lundi 15 janvier 2018

Mnemosyne status update, 5 years of use



Mnemosyne project status update

14 January 2018

I think it is time I summarize my experience with Mnemosyne again. It has now been approximately a year since I last blogged about this spaced-retrieval program.

I now have 5950 cards with 1-20 questions each (new cards usually contain 1-4 questions). They are spread out in the following main categories:

·         Anatomy and physiology (408 cards)
o   Contains pictures of diseases (Lyme, Mumps, Rubella), questions about hormones (serotonin, aldosterone, etc.), body parts and organs (bones, arteries, heart, muscles, etc.), systems (infectious, inflammatory, etc.), normal biological values (K, Na, Hb, TP-INR, heart rate, etc.).
·         Health (1438 cards)
o   Contains questions broadly related to health (e.g. definitions, public-health, motivational interviews, cyber health, Swiss health system, health promotion, history of medicine, perspectives, laws, etc.). Also contains cards about techniques, physiology, research and diseases which I didn’t bother to classify correctly. Admittedly all decisions are arbitrary and many cards deal with multiple domains.
·         Social work (262 cards)
o   Contains questions related to social work (e.g. sexuality and intellectual handicap, challenging behaviors, behavior management, conduct disorder, how to plan activities, health of migrants, how to behave with deaf-blind people, social work associations, etc.) and education (dyslexia, dyspraxia, ADHD, learning difficulties).
·         General knowledge (139 cards)
o   Contains questions that I couldn’t fit in other categories (e.g. what is copyright, gamification of activities, Chinese words to distinguish, English words to pronounce, DNA and crimes, etc.).
·         Education/teaching (165 cards)
o   Contains questions related to education (e.g. effective ways to use homework, inclusion, evidence-based learning strategies, educational institutions, spaced learning, dyslexia, thinking skills, autism, giftedness, growth mindset, Montessori, etc.).
·         Pathologies, symptoms and diagnosis (809 cards)
o   Contains questions related to diseases or diagnosis (e.g. diabetes, obstructive sleep apnea, schizophrenia, pneumothorax, addiction, heart failure, etc.). Most of them contain at the very least some key items such as: definition, prevalence/incidence (in general population), etiology, signs and symptoms, diagnosis (criteria, thresholds, diagnostic options, LR+/LR-), differential diagnosis, treatment (pharmacological, non-pharmacological, surgical, therapy, 1st/2nd line), prevention, prognosis and complications.
·         Psychology, sociology and culture (105 cards)
o   Contains questions related to psychology (e.g. emotional intelligence, coping strategies, cognitive development, adolescence, culture, validation, labelling, couple theories, neuromyths, Kahnehman’s ideas, etc.)
·         Scientific research (2006 cards)
o   Contains questions related to research and evidence-based medicine/practice (e.g. study designs, how to flaw/bias results, biases, how to read a meta-analysis, the GRADE approach, how WHO guidelines are made, steps in doing a systematic review, peer-review, reporting guidelines, key flaws/errors in the literature, blinding, randomisation, why pre-register studies, designing randomized trials, how to teach EBM, definitions, future of health research, criticisms, myths, managing trials, story of health research, etc.).
·         Drugs and substances (309 cards)
o   Contains cards on drugs and substances (e.g. paracetamol, lidocaine, warfarin, iron, melatonin, epinephrine, adenosine, pantoprazole, fennel, etc.). Some cards are meant to link names to substances (e.g. Sintrom -> warfarin) and others focus on contra-indications, side-effects, details on how to use, dosage or antidote mainly.
·         Techniques and protocols (204 cards)
o   Contains cards related to techniques (e.g. clinical examination of ear/nose/eye/etc., dressing a wound, wound care, transfusion, drawing blood, administering oxygen, catheter care, EKG, CPR, suicidal risk, etc.). Those may ask about the steps/how to, the equipment required, the indications/contra-indications or risks.
·         Untagged and retracted cards which are no longer used (238 cards)
o   Contains cards I no longer deem useful (e.g. psychanalysis, nursing theories, techniques I never use in practice and don’t feel will be needed, etc.), cards which were too complicated (20+ questions in a single card or so) and, I notice, cards which shouldn’t have ended up there (wrong decision in retrospective? Error?).
·         Other

It looks like I have roughly doubled the amount of active cards in a year, with about 1800 new cards on research, 500 new cards on diseases and 500 new cards about health.

I usually answer 60-80 cards per day (often before going to bed, as part of my routine), which takes approximately 1 hour. This is twice as many cards as I used to do a year ago in the same amount of time (but cards are now much shorter with less questions). I will get 50-80 cards per day in the following month (it seems). I do not use cramming functions.


I have added what looks like 30 cards on average (1-70) every 7-10 days (I often add cards during weekends, which explains the 7 days intervals) with a spike 6 months ago (a transition period where I had more free time). I usually write the contents of cards during the week on a Word document I keep open and add them to Mnemosyne in bulk afterwards. While Mnemosyne keeps telling me I probably shouldn’t add 15 cards at once, I never care about the warnings and haven’t had problems so far. It is sometimes quite funny to see how by the time I add the card in Mnemosyne I have entirely forgotten the answer.


My statistics tell me I have answered cards almost every single day (350/365 days?) in the past year with, on average, 90% retention/correct answers. It’s actually not entirely correct since there are often many questions on a single card and I may grade it as “correct” if I miss a question I don’t think is important or if I read the question differently than expected (at which point I also sometimes restart Mnemosyne since “skip” isn’t an option). I also sometimes offer different answers than those expected and may decide that both options are correct. I wish Mnemosyne displayed statistics over many years, which could perhaps allow me to see long-term patterns (decline in memory with age?).


