.jpg)
Thumbs inflation
category: general [glöplog]
For years I had a strong feeling that we're not thumbing down things enough anymore. Now there's a clear proof.
Methodology:
Latest 20260826 prods dump is taken (why is it gz, not xz?).
Unfortunately, dumps only contain aggregate number of votes, so it is impossible to get exact vote date. So an assumption is made: prods get most of their votes not so long after they're released, so we can proxy-attribute voting behavior to prod release date.
A total of all thumbs-up, pigs, thumbs-down is calculated from all the prods released in a given month.
These values are normalized by total per-month count of votes. An month-average vote is computed from all average prods vote. Percentiles are computed too to shape the overall average-vote distribution.
Chart:
There are clearly 5 different sections:
A conclusion is not made. As a person of eastern-european culture provenance, I am not happy.
Disclosure: analysis has been vibe-coded running DeepSeek-V4-Flash-0731 locally fully on solar power.
P.S.: I wish we had individual votes data in dumps. If those existed, we could also potentially have cool recommendation systems like "give me the first top-N prod in a given category that I haven't commented yet", and "give me a prod that I haven't commented, but which had been thumbed-up by people why mostly agree with my thumbs", etc.
Methodology:
Latest 20260826 prods dump is taken (why is it gz, not xz?).
Unfortunately, dumps only contain aggregate number of votes, so it is impossible to get exact vote date. So an assumption is made: prods get most of their votes not so long after they're released, so we can proxy-attribute voting behavior to prod release date.
A total of all thumbs-up, pigs, thumbs-down is calculated from all the prods released in a given month.
These values are normalized by total per-month count of votes. An month-average vote is computed from all average prods vote. Percentiles are computed too to shape the overall average-vote distribution.
Chart:
There are clearly 5 different sections:
- Pre-1986. There are not enough prods/data, so it's just noise.
- 1986-2000. 1000 votes per prod per month. Slight downward trend on thumbs-up, and upward on thumbs-down, but overall things are pretty stable regardless, average vote around 0.4.
- 2000-2010. Total number of votes is all time high, arount 3000/prod-month. And we're mostly honest, do not hesitate to call things that are shit shit, etc. Average vote stays the same, around 0.4.
- 2010-2020. A strong trend emerges: more thumbs-up, fewer thumbs-down and pigs. Average vote rises 2x: from 0.4 to 0.8. Total votes goes down to below 1500/prod-month.
- 2020+. Barely any thumb-downs at all. Avg vote stabilizes around 0.8. Total votes around 1000-1500/prod-month.
A conclusion is not made. As a person of eastern-european culture provenance, I am not happy.
Disclosure: analysis has been vibe-coded running DeepSeek-V4-Flash-0731 locally fully on solar power.
P.S.: I wish we had individual votes data in dumps. If those existed, we could also potentially have cool recommendation systems like "give me the first top-N prod in a given category that I haven't commented yet", and "give me a prod that I haven't commented, but which had been thumbed-up by people why mostly agree with my thumbs", etc.
Fascinating.
As a demoscene person I don’t thumb down anymore (or really occasionally).
When seeing prod I don’t like I just ignore it.
Maybe this is what happens when you get older?
Maybe a regression model between the age and the amount of 👎 would show us something, but age of Poueters is not easy to infer.
As a demoscene person I don’t thumb down anymore (or really occasionally).
When seeing prod I don’t like I just ignore it.
Maybe this is what happens when you get older?
Maybe a regression model between the age and the amount of 👎 would show us something, but age of Poueters is not easy to infer.
Thumbing down feels extremely discouraging; I want more demoscene, so it seems counterproductive to do.
Thumbing down is also for me something, that i very rarely do. This might only happen when absolute crap was ranked high on a demoparty. I prefer more a piggy and type then positive points and not so positive to keep people motivated. Tumbing up is that, what i do mostly as it feels simple better, to value peoples work as they invest often much time, love and a passion for our little demoscene world.
