Home « Forum « Author Hangout

Forum: Author Hangout

Blogs

red61544

The blogs are bitching about the scoring system again. Why doesn't someone develop an immunization that all authors are required to take that eliminates their ability to bitch about scoring? (They'd probably start bitching that it causes autism.) I propose a new scoring system that is reader friendly - if you finish the story in one month, you get a ten. More than a month rates a 1. That would piss everybody off and they could all blame me! C'mon guys, quit your bitching - we've heard it all before! No rating system is going to please everyone unless he is guaranteed top ratings! So live with it!

Dominions Son

@red61544

But then we'd have to find something else to bitch about.

Replies:   richardshagrin
richardshagrin

@Dominions Son

See the thread about an Australian named Hicks. Relevance to SOL or any story that might have been on SOL, if we were lucky, is very close to zero.

And then there are the threads about politics and religion.

Lets start one now. You hear about anti-heroes but why not about uncle-heroes?

Dominions Son

@richardshagrin


Lets start one now. You hear about anti-heroes but why not about uncle-heroes?


Anti, not auntie.

I'm writing a story with an anti-villian.

Ernest Bywater

@richardshagrin

See the thread about an Australian named Hicks. Relevance to SOL or any story that might have been on SOL, if we were lucky, is very close to zero.


The thread that discussion is on is about Military tribunals, which is why it got started, as there is a relevance in regards to the relevance of a military tribunal hearing a case against a civilian. That's how it got there, and then drifted a little, and the only time I've seen a thread not drift is heavy discussions on scores

I don't bitch about the scores because I understand them, their purpose, and how they apply to all evenly, so they're fair. What I don't like is the advice to over vote to pump a score, because that then devalues all the valid scores given up to that time.

Replies:   REP
awnlee jawking

@red61544

I find it fascinating to study readers' voting patterns and, dare I say it, the various ways an author can game the system to improve their stories' ratings. I'm a long way from understanding all the phenomena so I find the viewpoints of others interesting and sometimes educational.

AJ

aubie56

@red61544

@red61544

Why should I just live with it? I am sick and tired of receiving 5 votes less than a score of 6 out from a total of around 260 votes and having that give me a score of 6.93 instead of the 8.34 that the basic raw scores amount to. I don't give a shit about whether or not it happens to anybody else, I care that it happens to me.

I'm not trying to inflate my scores so much as I am trying to get Lazeez and crew to see that the scoring system hurts authors. Sure, I am supported by 5,000 or so readers, and what I am annoyed about is that I often get a message that says something to the effect that "Why are you getting such a low score when I gave you a 10?" Well, I wrote my blog for those people and not for the people who consistently get 8.0-8.6 ratings for what I consider boring drivel.

I would probably quit writing for publication on the web if I did not get some many emails asking me to keep it up. The scores are just damned depressing! So you live with it!

Ernest Bywater

@aubie56

The scores are just damned depressing! So you live with it!


Aubie, if they depress you, just ignore them. The scores are for the readers, not the authors. That's why I ignore them, except when checking something for a discussion about them.

red61544

@aubie56

You should just live with it because everyone is tired of hearing you and others just bitch about it. There is no possible scoring system that will please everyone! How about a system that asks: "How would you rate this story as compared to other stories in this genre which you have read?" That would put most of your stories up against Woodmanone and you'd still be pissed off! Instead of bitching, design a ratings system and submit it on the forum for the other authors to critique. The system wasn't designed to reward authors; it was designed to allow readers to discern which stories may interest them. If you want a system that rewards authors, let the reader award the author from one to ten "attaboys".

Dominions Son

@aubie56

I am trying to get Lazeez and crew to see that the scoring system hurts authors.


Lazeez as said he has modified the score system multiple times based on author complaints and he isn't going to entertain anymore changes.

Why should I just live with it?


Because you can't do anything about it?

Maybe this will help:

God, grant me the serenity to accept the things I cannot change,
Courage to change the things I can,
And wisdom to know the difference.

Replies:   Ross at Play
Crumbly Writer

@aubie56

Why should I just live with it? I am sick and tired of receiving 5 votes less than a score of 6 out from a total of around 260 votes and having that give me a score of 6.93 instead of the 8.34 that the basic raw scores amount to. I don't give a shit about whether or not it happens to anybody else, I care that it happens to me.

In my case (I never took Lazeez's pledge never to bitch about scores, because I knew the topic would continually reappear), I frequently evaluate sites to select which are best for posting stories to. When I see the same stories getting different scores (even rated differently), I tend to apply those results to the site, rather than to the scoring adjustments. Thus I, and others like me, might very well pull stories from a site because we no longer feel the stories are a good fit on the site.

I'm also a stat follower, however I don't evaluate the raw scores, instead like most economists, I look for changes in scores. Whatever the score, if the score dips for a particular chapters, chances are I did something wrong. If I understand there's something upsetting (or particularly exciting) about a given chapter, I can account for it, but if it's a mystery, then chances are I missed something about the story which I need to account for and possibly make changes to the story. That's something you can't do with book sales where readers either buy the book or they don't (and they often don't read the ones they do buy, too).

Hope that helps frame the discussions a little more.

Wheezer

I wish the scoring was more like it was when I first joined SOL. I would love to be able to look at a story score and know if it got low marks because it was a really poorly told story, or if it just contained terrible writing technique that made an otherwise good story unreadable.

Replies:   Ernest Bywater
Ernest Bywater

@Wheezer

I would love to be able to look at a story score and know if it got low marks because it was a really poorly told story, or if it just contained terrible writing technique that made an otherwise good story unreadable.


If the story has a review it does most of that for you.

Replies:   Grant
Grant

@Ernest Bywater

I would love to be able to look at a story score and know if it got low marks because it was a really poorly told story, or if it just contained terrible writing technique that made an otherwise good story unreadable.



If the story has a review it does most of that for you.

However the number of reviewed stories is even smaller than the percentage of people that score a story once they're read it.

grandad_rufus

I rarely give feedback scores now, I the options given from "You Call This A Story!?" through to "Most Amazing Story", are either meaningless, or highly personalized. A simple 1 to 10 would be better though 1 to 5 would probably be clearer to an author.

Replies:   Ernest Bywater  Grant
Ernest Bywater

@Grant

However the number of reviewed stories is even smaller than the percentage of people that score a story once they're read it.


True, but why not look for and use it when it's there?

