Search:

SIYE Time:15:23 on 28th March 2024
SIYE Login: no

News
October Dumbledore Silver Trinket Awards Poll
It is that time of month again; the time where we raise our esteemed authors upon a pedestal.
The nominations are in and it is up to you, the users of SIYE, to decide which stories are of a caliber that demand recognition.

Voting is open to registered users only. Please take the time to vote for your favourites.

The poll will be open until November 7th.

VOTE HERE

Congratulations to all the nominees and best of luck. Nomination results for October can be viewed in the Trophy Room.

Thank you for your support. Please don't forget to nominate your favourite stories for November.

EDIT: Please read comment inside for explanation of DSTA adjustments used.

SIYE Geeks
The Magic Geeks on 2006.11.01 - 12:59AM ()

Comments



hot48cricket came out of the woodwork on 2006.11.01 - 01:03AM to say:

Dang! Too many good stories this month!! Making a choice was very, very hard!!!



Keira Azul came out of the woodwork on 2006.11.01 - 01:12AM to say:

Wow guys! I'm really honored! Good luck to all of the nominees and thanks to everyone who nominated my story. One quick question to the admins though (on a totally seperate topic)- Is there a reason my stories never go up on the Recently Updated list? Just wondering, because my last chapter to the challenge was submited just before midnight and I want to make sure it got in. Thanks! :)



cwarbeck came out of the woodwork on 2006.11.01 - 02:25AM to say:

Thanks for all the nominations. Good luck to everyone who got nominated. So many excellent choices this month!



GinnyMarie came out of the woodwork on 2006.11.01 - 06:22AM to say:

good luck everyone!



melindaleo came out of the woodwork on 2006.11.01 - 02:46PM to say:

Ohh, thanks so much for all the nominations, and best of luck to all the nominees. I'm completely flattered, and it tickles me beyond belief to get nominations for both best comedy and best angst for the same story, lol. I love this site.



Chreechree came out of the woodwork on 2006.11.01 - 03:48PM to say:

Eeep. What an honor. Good luck to everyone.



fake a smile came out of the woodwork on 2006.11.01 - 10:45PM to say:

Okay, I can never remember which ones of these require me to use HTML tags and which don't, and it won't let me delete the previous post, so for the sake of your eyes, I'm going to repost it in a more readable form below. If an admin sees this, you can delete the massive paragraph above this.

Hello all,

Thank you to all those who nominated me. It is most greatly appreciated.

Now, I was looking at the Trophy Room, and I have to say that what I saw both surprised and somewhat disturbed me. I noticed that the Adjusted field was changed this month from previous months, and I can't say I'm all too pleased by the change. In the past, stories would be slightly adjusted down in total nominations when the same users continued to nominate them in consecutive months. This month, however, many stories have been adjusted higher than the total number of nominations they have, in one case to more than 6 times the actual amount of nominations received. Does that not seem a bit ridiculous to anybody else?

I understand the desire to make it easier for new stories to make it into the polls, but if things keep up at this rate I don't think I'll continue voting. Why? Because before long, none of my favorite stories will even appear in the polls any more.

It's already starting to affect some authors. Viridian is most notably absent from the polls this month, and looking in the trophy room, you can clearly see that he should have been in the running for more than one category. If I win a category like Best Overall that would be great and all, but it will feel a bit hollow knowing that Viridian was not included based on the new adjusted nomination system. Should Viridian or any other author be punished simply because he is one of the best writers on the site and consistently posts great chapters?

As happy as it would make me to see my own story or even a different story win certain categories, I don't mind seeing a story like Viridian's Nightmares of Futures Past win so much. It's one of the best stories in fanfiction and deserves all the accolades it has been given. So have all the other stories that have been dominating the voting for the past several months. That's why they are dominating the polls. If we didn't want to see those stories win so much, we'd stop voting for them.

Some of the suggestions to get fresh blood into the voting have made sense to me. The new Best One Shot category was a welcome addition this month. Other categories such as Best New Story and Best New Author would be welcomed as well, but I think we need to take a step back from the current course we are on before we elminate and alienate the best authors on the site.

I apologize to Viridian for using him as an example so much, but it really bugged me to see him excluded from the polls this month. I enjoyed competing with him in the past and looked forward to (hopefully) topping him in one of the categories one month. There are some other great stories in there that are pretty stiff competition, but I hate to see one of the best stories out there get excluded on a technicality.

Thank you all for listening to my grievances. I appreciate your time. And thank you again for all the nominations and votes I have received.

-Matt





MissV came out of the woodwork on 2006.11.01 - 10:52PM to say:

I have to say I agree with fake a smile. I'm new to the site and was pleased to see that you were honoring your authors and their stories in such a way. But the discrepencies between the number of nominations received and the adjusted number does cause one to wonder. I chose to vote tonight for the stories that I thought were best in each category,



MissV came out of the woodwork on 2006.11.01 - 11:10PM to say:

I accidentally posted that last message without the following: lucky for me the stories that I have most enjoyed on this site were available to be voted for. Had they not be I wouldn't have registered so that I could vote and you wouldn't be hearing from me now. Then again maybe I should have stayed in the woodwork.



lecook4 came out of the woodwork on 2006.11.01 - 11:19PM to say:

I think a quick and simple expalnation of what TOTAL, NEW and AVERAGE actually mean would help. I may be totally wrong, but I thought they meant Total = number of votes you received that month in that category. New was just to let you know how many of those were new voters for you in that category and Average was average number of votes you have received for that category, but not necissarily that month? Otherwise it does seem strange to only have gotten 2 votes total, but the average be 6 votes. I'm just looking at my own.

