs the suggestions it makes as a weighting factor in wether it suggests something or not. Jasonfward (talk) 19:33, 23 July 2011 (UTC)

Hi! Thanks for your feedback. I can understand the problem here, while it's great to find that we're able to find other TV series to suggest to you, it doesn't help when they're from the wrong continent. We do have the idea of looking at geographic information on our list of things we want to look at. I'll make sure that your comments are added to that, as getting this kind of thing right would probably improve SuggestBot a lot. Thanks again! Cheers, Nettrom (talk) 01:35, 26 July 2011 (UTC)

Watchlists

I just wondered if Suggest bot can view a user's watchlist - and whilst it's obviously not useful to suggest those articles - perhaps those articles (apart from user pages) have categories etc that could be used to refine the 'well matchness' of other suggestions. EdwardLane (talk) 19:11, 4 September 2011 (UTC)

SuggestBot cannot currently view a user's watchlist, it's one of the ideas we have on our list of things to do because, as you mention, they can be a good indicator of a user's interests and therefore provide us with better matching. Cheers, Nettrom (talk) 23:04, 5 September 2011 (UTC)

Feedback from Edward Lane

First column - can I see why it was suggested Y or N or ?
Second column - am I interested Y or N or ?

So

YY I can see why it was suggested and I have some interest in editting these articles
YN I can see why it was suggested but I'm not interested
NY I can't see why it was suggested but I am interested !!
NN I can't see why it was suggested at all and I'm not interested but nice try

Some places I have ? meaning I might be able to guess why it was suggested but it's a bit tenous, or I might have vague interest in editing the article.

Stubs YYMultiple trapeze ??Ethiopian xeric grasslands and shrublands ?NGreat Longstone for Ashford railway station YYTriple trapeze Y?Great Sheffield Flood YNZula YNWelcome to the Real World (Trapeze album) ?NJabot (neckwear) Y?Gwynfynydd Gold Mine ??Llangennith NNKaterina Stikoudi YYVolcanic field YNMel Galley N?Spinneret (polymers) YNSouth Sudan Liberation Movement NNRiver Rother, East Sussex YYCoffin bone NNRiver Rother, West Sussex Y?Lake Abijatta

Cleanup ?NMoney clip YNArok Thon Arok NNOsama bin Laden Merge N?Leather glove N?Effects of World War II YYStatic trapeze Add Sources Y?False-color YYTrapeze (sailing) NNMarlin Model 60 Wikify NNMaine Cottage NNAcademic degree N?Party hat Expand YNList of numbered highways in South Carolina Y?Index of Eritrea-related articles YNList of Nebraska numbered highways

so that's 18 out of 30 that are clearly using 'good' logic, I'd be interested to see why I've had the NN ones suggested. EdwardLane (talk) 09:25, 6 September 2011 (UTC)

Thanks for the feedback! I'm happy to hear that the majority of the suggestions made sense to you, and I'm of course also happy to see that there were a fair amount of very good suggestions.

I checked the suggestions you marked NN, and they fall into two categories. The first and largest is articles found through editors who have edited the same articles as you. River Rother, East Sussex, River Rother, West Sussex, Osama bin Laden, Maine Cottage, and Academic degree were all found that way. Except for Osama Bin Laden, which ranked 2nd, they were all fairly low-ranked, and that might come from the fact that we only suggest articles that are tagged as in need of work. The other category is articles linked from the ones you edited, and the suggestions rank fairly high on the list of articles found, Katerina Stikoudi was 6th, and Marlin Model 60 was 9th. I have seen feedback from some users where similarly ranked articles were really good suggestions, so it might just be that the edits we used this time around didn't find the most relevant ones that way.

Hope this explanation was helpful. Again, thanks so much for the feedback, and let me know if something was confusing! Cheers, Nettrom (talk) 15:38, 6 September 2011 (UTC)

Well at least that explains why they appeared - the N? ones are probably of a similar ilk. Muchos gracias for the suggestions I'll come back in a while and see if I can work out how often I ought to put myself down for regular suggestions EdwardLane (talk) 17:58, 6 September 2011 (UTC)

Where are my suggestions?