Replies:   Grant
Ernest Bywater

@grandad_rufus

"Most Amazing Story"


An appropriate score for what any politician on the electoral trail tells you about what they've done during their last term in office, and what they plan to do during their next one.

Grant

@grandad_rufus

the options given from "You Call This A Story!?" through to "Most Amazing Story", are either meaningless, or highly personalized. A simple 1 to 10 would be better

That's what they translate to.

Replies:   REP
Grant
Updated:

@Ernest Bywater

However the number of reviewed stories is even smaller than the percentage of people that score a story once they're read it.


True, but why not look for and use it when it's there?


I do, but they are very few & far between- particularly on the older stories. It is good though to see more reviews being done for the more recent/current stories.

Ross at Play

@richardshagrin

See the thread about an Australian named Hicks. Relevance to SOL or any story that might have been on SOL, if we were lucky, is very close to zero.

SURE, absolutely no relevance to SOL.
The significant difference with that exchange to the many about American politics is this has been a polite exchange of information and opinions about something of interest to a few here
And it will be wrapped up soon with no harsh words about other participants in the exchange.

Ross at Play

@aubie56

I am sick and tired of receiving 5 votes less than a score of 6 out from a total of around 260 votes and having that give me a score of 6.93 instead of the 8.34 that the basic raw scores amount to.

I think you are saying it is the 'outlyers' among scores that disturb you.
There are many sports where the top and bottom 10% of scores are eliminated before any averages, etc. are calculated.
I would VERY MUCH approve of that happening here, and I'm sure it would not be a particularly difficult fix for Lazeez to program into the system

Ross at Play
Updated:

@Dominions Son


Lazeez as said he has modified the score system multiple times based on author complaints and he isn't going to entertain anymore changes.

___Why should I just live with it?

Because you can't do anything about it?
Maybe this will help:

___God, grant me the serenity
___to accept the things I cannot change,
___Courage to change the things I can,
___And wisdom to know the difference.


... But Lazeez is God! ... so this must be one of those things that can be changed.

Replies:   Dominions Son
Switch Blayde

@Ross at Play

There are many sports where the top and bottom 10% of scores are eliminated before any averages, etc. are calculated.


That's part of the SOL scoring system, although I believe it's 5%. Lazeez put that in at the request of authors.

Replies:   Ross at Play
Ross at Play

@Switch Blayde

Great. I have NEVER whinged about the scoring ... said the "writer" who has yet to post a story.

Dominions Son

@Ross at Play


... But Lazeez is God! ... so this must be one of those things that can be changed.


But he has already decreed that it will not change again.

Replies:   red61544
red61544

@Dominions Son

But he has already decreed that it will not change again.

...and, according to Christian doctrine, God is immutable!

Replies:   Ross at Play
Ross at Play

@red61544

immutable

I know he never shuts up. What's that got to do with this?

Replies:   Dominions Son
docholladay

I always preferred my method of voting. I send a thank you note to the writer when I read a story.

Replies:   Crumbly Writer
Ernest Bywater

@Ross at Play

difficult fix for Lazeez to program into the system


It already happens, the top and bottom 5% (in whole numbers rounded down) are cut from the raw scores before the total is averaged for use in the score.

Dominions Son

@Ross at Play

I know he never shuts up. What's that got to do with this?


On a more serious note, Lazeez declared an end to changing the scoring system, because he was sick of the authors constantly whining about it.

That some authors think that continuing to whine about it will change his mind in a positive way seems just a tad delusional to me.

Replies:   red61544  docholladay  REP
red61544

@Dominions Son

That some authors think that continuing to whine about it will change his mind in a positive way seems just a tad delusional to me.


Amen!!!

docholladay

@Dominions Son

That some authors think that continuing to whine about it will change his mind in a positive way seems just a tad delusional to me.


There is no way any scoring system could ever satisfy every writer or critic. As a reader I am extremely grateful for the time and effort that has been put forth in making this a great story site.

Replies:   red61544  Capt Zapp
red61544

@docholladay

There is no way any scoring system could ever satisfy every writer or critic. As a reader I am extremely grateful for the time and effort that has been put forth in making this a great story site.


Again, I say Amen. Eventually, if the bitching continues, Lazeez will get disgusted enough to just quit on us. I appreciate this site and to bitch at its creator because an author's feelings are hurt by the scores he received is unconscionable. That's akin to an athlete blaming the referee when he has a bad game.

Replies:   Ross at Play
Switch Blayde

Some comments extracted from a Wikipedia article on grading on a curve. I'm not passing judgment. I just thought it was interesting and maybe relevant.

"The absolute values are less relevant, provided that the order of the scores corresponds to the relative performance of each student within the course."

"The ultimate objective of grading curves is to minimize or eliminate the influence of variation between different instructors of the same course, ensuring that the students in any given class are assessed relative to their peers."

"Curved grading can...affect their sense of faculty fairness."

Replies:   Ernest Bywater  REP
Capt Zapp

@docholladay

The only change I wouldn't mind seeing is a count of each score on the Detailed Info and Stats page. It already shows the number of the highest vote but trying to figure out how many of the other votes you have is almost impossible. If the bar graph included the count of each vote, it would make it easier for us to calculate the raw score average ourselves. Of course, that wouldn't change what the readers see, but it might make some authors feel better about their scores.

Ernest Bywater

@Switch Blayde

The ultimate objective of grading curves is to minimize or eliminate the influence of variation between different


At that's the basis behind most of the current system, to provide a valid relativity between the different scoring systems used in the past so they can be brought together to give one valid relative score across all the scoring periods and systems.

Replies:   Switch Blayde
Ernest Bywater

@Capt Zapp

I wouldn't mind seeing is a count of each score on the Detailed Info and Stats page


top right corner of most pages on SoL has a webmaster link, and it has an option for Feature Request - it's there to make these suggestions by.

Switch Blayde
Updated:

@Ernest Bywater


At that's the basis behind most of the current system, to provide a valid relativity between the different scoring systems used in the past so they can be brought together to give one valid relative score across all the scoring periods and systems.


That may be an ingredient to it, but the main reason was to not give so many As (or in our case 9s and 10s).

This is from another article on grading on a curve:


That fatal flaw is found in the requirement that any area in the positive half of the curve must be duplicated in the negative. A true bell-curve grading scheme requires that instructors produce equal numbers of "A"s and "F"s, equal numbers of "D"s and "B"s, with the largest area reserved for the middle "C"s.