Maybe one of the Admin's could explain the difference for us. Thanks. I love having the totals btw, just need to clarify what they really signify.



Keira Azul came out of the woodwork on 2006.11.01 - 11:34PM to say:

Yes, but how are they weighted? The adjusted totals do seem to be incredibly random - but perhaps that's just because I can't see a pattern. Hopefully there will be an answer soon, because I agree that it wouldn't be fair to exclude someone because they've won before - it just shows how masterful they are at their art.



The Magic Geeks came out of the woodwork on 2006.11.02 - 02:05AM to say:

The DSTAs are certainly one of the many things that make SIYE special, however if they lose their meaning they will become merely a sham. While undoubtedly many of you notice some anomolies in this month's DSTAs, there are reasons behind them. I will attempt to elucidate some of the reasoning and considerations that were factored into generating the current ballot.

To begin with, the simplest item: why are there only 3 stories listed for Best One-Shot, and for that matter, since when has there been a Best One-Shot category? The Best One-Shot category has been a much requested one, and attempts to give some recognition to stories that are One-Shots and while often excellent, do not attain the reader base or popularity of longer running stories. This category has been in place for half the month of October. The option to nominate a story for Best One-Shot will only appear for stories that are one chapter in length and are completed. While there were 9 stories nominated, 6 of these had 2 or less nominations, falling short of the 3 nomination minimum required for ballot eligibility, and therefore did not appear on the ballot. This 3 nomination minimum has been in effect for many months, and attempts to ensure that more than just the author and a friend nominate the story. In most cases this clause does not come into effect as the nominations are significantly higher than just 3.

Now for the complicated bit... why the adjustments? It has been noted by many that the same stories and authors tend to end up on the ballot each month and subsequently tend to win the DSTAs. While many of these authors/stories are of an unquestionable calibre, it seems appropriate that more than sheer numbers should be factored into determining the ballot. For one thing, longer running stories will always garner more nominations due to a larger readship, this does not necessarily imply however that they are superior to stories that reach completion in a shorter time.
Originally the intent was to simply prevent the popular stories from continuously winning, and for that simply disqualifying them would have been fine. However, if I was an author and one of my motivations was winning DSTA, I would be somewhat disheartened to learn that I had a great story, but because I had won a few times my story could never win again, no matter how long I worked on it or how much I improved. Straight-out disqualification therefore seems a little unfair. I think that the intent now is to elevate the level of competition. To place authors against themselves. An english major who has taken an interest in fanfiction is going to do much better than someone who doesn't speak English as their first language, however one might have reached their peak, while the other has scope for improvement. Ideally everyone should have a chance to win at all times, however, I think everyone wants to see stories that are good enough to win become even better: to improve the quality of their story, to gain more readers, and to keep the readers they already have.
The approach taken was a mathematical one. It is by no means perfect nor final, but it represents an attempt to consider more factors in a fair way, without outright disqualifying any story. I really want this to be more than just a fancy equation, I want each part to have a defined and clear cut purpose, both in theory and in practise. I think that the pattern of winning is less important than holding true to the original purpose of the DSTAs, to uphold the high quality that SIYE has, and to see even better stories in years to come; to encourage a high standard from authors, and to allow the readers to show their appreciation for those stories they deem most worthy.

So after that preamble, before I get to the adjustment itself, allow me to explain the columns in the Trophy Room. Firstly, the adjustment only applies to the current month. The Trophy Room displays the previous months as if they had the same adjustment, but no adjustment was actually employed in previous months.
The columns:
Total - this is the total number of nominations that a given story received in a given category for the month in question; it is an unadjusted number
Score - this is no longer an adjusted nomination, but rather a score; it is the result of an equation that attempts to consider multiple factors; stories are ordered by this number and for this month, it was this number that determined the stories that qualified for the ballot
New - this number represents the number of nominations (out of the Total) that came from users who have never previously nominated the given story for the given category
Max - this is the highest number of nominations that the story has received for this category in any month to date

