Feedback on Repo.OpenPandora.org


And on the whole the issues to address are:

Voting neutral is neutral.

Binary doesn't diversify

A value doesn't state (merely implies) opinion and doesn't diversify areas

Statistics are fun, feedback is great.

Don't get stagnant over time

What i mean by that is giving something an ok vote should be based on the fact that apps are all ok (0) to begin with, which begs for a system where bad apps are penalized (-) and good ones are rewarded. (+)

Mediocre apps are slightly better than apps without votes. Voting neutral is better than not, it does something, and its not detrimental.

A system where all the values arent activly used on equal merit isnt a diverse system. This is a human bias, but nonetheless those issues are most important, since this is a human interaction system with other humans. As such, the human factor to encourage improvement and keep the playing field level and open to new takers is crucial.

I think for aggregating statistics, voting needs to be down to that primary vote, with optional differentiating three areas, and an optional comment section with binary votes on the comments and answers.

Many upvotes to a comment will show it along with the other aggregated info.

If this is done in a manner that lets interaction happen, the issues can be addressed, answered, and if something is done, the voting should reflect this. Unvoted stuff gets its own category, and if one vote is all it takes for your app to dissapear from the attention span, all it takes is one fix or a new release to mitigate it.

A system of trust is a good system → the rewards outweigh the backlash.

Thoughts: In keeping with simple and generating as much non arbitrary info as possible i think the categories should be 

purpose

aesthetics

stability

The first one being the gameplay of a game, and the content of a program.

With programs and games, we have to keep in mind for a program that tries to be what it is, (simple editor) when it does it right, it doesn't make sense to complain about what it isn't. (whole suite)

It could be a beautiful game that plays well and looks great, but if it crashes 3 stars doesnt tell anyone anything about why. Instead of making it an investigation reading out of, rather than reading into, makes sense on this scale.
 
Last edited by a moderator:
The metric temperature system is based on water and it freezes at 0 and boils at 100 given normal atmospheric pressure.

Foxblock: Wouldnt that mean once it gets "meaningful" it also gets stagnant.
I'm well aware of that, I've got a natural sciences degree. It's just a really crappy standard for use outside of the sciences as it doesn't in any way shape or form reflect ones natural sense of temperature. Basically how much does the temperature differ from my body temperature and what I feel comfortable with. 
It reminds me of the year I spent under the tyranny of the metric system, it took a solid 6 months before the temperature system made any sense to me at all, because it's not based upon human experience and wasn't based on a system I was used to either.
 
What's wrong with the metric system? Phase transitions (freezing and boiling) of pure water make much more sense to me, than the defining points of Fahrenheit, which are based on the freezing temperature of brine (salty water) and some estimate of average human body temperature.

Anyway, back on topic: the main problem is that not enough people give ratings. One explanation could be that many people use PNDManager instead of downloading from the repo website directly, and there are three big obstacles to giving ratings in PNDManager:

1) it is not clear to most people that it is even possible to give ratings from within PNDManager

2) it is relatively difficult to enable your repo account in PNDManager (would be easier if there was a GUI for that)

3) AFAIK you cannot easily get a list of "recently installed PNDs", which is what you would need if you want to give ratings to stuff you tried out recently.
That's because you were born and have always lived under the metric system. Phase transitions make no sense outside of the realm of science. Boiling water is about double the temperature of a human body and you get burnt way before that point. Likewise, you get cold well before the point where you hit freezing.
Fahrenheit sets 100 degrees to be approximately the temperature of the human body, give or take a degree or so. Which is a much more reasonable place to measure things from for people. Ever had your mother put her hand on your forehead as a child to check your temperature? She was comparing your temperature with approximately 100 F, not 0 C or 100C.

Likewise when people decide whether to put on a jacket or take one off, they don't compare that with freezing or boiling, they do that compared with their body temperature regulation needs not phase changes.

What's more, it's only 0c and 100c for boiling at sea level, when you're farther from sea level those values change somewhat.

There's this whole mythology and rationalization that's been invented to prop up the metric system for every day living, but in reality the metric system doesn't make anything that people commonly do any easier. I know both systems, I've used both system extensively, and unless one is engaged in science the metric system is just not any improvement over the Imperial system of measure.