When looking at overall grades, almost all (4622) cards are grade 3 “correct but with much effort”, a decent amount (1498) are grade 4 “correct with some effort”, a few (58) are grade 2 “barely correct answer” and almost none (7 and 1) are grade 1 and 5. The numbers don’t add up to 5950 as untagged and retracted cards are also included in the analysis (there is no option to keep more than one card category). One hypothesis may be that I do not make very difficult or very easy cards. Perhaps I only try to learn things that aren’t too difficult. My personal opinion is that I simply almost never consider that I know something really well (grade 5, “correct without any difficulties”) and have a hard time considering I know things well (grade 4). Also, I may be afraid to grade cards incorrectly and not see them again and grade 3 seems like a safer option than grade 4 and 5.

Most of the cards feel very easy to answer with a couple rare exceptions (e.g. I keep confusing some drugs with very similar names such as alosetron/ondansetron/olansetron, Chinese words are still confusing and questions with many answers such as telling all the 10+ complications of a disease can be a pain). 

I am now trying to learn and remember the certainty associated with results of studies (such as “low certainty drug X reduces mortality with disease Y”) but haven’t yet found it satisfying as doubts still linger and I repeatedly make errors. Having read the full study doesn’t seem to make a difference.
Learning likelihood ratios associated with diagnostic tests is also quite challenging at times (was it LR+3, LR+3.4, LR+10?) and it is interesting to see how I now tend to assess my answers with margins of errors (e.g. the answer is LR+ 5, if I thought it was LR+3, LR+4, LR+7 I would still judge my answer reasonably close and correct as it wouldn’t make much of a difference in practice).

I wonder if re-reading research books would lead to more new cards (new insights, new/better/different understanding) or if it would be a loss of my time. I may have to try one day.

I still wonder how to best keep the cards up to date (e.g. new treatments for disease X, treatment Y now considered useless and risky) without spending too much time doing that. How often should I check if things have changed? I will likely run into trouble if one day I learn that something I have learned is wrong (e.g. pharyngitis is actually a cancer) since I may have used various wordings to talk about the same thing and related questions may be spread out over many different cards. Advanced searching options may then perhaps be of help, especially with 5000+ cards.

One may then think about creating a centralized system which automatically updates cards with new current knowledge (depository of knowledge?). Kind of like Wikipedia automatic updates when new Cochrane systematic reviews are published. Could be neat, my cards would be up to date all the time. Or super-dangerous if those managing the database are conflicted, biased or make errors. Also can be difficult with interpretations of questions/answers, we may then need basic templates which get re-used or training to all understand them the same way.

I remain a bit concerned about the future of the Mnemosyne project considering how few people seem to be working on it. If I plan to keep using such programs for at least the next decade I may have to make plans in case development is halted. 

An option to "divide" a card in two unique cards could be useful, particularly when there are, say, 5 questions in a card and I know 4 very well and 1 not at all. Making a fresh new card makes it look as if I had never seen the question before.

I do keep regular backups of my cards in case I lose access to my laptop (broken, stolen, lost, not working, etc.). The thought of losing years of effort isn’t pleasing but I’m not really afraid of that, it’s a possibility I acknowledge.

Some last thoughts... Am I using Mnemosyne in an efficient/effective way? I don’t know. Is it really helping me remember things? I don’t know. Is it harming me? I don’t know. Maybe one day I may have answers. In the meantime every day is an experiment.

samedi 7 octobre 2017

Kronos Scourge invasion

The Kronos Scourge invasion is on us and I couldn't resist giving it a look. Which means logging on, getting the 2x quest items from my mailbox (one of which is bugged and shouldn't have been sent) and running to the Eastern Plaguelands... Where I could clearly see a skull by browsing the world map yet couldn't find any flying ziggurat or scourge crystals. Also one of the Scourge invasion questgivers had a ? question mark even though I didn't have his quest (I couldn't complete it anyway). On Oct 6th I was told we had already killed 25 "waves" of scourge.

Good news is that I did end up finding the crystals in the Burning Steppes. And quite some bit of Alliance farming them (are you stacking oils for Naxxramas?).



Luckily most of them focused on the scourge and not on me. As described in a bugreport, I did not see any necropolis flying in the sky. All scourge creatures are missing their money loot. Killing scourge creatures does lead to flying spells hitting the crystal (which cannot be selected... but probably should) but it also keeps me in combat because reasons.

90% of what happens during this invasion

Scourge minions melee hit me for roughly 70/hit (reduced), which would mean 88/hit or so unreduced, which looks too low. Doesn't seem to be much for me with this event unfortunately... maybe I will farm some for reputation until revered (that'd mean less arcane crystals to buy).

Unfair world, mages can kill 50 or so of those in 20 seconds. Say hello to vanilla.
I have yet to see some scourge in Undercity and couldn't find a group to fight a Shadow of Doom. It's interesting to see how this event seems very small and unimportant to me now yet felt much different back in 2006.

... On a 100% unrelated note my roughly 2 years old report about Gahz'ranka's slam hitting for roughly 10x too much damage is now #confirmed, yey!

mardi 27 juin 2017

Picture sources of vanilla World of Warcraft

This is a depository of 2004-2006 pictures of World of Warcraft. More than 20 pictures or so must be present for a link to be included here.

Miscellaneous and limited amount of pictures

http://vilesmile.blogspot.ch/
http://pifactorial.blogspot.ch/2005_07_01_archive.html
http://www.lurkerlounge.com/forums/thread-5801.html (getting friendly with STV pirates)
http://z13.invisionfree.com/IBG/ar/f12.htm
https://www.gamespot.com/articles/world-of-warcraft-walkthrough/1100-6114179/

2005-2006, S8's depository, 500+ pictures

http://s8.org/wow/screenshots/

2005-2006, Coute's gallery, 200+ pictures

Coute's long URL

2005-Feb 2006, Novylen's/Rbo's gallery, 300+ pictures

http://gallery.novylen.net/v/rbos/warcraft/
http://gallery.novylen.net/v/rbos/warcraft/?g2_page=25