Finally, the adjustment... the adjustment, termed score from herein, has two parts, one to consider growth in readership and one to consider aquisition of new readers; while these items are definately related, they are by no means the same.
The score (S) is the product of the growth (G) and the adjusted weight of the nominations (W) (i.e. S=G*W) Growth is based on the following notion: the number of nominations a story receives is proportional to the number of readers times some quality factor. (e.g. it is plausible that a story with more readers will get more nominations than a story of equivalent quality but fewer readers, and likewise it is reasonable that of two stories with an equivalent number of readers the one with better quality will recieve more nominations; these are of course generalizations, nonetheless they are fairly representative). The absolute number of nominations receieved therefore is indicative of both the quality and the number of readers and consequently, an increase in nominations suggests growth. The question was what should growth be compared with? The possibilities were a) the number of nominations in the previous month, b) the average nominations to date, and c) the maximum nominations to date. Nominations in the previous month was simply too variable and did not represent an accurate measure. Average was originally used, however average is a poor statistical measure as it is easlier affected by extreme values, in this case low months. Maximum seemed both mathematically sound and logical. When you wish to know how you are doing you want to be compared to your best, and this is what maximum nominations represents. The simplest calculation of growth would be Nominations/Maximum, however this often yields too large a value and skews the score in an undesired fashion. In order to reduce the effect of this, the squareroot of the final value was taken. Furthermore, this does not explicity set an objective growth. The objective growth set will undoubtedly seem very high to many people, however, if you examine the nomination data you will find that it is quite reasonable. The objective growth was set at 15%; this means that if a story had a maximum of 20 nominations one month, it would need 3 more nominations to have demonstrated this growth. But you ask, what of a new story? It obviously has no maximum and infinite growth would be a useless quantity. It was deemed that a new story should be equivalent to the highest possible growth plus a slight bonus, after all, if a story that has only been out for one month can get the same number of nominations as one that has been out for two months, it merits a higher score. The 'maximum' for a new story was therefore set at 1 (the minimum number of nominations a story can receive and still be listed) and a bonus of 10%.
The final calculation of growth therefore is the following:
if (New Story){
G=1.1*sqr(N/1.15)
}else{
G=sqr(N/M*1.15)
}
Where:
G is the growth factor
N is the total number of nominations for the story in question for the current month in a given category
M is the maximum number of nominations for the story in question in a given category for any month to date

Growth on its own is not a complete indication of story merit; a story that grows from 1 nomination to 2 nominations has the same growth as a story that grows from 10 nominations to 20 nominations; It is essential therefore to consider the actual number of nominations received. One additional consideration however must be factored in and that is the number of nominations received from users who haven't nominated this story for this category previously. Growth inherently measures the number of new nominations, however it is both indirect, and down-weighted. While there are exceptions, many readers will always nominate a story they like, unless there is a significant drop in quality. While there is nothing wrong with this, it does suggest that a nomination from a user who has never previously nominated this story for the given category should have more weight. Allow me to point out two things before going any further. Firstly, despite what may be expected, nominations from new users make up a significant portion of the total nominations (typically around 50%). Secondly, the nominations of repeat users are not simply being discarded, they play a vital role in the calculation of the growth factor. The weighting is based on a decay function, designed to rapidly increase the weight of nominations from new users relative to that of repeat users. If the same user nominates the same story for the same category 4 times, the weight will be roughly 1/4 that of a first time nomination. This attempts to emphasize a stories ability to draw new readers.
The final calculation for weight therefore is the following:
w=0.85^(R^2)
Where:
w is the weight of a nomination
R is the number of times that this user has nominated this story for this category previously (repeats)

The weight for all nominations is summed to give a final weight:
W=sum(w)
Where:
W is the cumulative weight of all nominations
w is the weight of a single nomination

The final score is the product of weight (W) and growth (G)
S=G*W

As alluded to earlier, there are many other factors that could be considered to improve this adjustment. At this time however, the data required for many of these is not available.

Using this scoring system to factor in multiple attributes, does uphold the original objectives. Firstly, through the inclusion of the growth factor, a story is essentially evaluated against itself, while the weighted nominations evaluates a story relative to other eligible stories. Furthermore it makes stories that were previously assured a position on the ballot truly earn that position each month. This is by no means an impossible task, as many stories that are previous winners or have previously been on the ballot are yet again featured on the ballot. The change however is the significant number of new stories that have been given a place on the ballot (about 40% new stories, exluding Best One-Shot which will always have all new stories).
Action Adventure: 4 new/11
Angst: 5 new/10
Comedy: 2 new/8
Drama: 6 new/12
Romance: 5 new/10
Overall: 3 new/10
Author: 4 new/11
One-Shot: 3 new/3

Overall, there appears to be a decent balance of old and new on the ballot, and all the stories that made it are of an excellent calibre. Allow me to go somewhat off topic for a moment and congratulate all those who were nominated and especially those who made the ballot. Furthermore, the users of this site deserve a pat on the back for the support they have shown their favourite authors in nominating them. Each month, to date, has had more nominations than the last, with the nominations for this month exceeding those of last month by 9%. Futhermore, in a period of less than a day, over 100 users have cast a vote.

I hope this clears up some things, and if you have any questions, we will be more than happy to answer them. I encourage you to share your thoughts on this matter, whether positive or negative, as it is you, the users, that this site caters to.



Viridian came out of the woodwork on 2006.11.02 - 11:43AM to say:

If you want to maintain a balance of old and new stories, setting up divisions (like New Stories, up to 50k, 50-100k, 100-200k, 200K+) would probably serve as well, without the mathematical complexities and giving people the wrong impression. That way, the long-running stories would not be seen as presenting a barrier to new competition. It would also pit authors in roughly the same cohort against each other which would spur them on as well.



fake a smile came out of the woodwork on 2006.11.02 - 11:44AM to say:

Okay, while your goals are applaudable, I think some kinks still need to be worked out of the system. I don't hold that against the site as it's nigh impossible to get this sort of thing right on the first shot, but in reading through the explanation above I have a couple thoughts that I'd like to share.