Anyways, I think the current system of ratings is probably as good as we're going to get, anything else is going to make things convoluted and confusing, assuming it doesn't discourage people from voting outright.
 
Basically how much does the temperature differ from my body temperature and what I feel comfortable with.
0 is cold, 10 is cool, 20 is comfortable, 30 is bearably hot, 40 is uncomfortable.Works better for me than the approximately equivalent fahrenheit scale of 30, 55, 70, 85, and 100.
 
To use your own words, you think metric is bad

...because you were born and have always lived under the metric imperial system
---

Actually, I'll just point to what Stan and _wb_ said.

The phase transitions or water are extremely important in real life - and are also temperatures people can relate to.

0°C is when puddles are iced over and one must take care to not to slip up

100°C is the temperature of water coming out of a kettle

---

Also, the numbers associated with other imperial measurements are horrid to do arithmetic with.

For the normal user of metric:

Distance units --> divide or multiply by 1000

Mass units --> divide or multiply by 1000

Which is easier than inches, feet, yards, chains, furlongs, poles, miles, drams, ounces, pounds, hundredweights and tons (IMO)
 
Boiling water is about double the temperature of a human body
eh?

This depends entirely on what system you use.

I use Metric and say it's about triple

If I used Kelvin, I'd say boiling water is 16% hotter
 
Last edited by a moderator:
To bring it at least in line with the headline :) I'd once again like to propose two things for the website:

* The ability to "undownload" an app. That is, a small button on the "My Downloads" view that allows me to remove that app from my list of downloads. This means I won't get prompts of "updates available" for applications that I no longer use. 

* A filter on "My Downloads" that shows me the apps I've downloaded but not yet rated. This goes together with the main theme for this thread (how to get better ratings), but also with the one above - I don't want to rate apps I do not use.
 
Binary systems generally work even worse than the star system. There's no way of differentiating from a slightly below average app and one that's truly horrible. Plus, if it's broken that's reflected the same as the person didn't like it. Rather than getting mostly 1 and 2 stars for down and 4 and 5 for good, you just get 100% 1s and 5s, which is even less informative. Plus, some things are generally mediocre, and people should have the option of saying that they neither love nor hate it.
Well that simply is not true, since you can distinguish between an app which has 100 dislikes and 1 like (1%) and one that has 20 dislikes and 10 likes (33%) pretty well. Look at YouTube videos and the nice like/dislike bar, it's just as useful as a star-based indicator.

Mediocre thing generally will get less ratings, since they don't stand out and if you can filter for that it's all fine.

I agree though, a binary system might need a greater sample size to work well.

Also rating 3 or 4 instead of 5 does not give any additional information without a comment.

And, again, I don't think the voting process is the problem here, so let's not purely concentrate on that. People should get more incentive to leave a meaningful comment or feedback in general. (this could also include a system where people don't just vote on overall app quality, but genre specific things like "replay value" or "quality of port", etc.)
 
Last edited by a moderator:
I think a system like this could be good: You have the buttons "like"=1, "dislike"=-1 and "neutral"=0, but once you have clicked "like"/"dislike" you can click it again if you have a very strong opinion to make |<your opinion>|=a with a>1. Maybe a=2 or a=3.

For the votes -(aa1111111111111)0011111aaaaaaaaaa this could then be displayed in a bar like this: @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

With a purely binary voting system that score would most likely look like this @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ which is something completely different.

In my opinion that's better than a binary system because that way making very good content for a narrower target group is rewarded more than making something which is liked by a broad audience only just enough to "like" it.

But I agree that it's probably more important to get people voting than to make up sophisticated voting systems.

Edit: Replaced the "2"s in my example with "a"s because that's what I meant. I had first thought of this with a=2 and only then thought of the possibility to abstract this a little more because maybe another value than 2 might be good too/better.
 
Last edited by a moderator:
I like the notion of visualising the votes too, maybe it could show volume of votes better too.

