half-ratings

[under_pressure]

Why do folks constantly tout the fact that they "don't care about ratings" like it's something to be proud of? I mean, great for you if you genuinely find ratings irrelevant, but I feel like too many are adopting the "I wasn't caring about ratings before it was cool" mentality. Get off your high horse.

You know what? I happen to like ratings. I like getting ratings, ESPECIALLY good ones. They're not entirely accurate, but who ever demanded that they be as a collective? Ratings are a good way to individually quantify appreciation of a map. And goddamnit, I like having that option.

On a less tangential note, I liked the rating system with half-values, but since Arachnid has apparently adamantly expressed that they will not be reinstated, I'm not what point there is in me saying so.

[Skyling]

Nice rant! I'd give it a 3.5 out of 5.
Oh, wait. Hm. Well, make that a 3 out of 5, I guess. I hope you don't mind; it's just a rating.

[origami_alligator]

Having a 4.5 option was good. I used that a lot.

And now 4 includes maps that would have been rated anywhere from roughly 3.5 to 4.5. If you want to rate 4.5 use 4 instead. It's not that huge of a leap.

And if you disagree that a lot of users are complaining, just browse a few random maps. Chances are you'll see a "x.5 rounded up/down" comment. Also, allowing .5's for voting shouldn't be a problem because of the next section.

It's not a complaint, usually. I see a lot of the "x.5/5 up/down" comments, but it's not like they are making a statement about the ratings system. They are expressing the rating they truly feel about the map and then rounding up or down because of it. "3.5 up" and "3.5 down" are two very different things, as one shows that one person probably thought a map had something extra, while another person didn't. The variation of ratings now comes from personal preference rather than an uncertainty about what to give the map. There is no, "oh I thought this was pretty okay, so 3.5 because it's better than a 3 but not as good as a 4." Now it's, "This was lacking something (maybe an explanation of that something) so 3.5 down."

It's letting people comment with something other than what they rated, and saying something about the map that they specifically did or did not like. I don't think that's a bad thing at all.

[quote="

Are you saying "We made the ratings system worse, so that people are forced to comment"?
xD

I'm saying the ratings system now promotes people to comment as well as rate.
For example, saying, "3.5/5^ because I like how you used the thwumps, even though the mines were annoying as hell," or saying, "3.5/5 down. I didn't like the placement of the gold," is better than taking an average rating, calling it good and moving on to a new map, without commenting or saying something about the map that you did or did not like.

[Nexx]

Ratings are a good way to individually quantify appreciation of a map. And goddamnit, I like having that option.
...
Get off your high horse.

Win.

On a less tangential note, I liked the rating system with half-values, but since Arachnid has apparently adamantly expressed that they will not be reinstated, I'm not what point there is in me saying so.

I am personally continuing the crusade to have .5's in the drop down box because doing so would take a matter of minutes. Until Arachnid implements that or tells me "Google App Engine makes it impossible", I will push for it.

Nice rant! I'd give it a 3.5 out of 5.
Oh, wait. Hm. Well, make that a 3 out of 5, I guess. I hope you don't mind; it's just a rating.

You misunderstood the point, which relies on the context of NUMA.

It's not a complaint, usually.

Why does it need to be? It's a failure of the system to meet the needs of its users, and it's widespread. Not to mention that fixing it is super easy. That's not enough?

I'm saying the ratings system now promotes people to comment as well as rate.

...Are you serious? You're saying that it's a good thing for the system to be flawed so that users are forced to comment to be clear about their intentions?

Let's imagine something. Say Arachnid generates four random numbers in the range 1 - 9. Let's say the results are 1, 2, 4, 6. Then he divides each by two, adds 0 and 5 to that set, and *voila!* -- he has NUMA's new rating system: {0, 0.5, 1, 2, 3, 5}. By your logic, you would prefer this system over old NUMA's {0, 0.5, 1, 1.5, ..., 5} because the choices are a bit out of whack, thereby "promoting people to comment as well as rate."