The first relates to how the Growth is factored. The Growth has one major flaw that I can see at first glance. It can boost brand new stories to outrageous numbers as was clearly evident this past month. Something needs to be done to the equation for new stories to balance it out better. I also don't think that the Maximum value should be used to calculate the Growth. Or rather, I think it should take into account both the Average and the Maximum to better represent how well a story typically does. Perhaps something as simple as an average of the two would suffice.

My biggest complaint with this system is the weighting system. The theory behind what you're doing isn't bad, but in the long run you are going to hurt the nomination numbers you get from the most loyal readers. Take these numbers into account with R being the number of times you've nominated a story in the past, and w as the weight your current vote will get.

R=0 w=1 ; R=1 w=.85 ; R=2 w=.52 ; R=3 w=.23 ; R=4 w=.07

I don't know about anyone else, but it doesn't seem right to me that some votes mean more than others. And the fact that of the matter is that if you've already nominated a story 3 or 4 times in the past, your vote means nothing. What's the point of even nominating the story again in that case? So the more loyal readers stop voting, and that is going to ruin your whole growth system which is operating under the assumption that those readers will continue to vote even when their vote calculates to a weight of less than 0.1

Why should one user's votes count for more than another's? New users to the site will automatically have more say than the older users on which stories ends up in the polls, and that doesn't seem right to me. I don't know how others think, but I believe that each vote should receive equal weight regardless of whether or not that person has voted in the past; otherwise the polls will cease to represent the majority opinion on what the best stories on the site are.

I again thank you all for your time.

-Matt





The Magic Geeks came out of the woodwork on 2006.11.02 - 02:18PM to say:

@Viridian:
Your idea definately has merit. Dividing the stories up into groups retains previous winners without impacting upon newer stories. I think however that it may be preferable to simply segregate based on previous winners and others. Firstly, not that many stories of significant size are in progress. There are under 100 stories that are both in progress and exceed 50k. While many of these are updated each month (~30), considering that not every category applies to all stories the field will be rather sparce, and the ballot will be nearly identical each month. Additionally, if the DSTAs are subdivided into categories, then the number of votes each user has to cast increases (e.g. with 5 categories there would be upto 43 categories to vote on, with two categories there would be 15 categories to vote on)

One of the objectives is for the DSTAs to mean something, and that is greatly lessened if stories make the ballot by default. Either because there is no competition or because they have a very large readership.


@fake a smile:
I agree that the method implemented is far from perfect, and thank you for your understanding in this matter. With regards to the seemingly inflated values for new stories, I do agree that they are rather high, however I do not think that it is significant. For one thing, the score is a meaningless number and isn't designed for direct comparison but merely for ranking. If a story has a score 1 greater than another story or 10 greater, it still gets a higher position. I say this relating to the most notable score from this month which came from a new story. Considering the shift in places for other stories, I again agree that it is a bit much, but ask you to consider it from a slightly different perspective for a second. Firstly ignore the actual number used in the score and look simply at the positioning. Secondly think in terms of merit as opposed to popularity, and keep in mind that there are only around 10 spots on the ballot. Consider the following examples:
Story A, brand new gets 5 nominations
Story B out for 2 months gets 5 nominations
If a story is out for two months, it means that it had a chapter in the previous DSTA period and that people nominated it; this means it has already formed a reader-base. If a new story can outperform an established story I think it should rank higher.

Story A, brand new gets 10 nominations
Story B, out for 3 months gets 12 nominations, only 8 of which are new
Despite story B having more nominations, it has had more time to acquire them and it has fewer new nominations, and I would rank A higher than B. Perhaps one interesting thing would be to compare the new votes directly. While not always true, if you consider the October data, you will find a fairly strong correlation between the order based on score and the order based on new readers. I think however that I see your point. The correlation with new readers appears to devalue to the input of previous readers. I certainly agree that the long time users are what keeps this site going, however without new users this site would have no growth. Considering the situation from a purely analytical point of view, it is much more probable that a reader who liked a story and nominated it will do so again than it is that a new reader will nominate the story. One has already formed a liking for the story, whereas the other has yet to develop even a first impression; one has had time between chapters consider each update separately offseting a poor chapter, while the other ranks the story as a collective whole. I would have suggested 95th percentile, however that is a rather inefficient calculation on this data. Your suggestion of averaging the average and maximum is a good one, and the only other I can think of is average plus 1 standard deviation (because of the skew, adding two standard deviations would generally give a number higher than maximum)

The main contributer to order in this equation was not the weighting nor simply the growth of a story but the growth used for new stories. The original equation used 0.9 (instead of 0.85) and average instead of maximum, and had no capability to significantly alter the positions. Since many users are new, and over a short period, the decay isn't significant (e.g. one repeat is 0.85 vs 1.0); Changing the decay to numbers as low as 0.75 did not have a significant effect.