Voting twice for extra opinion is not intuitive, and is really a -1  0  +1 system where you can go put an extra in at either end, its more complex in not being able to counter ones one vote, and it doesnt account for extreme neutral opinion. (?)

How about just doing the last 30 votes in such a visual style?
 
Last edited by a moderator:
On a completely different tangent, when uploading a new PND it is telling me that I'm uploading a PND with an ID that already exists in the database and nothing else. It wants to know if I want to overwrite the original but I have no idea what the original is. Did I accidentally put JAVA as the ID? Is it because an earlier attempt failed but still added the ID to the database? Is it someone else's PND that I just coincidentally chose the same ID for somehow magically? I don't know, so I'm uploading again with a completely different ID that is guaranteed to be unique.

edit: oh, and the feedback is that it shouldn't be doing this. If there's an ID collision it would be nice to know more information about it. At least the ID itself if nothing else is possible.

edit edit: ok, something has got to be broken. It was indeed conflicting on Java, having just gone and told it to overwrite the old PND of whatever it was. There is no mention of "java" anywhere in the PND I just uploaded at all, the PXML is devoid of all things java. The string "java" in a case insensitive search does not appear in the PND. So now I've disabled the java.pnd download until I can figure out what the heck is going on here.

edit edit edit: and I just accidentally deleted the previously archived java.pnd instead of downloading it. It should have been obvious the icon was not a download icon, but there still should probably be a warning or confirmation before it just wipes it out.
 
Last edited by a moderator:
I like the notion of visualising the votes too, maybe it could show volume of votes better too.

Voting twice for extra opinion is not intuitive, and is really a -1  0  +1 system where you can go put an extra in at either end, its more complex in not being able to counter ones one vote, and it doesnt account for extreme neutral opinion. (?)

How about just doing the last 30 votes in such a visual style?
Maybe there should be seperate buttons for voting +/-2. Extreme neutral opinion doesn't make sense to me. It's supposed to be more complex, because after all the world /is/ complex and over simplification doesn't help in most cases. I think there might be another point in your post which I don't get and thus can't refute/consider.

Edit: Oh, I forgot this point which I think I get:

How about just doing the last 30 votes in such a visual style?
I think this works better the more votes you have, so probably better not. But obviously the bar shouldn't become longer the more votes there are, this was just supposed to clarify my point with some ASCII art. In an actual implementation that should rather be a line with shows which percentages of people voting for each options.

But what I mainly wanted to say was just maths, the general idea, not the visualisations.

Ideas on visualisation: The line thickness could be increased as more votes accumulate (probably best on a log scale). The votes could also be displayed as -sum(n-aa+n-1)|n0|sum(n+aa+n+1) or -n-aa-n-1|n+a+n+1 (so either add it before the user sees it or let the user see the summands. One gives more information, one is easier/shorter) with ni as the number of votes for i.
 
Last edited by a moderator:
Well that simply is not true, since you can distinguish between an app which has 100 dislikes and 1 like (1%) and one that has 20 dislikes and 10 likes (33%) pretty well. Look at YouTube videos and the nice like/dislike bar, it's just as useful as a star-based indicator.
You can never claim it is "as useful" since it contains less information. Therefore a star system can not be reduced to a like/dislike system, and it's pretty obvious most people's opinions are far from being binary.
 
That's technically true... but it's not how it works in practice.   The reality is that (in the current system) anything less than 5 stars is a bad rating, so all that extra information pretty much boils down to a like/dislike choice anyway.   Worse than that,  what some would consider a 'good' rating. (say a 4)  is actually usually coming down on the 'dislike' side of the scale,  so it's a misleading system as well.

If you want a multilevel rating system,  the rating levels have to mean something or there's no point.

Amazon's system has some good points.  It couples ratings with a comment feature and for each rated item the system shows a histogram of the ratings.    You can then click on the bars of the histogram to view the comments left by the people who rated it that way.  

I like that because the comment usually explains why that level was chosen, then I can decide for myself whether what was important to the reviewers is important to me.

- Neelix
 
I have never been able to register on the repo. It may require special characters to register. Spaces between words? Dashes? Capitalized? I have tried them all.
 
Back
Top