I asked for them over 6 weeks ago! Anyone else with this problem? onyx³²¹ 12:01, 7 September 2011 (UTC)

I'm sorry about that! I checked our logs, the history of this talk page, and also went looking for a userbox or template on your user/talk page, and couldn't find anything. I've told SuggestBot it should go post a set of suggestions to your user talk page, you should get them ASAP. If you like them and want to get suggestions regular intervals you can find instructions on how to do that here. Let me know if you have any questions and I'll help you out! Cheers, Nettrom (talk) 15:10, 7 September 2011 (UTC)

Much better, but most of these I couldn't see why they were suggested. Cool idea though. onyx³²¹ 12:27, 8 September 2011 (UTC)

I had a look at your contributions as of the time the suggestions were generated, and from what I could see there weren't a lot of articles, and therefore I failed to see a strong common theme among them. The suggestions tend to improve with the number of articles edited, particularly if they have a common theme among them (e.g. RPGs or computer games), which is a typical issue with recommender systems (for instance you'll see the same with Amazon.com's system, it gets better at predicting what is relevant to you as you search for or buy more stuff). While I understand if you choose not to, I'd encourage you to stop by again a few weeks from now when you've contributed to more articles, and see if it improves. Thanks again for trying it out! Cheers, Nettrom (talk) 15:53, 13 September 2011 (UTC)

Sponge toffee

How did you know i would like this article? i didnt even know it was called this. i ate some just 2 days ago, and have been periodically fascinated with it. seriously, what algorithm chose this, cause i dont edit many articles like this, except Annabelle Candy Company, which doesnt make this product.Mercurywoodrose (talk) 02:44, 17 September 2011 (UTC)

Glad to hear you liked that suggestion! I had a look at our logs, and Sponge toffee was found by looking at editors who are similar to you (meaning they edited some of the same articles as you). This particular article was also fairly low-ranked, which corresponds to you mentioning you don't edit many articles like this (and the fact that we only suggest articles tagged as needing work by the Wikipedia community). Let me know if this was confusing, and I'll try to explain it better. And again, happy to hear we found it, thanks! Cheers, Nettrom (talk) 19:40, 18 September 2011 (UTC)

Feedback from Jinnai

• First column - can I see why it was suggested Y or N or ?
• Second collum - interested Y or N or ?
• ? for the first collum means I could see possibly, but not sure why it took pecidence over others. For the second collum, it means I'd be unsure whether I'd like to work on it, possibly due to some specialized knowledge requirement.
• For stubs a N* may mean I see no reason to bother as I don't believe it can be improved. This is if the first collum has a ? or Y as I have enough knowledge to understand that.

Stubs YN*Mr. Ho YNToccio the Angel YYAlbert Odyssey Y?Heartbeat (company) ?NChameleon (manga) ??Sakura Asagi ?NLittle Master ?N*Chiaki Kawamata YNAlexis Tipton YN*Dragon Quest: Shōnen Yangus to Fushigi no Dungeon ?N*Go! Comi NNTake a Break! Pinball Y?Cross Gate NNPro Wrestling Superstar Retsuden NNSonobe, Kyoto ?NPower-Up: How Japanese Video Games Gave the World an Extra Life NNTokyo Metropolitan Symphony Orchestra ?N*Eiichi Fukui ?NOnimusha Tactics

Cleanup YYDragon Warrior VII N?King's Quest IV: The Perils of Rosella NNOsama bin Laden Merge NNReactions to the death of Osama bin Laden ?NSengoku Strays YNFinal Fantasy character jobs Add Sources YNDragon Ball Y?Yuji Horii ??Chibi Vampire Wikify NNMidicronica ??List of Sengoku Basara episodes YNJump Ultimate Stars Expand Y?Nabari no Ou ?YRosario + Vampire ??To Love-Ru

As you can see most of the topics I can see to some extent why they were listed, just not why those particular ones. There are a few oddballs out there though. Specifically, Sonobe, Kyoto stands out as its a small town in a country I don't live in. True I do edit Japanese articles, but not really about such small geographical locations where one would have to know the area.∞陣内Jinnai 05:58, 27 September 2011 (UTC)

Thanks for taking the time to do this, much appreciated! Great of you to also let us know which stubs are probably not going to improve much, that's useful to know too! Regarding how SuggestBot selects articles, we currently only suggest articles that are tagged as needing work, so the suggestions are a combination of being related to what you've already edited and being in need of improvements. In some cases, that leads to somewhat confusing results.