But it is what it is. Now that I can see the raw scores I'm satisfied.

EDITED TO ADD:

The article with the above quote was written in response to someone complaining they got a 96% on a test and received an F (everyone else in the class scored higher).

Replies:   Ernest Bywater
awnlee jawking

@Capt Zapp

The only change I wouldn't mind seeing is a count of each score on the Detailed Info and Stats page.


Me too. I thought I was missing something because some authors already seem to know this information, but I never got around to asking how.

AJ

Switch Blayde
Updated:

@awnlee jawking


I thought I was missing something because some authors already seem to know this information, but I never got around to asking how.


It's on a case by case basis. If an author agrees not to complain about the scoring system, Lazeez will provide that author the ability to see the raw scores.

Replies:   Ernest Bywater
Dominions Son

@awnlee jawking

Me too. I thought I was missing something because some authors already seem to know this information, but I never got around to asking how.


From the home page, click on the Authors/Editors link then on Manage Stories.

If you click on one of your stories in the list on that page, you get a page of statistics for that story.

That stats page includes a histogram of the raw scores.

Granted, it's not exact for every bar, but it gives you a good feel for the distribution of your votes.

Ernest Bywater

@Switch Blayde

That may be an ingredient to it, but the main reason was to not give so many As (or in our case 9s and 10s).


Actually, from what I remember of all the past blog posts Lazeez had on this before they got eaten by a dying server is a little different. If I've got this wrong, I'm sure I'll get jumped on and corrected by someone. Also, I'm sure the reality is a lot more complex, but this is really a simplified explanation.

From what I remember the history of the situation is what shaped the current system and the key factors are:

1. Provide the readers with a measure of relative worth to other readers about how they liked the stories.

2. Different scoring systems in the past, due to the system evolving since the site started. Of which there has been four or five or more - I'm not sure.

3. Retian the old scores from the earlier systems and create a way to make them relevant to each other.

4. A need to eliminate most of the fanboi and hate scores.

The way this was done, from what I remember, was to select a value and normalize all the scores in each period using that value as the mean for all the scores in each period. This was applied to every score given to a story. Also, when the three aspect scoring system was killed only the Appeal score was used. I think the chosen mean was 6.

So each scoring period was processed by itself to rate all the scores given in that period relative to each other, then the mean of those scores calculated. From which a value was calculated to have a mean of 6, and that value applied to all the scores to adjust them to have a mean of 6. This was done to each the scores of each scoring period as a separate calculation so each period now had a mean of 6. This allowed all the scores given to now have a valid value relative to each other, and meant they could now all be used in a single calculation. In short - this action took a bunch of mixed fruit scores and made them all apple scores on a grade curve for a fair comparison.

The next stage was to calculate the individual score for each story. That is done by removing the top and bottom 5% of the scores (but only in whole numbers, so they get rounded down to whole numbers), and an average for that story is calculated from what's left - and that's the official score.

There are two aspects of the system that do cause some temporary anomalies.

First the rounding down means that when the votes get to a certain point they do not cause another score to be dropped until the number of votes makes it another whole number. Example: 80 votes means 4 votes are dropped at each end, this continues to happen up to 99 votes, and only when 100 votes are in does the number dropped move over to 5 votes being removed at each end. This will see a sudden change bigger than expected by one vote.

The other is a vote at either end of the spectrum results in a change in the value of the votes being dropped. Example: Most of the scores are grouped in the middle or top end. there are 87 votes so the top and bottom 4 are dropped. The story has 3 votes of 1 and the next is a vote of 5. This means the 3 x 1 and the 1 x 5 are dropped. A new vote of one means the 4 x 1 are dropped and the 1 x 5 is now counted. Thus a 1 vote can put the score up by moving the cut off further down the curve. Similar anomalies can happen at both ends.

An example of a combined application to see a score drop is: 99 votes in place that includes 7 votes of 10, 4 votes of 1, and 2 votes of 6. Thus 4 x 10 and 4 x 1 are dropped. A new vote of 10 is made for the 100th vote, now 5 x 10 are dropped along with 4 x 1 and 1 x 6. The total to be averaged goes up by 10 from the new vote, but loses another 10 from the cut, so no change to the total for that, but the drop of the low end now includes a 6 so the total drops by 6 while the average is now done by 90 and not 89. Thus the official score will drop due to a vote of 10 at an important time. Say the story raw scores are 99 votes for 580 giving a score of 5.89 (since the lower decimals aren't rounded) calculated by 536 divided by 91 due to 4 scores at each end being dropped - 4 x 10 and 4 x 1 removing 44 points. The new vote of 10 makes the raw score 100 votes for 590. Now ten votes are dropped: 5 x 10, 4 x 1, and 1 x 6 removing 60 to give 530 divided by 90 to give 5.88 (since the lower decimals aren't rounded so what comes after the 2nd decimal doesn't show.) The anomalies are always short lived and usually vanish at the next vote.

The two key things to remember are the scores are there solely for the readers and they affect all the authors in the same way, so the system is fair.

Ernest Bywater

@Switch Blayde

It's on a case by case basis. If an author agrees not to complain about the scoring system, Lazeez will provide that author the ability to see the raw scores.


My understanding of the offer was they would be given access to all the raw scores in their full glory. While any author can see the bar graph located by going to the Stats page and clicking on the story.

Replies:   Switch Blayde
Switch Blayde

@Ernest Bywater

My understanding of the offer was they would be given access to all the raw scores in their full glory.


Not exactly. I don't see all the raw scores; only the average of those scores. They appear in a column called "VA Scr" just to the right of "Score" (the SOL score given to the story).

The bar chart referenced before tells you how many 1s, 2s, etc.

Replies:   Capt Zapp
REP

@Ernest Bywater

What I don't like is the advice to over vote to pump a score, because that then devalues all the valid scores given up to that time.


Not sure what you mean by 'over vote'. My impression is that submitting a 'second' vote just changed the score you entered on your first vote.

Or did you mean 'submit an excessively high score'?

Replies:   Ernest Bywater
REP

@aubie56

From my understanding of Lazeez weighted scoring system, it seems very subjective.

1. I understand he tries to adjust the score for readers who don't vote. I don't understand how this can be done fairly since he doesn't know the opinions of those readers. Besides, they had the opportunity to register their vote and decided to not do so.