I think that what would be reasonable is to have four attributes to the equation in this way the different factors would be considered without devaluing any nomination:
1) Growth (G)
Basically as it is now, although with less emphasis on new stories

2) New Nominations (N)
Some variant of the number of new nominations

3) Total Nominations (T)
Some variant of the total number of nominations

4) Penalty for Nominations lost (L)
This is something that hasn't been mentioned before. It is a measure of the ability to retain readers. If a good story begins to lose quality after 10 chapters, it will have acquired a good following, and will continue to get new users, however old users may be turning from the story.

Overall, it may look something like S=G*N*(T-L)

Keep in mind that the DSTAs represent more than just a popularity contest. They seek to improve story quality by motivating authors and rewarding improvement. Additionally, the weighting of nominations in one step of the process does not mean that other nominations are unimportant. The other nominations are factored into growth which is at least as much of a determinant of final position as weightage, consider the overall result more than the effect than any given stage has.

I have added some data to the Trophy Room that may be beneficial to one considering this adjustment.
Please continue to share your thoughts on the matter, we strive to consider as many points of view as possible and to help ensure that SIYE stays the best HPGW site, and continues to grow.



Sovran came out of the woodwork on 2006.11.02 - 02:55PM to say:

I see where this effort is directed, and I think that's laudable, but I'm not sure the results are exactly as planned. Luckily, I have two of my own stories to use as examples in the same category, and I don't want either of them to win, so it's a safe comparison.

This month, in Romance, Meaning of One got 12 raw nomintions, of which 8 were new. A Dream Deferred to 5 raw nominations, of which 5 were new (because it's a one-shot). However, because of the scoring system, ADD placed 4 slots higher than MoO, and ADD made the ballot while MoO did not. It seems counterintuitive that a story with more raw nominations and more new nominations should have a lower score.

The reason, of course, is the comparison of a story's current nominations to its previous max nominations. That creates an effective penalty on stories that get fewer nominations than they have previously. (In my case, I think that's probably because September contained a pretty big moment, while October was slightly more tame. I hope the differences aren't because the quality of my writing has gone downhill.) Now, I think that penalty is too harsh because the new stories are already getting an effective bonus because all of their nominations are new. I think you should do one or the other, but not both, of those things. With both, you're creating a double penalty for existing stories and making it really, really hard for an existing story to get ahead of a new story. In many cases, you're right that it doesn't matter because the same group gets onto the ballot. But do you want to have to face the author that didn't get on the ballot because of that double penalty? We need a system that works all of the time, with provisions or allowances for even the unlikely cases. My two stories are a good example of an unusual case, I think.

I like the idea of a 'bonus' for getting new nominators. But can we get rid of the penalty for having fewer nominations than in previous months? That seems unfair, given the up-and-down nature of various chapters in a multi-chapter fic.



lecook4 came out of the woodwork on 2006.11.02 - 03:54PM to say:

Okay, I feel pretty stupid, and maybe it's becuase I haven't had time to sit down quietly and try to understand the math here, but I have a silly question.

You're probably going to laugh at the simplicity of the question and you've probably already answered it somewhere above in that long detail, but how does it affect those of us with multi chaptered stories. The total votes seem to take into consideration the story as a whole for that month without seperation of chapter. And by that I mean, there are those who submit 3,4 and even 5 chapters in one voting month. I have been only able to submit a chapter every two to three weeks or maybe one chapter in a voting month. So does this penalize me because I only had one chapter while it compounds votes together for each chapter someone else may have written. They may be on an even par with me PER chapter but way ahead because they submitted more chapters and it totalled them together. Is this how it works. If so, I think it should be broken out by chapter and listed by chapter in the polls. It might make a difference. I'm probably not making any sense, but I hope you understand what I mean. Your calculations may already take this into effect, but it doesn't look that way at the moment. I'm not complaining, just curious. If I had four different chapters submitted in October and they each garnered 3 votes, that would give me twelve total fpr the month, whereas someone else might only get 3 votes for thier one and only chapter that month. Although they are even in votes per chapter, I'm way ahead on a total basis. It seems each chapter should be voted on its own merits. Is this the way its done because the polls do not list specific chapters.

Sorry to drag on here, it's late and I'm still feeling the effects of too many toasts at a farewell party last night. Spensers Burundian Ale is nothing compared to the stuff I raised my glass with last night. Ugh! :-)



fake a smile came out of the woodwork on 2006.11.02 - 05:44PM to say:

lecook,

As I understand the system, each chapter is not nominated separately in any given month. Even if I post 5 chapters in a month, my story can only get one nomination from each user (rather than 5 nominations from a user at one per chapter). I agree that if each chapter were nominated separately that they should be treated as separate entities entirely in the polls, but as that is not the case, I believe that matter is a non-issue. I hope that clears up any confusion you might have had. If I am incorrect on any of the above, I apologize. I'm sure the admins will be kind enough to set the matter straight should that be the case.