You're absolutely right about Sonobe, Kyoto, and that is the kind of detail that we currently cannot catch. It's a good point that I will make sure to bring along when we work in improving the bot and see if we can make it understand those kind of issues somewhat better. And you're also right about the oddballs, I see that you also get a couple of Osama bin Laden-related articles, that might also be something we can figure out how to improve. Thanks again for your feedback! Cheers, Nettrom (talk) 15:11, 6 October 2011 (UTC)

please

Suggestbot, can you please give me suggestions? I have had your template on my page for a month or two now, but you continue to ignore me. Please, I just want some suggestions. Shakinglord:Kudos, Mailbox, ??? 00:04, 8 October 2011 (UTC)

Hi, sorry to hear that! I had a look at your userpage and noticed that you had the code for the userbox there, instead of using it as a template. That doesn't work with the way SuggestBot finds the people who are interested, so I went ahead and corrected it, and you should then get some suggestions when it runs the next time (in about ten hours). If you want the suggestions more often than once a month, you can change the frequency parameter in the template, e.g. "frequency=twice a month" will make the bot send them to you every other week, and so on.

Hope that helps! Let me know if you have any questions and I'll do my best to help you out, and again, sorry to hear you've been waiting. Cheers, Nettrom (talk) 01:45, 8 October 2011 (UTC)

Feedback - includes feedback on stars

3 columns of Y or N or ?

1. has the article been 'well' picked (do I know why?)
2. am I interested in it
3. are the star ratings correct (I like them but I'm not sure if they are accurate yet)

if the Star rating is out by 1 star (from the well picked even if I'm not interested) I've given it a ? if it's spot on it gets a Y if it's two out it gets a N - I'm assuming it must give 1, 2 or 3 stars (so can't pick zero stars)

You explained previously for Osama that someone else edited other articles I did then went on to edit him but I've still marked that as 'not well picked' for consistancy.

The ones that probably need the star rating looked at most are

• those that have Y something N (which are really not rated high enough)
• and those others that have something something N (which are rated much too high)

and obviously anything has NNY is somehting correctly rated low that I'm not really interested in and don't know why it was suggested :)

Hope you find that constructive EdwardLane (talk) 15:18, 4 October 2011 (UTC)

Stubs			Cleanup
	Y??Bennett Lake Volcanic Complex			NNY List of Preston North End F.C. managers
	Y?N Klaus Lackner			Y?Y Shale gas by country
	YYNCaerphilly mountain			YY?Llandaff North
	?NYDwight Foster (ice hockey)		Merge
	Y??Kunama language			YYNEnhanced weathering
	YYNSubaerial eruption			N?NReactions to the death of Osama bin Laden
	NNYSoviet First League 1963			NNYPlus fours
	NNYSoviet First League 1978		Add sources
	NNYCoroners Court of Queensland			Y?YPlayStation Network
	NNYMundos Opuestos			?NNBrink (video game)
	Y?YSony Computer Entertainment America v. George Hotz			YYNCardiff East (geographical area)
	Y??Subaerial		Wikify
	Y?N Danakil Depression			Y?? Weyburn-Midale Carbon Dioxide Project
	NNY Soviet First League 1984			NNY Expecting Models
	???Wassonite			N?Y Machars
	?NYBeyond the Sea (K album)		Expand
	Y?NRheid			NNYJerry Vale
	YYNLateral eruption			NNYItalian general election, 1996 (Sardinia)
	Y?YCarbon dioxide scrubber			NNYSoviet Second League B 1970

This is awesome, thanks so much for the feedback! When it comes to the suggestions I went through our logs and noticed a similar pattern that I believe I mentioned last time; the articles that we believe are most relevant to you are also the ones you tend to see as relevant (with the exception of Osama bin Laden, which I think we might be suggesting to lots of people although it's fairly irrelevant to many, it's on my list of things to look at). I also see that many of the uninteresting articles are found through text searches, so there's probably one (or a few) keyword(s) that led SuggestBot astray, that tends to happen (it's also on my list of things to look at).

I am of course glad to hear you like our stars, and I am very happy you took the time to look at them and comment on whether they are correct or not. I checked many of the low scoring ones you thought should have been rated higher, and based on the subject and the content I believe I understand why they could have been rated higher. The way our system currently works they'll be labelled as stubs though, and thus get a one star rating.

Could you let me know more about why you think Brink (video game) and Reactions to the death of Osama bin Laden should get fewer stars? I looked at those articles and they seemed fairly good to me, so I'm very interested in learning what I am missing, as that might also help improve how we calculate our stars. Thanks again for the feedback, it's much appreciated! Cheers, Nettrom (talk) 15:27, 6 October 2011 (UTC)

sure I can try and explain brink and osama :)

Inserting ito my own comment - figured out the confusion, I thought the star rating was your attempt to measure how good a suggestion it was, not how well made the article was.