2. I understand that readers often give a story a score that is higher than it deserves, but there are also readers who give the story a score lower than it deserves. I can't think of a fair way to adjust a story's score for inaccurate scoring.

Personally, I think the average of the raw scores is the only fair way to rate a story. At least it will be an accurate reflection of the scores given to all writers' stories. So what if a story's average come out high, it is just an indication of reader satisfaction.

REP

@Grant

That's what they translate to.


Did you happen to notice 'Good', which is normally considered the average value is rated a '7'. My personal thought is that 'Good' should be a 5 on a scale of 1 to 9 or 0 to 10.

Switch Blayde

@REP

1. I understand he tries to adjust the score for readers who don't vote.


I don't want to speak for him (or is it Him?), but I believe it's because when you like a story you give it a high score; but when you don't like a story you probably don't finish it so you don't give it a (low) score. That skews the scores to the higher end.

As I said, I can now see the real average and distribution of scores given. I hardly ever look at scores after the story has been out there for a week or so, but if I do that's what I look at. I like sorting my stories by the "VA Scr."

REP

@Dominions Son

continuing to whine about it will change his mind in a positive way seems just a tad delusional to me.


Yeah! Add in that the positive whining probably wants him to change the system in two or more incompatible ways would explain why he said No More Changes!

REP

@Switch Blayde

Using the curve can be unfair also.

Assume 100 students earn raw percentage grades within the range of 90-99%. The students receiving 90% would fail the test/course.

Bondi Beach

@richardshagrin

Lets start one now. You hear about anti-heroes but why not about uncle-heroes?


There's a very simple reason: auntie-heroes are a staple of Indian porn. There are no uncle-heroes in that genre.

bb

Capt Zapp

@Switch Blayde

The bar chart referenced before tells you how many 1s, 2s, etc.


Actually, the only number you know for certain is the vote with the highest count. Once your vote count gets up over about 100 (depending on the distribution), you have to 'guesstimate' and the bars become meaningless for the other stats.

Replies:   Switch Blayde
Ernest Bywater

@REP

Not sure what you mean by 'over vote'


at one time, well in the past, no information was kept about who voted on what story beyond a few weeks. Thus it was possible for someone to vote on a story a number of times, as long as they left a long enough period between votes - so one person could vote multiple times on the same story. That got changes some years ago, and is no longer possible.

Replies:   REP
Ernest Bywater

@REP

scoring system, it seems very subjective.


All votes are subjective by the reader. The problem was coming up with a way to relate scores given under different scoring systems to each other. It was either find a way to bring them all to the one norm or dump all the older scores and start again. he chose to normalize them via the system explained earlier.

The funny thing about this is I'm one of the few who gets consistently high end scores on most of their stories, but don't mind the system cutting them back a bit to normalize them. That's because the system affects everyone in the same way and all the scores are in the same relative position to each other, regardless of what they actually read.

Ernest Bywater

@Switch Blayde

That skews the scores to the higher end.


That's not always the case. I've had people email me to tell me they voted the score down without finishing it because they didn't like either present tense or vernacular writing. But I'm sure there are many people who don't vote when they dump a story partway through - although I'll vote a story I dump as to how I rate it.

Crumbly Writer

@Grant

However the number of reviewed stories is even smaller than the percentage of people that score a story once they're read it.

Especially since many reviewers have a policy of "never review the same author twice", despite how many books they've written or how good his/her writing is.

Crumbly Writer

@docholladay

I always preferred my method of voting. I send a thank you note to the writer when I read a story.

And thank you for that. We appreciate it. Scores are open to interpretation, but thanks are fairly universal.

Crumbly Writer

@REP

1. I understand he tries to adjust the score for readers who don't vote. I don't understand how this can be done fairly since he doesn't know the opinions of those readers. Besides, they had the opportunity to register their vote and decided to not do so.

He's never claimed this. Instead, most scores tend to be overinflated (too many 10 votes for average stories) so he downgrades most 'typical' stories to produce meaningful scores.

Authors and a few readers object when they vote 10s and see a resulting score of only 7. Personally, my objection is not knowing whether a lower score on a particular site (say FS) is because the readers there don't appreciate a particular story, or the one story got graded on a curve at that particular point in time.

Replies:   Dominions Son
Crumbly Writer

@Switch Blayde

As I said, I can now see the real average and distribution of scores given. I hardly ever look at scores after the story has been out there for a week or so, but if I do that's what I look at. I like sorting my stories by the "VA Scr."

It almost becomes a game. Once a story is initially scored, it rarely changes much beyond that initial score, so an author's best bet is to start off with a bang, presenting their most exciting chapters first--which is a traditional approach for most literature--rather than waiting for the story to be scored at completion.

I've also noted that scores drop significantly when the story is done. Fans will vote a 'final' score just after a story is completed, but those who never felt compelled to read the story will read it once it finishes, and generally post lower votes, bringing the score down. It will generate up over time, but those initial scores are essential to the ultimate score.

Replies:   Ernest Bywater
Dominions Son

@Crumbly Writer

my objection is not knowing whether a lower score on a particular site (say FS) is because the readers there don't appreciate a particular story, or the one story got graded on a curve at that particular point in time.


Shouldn't you be able to figure that out from looking at the score histogram?

Replies:   Crumbly Writer
Crumbly Writer

@Dominions Son

Shouldn't you be able to figure that out from looking at the score histogram?

That doesn't account for why a single story does better or worse. It may be that the story wasn't appreciated, or it may have been the site's normalizing at that particular time (say when Lazeez dumped a bunch of professional historic stories to boost the sites content).

richardshagrin

@Crumbly Writer

I could be wrong, but I think my job as reviewer, in addition to bringing good stories that haven't been reviewed, at least fairly recently (the last 5 or 10 years), to the readers of the review (not sure how many do, but I am reasonable certain its far less than 100% of everyone who looks at the front page that day) but also by doing so motivate a few to look at the author's page(s) where the story appears and perhaps select another or more than one.

How does it do most of the SOL audience any good to review a story on the 50 all-time greats list? I need to find good ones to review that otherwise might not be read. It doesn't have to be from 1999 with no story description, although there are some. Take a look at my reviews as Barron of Ideas (aka Barren of Ideas) to see that sometimes I have reviewed two stories by an author, and sometimes in my review I indicate there are lots of other good stories by the same author. Just to prove I am inconsistent, I have reviewed a story by Don Lockwood, the once and future Frank (redacted).