2005-2006, Project JJ's gallery, 150 pictures

http://gallery.projectjj.com/v/wow/

2005, Denvb's depository, 100+ pictures

http://denvb.eu/screenshots/wow/

2004-2005, QS's depository, 50+ pictures

https://www.rps.net/QS/SCREENSHOTS/WoW/

2005, Wowhelland's 100+ pictures

http://s299.photobucket.com/user/wowhelland/library/Vanilla%20WoW%20screenshot%20collection?sort=9&page=5

Demsix's 1900+ pictures

http://demsix.imgur.com/

2004, level 0-19 Alliance, 50 pictures or so

https://www.gamespot.com/articles/exploring-the-world-of-warcraft/1100-6093288/

2004, level 10+ Alliance warrior (mainly), 50+ pictures

http://demolishwow.blogspot.ch/2004/05/
http://demolishwow.blogspot.ch/2004_06_01_archive.html

2005, level 35-60 shaman, 50+ pictures

http://imamuwow.blogspot.ch/

2005-2006, Alliance rogue/warlock/warrior, mainly level 50+, 200+ pictures

http://www.afn.org/~afn61198/WoW/

June 2006-Dec 2006, Alliance warrior 0-60, 100 pictures

http://shield-block.blogspot.ch/2014/03/wow-memories-in-screenshots-2006.html

2005, Alliance warlock/rogue level 0-57, 50+ pictures 

http://www.ttoprpg.com/wow.htm

2004-2006, Alliance warrior (mainly), 200+ pictures

http://www.skav.org/wow/

Nov 2004-Dec 2006, level 0-40+ Alliance druid/rogue (mainly),

http://brezwow.blogspot.ch/2004/11/
http://brezwow.blogspot.ch/2004/12/
http://brezwow.blogspot.ch/2005/01/
...

Feb 2005- , level 0-60 Alliance paladin, 300+ pictures

http://chordian.net/2017/05/18/my-endeavors-in-world-of-warcraft-part-1/
http://chordian.net/2017/05/19/my-endeavors-in-world-of-warcraft-part-2/
http://chordian.net/2017/05/20/my-endeavors-in-world-of-warcraft-part-3/
http://chordian.net/2017/05/21/my-endeavors-in-world-of-warcraft-part-4/

Feb 2005, 100+ miscellaneous dungeon/raid pictures

http://www.descendantsofdraenor.com/eqdkp/plugins/gallery/picview.php?pid=19&cat_id=1

mardi 23 mai 2017

Brief experiment, chance to get 50 gold for participating in a survey versus nothing

This is a close-replication of my previous experiments on the effects of giving a 50 silver incentive to increase answer rates to a in-game World of Warcraft survey.

[Methods] The methods were similar as those previously used [1]; 50 Kronos players were identified with the first /who query and half of them were offered a random chance to get 50 gold. The reward was increased by 1000% as it was thought that it was too low to get players interested in the previous experiments.

The 2nd, 4th, 6th, ... players on the /who list (unordered) got the intervention message (incentive) and the 1st, 3rd, 5th, .... got the control message, as I previously suggested doing [2]. All messages were sent from my level 60 warlock character, "Tekai", as in previous experiments. Odd players were whispered first.

To select the player getting the 50 gold reward, all intervention players were put in a list going from 1 (first player answering) to X (last player answering) and a /random X was used once. The player with the corresponding number was then either sent the gold with in-game mails or it was traded directly.

The experiment was done on Tuesday 23rd of May 2017, 17:00 Server Time.

Control group message : "Greetings, I am making a survey to better evaluate how much latency Kronos players have and I would appreciate your participation. All answers will be kept anonymous."

Intervention group message : "Greetings, I am making a survey to better evaluate how much latency Kronos players have and I would appreciate your participation. All answers will be kept anonymous and a single randomly selected participant will be sent 50 gold by mail."

Follow-up message : "Great! How much latency (ms) do you currently have? Are you in the EU/RU/US/NA/Asia/Africa/Other?"

End message : "Thank you very much for your participation! Have a good day!"







Intervention end message : "Thank you  very much for your participation, the winner will be described in the world channel in the next 30 minutes. Have a good day!"

Further details message : "New players often wonder about the latency of the server, this survey would allow them to have an overview. Please send me a in-game mail if you want to look at the results."

Explanation message : "All participating players will be given a unique number (eg. 1 for the first to answer, 2 for the second); a single /random will then be used to define a winner when the survey ends. This player will get the 50 gold by in-game mail or trade."

[Results]  The experiment was stopped early without going on with the control intervention as the response rate was deemed too low for the intervention to be effective. A total of 21 players were whispered in the intervention group. Three (3) players were AFK and four (4) were offline.

8 players answered in the intervention (incentive) group (38.1%). The /random excluded a player due to the content of his answer and ended up with a "2", the second player in the list was thus given 50 gold, which were traded directly near the Orgrimmar bank.

No explanations or futher details messages were sent and they were not requested.

The reported latencies (ms) were as follows :
  • EU: 20ms
  • US: 161ms
  • Unspecified location: 349ms, 27ms, 67ms,  "0ms but usually 65ms", "195-205 (best) or 95-205 (average)
[Harms] One player expressed being very unhappy about my whispers as he answered "fu".

[Conclusions]  Previous experiments ended up with slightly lower response rates (about 34%), which means the intervention (incentive) may have had a small effect. Nevertheless the potential increase was very small and unlikely to be of interest to players making in-game surveys. The latencies reported were similar as those previously identified.

Increasing the reward for participating in a survey to 50 gold and adding a chance element was not associated with an arguably important increase in participation. This strategy is thus not recommended.

Perhaps whispered players did not believe the messages were honest, or they may have thought their chances of winning were too low. The whispers might be too formal, the demand too complex (for example, very few players correctly report both country and latency) or it could be that in-game gold rewards are not as attractive as I think. A reverse approach asking players what incentives they would like or find interesting enough or what wording would be most appropriate could be helpful.