I am pointing out that what you offered was a side note, not an argument. Please defend it as such.

The variation of ratings now comes from personal preference rather than an uncertainty about what to give the map. There is no, "oh I thought this was pretty okay, so 3.5 because it's better than a 3 but not as good as a 4." Now it's, "This was lacking something (maybe an explanation of that something) so 3.5 down."

So you're saying that old NUMA's system relied more on uncertainty to arrive at a rating, while new NUMA's system relies more on personal preference? Sounds like hand-waving to me. If you want to use this as a legitimate argument, you're going to have to elaborate.

I mean, it's true that I saw a few "well it was better than a 3 but not quite a 4, so 3.5" comments on old NUMA, but I would hardly call that rating based on uncertainty. They're using reason to arrive at 3.5, just as they now have to use more reason to decide between 3 and 4 on that 3.5. If anything, I would say it's arguable that the latter decision involves a bit more uncertainty than the former.

My summary of arguments for .5's for voting
Against:
- They're not necessary
- It's (generally) not a big difference in score

For:
- A lot of users would use them
- Wouldn't change viewing of averages
- Easy to implement

Have I got it about right?

[a happy song]

My summary of arguments for this thread
Against .5's:
- They're not necessary
- It's (generally) not a big difference in score

For .5's:
- A lot of users would use them
- Easy to implement

Have I got it about right?

Pretty much.

As mentioned before: The main flaw with the current system - for me at least - is the lack of an average score.

[Nexx]

Pretty much.

:)

As mentioned before: The main flaw with the current system - for me at least - is the lack of an average score.

As I said before, you don't have to go by what's written on the site (about 2 = below average, 3 = above average). For starters, what exactly does "average" mean and why are you so intent on using it to determine a rating? Rating out of 5 is an incredibly common system, and the lack of 2.5 really shouldn't be a problem. If you rent an "average" movie from Netflix (which uses 0 - 5 integers only), what would you rate it when you return it? I strongly believe that if you give your personal process of evaluation a bit of thought, you should be able to reconcile it with the 0 - 5 system. ...though perhaps that feeling comes from my personal rejection of the use of "average" for grading things.

[a happy song]

As I said before, you don't have to go by what's written on the site (about 2 = below average, 3 = above average). For starters, what exactly does "average" mean and why are you so intent on using it to determine a rating? Rating out of 5 is an incredibly common system, and the lack of 2.5 really shouldn't be a problem. If you rent an "average" movie from Netflix (which uses 0 - 5 integers only), what would you rate it when you return it? I strongly believe that if you give your personal process of evaluation a bit of thought, you should be able to reconcile it with the 0 - 5 system. ...though perhaps that feeling comes from my personal rejection of the use of "average" for grading things.

That's actually a very good point, thanks :)

[SkyPanda]

You're saying that it's a good thing for the system to be flawed so that users are forced to comment to be clear about their intentions?

Let's imagine something. Say Arachnid generates three random numbers in the range 1 - 9. Let's say the results are 1, 2, 4, 6. Then he divides each by two, adds 0 and 5 to that set, and *voila!* -- he has NUMA's new rating system: {0, 0.5, 1, 2, 3, 5}. By your logic, you would prefer this system over old NUMA's {0, 0.5, 1, 1.5, ..., 5} because the choices are a bit out of whack, thereby "promoting people to comment as well as rate."

I am pointing out that what you offered was a side note, not an argument. Please defend it as such.

Thanks Avarin, you said it better than I could :)
I'd like to add to that by saying that you're proposing a benefit that I don't think will actually happen. I've seen plenty of "3.5 down" comments with nothing else said whatsoever about the map. Many people rate without commenting, giving them different numbers to play with isn't going to change that.

[mintnut]

I don;t get it, what's so special about these half ratings. I mean, your trying to rank all the maps on numa, with a subjective scale. The ratings are never going to be a definitive or even accurate representation of map quality, as it all depends on who you are and what you want from the maps on offer at numa. Also, comments for the win.