-Matt





AdminQ came out of the woodwork on 2006.11.02 - 05:58PM to say:

@Fake a smile: for the records, so if anyone reads this and follows along, you are correct, in each point.



rummz came out of the woodwork on 2006.11.02 - 07:26PM to say:

ummm... i am just happy being nominated... thanks.. :D good luck to everybody



lecook4 came out of the woodwork on 2006.11.02 - 07:51PM to say:

fake a smile Thanks for the info. I was afraid that votes were getting compounded due to multi-chapters. I knew you couldn't vote for the same chapter twice, but I didn't realize it carried on over to additional chapters. COOL! That helps a lot. :-) Now I can get back to tending my hangover.



The Magic Geeks came out of the woodwork on 2006.11.03 - 12:47AM to say:

@Sovran:
Firstly, I sincerely appologize for using your stories as my example, however, they do demonstrate a less than ideal outcome, which would be nice to fix.

                                                       Total Score   New  Max    Avg   Months  SD
Oct | A Dream Deferred                                  | 5  | 11.47 | 5  | New | New   | New | New |
Oct | Meaning of One, Part One: The Philosopher's Stone | 12 | 7.76  | 8  | 20  | 12.67 | 3  | 6.6 |
Sep | Meaning of One, Part One: The Philosopher's Stone | 20 | 20.02 | 13 | 14  | 9     | 2  | 5   |
Aug | Meaning of One, Part One: The Philosopher's Stone | 14 | 23.64 | 11 | 4   | 4     | 1  | 0   |
Jul | Meaning of One, Part One: The Philosopher's Stone | 4  | 8.21  | 4  | New | New   | New | New
|

Considering your stories from a purely analytical point of view, the following points arise:
MoO has shown good growth in the months of Aug and Sep
MoO was of sufficient quality, in Jul, Aug, and Sep, to persuade individuals who had not previously nominated it to nominate it, demonstrating that it is capable of attaining new readers
There was a slight drop in the proportion of new users between the months of Aug and Sep, possibly suggesting that the rate of growth is beginning to slow, either due to user constraints (i.e. the site doesn't have enough users to maintain the previous growth rate) or due to story changes (i.e. a change in style, an addition of an undesireable element, a decrease (or lack of improvement) in quality, or simply not delivering a scene in the manner it was built up to be)
In Oct, a marked drop in nominations is evident, and moreover, many of those nominations were from new readers. A story that previously attained 20 nominations in a single month now attained only 60% of that. Over the months, MoO has obtained a total of 50 nominations, from 36 different users. Only 9 (4*2+5*3) users have nominated MoO more than once. The data suggests that the readers were less pleased with the updates of this month than with previous months.
For ADD, there is a limited data set, and a story that attained 5 nominations in its first DSTA month, is quite good.
In comparing the two, while it is true that MoO did receive both more total and more new nominations, it is expected. Stories over time gain a larger reader base, and consequently should receive more nominations, they also increase in popularity attracting new users; this downplays the significance of MoO larger numbers. Basically, MoO performed better than average in Aug and Sep and therefore deserved recognition, in Oct however it performed as expected (relative to other stories, but below expectations relative to its previous record), and was less deserving of recognition than ADD which performed above expectations.

Now that I am done justifying the results, let me plainly state that a) I am quite fond of your writing and b) I agree with your assessment that it is unusual to see a story with more nominations, on both counts, score lower than another story. I still do believe that a penalty is required however to somehow identify stories that are not performing upto the expectations that have created for themselves and c) I agree that a double penalty exists and should be removed.

A modification to the equation of S=G*N*(T-L) proposed above, would be to eliminate the growth factor since that is implicitly considered in the N and L parameters. Perhaps: S=1.1^(N-L)*T, with a 10% bonus for new stories (basically adding one extra new nomination) (Where S is score, N is new nominations, L is lost nominations (those from users who previously nominated but did not repeat), T is total nominations). The simplest form would of course be S=1.1^(T-L) however I do not believe this accounts for new nominations appropriately.

Once again, sorry for using your stories as examples, but hopefully some good comes of it.



Sovran came out of the woodwork on 2006.11.03 - 01:04AM to say:

You're more than welcome to use my stories. They make for good examples, and neither one of them deserves to win Romance, so I'm not attached to the outcome of the nominations or voting in this case.

What I think really needs to be addressed is your statement B. Multi-chaptered stories, like MoO or any other, will have highs and lows in their various categories. Some of my chapters will be more angsty than others, while some will be more dramatic, and so forth. This is true for all chapters of all stories... if it were not, each chapter would contain a climactic moment in all categories, and we'd all go insane trying to make sense of them. By using a comparison between the current month's nominations and the previous max nominations, you're forcing a comparison between the current chapter and a previous climactic moment. My story, for example, had a HUGE climax in September, and October has dealt with the aftermath thereof. That doesn't necessarily mean that October chapters wern't good in whatever categories. . . it just means that they were less striking than September chapters. It's not (I hope) that I'm not performing up to the expectations that I have created for myself. It's that the story is at a different point in its natural cycle.