A player expressed being unhappy about my whisper. Considering the content of the whisper I sent, I suppose this reaction was due to the perhaps intrusive or disruptive nature of whispers which may break immersion or lead to a lower attention to the current activity the player was doing. While a relatively rare event (this is the first time a player answered in such a way among the 125 or so players whispered in those experiments), it is possible that other players were equally unhappy yet did not tell me about it. Strategies to prevent such events (or better notice them) could be sought, such as using different communication strategies (e.g. in-game mails), although they may further reduce participation, increase costs and bother players even more. Alternatively one could think that whatever one does in game some players will be bothered by it and this is not a good enough reason to stop. It is also conceivable that in-game surveys (all of them or some of them) end up bothering players more than they benefit them and this question could deserve future attention.

Acknowledgments

I (Tekai) used to play on the Kronos I server as a level 60 warlock for months in a progression raiding guild, I wrote extensively about it, contributed to bugreports, quality improvements, discussions and have published the (now completed) Kronos WoW Comics (humoristic drawings inspired from the server) on the official Project Kronos Facebook page. I don't have any financial interest in this survey and its results and I am not financed in any way by the Kronos team or anyone else.

This is mostly an exercise of curiosity and for fun (yes my fun is wierd), please don't take it as a scientific study, but go ahead and do refer to it if you find it interesting.

Data :

Intervention group, 21 players whispered, 3 AFK, 4 offline

17:05:37 [W From] <5:_>: 349ms
17:05:52 [W From] <47:_>: fu
17:05:57 [W From] <32:_>: 27ms atm
17:07:41 [W From] <60:_>: 67
17:08:32 [W From] <60:_>: 161 US
17:09:20 [W From] <14:_>: EU, 20ms lag now
17:10:44 [W From] <59:_>: 0ms it says, but usually its 65ms
17:13:28 [W From] <AFK><60:_>: 190 is the best i get, 95-205 is average, 195-205*

Whispers were not sent to the control group.

dimanche 9 avril 2017

Thinking about Kronos unknown long-term future

I have been wondering for some time about the Kronos vanilla server future. More specifically, what will happen to it in the long-term (months/years) when it will reach the last vanilla patch, with Naxxramas released?

From what I can remember, a large amount of players complained on Blizzard's servers when there were no patch or updates for a long time (with the caveat that patches were expected). Is this what will happen? Would players organize events or something to keep new things happening?

Are there going to be players so over-geared with best in slots items that premades will once again cause troubles in Battlegrounds? The difference between new and old players in PvP will be even greater than it has been before. What will happen if this is the case? Special queues for premades, just like Blizzard did with twink players that have decided to stop getting experience gains?

How about bind on equip items coming from high-level raids? The more players raid Naxxramas, AQ and BWL, the cheaper the epic items coming from those raids will become, as will epic crafted items. At some point, from a gear point of view, Molten Core or Ahn'Qiraj 20man can be skipped. Is this a problem for new players? Are we going to see a decline in people running some raids?

When players have the best items and mounts, what do they then spend their gold on? Are we going to see some players with so massive amounts of gold that they could get a monopoly over large parts of the Auction House? Will this raise prices to a large extent instead of seeing them go down?

With all those players with max-profession skills, will the price of Moonclothes, Transmutes and Refined Deeprock Salt go down due to large amounts of alt-moneymaking-machines?

Are PvE players going to quit once done with Naxxramas and all previous raids? Will they continue until all their characters have all possible best-in-slots? What will keep them motivated to run AQ40 again for the 40th time?

With all the PvP players already having a Warlord title, is it going to lose its prestige? Are PvP players going to lose interst in trying to get it? Will they keep doing PvP ? For how long?

Assuming Kronos continues to stand up among the constant chaotic activity of WoW private servers, I will certainly be very happy to follow how things go on. *Curiosity*

vendredi 24 mars 2017

Videos of retail sources for vanilla WoW - Searches done

[Last updated 15th May 2017]

This is a list of places I have extensively (from beginning to end) searched for relevant vanilla WoW videos. The goal is to not make the same searches too many times. Our time is finite.

Google video searches :
  • [2005-2006] "wow night elf"
  • [2005-2006] "warcraft night elf"
  • [2005-2006] "darkshore wow"
  • [2005-2006] "mulgore wow"
  • [2005-2006] "loch modan"
  • [2005-2006] "warcraft quest"
  • [2005-2006] "tauren"
  • [2005-2006] "redridge"
  • [2005-2006] "stonewatch"
  • [2005-2006] "lakeshire" 
  • [2005-2006] "deadwind" 
  • [2005-2006] "silverpine" 
  • [2005-2006] "ilgalar"
  • [2005-2006] "wow level" 
  • [2005-2006] "teldrassil" 
  • [2005-2006] "warcraft level"
  • [2005-2006] "西部荒野" (= Westfall)
  • [2005-2006] "洛克莫丹" (= Loch Modan)
  • [2005-2006] "泰达希尔" (= Teldrassil)
  • [2005-2006] "银松森林" (= Silverpine Forest) 
  • [2005-2006] "wow aoe"
  • [-2017] "wow aoe farm 2006" 
Archive.org searches :
Youtube author videos handsearched :

vendredi 10 mars 2017

Replication 1 : Brief experiment, 50 silver for participation in a survey versus nothing

This is a direct replication of my previous experiment on the effects of giving a 50 silver incentive to increase answer rates to a in-game survey. The methods were the same as those previously used [1], with the exception that I whispered players from my /who list in reverse order this time. The players from the bottom of the /who list were first whispered the incentive question and only then were the 25 players in the upper part of the /who list whispered the question without incentives. The experiment was done on Friday 10th of March 2017, 17:17 Server Time.

This change was done because I hypothesized that the order could affect my results and reversing it in the replication would end up balancing the first experiment. For example, there could be more players willing to answer in zones beginning with different alphabetical letters. Should I try this experiment again, I would instead encourage whispering players in a different manner, like this : player 2 on the list, player 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24. And then giving out the second intervention to players 1, 3, 5, etc.