[Nexx]

That's actually a very good point, thanks :)

Actually, it's funny how I go to lengths to explain my argument, but I don't actually explicitly state my argument sometimes. Not that anyone cares at this point, but my argument was that if you're fine with the other .5's being gone, you should be fine with 2.5 being gone, and if you give it some thought and you really aren't fine with 2.5 being gone, then you would probably use the other .5 ratings as well.

I don;t get it, what's so special about these half ratings. [...] Comments for the win.

Yes, this is the "they're not necessary" argument.

I already laid it out for everyone:

My summary of arguments for .5's for voting
Against:
- They're not necessary
- It's (generally) not a big difference in score

For:
- A lot of users would use them
- Wouldn't change viewing of averages
- Easy to implement

-----

I mean, your trying to rank all the maps on numa, with a subjective scale. The ratings are never going to be a definitive or even accurate representation of map quality, as it all depends on who you are and what you want from the maps on offer at numa.

True, but no one said it should be and, more to the point, average ratings are not the issue in question here. We're talking about voting only. Putting in .5's allows the people who want to vote more accurately to do so. For people who vote more generally, they can continue doing so.

-----

Also, I would like to amend something I said earlier:

Viewing of averages should use whole numbers only. [...] There's two reasons for this:
- NUMA is imperfect
- Community consensus != your consensus

[Explanation for NUMA's imperfections, namely that rates vary a lot and average # of rates in nowhere near a community consensus]

Which brings me right to the second point, which is that even if NUMA provided a perfect community consensus of all maps, that would not necessarily make the ratings any more useful for users because personal tastes vary considerably.

After some more thought, I disagree with my second point. The ratings system does not exist for the purpose of being an indicator of the map's value for most people. Its purpose is to generate a quantifiable community consensus about a map. Because of that, it can function as a rough indicator of the map's value for most people.

So now I argue for whole number averages simply because there's too much chance error in the ratings of most maps, which is a result of the community being rather small. If maps got something like 50 rates on average, then I would push for .5's for averages, and I don't think there would be many in disagreement.

Thus, I would like to complete the sentences of several people (including mods and NUMA admins) who have used an incomplete argument to justify the removal of .5's from averages: "you just can't be that precise with so few rates".

[mintnut]

Avarin, way to go and manhandle what I said.:

"I mean, your trying to rank all the maps on numa, with a subjective scale. The ratings are never going to be a definitive or even accurate representation of map quality, as it all depends on who you are and what you want from the maps on offer at numa."

I never once said that this was about the average scale. Are you trying to say that you could subjectively rate every map on numa, and that it would mean anything. No matter what scale you are using, that is an awful assumption. A rating is basically just a quantified, neutered opinion. Adding .5's will not rectify that.

This issue needs to be put to bed. There isn't a vast difference between a .5 scale. But arachnid is a busy man, and there are much more useful things he could be spending his time on, things which aren't as trivial as to whether you can make a vague number masquerading as feedback any more meaningless.

(Also, you would probably do better to start using the entirety of the rating scale we have anyway. If more 1's, 2's and 3's were banded about, maybe the constant stream of 4's and 5's given to everything would mean more.)

[notsteve]

i like whole numbers.
but i do feel we need an average...wait a second!

0 worthless
1 pretty bad
2 below average
3 average
4 above average
5 pretty awesome

those ratings are much more fun! and we didn't even change anything
i do feel that those would reflect how people rate much better

[Condog]

i like whole numbers.
but i do feel we need an average...wait a second!

0 worthless
1 pretty bad
2 below average
3 average
4 above average
5 pretty awesome

those ratings are much more fun! and we didn't even change anything
i do feel that those would reflect how people rate much better

But the very nature of that order means that 3 in inherently not average. It has two options above, and three options below. For it to be average, there needs to be an equal number of options on both side of the centre.