If the article has a wikiproject assessment then using the stub/start/C/B/A/GA/FA class graphic would be clearer - and ??? if it is unassessed might be easier to do the second task.EdwardLane (talk) 16:41, 6 October 2011 (UTC)

I think I'm getting suggested brink because I made an edit or two on the 'playstation netword outage' article. I was writing about the change in the terms and conditions sony have introduced - nothing to do with playstation games themselves. I do occassionally make a contributino to a boardgame article. but I don't play any first person shooters and I'd never heard of brink. I could probably go search for references for it - but I've no interest in the subject. I don't see any obvious 'citation needed' tags in the article either (that would mean I could run a specific search without needing to read the whole thing). All in all that means I think my possible contributions and chance of contributing is low - so as a suggestion it ought to rate 1 star.

Osama is suggested based on someone else editting it taht also editted other articles I edit, but I've got little interest in mainstream media hype and terrorism. So 3 stars of suggestion for an article I've not touched a connected link to seems a bit enthusiastic - 2 stars perhaps if it's a huge article with lots of tags, 1 star would have been plenty.

You seem to suggest that stubs will only give me a 1 star 'you might actually want to edit this rating' but I quite like playing with stubs - especially when they are connected to articles I'm already interested in. Perhaps it's possible to check my edit history to see what size articles are/were when I editted them and use that as an indicator of preference for (or against) editting small articles? and if the modal article size I edit is below a certain threshold give all stubs suggestions a plus 1 star ? EdwardLane (talk) 16:34, 6 October 2011 (UTC)

Hi again, sorry I forgot to respond to this sooner, my bad. I'm glad to hear the confusion got sorted out, as a researcher it was interesting to learn that the stars were confusing (we've had a few design attempts and these stars were the best we came up with). Your suggestions for using the class graphics is a good idea, if we end up keeping the quality information available I'll see if we can make that work as it might make it easier for our users to understand the information. And thanks for letting me know more about your thought process on finding out if/how you can contribute to the articles! That kind of information is really valuable to us since it can help us understand if we can make SuggestBot do parts of it and improve the suggestions, for instance your idea of looking at the size of articles you contribute to is a good one which I'll add to my list of ideas. Thanks again! Cheers, Nettrom (talk) 15:04, 13 October 2011 (UTC)

Feedback from SA

Firstly I would just like to say that I can't believe SuggestBot works so well! It hit the nail on the head for 31 out of 34 suggestions for me. The ones it got wrong are just off subject. I am a car fan. Almost anything automotive is intresting to me. But it gave me these:

• American College of Switzerland
• I'm On One
• FinePix S5800

These choices are a little odd. The college I have nothing to do with, I'm Australian not Swiss. The song couldn't be further from my intrests in music. As for the camera... well this is spooky. I have been thinking of getting a camera. A 2mpx moblile phone is not good enough for bird watching or train spotting, which I'm thinking of getting into. But last time I checked SuggestBot didn't read minds. I hope this is helpful. I have one question: How does one get into the study thing with SuggestBot? Oh and I have SuggestBot leaving me my suggestions on Starfleet Academy/SuggestBot if your intrested. Well see you 'round :) Starfleet Academy "Live long and prosper." 03:37, 13 October 2011 (UTC)

Hi, thanks for your feedback! I'm of course very happy to hear that you got such good suggestions, that's what we aim for! You mentioned you're a car fan, and when you edit articles within a fairly narrow field of interest a system like SuggestBot tends to get less confused, and so the quality of the suggestions is generally good.

As you've noticed we're not perfect, the first two suggestions you listed were found by looking for editors who "look like you", and then picking articles they've edited that you haven't. In this case the articles we picked weren't on topic, maybe because we only choose articles that the community has tagged as needing improvements, or it can also be that several of the other editors edited these other articles. The camera article was found by looking at articles with similar text, and while I'm amused by thinking SuggestBot can read minds, as you see it cannot. There might be a keyword or two that triggers the similarity, and in this case the camera article appeared to be the most relevant article that fit in the stub category.

Regarding the research study: there's no need to sign up or anything to be a part of it. We automatically sign up our users who get suggestions regularly, and of course allow those who do not wish to participate to withdraw. As you've probably noticed, you got some information about the quality of the articles, some other users get other information. Then we'll see what works and what doesn't, and add the things that work to SuggestBot to improve the experience for our users.

Let me know if you have any questions, I'd be happy to help you out, and thanks again for your feedback! Cheers, Nettrom (talk) 14:54, 13 October 2011 (UTC)

Well... I thought that might be the case, So I guess I should say that I think the rating system works very well.

PS I guess I will confuse SB if/when I start editing historical stuff <laughter /> :) Starfleet Academy "Live long and prosper." 03:45, 14 October 2011 (UTC)