[Results] A total of 44 players were whispered, 24 in the intervention group and 20 in the control group. Players in the intervention group had a mean level of 44.4 and players in the control group a mean level of 44. A total of five (5) players were offline when whispered (1 in the intervention group and 4 in the control group) and a total of four (4) players were away from keyboard (AFK) when whispered (2 in the intervention group and 2 in the control group).

One (1) player in the control group didn't talk english and couldn't participate. Two (2) players, one in each group said I had already whispered them and decline to participate twice. One (1) player in the incentive group didn't want to get the reward twice when I whispered him on his other character, and one (1) player in the intervention group said money wasn't needed "I'm happy to help, No money neccessary, :)".

9 players answered in the intervention group (37.5%) (the player answering twice wasn't counted) and 8 in the control group (33.3%) (the player that didn't speak english wasn't taken into account, neither was the one answering twice).

Relative risks can be calculated by blindly (I know, I know...) using Medcalc [3].  Incentives were associated with a 17% (RR 1.17, 95% CI 0.52 to 2.63) higher chance of getting an answer, although the confidence interval crossed 1, which means there could be no real difference. The absolute difference was +4.2% (37.5%-33.3%).

When results from the previous experiment were combined with those new ones* (this means I assume they are similar, which could be wrong), the incentives were associated with a 7% (RR 0.93, 95% CI 0.54 to 1.63) lower chance to get an answer, but again, the results could be due to chance. The absolute difference was -4% (41%-45%).

The reported latencies (ms) were as follows :
  • EU : 99ms, 53ms, 45ms, 167ms, 47ms, 73ms, 50-100ms, 15-30ms
  • US : 170ms, 166ms, 100-175ms, 
  • RU : 50-60ms
  • Unspecified location : 18ms, 195ms, 16-31ms
  • One (1) player couldn't see his latency due to addons "I cant see latency, sorry" and one (1) gave a narrative answer "hi, today I have experienced some heavy lag spikes, but since tha made the network changes, latency have been good overall" and players added details "its usually good, idk what the average latency is or whatever", "there is always the weird server lag especially if you encounter ppl from NA as European".
I ended up giving out a total of 4 gold (8x50 silver).

Perhaps of interest is that no single player asked me if and where the results could be seen or what I wanted to do with them. The "further details" message was never sent to anyone.

Some players seemed particularly happy to participate and friendly "I'm happy to help", "You are doing gods work :)", "no problem thanks you too", "Greetings sir, ofc i can help out with that". I did not get negative comments about the amount being too low.

[Harms] I had to apologize twice for sending the same whispers to characters from the same players and none complained of this being a problem "hi i already said around 60 on another char, no need to send again :p, no problemo", "I have already answered on another character and answering twice would provide lackluster results, right?, No worries, You are doing gods work :)".

[Conclusions] The overall results suggest that a 50 silver incentive is associated with a lower chance of getting an answer when players are whispered to participate in an online survey. To get a definitive answer I would need a much larger sample, especially considering how the effect, if any, appears to be small.

In the very best possible scenario, with the maximum relative increase of 63% (RR 1.63), and a baseline answer rate in control groups of 34% (33.3% to 34.8% in my experiments), 50 silver incentives would lead to a 55% answer rate (0.34*1.63). This is a 21% absolute difference, which means I would have to pay about 5 players (5x50 silver = 2 gold and 50 silver) to get one more answer. And in the worst case scenario, giving the 50 silver incentive would lower the amount of players answering. I don't think that 50 silver incentives is worth it for surveys about latency. Perhaps results would be different with another kind of survey or question, or wording.

Latency appeared to be stable between experiments, even if the hour of the day and the day of the week weren't the same. I found players to be quite friendly and patient in their answers.

Acknowledgments

I (Tekai) used to play on the Kronos I server as a level 60 warlock for months in a progression raiding guild, I wrote extensively about it, contributed to bugreports, quality improvements, discussions and have published the (now completed) Kronos WoW Comics (humoristic drawings inspired from the server) on the official Project Kronos Facebook page. I don't have any financial interest in this survey and its results and I am not financed in any way by the Kronos team or anyone else.

This is mostly an exercise of curiosity and for fun (yes my fun is wierd), please don't take it as a scientific study, but go ahead and do refer to it if you find it interesting.

Data :

Intervention group, 24 players whispered, from bottom /who to top, 2 AFK, 1 offline

17:17:40 [W From] <49:_>: 99ms in EU (_)
17:18:28 [W From] <57:_>: 53 ms in EU.
17:18:33 [W From] <22:_>: dont know how to check, 45, EU
17:20:14 [W From] <17:_>: hi i already said around 60 on another char, no need to send again :p, no problemo
17:20:10 [W From] <60:_>: USE 170 ms on avg.
17:20:35 [W From] <17:_>: my average ms is 166, I live in the US on the East Coast.
17:22:09 [W From] <15:_>: I'm happy to help, No money neccessary, :), 167 UK/EU
17:23:37 [W From] <60:_>: I'm EU (_) and have 47 MS
17:24:38 [W From] <60:_>: 18ms atm
17:24:23 [W From] <60:_>: idk what to tell u its usually good, idk what the average latency is or whatever, ya i knnow mine 195 right now

Control group, 20 players whispered ,  2 AFK, 4 missing

17:26:43 [W From] <60:_>: I cant see latency, sorry
17:28:18 [W From] <60:_>: 100-175ms, usa based
17:30:05 [W From] <60:_>: I have already answered on another character and answering twice would provide lackluster results, right?, No worries, You are doing gods work :)
17:30:34 [W From] <60:_>: hi, today I have experienced some heavy lag spikes, but since tha made the network changes, latency have been good overall
17:30:59 [W From] <60:_>: hey, im from Russia and i have 50 ms atm. Usually its around 60ms
17:33:12 [W From] <2:_>: EU and right now its 73, no problem thanks you too
17:33:28 [W From] <60:_>: all from 50-100ms, but there is always the weird server lag especially if you encounter ppl from NA as European
17:33:48 [W From] <42:_>: 16-31ms
17:35:07 [W From] <8:_>: Greetings sir, ofc i can help out with that, Im in _ and my latency varies from 15ms-30ms, same to you!
17:33:32 [W From] <60:_>: sry i dont speak english

*For a total of 16 answers in the incentive group for 39 whispers and 19 answers in the control group for 42 answers.

mercredi 8 mars 2017

Brief experiment, 50 silver for participation in a survey versus nothing

[Objectives] To find out how much of an effect on the percentage of players answering a survey a 50 in-game silver incentive can have when given to players agreeing to participate in a Kronos survey.