[origami_alligator]

For:
- A lot of users would use them
- Wouldn't change viewing of averages
- Easy to implement

On the contrary, my dear Watson!
Changing the scale of the rating system would force the display to be taken out to the same decimal place or increase the number of ninjas.

An average 4.5 rating after 5 votes with the current display of the maps average would show a perfect 5. This is obviously not a true showing of the average rating of the map based on the vote options.

[Evil_Sire]

Having whole numbers helps me catogorize maps easily.

I'll usually play most maps maps that have a rating of a 4 or a 5, and they'll mostly be of that rating. Having half-ratings would screw everything up, I like to play popular maps (normally the top two ratings).

So in conclusion:

Half ratings means that you are basically rating out of ten, which is alot of numbers to choose from and you could be left waiting to decide. Half-ratings are uneccassery and awkward, whilst it is true they provide a more accurate rating, it is easier to group maps with whole numbers.

Reason for grouping maps:

Grouping maps helps you to decide what maps to play, and is a good system. Half-ratings will also kinda mess things up.

Why it would mess things up:

The system (as this has happened on many of my maps) automatically rounds up, therefore we are not in need of half-ratings, it would complicate things. Also, the hot maps page is easier to manage with whole-ratings.

I am against half-ratings. Though I still use x.5 up/down.

[b3njamin]

I have a Idea!
what about this:


You can Vote the half-ratings (0 1 2 3 4 5 and 0.5 1.5 2.5 3.5 4.5)
But it only shows the normel ratings. (full ratings)

example:

Some map gets the following ratings, 1.5 and 3.5. Normally this would be 2.5, but now it would show a 3.

[origami_alligator]

I have a Idea!
what about this:


You can Vote the half-ratings (0 1 2 3 4 5 and 0.5 1.5 2.5 3.5 4.5)
But it only shows the normel ratings. (full ratings)

example:

Some map gets the following ratings, 1.5 and 3.5. Normally this would be 2.5, but now it would show a 3.

Already explained above

[Nexx]

Sorry to semi-resurrect this thread, but I've been meaning to respond to it for awhile now.

I have a Idea!
You can Vote the half-ratings (0 1 2 3 4 5 and 0.5 1.5 2.5 3.5 4.5)
But it only shows the normel ratings. (full ratings)

Ummmm, did you even read the thread? That was suggested in the 6th post, and has been a major topic of discussion since.

Avarin, way to go and manhandle what I said.

Sorry, but I'm having difficulty understanding what you're trying to say. In my previous post I tried interpreting you words as best I could and responded to that. I'll try to interpret your new post more strictly.

"The ratings are never going to be a definitive or even accurate representation of map quality, as it all depends on who you are and what you want from the maps on offer at numa."

It sounds to me like you're saying that averages should reflect how much people will like each map for every person. If that's what you're saying, then you need to rethink how communities work. Furthermore, do people even bother reading my posts? Because I explained this already:

The ratings system does not exist for the purpose of being an indicator of the map's value for most people. Its purpose is to generate a quantifiable community consensus about a map. Because of that, it [has the potential to] function as a rough indicator of the map's value for most people.

Let me know what you think about that.

----------

Are you trying to say that you could subjectively rate every map on numa, and that it would mean anything. No matter what scale you are using, that is an awful assumption. A rating is basically just a quantified, neutered opinion. Adding .5's will not rectify that.

That's the second time you've said that, and I have to admit I have no idea what you are saying. It would mean that I gave my quantified opinion to every map on NUMA, no more, no less. Please clarify if you want a better response from me.

----------

There isn't a vast difference between a .5 scale. But arachnid is a busy man, and there are much more useful things he could be spending his time on, things which aren't as trivial as to whether you can make a vague number masquerading as feedback any more meaningless.