What do you guys think? Or ladies. Does it make sense to review a lot of stories by the same author, pointing out a lot of the same things every time. Or should my time, and your time reading reviews point more to the less known stories that you might otherwise never see.

Dominions Son

@Crumbly Writer

It may be that the story wasn't appreciated, or it may have been the site's normalizing at that particular time (say when Lazeez dumped a bunch of professional historic stories to boost the sites content).


The histogram appears to be the raw votes. If it were of individual votes but after adjustments, then the score histograms on my stories make no sense.

Take my story: http://storiesonline.net/s/73034/son-of-chronos-book-1-origin

Votes 554
Score: 7.84

Scores from histogram in declining order 10(181), 8, 9, 7 6, 5, 1, 2, 4, 3

Given how you describe the scoring adjustments, if the histogram was based off of already adjusted individual scores, I wouldn't expect it to be possible for any story to skew all the way to the top end like that.

Ernest Bywater

@Crumbly Writer

"never review the same author twice",


I thought it was 'never review the same story twice' not author. I know a few reviewers who review a number of stories from the same author.

Replies:   Crumbly Writer
Ernest Bywater

@Crumbly Writer

I've also noted that scores drop significantly when the story is done.


That's odd, mine tend to go up over time, as more people read them and the numebr of votes goes up..

Replies:   Crumbly Writer
Ernest Bywater

@Dominions Son

The histogram appears to be the raw votes.


that's my understanding of it.

Crumbly Writer

@richardshagrin

I could be wrong, but I think my job as reviewer, in addition to bringing good stories that haven't been reviewed, at least fairly recently (the last 5 or 10 years), to the readers of the review (not sure how many do, but I am reasonable certain its far less than 100% of everyone who looks at the front page that day) but also by doing so motivate a few to look at the author's page(s) where the story appears and perhaps select another or more than one.

I wasn't necessarily objecting to that position, merely stating it as a semi-official standard denoting just how few reviews there are. With so few reviewers and so many stories, you've got to pick and choose. The point was, the reviews are not necessarily apropos of what's out there.

Crumbly Writer

@Ernest Bywater

I thought it was 'never review the same story twice' not author. I know a few reviewers who review a number of stories from the same author.

In that case, I stand corrected. That wasn't my understanding, probably because it's not an official policy but just the decision by individual reviews to act in concert.

Crumbly Writer

@Ernest Bywater

That's odd, mine tend to go up over time, as more people read them and the numebr of votes goes up.

It's an interesting dynamic. Initially, after the story completes, the scores jump for one to two days, then they descend as new readers discover them to be not to their liking, but over time they tend to drift back up again.

What I'm suggesting, is that stories that score low tend to remain there, rather than they might if posted at the same time. It's like in recent politics. Despite polls consisting of only a tiny fraction of any given representative: those still owning home phone who don't immediately hang up (i.e. old people eager to talk to anyone), the polls remain highly accurate. That doesn't speak well for the polls, instead it suggests that the polls influence voters more than it reflects how voters feel at any given moment. Voters and readers both want to fit in, so if everyone else thinks a story isn't their cup of tea, most others will agree, rather than sticking their neck out and saying "No, this is a 10" or "Shit, this is only a 4 at best".

That's not to say that poor scoring stories don't belong there, it's just my view that polls and scores are subject to what's known as 'observational bias', a factor that's rarely taken seriously by the press or studied enough to document it.

Switch Blayde

@Capt Zapp

Actually, the only number you know for certain is the vote with the highest count. Once your vote count gets up over about 100 (depending on the distribution), you have to 'guesstimate' and the bars become meaningless for the other stats.


Not anymore. There's a number at the top of each bar. Maybe that's only for those who can see the "VA Scr" on the Stats page.

Ernest Bywater

@Switch Blayde

Not anymore. There's a number at the top of each bar.


Nope, there's only the value of the highest on my graphs. But I never promised not to discuss the scores, saw no need to.

Replies:   Switch Blayde
Switch Blayde

@Ernest Bywater

Nope, there's only the value of the highest on my graphs. But I never promised not to discuss the scores, saw no need to.


Then you can't see the "VA Scr" on the Stats page either. They must go hand-in-hand.

As I said, Lazeez made me happy by providing it. I don't complain about the scoring system anymore.

Replies:   Ernest Bywater
Ernest Bywater

@Switch Blayde


Then you can't see the "VA Scr" on the Stats page either.


No I don't see that. I never complained about the scores or the stats page, I only ever notice them during these discussions. So i never saw a need to worry about it.

samuelmichaels

@Crumbly Writer

That doesn't account for why a single story does better or worse. It may be that the story wasn't appreciated, or it may have been the site's normalizing at that particular time (say when Lazeez dumped a bunch of professional historic stories to boost the sites content).

Don't know about FS since it has a smaller volume, but if I recall correctly SOL uses a 1-year interval for "normalization periods", which translates to thousands of stories. It would take a lot of very-high-scroring posts to affect the curve on SOL.

Replies:   Crumbly Writer
Grant

@richardshagrin

Does it make sense to review a lot of stories by the same author, pointing out a lot of the same things every time.

Why not?
If you've read a story & feel it's worthy of a review, review it.
To me it's not about doing a review of a story by an Author, but reviews of stories, by many authors; or even the same author if they have written many stories.

Capt Zapp

@Switch Blayde

Maybe that's only for those who can see the "VA Scr" on the Stats page.


Must be because I don't see them.

Ross at Play

@red61544

Again, I say Amen. Eventually, if the bitching continues, Lazeez will get disgusted enough to just quit on us.

I suggest Lazeez into a "Scoring Bitch Box", like a "Swear Box" some people use.
He'd have made a fortune last night.

Ross at Play

@richardshagrin

What do you guys think? Or ladies. Does it make sense to review a lot of stories by the same author, ... or should my time, and your time reading reviews point more to the less known stories that you might otherwise never see.

What do readers want from reviews? I think that's CAN this guy or gal spin a good yarn, and how well do they write. A reviewer can usually assess those things from just one SHORT story.
I'd like reviews of one story each from as many authors as possible.
Ideally choose one that is relatively short and recent, representative of the author's usual genre, among their better stories based on score and downloads - however conflicting those will often be.
If in doubt, I'd go for short and recent.

Replies:   Grant
Grant

@Ross at Play

A reviewer can usually assess those things from just one SHORT story.