The secondary objective was to gather data on the latency players from different countries have.

[Methods] 50 players were identified with the first /who query sent to the Kronos I (K1) server on the 8th March 2017 between 12:50 and 13:10 Server Time. They were all sent whispers from my main character, a level 60 undead warlock named "Tekai". The first 25 players in the list were the intervention group and received an invitation to participate in a survey with a 50 silver reward for participation. The next 25 players in the list were the control group and received a standard invitation to participate in a survey. Participant's answers were copy/pasted on this blogpost along with their level and names were removed.

Control group message : "Greetings, I am making a survey to better evaluate how much latency Kronos I players have and I would appreciate your participation. All answers will be kept anonymous."

Intervention group message : "Greetings, I am making a survey to better evaluate how much latency Kronos I players have and I would appreciate your participation. All answers will be kept anonymous and participants will be sent 50 silver by mail."

Follow-up message : "Great! How much latency (ms) do you currently have? Are you in the EU/RU/US/NA/Asia/Africa?"

End message : "Thank you very much for your participation! Have a good day!"

Further details message : "New players often wonder about the latency of the server, this survey would allow them to have an overview. Please send me a in-game mail if you want to look at the results."

I had previously obtained 57% (survey about forum usage) [1] and 39% (survey about world channel) [2] response rates with similar methods (in-game whispers) and expected similar numbers in the control group. This experiment will be underpowered and should be interpretated as such.

[Results] A total of 45 players where whispered, 23 in the intervention group and 22 in the control group. Players in the intervention group had a mean level of 46.38 and players in the control group a mean level of 46.64. Five (5) players were offline when whispered and a total of five (5) players were away from keyboard (AFK) when whispered; I did not keep note of in which group they were. One player in the intervention group probably talked russian "Chto?" and couldn't understand my message.

8 players answered in the intervention (incentive) group (34.8%) and 11 in the control group (no incentive) (50%).

Relative risks can be calculated by blindly using Medcalc [3].  Incentives were associated with a 30% (RR 0.7, 95% CI 0.35 to 1.40) lower chance of getting an answer, although the confidence interval crossed 1, which means there could be no real difference. The absolute difference was -15.2% (34.8%-50%).

In other words sending 100 whispers with an incentive would lead to 35 answers and sending 100 whispers without it would lead to 50 answers.

The reported latencies (ms) were as follows :
  • EU : 47ms, 48 ms, 88 ms and 81 ms
  • NA : 140-200ms
  • US : 120ms
  • Oceania : 340 ms
  • RU : 129ms and 65ms
  • Unspecified location : 20-30ms, 60-65ms, 90-119ms, 420ms, 78ms, 81ms and 46ms
  • Latency and location wasn't clearly defined for three (3) other players : "no trouble so far, well maybe sometimes but very punctual moments and didn't last more than 2 or 3 seconds, and it doesn't happen often", "latency is good" and "not sure how low my ms is atm"
I ended up giving out a total of 4 in-game gold (8x50 silver). 16 minutes was the longest amount of time before a player answered my whisper. 30 minutes after the last whisper no further answer was received and I disconnected my character.

[Conclusions] The use of a 50 silver incentive to increase answer rates to a in-game survey on Kronos I was NOT associated with a higher participation. The players NOT offered a reward for participation participated more often (11 answers in the control group vers 8 in the incentive group) but this could be due to chance due to the low sample size.

If three more players had participated in the intervention group, the conclusions could be very different. Perhaps incentives do not work, perhaps it was a too easy to answer topic for incentives to make a difference, perhaps more AFK players were whispered in the intervention group. Or something else.

On a younger server a 50 silver incentives might have a stronger effect as players haven't yet accumulated much gold. The /who queries might not offer truly random players. Players that know me by name might have a higher chance to answer the survey. Players of higher level might be less interested by money (although this didn't appear to have an effect in this experiment).

A handful of players made comments suggesting the amount of money was too low, "lol", "only 50 silver :P". Others seemed enthusiastic "lets do this", "alright", "i am listening" and happy to have participated "=)", "same!", "you too", "Np have a good day".

Acknowledgments

I (Tekai) used to play on the Kronos I server as a level 60 warlock for months in a progression raiding guild, I wrote extensively about it, contributed to bugreports, quality improvements, discussions and have published the (now completed) Kronos WoW Comics (humoristic drawings inspired from the server) on the official Project Kronos Facebook page. I don't have any financial interest in this survey and its results and I am not financed in any way by the Kronos team or anyone else.

This is mostly an exercise of curiosity and for fun (yes my fun is wierd), please don't take it as a scientific study, but go ahead and do refer to it if you find it interesting.