Again, are you not reading my posts? In the first place, I summed this stuff up already:

My summary of arguments for .5's for voting
Against:
- They're not necessary
- It's (generally) not a big difference in score

For:
- A lot of users would use them
- Wouldn't change viewing of averages
- Easy to implement

Furthermore, you'll notice the last point on that list: easy to implement.
To implement this, Arachnid merely needs to edit the drop down box code to put in the new display strings and then the values they would generate for the map. Done. I imagine total coding time would be 10 minutes max, and that would be if it's super-hard code to navigate.

---------

(Also, you would probably do better to start using the entirety of the rating scale we have anyway. If more 1's, 2's and 3's were banded about, maybe the constant stream of 4's and 5's given to everything would mean more.)

I agree that there are a lot of 4's and 5's given out, seemingly for no good reason. Regardless, don't make the assumption that because I'm championing .5's for voting means that I want to use them. If they came back, I wouldn't touch 'em. But lots of others would use them constantly, and that's the point.

----------

Half-ratings will also kinda mess things up.

No. We're not talking about changing the averages to show half-ratings, just the voting options.

I am against half-ratings. Though I still use x.5 up/down.

That nullifies your whole argument. Unless you're saying that .5's would mess up everyone else.

----------

On the contrary, my dear Watson!
Changing the scale of the rating system would force the display to be taken out to the same decimal place or increase the number of ninjas.

An average 4.5 rating after 5 votes with the current display of the maps average would show a perfect 5. This is obviously not a true showing of the average rating of the map based on the vote options.

I'm sorry, what? Are you serious?

4.5's can currently exist. The current way of showing averages is just as inaccurate for an even number of votes, half being 4 and half being 5. More to the point, as long as the site rounds averages at all, it's not a true showing of the average rating.

And there is absolutely no other reason why .5's would need to be added to the average if they were added to the options. Furthermore, I made a poll of this on the old forums, and people want the ability to vote with .5's much more than the ability to see those .5's. So while people would still complain, there'd be less complainers.

Why are people so resistant to this? What am I missing?

[wolfgang]

I completely doubt that removing the half ratings has accomplished any of the things people are claiming (more emphasis on commenting, easier to group?). But by the same token, it hasn't accomplished anything because the change is so trivial. It really is a non issue, you should just accept the decision. I think, in the end it just comes down to the admins preferring whole numbers for somewhat aesthetic reasons.

[Nexx]

I completely doubt that removing the half ratings has accomplished any of the things people are claiming (more emphasis on commenting, easier to group?). But by the same token, it hasn't accomplished anything because the change is so trivial. It really is a non issue, you should just accept the decision. I think, in the end it just comes down to the admins preferring whole numbers for somewhat aesthetic reasons.

It's hardly a trivial change:

And if you disagree that a lot of users are [having to work around the current system], just browse a few random maps. Chances are you'll see a "x.5 rounded up/down" comment.

Maybe it's a non issue for you, wolfgang, but the result of my poll on the old forums showed that the vast majority want their .5's back for rating.

But you make me realize something, wolfgang. Namely that I took up this fight so readily because it's one of the embodiments of the whole "NUMA architects are not listening to NUMA users" issue, particularly since it could be solved so easily. Really I just want some flippin' transparency, and an open ear to constructive feedback. If you're going to make major changes, at least inform everyone of what and why, but at best, make sure there's a community consensus first, rather than basing the decision on the opinions of a just a few members. *sigh*

[mintnut]

Not listening to numa users? We aren't actually unanimous over this issue are we. There is just as big a camp who like the ratings as they are, as want the .5's back.

As for what you said earlier regarding my post, as far as I can tell, you have again misinterpreted everything I said, not least because you cut it up into short statements, rather than acknowledging what I said in it's entirety. I don't think I can explain my stance any better than I already have, especially when you just threw what you'd already said back as an answer to my counter arguments. I suggest we just leave the matter.

[Nexx]

Not listening to numa users? We aren't actually unanimous over this issue are we. There is just as big a camp who like the ratings as they are, as want the .5's back.