I don't think it's a good idea to put more negative energy into the internet.
A question about the "Total votes per release month" graph:
I don't think people could vote on Pouet before 99/2000.
So a value from 1988 is probably a vote for a production from 1988 but we don't know when this vote was casted?
I don't think people could vote on Pouet before 99/2000.
So a value from 1988 is probably a vote for a production from 1988 but we don't know when this vote was casted?
^ exactly. There's so much negativity online as a whole and the heyday of Pouet was really toxic in many ways. I participated in it too which is something I kind of regret these days. Now I'm old and I'd rather enjoy things than pour hate on them. If I don't like something, I'll just be quiet and reserve the thumb downs for something that I _really_ dislike.
Dudes talk about negative energy and toxicity, but it's actually just biasing the results if some vote prods with 0/1 and some with -1/0/1. (thumbs = votes to be clear)
Probably would make to sense to add some normalization to the vote calculation. Maybe for example (per each voter) calculate average of votes, then subtract the "DC component" and then even calculate some variance of the votes and scale with that too. But of course that would still require that people vote good and bad prods equally, which I don't think is happening.
Probably would make to sense to add some normalization to the vote calculation. Maybe for example (per each voter) calculate average of votes, then subtract the "DC component" and then even calculate some variance of the votes and scale with that too. But of course that would still require that people vote good and bad prods equally, which I don't think is happening.
yet also thumb deflation by the dropped number of active pouet users! you can release the best demo ever tomorrow but never get the same popularity numbers as 15 years ago :)
Quote:
Thumbing down is also for me something, that i very rarely do. This might only happen when absolute crap was ranked high on a demoparty. I prefer more a piggy and type then positive points and not so positive to keep people motivated. Tumbing up is that, what i do mostly as it feels simple better, to value peoples work as they invest often much time, love and a passion for our little demoscene world.
This is similar to what I've been doing, but it feels dishonest and unhelpful. We kinda end up in a toxic positivity echo-chamber, where everyone's exceptional, everyone is a princess, everyone deserves a first place prize.
Quote:
I don't think it's a good idea to put more negative energy into the internet.
I get your point. But I think that by that approach we also lose objectivity and valuable feedback. A sane critique is dampened, and instead the feedback gets polarized and bimodal: it's mostly either "you're doing great nice good job!" for something mediocre, or "it's an utter shit gtfo demoscene kill yourself" from someone who just doesn't feel like not putting more negativity.
Hence the inflation -- what was objectively average = 0 before, is now offset way higher, somewhere closer thumbs-up.
And if people didn't like your prod, but also felt discouraged to thumb it down and did nothing instead, you don't get any feedback signal at all.
Quote:
A question about the "Total votes per release month" graph:
I don't think people could vote on Pouet before 99/2000.
So a value from 1988 is probably a vote for a production from 1988 but we don't know when this vote was casted?
Ha! This is a good point! I just vibe-switched-the-intellect-of and failed to notice this. Thanks!
So, yes, all the data before 2000 is bogus wrt the point of this thread. Still interesting to see that many old prods get the whole range of average votes.
Quote:
Probably would make to sense to add some normalization to the vote calculation. Maybe for example (per each voter) calculate average of votes, then subtract the "DC component" and then even calculate some variance of the votes and scale with that too. But of course that would still require that people vote good and bad prods equally, which I don't think is happening.
Public pouet data dumps do not have neither per-user vote data, not individual vote events. The granularity is per-prod votes counts.
If we had that data, It would be also interesting to see average per-user votes across time.
Quote:
This is similar to what I've been doing, but it feels dishonest and unhelpful. We kinda end up in a toxic positivity echo-chamber, where everyone's exceptional, everyone is a princess, everyone deserves a first place prize.
I think as far as this site goes, it's become way less of a cesspool but there's still a loooong way to go until we reach anything resembling "toxic positivity".