Data :

Intervention group, 23 players whispered :

12:50:05 [W From] <32:_>: EU, 47 ms latency
12:50:16 [W From] <60:_>: 20-30
12:50:24 [W From] <60:_>: Hi sure, im usually playing around 60-65 ping
12:50:29 [W From] <60:_>: only 50 silver :P, go ask, eeeh !@#$ I have addons so i dont see it sec I disable it, i have 48 latency and I am in ___ eu
12:50:42 [W From] <28:_>: Current latency is 119ms thou average is around 90ms
12:51:19 [W From] <51:_>: RU 129
12:56:52 [W From] <60:_>: EU, not sure how low my ms is atm
13:09:35 [W From] <20:_>: oh, i had no trouble so far, well maybe sometimes but very punctual moments and didn't last more than 2 or 3 seconds, and it doesn't happen often

Control group, 22 players whispered

13:00:05 [W From] <60:_>: 420 ms, sometimes i spike to 1337
13:00:10 [W From] <60:_>: 120ms New York, USA
13:00:15 [W From] <60:_>: 65 ms, RU
13:00:36 [W From] <60:_>: 78ms
13:02:48 [W From] <38:_>: ocenia 340ms
13:03:35 [W From] <60:_>: 88MS/EU
13:05:20 [W From] <47:_>: 81
13:05:25 [W From] <9:_>: latency is good
13:05:55 [W From] <55:_>: 81ms right now, i'm in EU
13:07:13 [W From] <4:_>: Just started an hour ago, don't really know if I can do anything, 46ms
13:10:50 [W From] <60:_>: I usually have 140 to 200 MS.. as NA

samedi 4 mars 2017

Vanilla WoW Scourge Invasion 2006 references

As finding vanilla videos for the scourge invasion has proven difficult, I have decided to compile a helpful list of all references I can find about it in a single blogpost.

When?

During the 2006 vanilla Scourge invasion (patch 1.1, in June/July 2006), several regions come under attack from the scourge (nov 2006 Wowwiki).

What and where?

Scourge crysals spawn in the Eastern Plaguelands (6 spots), Burning Steppes (6), Blasted Lands (6), Tanaris (9), Azshara (4), and Winterspring (3) (nov 2006 Wowwiki). Locations of the crystals can be seen on the following screenshot : 2006 picture from Wowwiki.

Scourge crystals are floating crystals with a purple rune underneath and 5 purple fires surrounding them. About 15-45 scourge creatures spawn around it. Usually with two different types of creatures, evenly distributed in a circle around the crystal : Skeletal Shocktroopers/Spectral soldiers, Skeletal shocktrooper/Ghoul berserker (June 2006 picture, July 2006 video).

Some scourge crystals have necropolis floating over them (June 2006 video). During the Scourge Invasion, numerous high-level zones are under attack by one or two flying Necropoleis each. The zones under attack have purple skull markers on the continent-level maps (Nov 2006 Wowwiki).

There are also Necropoleis over the major cities. The ones over the major cities seem to be permanent but they only spawn low level scourge for the introductory quest (Nov 2006 Wowwiki).  Further, they are attacking the two 'Human' cities; Stormwind and Undercity (June 2006 post).

When the scourge creatures/minions around a crystal are killed, a blue spell gets cast and travels towards the crystal (July 2006 video) and weakens the necrotic shards (Nov 2006 Wowwiki). You have to kill about 100 [minions], if not more, of the undead to fully destroy it... When the Necrotic Crystal is 'destroyed', it is replaced by a 'Damaged Necrotic Crystal' that four Cultist Engineers appear around and begin to repair. If you leave the Damaged Crystal being repaired by the Cultists alone for a time, it will be repaired and the event starts all over again. (June 2006 post).

This guys [minions] spawn SUPER fast too 4 mages was at 1 camp lastnight and we all had no troubles with pulls (Sept 2006 guide). How long do you estimate it takes (on average) to make one camp disapear? You are looking at 20-30 minutes with a large group, faster if you have more people, slower if you have less. It is ENTIRELY based on how fast you kill the undead that spawn around it, and then how fast you take out the Cultist/Shadow (Aug 2006 guide).

Each time a scourge crystal is broken, the Argent Dawn offers increasingly more powerful blessings (nov 2006 Wowwiki). Destroying all four Shadows of Doom will destroy the Necrotic Shard, heal nearby players for 10,000, grant them a powerful buff, and stop new Scourge from spawning there (at least for a short time) (Nov 2006 Wowwiki). Destroying all the Necrotic Shards attached to a Necropolis will destroy the Necropolis. Destroying all the Necropoleis in a zone will cause that zone to no longer be under attack. The Scourge will eventually deploy more Necropoleis however, so clearing a zone does not mean it is safe forever (Nov 2006 Wowwiki). But alas, our successes were short lived as the ziggurats were again replaced within just a few hours of their destruction (June 2006 blog).

At any rate, the clearest indication of a new batch of beasties comes when "The Mouth of Kel Thuzad" shouts to his minions to push forward once more (June 2006 IGN).

Those who wish to take up arms against the undead invaders should speak with a representative of the Argent Dawn to learn what regions need help and how the defense is holding up (nov 2006 Wowwiki). NPCs near Argent Dawn tents in the major cities will note where the Necropoleis are attacking, and how many are in each zone (Nov 2006, Wowwiki).

Creatures and bosses

Argent Dawn?
NPCs near Argent Dawn tents in the major cities will note where the Necropoleis are attacking, and how many are in each zone (Nov 2006, Wowwiki).
The npc said we had completed 125 battles...so 25 left (June 2006 post).

Argent Quartermaster (ID)

Argent Dawn paladin for Alliance (ID) and Argent Dawn Initiates for the Horde (ID)
Text : "The first steps of our fight against the Scourge go well. We have had some successes, and hopefully my assistance can aid you in bringing future success to the battle." And offers the option "Give me one of your magic items.", this gives the player 3x Lesser Mark of the Dawn - Although this option will vary depending on the server progress with the invasion (June 2006 blog).

Argent Recruiter (ID)

Argent Outfitter (ID)

Argent Messenger (ID)

Argent Emissary (ID)

Cultist engineer
Neutral.
Can be talked with, require 8 necrotic runes to convert into a shadow of doom, the cultist then disappears, a shadow of doom appears and can be attacked after 5 seconds or so, it does not attack the player on spawn (2006 video).
"Use 8 necrotic runes and disrupt his ritual." is the text option (2006 video).
Spawn in groups of 4 around a crystal when the crystal is damaged.
Cast a purple channeling spell towards necrotic crystals when spawned (June 2006 video).

Scourge crystals (Necrotic shards, ID 16136)
100 health.
Can be targeted, level 60, agressive (June 2006 video).
Bolts of energy regularly travel from the crystal to the necropolis over it (every 30 seconds or so) (June 2006 post).
The shard itself cannot be directly damaged (June 2006, post).