This could be troublesome. Imagine, for example, that someone writes a chapter showing Harry and Ginny's wedding. Then the next chapter (posted the following month) is the first day of their honeymoon. The honeymoon chapter may very well be the most romantic content submitted to SIYE during that month. But it might also be not quite as romantic as the wedding chapter. Should the honeymoon chapter, even though it's the best Romance on the site for that month, be penalized because it's not quite as romantic as the wedding?

I offer two solutions. Firstly, you could keep the previous G factor, modified as you mentioned to fake a smile, and get rid of the T&L factor completely. Secondly, you could ditch the G factor as you mention above, but use the average number of nominations instead of the max number of nominations when comparing to the current month. Using an arithmetic mean can often be problematic, as I think you said, but in this case it seems like a good factor. Putting out a climactic chapter will increase the average and keep the element of competing against oneself, without forcing new chapters to try to match previous chapters when their content is completely different.

As always, this is just my 2 cents. I have zero problem with out the ballots came out this time in relation to my stories, but I wonder what the reaction would be if the same thing happened to other stories written by two different authors.



Sovran came out of the woodwork on 2006.11.03 - 01:11AM to say:

Ugh... in my last paragraph, the first 'out' should have been 'how'. As Gary Larson once wrote, "It was late and I was tired."



Viridian came out of the woodwork on 2006.11.03 - 09:21AM to say:

" I still do believe that a penalty is required however to somehow identify stories that are not performing upto the expectations that have created for themselves and c) I agree that a double penalty exists and should be removed." There are other factors, aside from story quality, that can influence that to an even greater degree. For example, posting later in the month than you have done previously can create a 'nomination deficit' relative to previous months. If you want to encourage behaviors such as withholding a chapter until the first of the month, that is the way to do it. (I agree that going to quarterly awards for some categories may be a good idea,) As for the 'divisions' suggestion, I didn't mean you had to literally use those breakdowns, just something like them. Do a scatter plot of word counts for works in progress and see where the clumps are. Or, if you want a simpler way to create some artificial diversity, just institute a rule where the winner in a given category is barred from nominations for the next month or so.



hjp74 came out of the woodwork on 2006.11.03 - 01:28PM to say:

ive just read through the comments & they ake sense in general however i may have been nominating wrong as ive voted by chapter for best action/adventure , agnst & so fforth by chapter but for best author & best ovrall i vote for the story as a whole as i fell that give a better impresion of the authors qualaty should i change this. Having said that in the end i belive that quality will till shine through & it will be intresting to see when it comes to actuall voting wether the new store give the more established ones a run for there mony.



AdminQ came out of the woodwork on 2006.11.03 - 04:19PM to say:

@hpj74:
You are safe. Doing so didnt hurt, as no one can vote for the same story more than once.



St Margarets came out of the woodwork on 2006.11.03 - 05:16PM to say:

As JKR would say, "Oh, dear Maths." Since I can't begin to understand the methodolgy behind the penalties and rewards of these Silver Trinkets, I'll just address the primary issue at hand and that is : what will reward/motivate the writer to improve? The short answer is to keep on writing. If the answer is to keep on writing, then monthly awards make sense up to a point. Where the system breaks down is trying to mathematically decide what is worthy and what isn't. Penalizing someone for "not performing to expectations" is a hurtful concept for anyone is who is in the middle of WIP - talk about undermining confidence! Who knows why a fic. doesn't get as many votes from month to month? Maybe their readers forgot to nominate. Maybe they read a chapter at another website. Who can say? Because a writer doesn't know *why* their rank has changed, it's all the more confusing. (and discouraging) Statistics shouldn't be the first or last word on the worthiness of a story. So perhaps we don't need to know how many people nominated a story or what that story's ranking is. Why not put the nominees in alphabetical order and then let the voting begin? After the voting, there will be one winner for that category and that will be that. Yes, there will probably be repeats of the winners from month to month, but that's the way fan fiction goes. There are "hot" fics. that capture everyone's imagination and they will end up with a ton of reviews and awards. That's okay. The authors deserve them since there's a lot of pressure in writing a WIP that is being scrutinized by so many. In the end, however, I'm opposed to competition among writers - unless it's an actual challenge that we knowingly enter. I write my stories because I love H/G and I love to share with my readers. I don't submit a chapter thinking I'm a better or worse author than anyone on this site. I think that's why I hate those ranking so much. When I see my name at the bottom of a list, my first thought is to think - boy, I must suck. Then I have to stop and remind myself that I don't suck as a writer. (And neither do any of the people down there with me) What I'd really like to see at this site are some staff choices or readers recs. each month - sort of like the Nifflers on Fiction Alley. This casual "check it out" kind of thing could really help the newbies out there. Just a thought. Anyway, I'm happy to see SIYE trying so hard to support writers. It's wonderful to see so many new things being tried out .