Also, in my experience, criticism delivered with a piggie usually hits better than with a thumb down. There's something about that "I'm not mad, just disappointed" energy :D (but also seriously)
Tldr: we were assholes but grew out alright.
I would even go a step further (but I don't have numbers) and say that the thumbs have become more positive *even though* the perceived quality of the releases has declined.
Discuss! :)
Discuss! :)
I'm not sure we should worry too much about the lack of thumb-downs when even the top releases rarely get more than a couple dozen comments.
@hfr: I don't think it has (declined). In "the old days" there weren't many releases (if any) where people would spend 5 or even 10 years on a single prod. The bar is quite high today. We also will, most likely, see more vibe-coded stuff in the future, i.e. the *perceived* quality will go way up. Most likely.
But yeah, I for one find thumb-downs rather pointless. Any kind of rating system, actually (fortunately the BBS is still "old school").
In forums where only thumb-ups are allowed, this is often used to create "negative spaces" (besides, "negative" content is most popular on social media and therefore gets the most "positive likes", ironically).
In general, "likes" and "thumb-ups" do not actually reflect how good something is, more like many people clicked the button (otherwise e.g. the best music on YT would have billions of likes, which it surely has not) (it's more like in "my buddy or famous person X liked it, let's just follow the leader to be on the safe side").
Constructive criticism is most valuable, and you don't get that very often (however, criticism potentially disturbs the "positive bubble" so many will avoid it just because of that. especially true in other environments, btw).
(sry for the long text but I had to take a break)
But yeah, I for one find thumb-downs rather pointless. Any kind of rating system, actually (fortunately the BBS is still "old school").
In forums where only thumb-ups are allowed, this is often used to create "negative spaces" (besides, "negative" content is most popular on social media and therefore gets the most "positive likes", ironically).
In general, "likes" and "thumb-ups" do not actually reflect how good something is, more like many people clicked the button (otherwise e.g. the best music on YT would have billions of likes, which it surely has not) (it's more like in "my buddy or famous person X liked it, let's just follow the leader to be on the safe side").
Constructive criticism is most valuable, and you don't get that very often (however, criticism potentially disturbs the "positive bubble" so many will avoid it just because of that. especially true in other environments, btw).
(sry for the long text but I had to take a break)
I found the last years too positive but that’s me.
On csdb it got so bad that now basically everything gets the highest votes, and 8/10 is called "downvoting" by some.
Not sure if that helps anyone
Not sure if that helps anyone
Side effect of CSDB maintaining an average score I guess - I like pouet's system more in that regard, as in counting how many people liked a prod (and the avg mostly showing how polarizing it is).
In fact often enough I find the # of thumbs up after a party more reliable than that party's results, especially when there was something weird going on in terms of rankings. Or that just means that I'm a bland average mainstream enjoyer, who knows ;)
But yeah - we _do_ have to keep the fuck you, but I'm also for reserving it for prods that we actively dislike and not just think are meh, or ranked too highly, or that crashed on our machine.
In fact often enough I find the # of thumbs up after a party more reliable than that party's results, especially when there was something weird going on in terms of rankings. Or that just means that I'm a bland average mainstream enjoyer, who knows ;)
But yeah - we _do_ have to keep the fuck you, but I'm also for reserving it for prods that we actively dislike and not just think are meh, or ranked too highly, or that crashed on our machine.
thümbing down prods that ranked too high in your opinion is pretty lame as it hits the creators and not the voters...
but yeah times changed and a thumbdown is now regarded much more hostile than it was, lets say, 20 years ago. somehow things got more personal i guess?
but yeah times changed and a thumbdown is now regarded much more hostile than it was, lets say, 20 years ago. somehow things got more personal i guess?
btw, speaking of constructive criticism: So pouet got these "lists" a while back. In principle already a good idea but what would be even better (in addition) would be "commented curation":
Allow users to create BBS-like pages that showcase e.g. a list of demos (screenshots + link to the prod page), along with some comments / background info. These should not be like BBS threads and allow only one post, which should also remain editable for the user after posting, e.g. for extending the list.