Damaged necrotic shards (ID )
Can be targeted, level 60, agressive (2006 video).
The shard itself cannot be directly damaged (June 2006, post).

Skeletal shocktrooper (ID 16299)
Level 58-60.
3965 health according to addons (2006 video).
Drops between 2 silver 40 copper and 5 silver 20 copper (July 2006 video).
Can be slowed with frost, frozen (July 2006 video).
Hit between 140-285 / melee hit, unreduced damage (July 2006 video).
The undead are non-elite and grant 5 Argent Dawn reputation until the end of Honored, unlike normal undead (Nov 2006, Wowwiki).
Can be kited very far away from his spawnpoint (June 2006 video).

Ghoul berserker (ID 16141)
Hit between 140-265 / melee hit, unreduced damage (June 2006 video).
Can be slowed with Wing Clip and Concussive shot (June 2006 video).
Enrages at about 30% health (June 2006 video).
The undead are non-elite and grant 5 Argent Dawn reputation until the end of Honored, unlike normal undead (Nov 2006, Wowwiki).
Can be kited very far away from his spawnpoint (June 2006 video).

Spectral soldier
Can be slowed, frozen (July 2006 video). 
The undead are non-elite and grant 5 Argent Dawn reputation until the end of Honored, unlike normal undead (Nov 2006, Wowwiki).

Shadow of doom
Level 61 elite.
48'000 health and 2434 mana according to addons (2006 video).
Hit between 1075-? / melee hit, unreduced damage (June 2006 video). 2612 maximum (crit?) according to addons (2006 video).
Cast a drain mana/health spell that hits for roughly 142 health / hit and slows movement, roughly every 15-45 seconds (June 2006 video).
Cast Fear on a single target, roughly every 15-30 seconds (June 2006 video, June 2006 2nd video).
Can be rooted, frozen (June 2006 video).
Can be slowed with frostbolt (June 2006 video).
Cannot be feared but not immune to death coil (2006 video).
Can be kited very far away from his spawnpoint (June 2006 video).
Drops low amounts of silver, around 1-5 silver (June 2006 video) (2006 video).
Upon its defeat, the Shadow of Doom will yield a Corrupter's Scourgestone, 50 reputation with the Argent Dawn, 30 Necrotic Runes, and has a chance to drop a chest piece of the Undead Slaying set. The drop rate on the chest isn't great. I've seen a drop rate of around 1 in 4 in the past, but currently have killed 15 in a row with no chest drop (none at all, not just my armour type) (Nov 2006 Wowwiki).
Kill it and the Crystal takes 25% damage (May 2006 beta testers).

Pallid Horror
Level 60 elite.
Neutral (July 2006 video).
Spawns in cities.
Casts a mass fear against close enemies, roughly of 2 second duration (July 2006 video).
Can be slowed with Wing Clip and Concussive shot (July 2006 video).

Patchwork terror (ID 16382)
Level 60 elite.
Neutral (July 2006 video).
Spawns in cities.
Can be slowed with Wing clip, concussive shot (July 2006 video).
Casts a mass fear against close enemies, of short duration (July 2006 video).
Can be stunned (July 2006 video).

Lumbering horror
Level 61 rare, random spawn around crystals.
Agressive.
8435 health according to addons (2006 video).

Flameshocker
Level 60.
Neutral.
Spawns in cities.

Skeletal trooper
Level 9 (June 2006 blog).
I go outside, see the bright purple circle on the ground there and kill dozens and dozens of level 6-8 Undead that pop up, but never got a single shard to drop. What do I need to do to get those? "   I probably had to kill 30-35 of them before I got 3 of the shards to drop (Aug 2006 guide).

Spectral apparition
Level 7 (June 2006 blog).
I go outside, see the bright purple circle on the ground there and kill dozens and dozens of level 6-8 Undead that pop up, but never got a single shard to drop. What do I need to do to get those? "   I probably had to kill 30-35 of them before I got 3 of the shards to drop (Aug 2006 guide).

Balzaphon
The Stratholme Boss, found by Harmonia, is a Lich that patrols around the fountain in the main area next to the small chapel and the Side Gate to Stratholme.His name is Balzaphon, and he kicks major butt. Serious frostbolt spammer that does a ton of damage. Drops any of three +damage and healing blues, including an amazing staff and a +22 dam necklace called 'Chains of the Lich' and TWO! TWO! Corrupter's Scourgestones (July 2006 comment).
VERY EASILY farmable, right next to the service entrance (June 2006 comment).

Lord Blackwood

Revanchion
Revanchion is found in DM West patrolling the hallway after Tendris Warpwood (you don't have to kill him).  He does a frost nova and has frost armor (slows your attack and movement). The strange thing was that he dodged/parried/blocked around 80% of all my attacks. Not sure if it was a bug or not. He hit me on average for 300 damage with 28.3% DR so he's not exactly a heavy hitter. And he gave 25 AD rep when killed (July 2006 comment).
Note : If someone can find the video "Hunter.Solo.Revanchion.Divx.avi" that would be incredible.

Scorn
He hits normally for his level(32 Elite), and he has some nasty Frost AE Spells for even level groups trying to tackle him (July 2006 comment). 
Location: Scarlet Monastery Graveyard, after Bloodmage Thalmos has been slain he comes walking to the crypt (June 2006 post).

Sever
The new one in Shadowfang Keep is a Scourge Abomination. His name is Sever, and he is found in the area formerly patrolled only by Ghosts near where the Butcher and the Baron are. There are no prereqs to his appearance. He is level 25 (July 2006 comment).  

Lady Falther'ess
Location: Razorfen Downs, cell area, only after Amemnar has been slain (June 2006 post).

Quick personal comment : 150 battles seems super short assuming hard-working players with better knowledge of the mechanics and well prepared for this invasion. There is a risk it might be very, very short-lived on private servers.