AdminQ came out of the woodwork on 2006.11.04 - 03:59PM to say:

St Margarets - Want a task? Email me with how you would do the contest, and we talk over some points. Thanks for adding your thoughts to this. Steve



lecook4 came out of the woodwork on 2006.11.04 - 04:45PM to say:

I have to agree with St Margarets in sentiment. I still have question about prolific authors being given more of a chance in winning over those who have longer times between posts. Each new chapter has the possibility of being read by a whole different crowd and garnering new votes, thus giving the more prolific writer a chance at winning with more votes. This is because the votes go into a tally for the whole story for the month and not just that one chapter. I would like to see each chapter stand on its own merits and not get combined with others for the monthly total. Each chapter submitted in a month could be voted on by any reader, but each vote would only go toward the chapter and not the story in total. The only ones that should be for the story combined should be Best Author and Best Story. This could be annually and be the top winners of each month gathered and voted on. Each chapter should be submitted and voted on its own merits alone. Just a thought and probably in the minority. LOL! I'm not thinking for me, heck I'm honored just to get a review, but just trying to put ideas out there for brain fodder.



hpmaniac666 came out of the woodwork on 2006.11.07 - 10:57AM to say:

Ok, this idea only just came to me so bear with me. Why not use a sort of league system. This will acomodate for both giving frequent winners recognition and allowing new athours to be exposed. For example, athours which win one month in the awards are promoted to a 'premiere league' where they are set against other frequent winners. Each month, the three stories receiving the most votes are promoted to the premier league while the three stories/athours with the least votes in the premier league are demoted and are not eligiable in the next months awards, after which they once more have the opportunity to work theire way up to the premier league. Obviously its a rough idea that needs a lot of fidding, but I think it could work. I also think more categories is a good idea, particularly in helping one shots gain exposure. Perhaps sub categories of one shots would be a good idea. Just a thought, anyway. xx congrats to the nominees



intromit came out of the woodwork on 2006.11.07 - 06:45PM to say:

Wanting to attract new blood into the DSTA is a laudable goal and I applaud your efforts - something does need to be done. However, a vote counting less each time it is made is simply insulting regardless of how good the math is. Further, the fact that a story does not receive the same of nominations per month somehow is directly related to quality is also false. Some chapters, no matter how well written, simply do not spark the same emotional response as the previous - and are not meant to. Mark Twain said it best, "There are three types of lies: lies, damn lies, and statistics." Inflating a story's position because it receives new nominations is a bad choice - it speaks in no way to quality or necessarily to relevance to that particular category (there are some odd people out there with equally odd ideas as to what romance, action, or comedy are). Many recommendations have already been made from a story not being able to win twice in a row to having quarterly, biannual, or annual awards, and more categories - there are many other options. No matter your choice, not everyone is going to like it, but disenfranchising is not the answer.
I would like to congratulate you on adding the Best One-shot category. I enjoyed being able to vote for One-shots separately - they are easy to forget if posted early in the month and it gave me a reason to read some fantastic works. I'm looking forward to seeing what solutions you come up with.



Mojomig came out of the woodwork on 2006.11.30 - 06:24AM to say:

I'm new to posting fics on SIYE and having been extremely fortunate enough to get nominated for a DSTA (Many thanks for that btw). I've looked at the trophy room data to find out how it all works. Even though I am an analyst in the Financial Services industry, I couldn't make much sense of it. All the explanations above help a little but it still seems way too complicated. I think I agree with St Margarets point of view. Anyone who gets 3 or more nominations for a category in a month goes down on the list. The voters then vote for one person in each category and whomever gets most wins. If the same stories win, its because they're the best stories. The motivation for me as a new writer with no fan base is to keep writing until I get a big fan base and maybe then I might win an award. I don't expect to be able to compete with the established, popular authors as that's just unrealistic, for someone with only a few weeks experience. Well anyway, must be getting back to work...



Quees came out of the woodwork on 2013.07.19 - 05:53AM to say:

qualified as professional instruments for diving and aviation purposes fake breitling chronomatic watches.bull.The smart watch will have quirky functions, which in some cases have nothing to do with timekeeping, fake rolex at all. These are for the guy who loves his gadgets.bull.And finally, the pocket watches cheap rolex watches . fake rolex A pocket watch is a rare sight these days cheap watches , but the man who carries one is, fake tag monaco watches at the very least, nostalgic. These are traditionally very elegant and charming timepieces, and are still quite fashionable as well as very reliable replica watches for sale . They are reminiscent of antiques, yet are currently in production.So, what type of watch has caught your eye. replica audemars piguet automatic watches Are you looking for a perfect watch for your kid. No matter your kid is in the process of learning to tell time or your kid is an independent boy or girl, a watch is a necessity in their daily life. As is wellknown, the


../back
‘! Go To Top ‘!

Sink Into Your Eyes is hosted by Grey Media Internet Services. HARRY POTTER, characters, names and related characters are trademarks of Warner Bros. TM & © 2001-2006. Harry Potter Publishing Rights © J.K.R. Note the opinions on this site are those made by the owners. All stories(fanfiction) are owned by the author and are subject to copyright law under transformative use. Authors on this site take no compensation for their works. This site © 2003-2006 ALL RIGHTS RESERVED. Special thanks to: Aredhel, Kaz, Michelle, and Jeco for all the hard work on SIYE 1.0 and to Marta for the wonderful artwork.
Featured Artwork © 2003-2006 by Yethro.
Design and code © 2006 by SteveD3(AdminQ)
Additional coding © 2008 by melkior and Bear