?
Some people write good short stories, others write good long stories.
Some first efforts are horrible at best, but over time the authors become very good at writing.
Some authors are very good at writing many different types of story, others are only good at the one genre.

If you read a story, and consider it's worth a review then review it- regardless of whether you've reviewed one for that author previously or not.

Replies:   Ross at Play
Ross at Play

@Grant

I would not assert you are wrong.

Ross at Play

@Ross at Play

I suggest Lazeez into a "Scoring Bitch Box", like a "Swear Box" some people use.
He'd have made a fortune last night.

... at the risk of being the first to be hit with two penalties for my suggested "Scoring Bitch Box" ...
(1) There is already an attempt to disregard outlying scores with the top and bottom 5% ignored. I suspect the "problem" is much bigger at the bottom end of scores than at the top, i.e. many less 10 scores by fans of authors than 1-bombs by readers seeking out story codes they disapprove of. Ideally the number discarded would be something like 5% at the top, and 20% at the bottom. The result if the overall objective (the median story scores about 6.0), the results would be much fairer to virtually every story with Mm, ws, and some other codes.
HOWEVER, (if the percentage of top and bottom scores ignored is different) that can only be achieved with another 'normalisation' process AFTER the elimination of the outlying scores.

(2) ... risking a huge "bitch box" fine ... Lazeez could do this quite easily ...
WHY IS THERE NOT - ONE story (at least) with a 9.99 'score'; and ONE with a 0.01 'score'?
The wording used would need to be changed from 'score' to 'ranking'. Trivial.
I agree with everything Lazeez does seeking higher scores for stories most readers like most (bar a minor quibble for my first point).
WHY do readers not then see a number from 9.00 to 9.99 for the stories in the top 2% of stories, scores from 8.00 to 8.99 for the next 8% of stories.

SURELY readers would understand that much better than a number like 8.54 for a story almost every reader scores as "Fantasic"?

Replies:   red61544  richardshagrin
awnlee jawking

@Dominions Son

Scores from histogram in declining order 10(181), 8, 9, 7 6, 5, 1, 2, 4, 3


Thanks for publishing those figures. They corroborate my theories about readers' voting patterns.

AJ

Replies:   Crumbly Writer
REP

@Ernest Bywater

at one time, well . . . . and is no longer possible.


Thanks for the clarification. I haven't been around SOL that long.

I also haven't tracked the basis and changes to Lazeez's scoring algorithm. I do remember prior discussions regarding what his algorithm was attempting to do. I personally disagreed with what some posters attributed to him as his goals, but it is his system and I am willing to accept what he implements. Overall, I don't concern myself with scores because as you say in your second post to me, 'All votes are subjective by the reader.' The only time scores seem important is when I allow my ego to take control of my common sense.

red61544
Updated:

@Ross at Play


WHY do readers not then see a number from 9.00 to 9.99 for the stories in the top 2% of stories, scores from 8.00 to 8.99 for the next 8% of stories.


If you look at the Home Page under All Time Long Classics, all fifty stories listed have a score above 9.0. The highest is 9.7. All Time Short Classics do not fare as well - only six are above 9.0. That's simply because short stories seldom get high scores. When you read stories like "Greenies" or "Ed Biggers" or "Oscar Meyers", the difference between a 9.0 and an 8.0 becomes very obvious. It's the same reason why some authors win a Pulitzers Prize and others settle for a Pullet Surprise!

richardshagrin

@Ross at Play

I hate to try to answer why no story is rated 1.00 or 9.99. I'd need to go over the normal curve and standard deviations from the mean. If the mean is 6 and the standard deviation is 1 or fairly close to it, about 68% of all scores will be within one standard deviation of the mean, or between 5 and 7. Of the other 32% in round numbers 16% will be over 7 and 16% will be under 5. Its likely for a variety of reasons that scores out past two standard deviations won't be truly normal, but still if you get one percent or so over 8 and under 4 it would be a surprise. The percentages over 9 and under 3 will be very small. I think there are more stories over 9 and fewer under 3 than would be predicted by a true normal curve. However that are very few that all get 10s or all get ones. Probably none, or even the way scores get adjusted here there might be a 1.00 or 10.0 There won't be any score under 1 as no reader can vote zero. There might not be a score shown as authors can reject having their stories rated. I suspect they might be calculated but hidden, when stories are displayed in score order, some stories with no score reported are between other stories with scores.

I hope I answered why there are no stories with scores of 0.01, at least. They can only get votes as low as one.

Crumbly Writer

@samuelmichaels

It would take a lot of very-high-scroring posts to affect the curve on SOL.

Not so long on Sci-Fi, slightly longer on FS.

Crumbly Writer

@Ross at Play

I suggest Lazeez into a "Scoring Bitch Box", like a "Swear Box" some people use.

What, you're going to stuff poor ol' Lazeez into a box? What's worse, a box full of scoring complaints? Sounds like a fate worse that ... supporting complaining authors!

@Ross at Play

I'd like reviews of one story each from as many authors as possible.

Personally, I'd reserve new reviews for when an author writes an entire new type of story which are unlike their other stories (ex: changing from "Stroke" to "Mystery", or "Romance" to "High Fantasy").

I would not assert you are wrong.

I assert you're wrong about not being wrong! I'm not wrong, am I?

Replies:   Ross at Play
Crumbly Writer

@awnlee jawking

@DS

Scores from histogram in declining order 10(181), 8, 9, 7 6, 5, 1, 2, 4, 3

@Awnlee Jawking

Thanks for publishing those figures. They corroborate my theories about readers' voting patterns.

There are a lot more people voting 76 than I'd thought! :)

docholladay

The problem is that regardless of which formulas are used. Some writers among others will be unhappy with the results. As I recall the present calculations are a result of trying to please as many people as possible with the results.

As a result no matter which formula or methods that are used someone will always be unhappy with the scoring results. Those will always be imperfect no matter which methods are used to give the final scores.

So instead of trying to modify something which will never be perfect. How about actually coming up with something which will have a positive result, thus making the sites better.

Replies:   Wheezer
Wheezer
Updated:

@docholladay


The problem is that regardless of which formulas are used. Some writers among others will be unhappy with the results. As I recall the present calculations are a result of trying to please as many people as possible with the results.