As for what you said earlier regarding my post, as far as I can tell, you have again misinterpreted everything I said, not least because you cut it up into short statements, rather than acknowledging what I said in it's entirety. I don't think I can explain my stance any better than I already have, especially when you just threw what you'd already said back as an answer to my counter arguments. I suggest we just leave the matter.

Oh hell no. I'm not going to let you push a legitimate issue aside, especially not with an attitude like that.

(1) I don't know how you don't understand my point yet, but here it is, clear as day:

There are two camps, yes, but they aren't in opposition on the matter that counts! One camp says the .5's are useful (for themselves), while the other camp says they aren't (also for themselves). That's where all the disagreements lie. My point: We can satisfy everyone. Adding .5's back in (which is easier than fixing pretty much any other issue that's been brought up, I would imagine) satisfies the first camp without going against the second camp. Being part of the second camp myself, I am imagining .5's in the drop down box and not caring. If you like the whole number system, you won't touch them, and they won't affect viewing of averages. I see literally no significant reason against making this fix.

(2) I cut up what you said instead of "acknowledging it in its entirety"? What does that even mean? Chopping is perfectly acceptable if done appropriately, and I sure did it appropriately. In the first place, your post says different things at different times, not all relating to the same argument. In your first counter-response, you have 5 paragraphs. The first three are all on the same topic, which I suppose would be best summed up with your line: "I mean, you're trying to rank all the maps on numa with a subjective scale." The fourth is a separate argument that I can respond to separately (it said: it's not a big deal and Arachnid's time is better spent elsewhere). The fifth is about using the rest of the rating scale, which can again be dealt with separately.

Furthermore, the first three paragraphs can still be broken up, because that's how debating works: you make an argument, then you back it up. If I can't understand the wording in your argument, I can look at the reasoning you're using to back it up. That is, I can "chop up" that paragraph and say "I don't understand" to the first sentence (your argument) yet reply fully to the second (your reasoning). And this is exactly what I did, so what, precisely, are you complaining about?

(3) Since I've explained why chopping is acceptable when done appropriately, I'm now going to chop up your new post a bit to deal with a new point.

I don't think I can explain my stance any better than I already have

As a matter of fact, you sure can. As I just explained, I understood the point you made in support of your argument in your counter-response, and I countered it. You could start by telling me why my counter isn't responding to the right idea, as you claim it is. Furthermore, I responded to several other points in your post that were separate from the argument I didn't understand. Are you saying I missed the point on those too? I should I assume implicit agreement there?

Now I could be wrong (-_-), but I imagine you probably meant that this is a small issue and that you could explain it and go from there, but you didn't really think it was worth it. Fine. Say that. I still would have responded with my (1) argument in that case.

Sorry if I sound ill-tempered. It's not against you as a person, just your post. :)

[mintnut]

Whatever. Ho hum.
x

[bobaganuesh_2]

condog's idea is good enough

[AsTimeFliesBy]

first and foremost, forgive me for reviving this topic, but i didnt want to create a new one

nexx, you write way too much. please be more concise as to not bore your audience. i read the whole first page and it took a long time, but i saw all your (and other's) long-ass posts on the second page and i said hell no

anyways i saw some posts on the first page i'd like to respond to. it seems like most people dont care to change ratings back, but i think we should because it's more precise. even many who dont care said there should be an average 2.5 rating. so we should just bring all half ratings back. someone also said people are now more inclined to comment, but i diagree. i feel those who rated w/o commenting before are still doing the same (at least on my levels). as mentioned before, ratings are inaccurate due to variances in personal opinion, but taking away half ratings made them even more inaccurate. also mentioned before, im one who finds myself having to decide whether to round up or down. sometimes i feel a level is better than a 3 but not good enough to be a 4. i'll say 3.5 in my comment but idk what to do about the rating.

im sure my view will have no impact whatsoever, but i just wanted to put it out there. im not sticking around for long so it probably wont matter much to me anyways, but i'd still like to have half ratings back

edit: sorry my post is pretty long too but it's not as long and it's pretty much all i have to say