rid of transient user's charts

[purple]

don't know if this has been suggested before, but it would be nice to rid the site of the charts of more transient users e.g. those who come onto the site, post a chart, and never return. What's the use of rating and commenting on a chart if you're sure not to get a response?

This would be a topic for debate. I'm sure Naples has some ideas. Obviously the cutoff would be subjective but I would say a chart should be deleted if it's user hasn't visited the site in four-ish months. I don't think the users ratings should be deleted though. Also, there are some charts that would be nice to keep because of their history on the site e.g. joannajewsom, johnnyrocketfingers, bc1991 and I think these charts should be voted on.

A difficult process I know, but it would really tidy up the site.

Love you, albummaster

[RFNAPLES]

You know they say ideas are a dime a dozen and mine are probably worth about 2 cents each (24 cents a dozen) if I may boast

I do not recommend purging recent charts but would like to see listed the last date a member was on site. Knowing the date of the last chart change and visit date would then allow each viewer to make their own determination of relevancy and usefulness.

We should consider eliminating all stagnant charts and ratings over 10 years old (including “recognized” charts). Those charts and ratings may not be relevant, certainly not topical.
_________________
Top 100 Greatest Music Albums by RFNAPLES
Bubbling Under The Top 100 Greatest Mus...y RFNAPLES

[videoheadcleaner]

I think the main focus of this should be on those charts that are deliberate attempts to skew stats. For example a chart might only have 1 album on it and that user has not returned or commented on anything for a year. It is more likely that this is a 'ghost' user.

Maybe there is a cutoff that blocks the ratings from entering the overall chart but still remains. Similar to a user's choice for no member feedback, maybe it could work with no user input.

[albummaster]

Personally, I think it may not be a great idea to get rid of any user charts. I don't think many people would appreciate their charts being deleted (and some users do return even after more than a year of inactivity). We'd have to pick some arbitrary time period, which would defeat the object if it was set too long. In terms of greatest albums of all time, one or two years is still a tiny amount of time in the greater scheme of things. How many albums released in 2009 or 2010 are on people's charts?

If a member chart allows feedback, providing the email address is valid, the original user would still receive an update to say their chart has been rated or commented (if they subscribe to these notifications). If they decide not to respond, it's up to them (nobody is forcing them to).

I think allowing to see when somebody last visited the site is a bit of a privacy issue. Some people might not want others to know when they are onlne, or when they last visited. We could allow people to opt in or out of publishing this information, but then it defeats the reason for wanting it in the first place. We show when a chart was last updated and that should be enough information by itself to know how recent or relevant a chart is.

I do agree with perhaps getting rid of recognised lists over 10 years old (as they miss out an entire decade). I think we should still list them on the site, but perhaps not allow their rankings to contribute to the overall rankings. We could then apply the same rule to member charts, once the site reaches ten years old (in five year's time!) i.e. purge all member lists over ten years old. Perhaps this would be too long, but at least it would be consistent.

We do focus quite a lot of energy in stopping people gaming the system. We now publsh IP addresses and the last update time. These are both strong indicators of whether somebody is trying to submit multiple charts. I agree charts with one entry are not particularly useful, but they only contribute 100 rank points whereas a chart with 50 entries contributes 2,550 points, so the skewing is very minor and favours larger charts.

In the next couple of weeks, we will publish our roadmap for 2011 and I think this will make some of the issues raised here seem a bit less central.

EDIT: The roadmap has now been published.

albummaster

[videoheadcleaner]

In a similar vain, anyone looked at Doctor Squid?

12 charts with a majority of them the same IP address.

[purple]

yeah, this is one of the more hilarious cases of trying to "game the system"

[albummaster]

I think the problem with Dr. Squid might have been raised before and the problem is that college campuses, large offices etc sometimes connect to the internet via a single proxy server which means that everyone from the same location wll have the same IP address.

Each of the charts featuring Dr. Squid was created using a very different email address (usually there is some sort of pattern in the last name, first name etc) and I can't believe that one person would want to go to all of this effort to create this many charts by themself.

There are nine charts featuring Dr. Squid from the same IP address which have been created over a period of three years:

User - Top x - Chart created - Chart last upated
doop - Top 21 - 2007-01-15 23:14:45 - 2010-04-06 14:36:57
godsavethequeen - Top 21 - 2008-02-11 16:17:07 - 2010-04-06 18:50:29
vidinsky - Top 10 - 2008-07-30 04:35:55 - 2010-04-13 20:44:22
powrtoch - Top 26 - 2010-04-01 19:58:59 - 2010-04-08 14:20:47
Desh1 - Top 2 - 2010-04-06 18:21:57 - 2010-04-06 18:21:58
Nara - Top 3 - 2010-04-06 18:46:45 - 2010-04-06 18:46:45
roderickfabndr - Top 2 - 2010-04-07 15:50:53 - 2010-04-07 15:50:54
craftybtch - Top 4 - 2010-04-07 19:02:50 - 2010-04-07 19:02:50
karinaamariodt - Top 5 - 2010-04-13 20:41:15 - 2010-04-13 20:41:16

The email addresses of these users are all very different and were not from disposable accounts (such as mailinator.com etc). No other albums in these charts are duplicated apart from Weezer - Blue Album (x3). What do people propose that we do?

[RFNAPLES]

All of the mentioned charts are small. Only a couple of the "members" have made comments or posts. I would make a direct inquiry to each email. If no response or bounce back, eliminate. They could be legitimate; I often share my IP through Wi-Fi. Perhaps friends were shown BEA and tried it out while on the same IP at or following the same concert/festival. But check it out-smells fishy.
_________________
Top 100 Greatest Music Albums by RFNAPLES
Bubbling Under The Top 100 Greatest Mus...y RFNAPLES