Just my own cantankerous opinion, but scores should be less about making writers happy and more about providing useful information to readers when they are trying to find something to read. Can a sci-fi story with a 6.5 score be a better read than a similar one with a 7.0? Does a story with a 5.2 score that low because the author is a terrible storyteller, or because his story is unedited and his command of basic English is atrocious...or both - making it pure crapo?

docholladay

@Wheezer

The problem still remains. There is no way everyone will be happy with the scores or the methods for calculating them. It just seems to have reached a point where they are actually creating more problems for everyone.

awnlee_jawking

@Wheezer

Can a sci-fi story with a 6.5 score be a better read than a similar one with a 7.0?


Yes, for quite a lot of reasons. For example, the 6.5 story might have been published in one go by a new author, whereas the 7.0 story was published at intervals by an established author.

Does a story with a 5.2 score that low because the author is a terrible storyteller, or because his story is unedited and his command of basic English is atrocious...or both - making it pure crapo?


No, it can just mean that the story's content was unpopular. For example, if someone were to write a story about Mother Teresa being gangbanged, the story would struggle to reach a score of 4.0 no matter how well written.

AJ

Switch Blayde

@Wheezer

Can a sci-fi story with a 6.5 score be a better read than a similar one with a 7.0?


Absolutely. I've found a 6.5 better than a 9.0.

Does a story with a 5.2 score that low because the author is a terrible storyteller, or because his story is unedited and his command of basic English is atrocious...or both - making it pure crapo?


Or none of the above. Readers' scoring is fickle to say the least.

Replies:   docholladay  Grant
docholladay

@Switch Blayde

Its fairly simple. Either you enjoy the story or not regardless of what the cumulative scores are. The scores vary so widely from one to the next and some voters deliberately vote stories down or up, for who knows what reasons.

When I enjoy something I always say thank you to the writer (hopefully they pass that on to the editors). I would probably do the same to the editors but there are no links for them at the end of a story like there is for writers.

Replies:   Crumbly Writer
Crumbly Writer
Updated:

@docholladay


When I enjoy something I always say thank you to the writer (hopefully they pass that on to the editors). I would probably do the same to the editors but there are no links for them at the end of a story like there is for writers.


Lazeez, if you're listening in: I typically list editors, is it possible to include active links to my editors in my end-of-story notes? If not, could we add that as a feature to help highlight authors. Not only could readers congratulate the editors, but it would be easier for authors to contact the editors of the better edited stories for consultations. Seems like a win-win to me.

Update: Just tried adding active links to my editors on Sci-Fi. It doesn't fly. It strips out the links entirely.

docholladay

@Crumbly Writer

Lazeez, if you're listening in: I typically list editors, is it possible to include active links to my editors in my end-of-story notes?


It could be just a clickable option to include the editors in the feedback message without actually giving out their addresses. Either way would work for me as a reader.

Of course I have no idea of how many other readers would take advantage of that feature.

Replies:   Crumbly Writer
Crumbly Writer

@docholladay

Of course I have no idea of how many other readers would take advantage of that feature.

Authors only receive feedback from a small percentage (it tops out at less than 5%) of readers, so I wouldn't imagine it'd be large. Still, reader feedback is a major feature of the site, and a primary motivation to posting stories.

I don't typically pass on each response to a story to my editors (less it sound like bragging on my part), so having a button to include them in the email responses would be a nice touch (though it would involve a new mechanism to list each story's editors).

Replies:   docholladay
docholladay

@Crumbly Writer

Well its just an idea, no way for me to know if its a workable idea or not. Depends on how much trouble it would be for Laz and everyone else to utilize. At least I mentioned it when I thought of the idea.

richardshagrin
Updated:

@Wheezer


Does a story with a 5.2 score that low because the author is a terrible storyteller, or because his story is unedited and his command of basic English is atrocious...or both - making it pure crapo?


There are stories with scores below 6 or any other number you chose that are still worth reading. The subject matter may not be one easily accepted by many SOL readers. MM is one example and there are others. I can think of a Rachael Ross story with a score in the 4s that is every bit a good technically as any of her others. And as well told. People just hate the subject discussed.

Allah Akbar
Drama
Did we win the War on Terror? How would we ever know? Merry Christmas, world.
[More Info]
Tags: Caution, Violent
Sex Contents: No Sex
Posted: ‎12‎/‎20‎/‎2012‎ ‎5‎:‎32‎:‎28‎ ‎PM 6 KB 1167 59
4.85

red61544

None of the "theories" on scoring take into account the great unknown: the authors who have turned scoring off. I suspect that a great many of those are authors who are pissed off at the scores they received. That's probably why we seldom see scores below 5.0. There are also a lot of readers who only rate a story that they enjoy. I'm one of those. I've started numerous stories that, after the third paragraph, I've decided were trash and not worth reading. But I don't bother to rate them. Both of those factors actually skew the statistical scores upward and result in higher ratings overall.

Ernest Bywater
Updated:

@Crumbly Writer


I typically list editors


So do I, and I've had some authors ask about the editors, and sent the emails on to the editors, and let them decide. I've got one editor who would like a bit more work, but he doesn't accept a story unless it's no sex or minimal sex so he ends up turning down a lot of authors.

Replies:   docholladay
docholladay

@Ernest Bywater

I admit my idea was to be able to thank them for their work. The same as I try and thank writers when I enjoy one of their stories. Of course it would have to be up to them whether or not they wanted to receive reader feedback however. I for one will respect their choice.

Grant
Updated:

@Switch Blayde

Readers' scoring is fickle to say the least.

Hardly surprising as everyone is different.

Many people rate a story purely on how much it does (or doesn't) appeal to them. I try to rate it also on how well written it was, the plot, spelling & the like. Of course a story that is well written & I enjoyed will score better than one that is well written & doesn't appeal to me.
And I try to rate it on the story overall- there may be parts of a story I don't like, but I personally don't mark down a story because there are one or 2 parts that don't appeal to me.
As the Brexit vote & the current US election show, many people these days seem to entirely base their opinions of things on one particular topic. It doesn't matter what a persons opinions might be on other topics- if they don't support the this one particular view, then they are the spawn of Satan & to be reviled at all costs.

And it's also why I don't score stories I don't finish. It's possible that it does improve significantly by the end, but since I can't get through what precedes that dramatic improvement I don't score it as it stands at the point I bail out- I wouldn't be reviewing the story, just the part of it I read.

Replies:   Ross at Play
Ross at Play

@Crumbly Writer

From Me: I would not assert you are wrong.
Your: I assert you're wrong about not being wrong! I'm not wrong, am I?