This could cover all kinds of topics (e.g. specific platforms, themes, techniques) and even things that are not allowed as prods, e.g. executable music.
Every now and then I watch ps' demo news and also read his thread on demozoo and there is so much that is so easy to miss.
Allow users to create BBS-like pages that showcase e.g. a list of demos (screenshots + link to the prod page), along with some comments / background info. These should not be like BBS threads and allow only one post, which should also remain editable for the user after posting, e.g. for extending the list.
This could cover all kinds of topics (e.g. specific platforms, themes, techniques) and even things that are not allowed as prods, e.g. executable music.
Every now and then I watch ps' demo news and also read his thread on demozoo and there is so much that is so easy to miss.
IMHO the voting is just useless and should be removed alltogether - in favour of the comments, which give a much better idea of what people like (or not) and why (which is the important bit afterall)
I'm sure that the thumbdown ratio would normalize to a considerable degree if Adok and BITS would start releasing again
Quote:
IMHO the voting is just useless and should be removed alltogether - in favour of the comments, which give a much better idea of what people like (or not) and why (which is the important bit afterall)
Respectfully disagree :).
Voting and ratings are extremely important for discoverability at the very least. Compo placing has been known to misfire sometimes, with great prods getting discovered by the community some time afterwards.
Amount of comments, clicks, or views are also a bad proxy. Is having flame wars in comments good? Is visiting prod page a click? Is downloading file? Clicking YT link? Views are also too sticky (look at the "all-time top" list on this very website).
You also evoked a good point. What do we even use voting for? There are multiple dimensions that I've noticed, e.g.:
- Feedback for authors: "you did great", "it's ok, but you need to work on ...", "what even is this garbage".
- Note-to-self: "I watched this already and liked it".
- Virtue signaling: "I liked this and want to resonate with fellow emotion enjoyers"
- Rating manipulation: "it was ok, but i feel it need more likes"; or the opposite "why is this getting too many likes, its unfair, i'll downvote"
- Recommendation: overall rating, and also sometimes you start noticing certain people in comments agreeing with you too often, and then get to their comment list, and start discovering amazing things that you somehow missed before.
I wonder if there's a value in untangling these ones, or whether i'm just overthinking it :)
The incentive for thumbs / likes on social media sites is to keep people busy so the company can rake in more money (selling adverts, selling your info, selling 'chart' positions / opinions / suggestions, ..).
On a non-profit site like pouet, this obviously has no meaning. You could remove the thumbs, just keep the comments, and nothing much would change (just generate the "top of the pops" from the number of comments).
What I am also curious about is how the OP's charts would look like if you'd filter out all users that have commented on less than 50 prods.
According to the stats this site has ~28k users. In reality it's probably more like 500, maybe a thousand more or less active ones.
On a non-profit site like pouet, this obviously has no meaning. You could remove the thumbs, just keep the comments, and nothing much would change (just generate the "top of the pops" from the number of comments).
What I am also curious about is how the OP's charts would look like if you'd filter out all users that have commented on less than 50 prods.
According to the stats this site has ~28k users. In reality it's probably more like 500, maybe a thousand more or less active ones.
Interesting statistics. But I don't agree on the conclusions. First of all, the whole thumb up / thumb down idea is rotten to begin with. A point system, like e.g. on asciiarena, would do the prods more justice and would be much more constructive. Secondly, I hate the discussion of presumably "national" traits, but because you mention your eastern European background: the "seriousness" with which demos are discussed in Russian scene media always struck me as often absurdly aggressive and completely out of place. On Hype blog or in diskmags, Russian-speaking sceners talk about demos as if they would be some real-life, proper-job tasks and not results of a hobby by computer people having fun. I don't like this vibe at all. So, instead of pushing for more thumb downs, I would rather argue for the abolishment if the whole thumb system, and for writing detailed (and constructive) critiques using words.