Meaning ... my opinion is different, but the basis for your opinion seems reasonable.
Basically just: "Live and Let Live".

Ross at Play

@Grant

I don't score it as it stands at the point I bail out- I wouldn't be reviewing the story, just the part of it I read.

Why Not? You've finished your reading of the story.
Risking another huge "bitch box" penalty, I would like to see an rank offered to readers when selecting their score: Too awful to continue reading!

Lazeez should then allocate a score of the number of chapters the reader has downloaded, up to a maximum of six!

technomage

@grant

Many people rate a story purely on how much it does (or doesn't) appeal to them.


I tend to look at things more objectively. I've read many stories that don't particularly appeal to me, but if the story moves me, then it gets a vote. That vote is solely based on how good of a job it elicited an emotional response. This is reflected in the fact that if I have to step out of the story to mentally correct typos, homophones, name-changes of characters, etc, then naturally I am no longer attached to the storytelling and thus receives a lower score (if scored at all as depending on how many reader-asides I have to make, I might not go back to the story at all).

As a brand-spanking new writer here (2 submissions) my scores are extremely limited(4 votes for the first story, 14 for the second, including my own). Now both are poems. One was a story, the other was what I call a "strobe-light effect on a window in time". While the first story has only been voted on 4 times, none were less than 6 (7, 9, 9, 10).

The second story, I make no excuses for. In fact I even mentioned that there were very little tags as it was open to the reader to mentally apply his own tags, as the story itself is ambiguous. Interestingly enough, this one has received 2 '1' votes. (though no feedback as the why so I am not that concerned about it).

I expect to get TONS of '1' votes whenever I get around to submitting a longer story as I already know I am extremely weak in dialog ( in real life too, I'm a computer programmer/hermit ).

Replies:   Capt Zapp
Capt Zapp

@technomage

...I am extremely weak in dialog ( in real life too, I'm a computer programmer/hermit...


You're not weak in dialog, you just speak a different language. BASIC? C++? Cobol? Fortran? LISP? Pascal?

:)

Replies:   technomage
technomage

@Capt Zapp

C# and MSSQL and from the web design jobs in the past (volunteer), php and MySQL, and thanks to Minecraft, LUA..

Not to mention in real life, a smittering(less than a smattering, which is less than a pittering, which is less ... you get my point) of English, Spanish, Russian, Japanese, Chinese, Korean, Yugoslavian, Romanian, and finally Dog! :)

All of those, except Dog, was so I could flirt with waitresses...ok maybe Dog too..

Replies:   Switch Blayde
Switch Blayde

@technomage

Ah, Basic. I wrote a golf program in Basic that keeps my handicap plus all kinds of statistics. I guess when I move to a Mac I'll lose that. :(

When I programmed, it was in BAL (Assembler language on a mainframe — Basic Assembly Language). When I bought my first PC in 1980 I bought DOS's assembler. Never wrote a single program with it which is why I used Basic.

Replies:   Capt Zapp
Capt Zapp

@Switch Blayde

When I programmed, it was in BAL (Assembler language on a mainframe — Basic Assembly Language). When I bought my first PC in 1980 I bought DOS's assembler. Never wrote a single program with it which is why I used Basic.


I began programming in BASIC on a TRS-80 M1/L2/32k machine. It wasn't mine, mind you. I learned on the demo in the local Radio Shack. When customers came in with questions about the TRS-80, the salesman would usually tell them to see me. I didn't even own one at the time.

A few years later, I took a Basic Programming course through a military extension course and they had us learning to use PUNCH CARDS. Talk about a pain in the rear. Nobody used punch cards in the 80's except this class.

Switch Blayde
Updated:

@Capt Zapp


Nobody used punch cards in the 80's except this class.


I used punch cards, though it was in the 1970s. Before mainframes had TSO (Time Sharing Option) and then VM (Virtual Machine), we coded our programs on coding sheets and the data entry gals typed it to create punch cards. Then we'd put the stack of cards in a tray for the computer room guys to run it. When we got our compile listing back we were able to use keypunch machines to change the bad cards (changes to the program). Man, I can't believe we ever got anything done.

Assembler was not a high level language. It was just up from machine code. The high level language at the time was COBOL and FORTRAN, mostly COBOL for business.

Replies:   REP
Ernest Bywater

@Capt Zapp

Nobody used punch cards in the 80's except this class.


I job I was doing for an international communications group in 1979 was cutting the punch cards for the proposed routing changes, then taking them and the program cards over to be run by the computer company. Get the print outs, look for issues, cut new cards as needed, and do it all again. Those some cards and program were still being used in 1983, don't know if they were still in use after that, but I know they were still in use until then as a minimum. And that's a real world usage to work out where to lay cables on the sea floor and set up earth stations and cables for the sat comms.

Replies:   Grant
Grant
Updated:

@Ernest Bywater


Those some cards and program were still being used in 1983, don't know if they were still in use after that, but I know they were still in use until then as a minimum.


Back then, for that type of work, those existing systems were still powerful enough to do the job, and a lot cheaper to keep running than replace with what was available at the time- other mainframes & mini-computers (PDP-11s etc).

REP

@Switch Blayde

we coded our programs on coding sheets and the data entry gals typed it to create punch cards. Then we'd put the stack of cards in a tray for the computer room guys to run it. When we got our compile listing back we were able to use keypunch machines to change the bad cards (changes to the program).


I had forgotten all about punch cards. I wasn't a programmer, but I wrote technical manuals on computer controlled systems. I had to be basically familiar with programming languages, so I took a few classes in DOS, Fortran, Pascal, and other programming languages and subjects. During labs, I had to create and then submit the punch cards for my programming assignment. Then I had to find evaluate the results to see why my program failed to run as expected.

Replies:   Crumbly Writer
Crumbly Writer

@REP

I had forgotten all about punch cards. I wasn't a programmer, but I wrote technical manuals on computer controlled systems. I had to be basically familiar with programming languages, so I took a few classes in DOS, Fortran, Pascal, and other programming languages and subjects.

I came of age a little later. We'd (largely) abandoned the use of punch cards, but had only advanced to 'punch tape' (small rolls of tape which we used a keyboard which produced a series of binary dots other machines could read into the computer.

I first used that archaic device in high-school, and by the time I reached college they were moving to full magnetic tapes.

During that time, I was also involved with DOS, Fortran, BASIC and Cobol (learned Pascal and C years later).

Back to Top