Popularity perpetuation  edit   (Category  edit)

Aliases: What’s popular becomes even more popular, Self-perpetuation of popularity, Popularity perpetuation tendency, The amplification of popularity, Popularity polarization, Popularity permanency, Popularity establishment, Popularity amplification loop

See also: Deviancy amplification spiral, He who comes first wins

This is a big, unfortunate, difficult problem, in my opinion.

(Not strictly popularity; also includes similar shades of meaning, such as awareness of a topic, …)

UserVoice

Make it easier for newer suggestions to rise in popularity / Make it easier for new suggestions to attract more votes https://uservoice.uservoice.com/suggestions/6356. Retrieved on 2007-05-11 11:18.

Right now the order in which suggestions are displayed is based solely on how many votes they have ACCUMULATED. While I can see why this would be the ideal view for ADMINS to see, I’m concerned that if that’s the way USERS see it too, it will inevitably result in “POPULARITY AMPLIFICATION”, where those suggestions that are ALREADY popular end up receiving the bulk of users’ votes while those suggestions at the bottom of the list end up being overlooked. Here are some possible solutions… 1. 1 TylerRick I suggest that the ordering of the list take into account other factors, including possibly the “freshness”/”staleness” of the suggestion, whether there have been RECENT votes, or even randomness… 2. 1 TylerRick As it is, I’m afraid all the suggestions that are on the top of the list (the popular ones), will end up attracting the majority of new (or “transfer”) votes. People will likely use up all their votes on the first 5-10 ideas they see — the ones that ALREADY have the most votes and hardly “need” more votes — and NOT EVEN READ DOWN FAR ENOUGH to see the newer ideas at the bottom… 3. 1 TylerRick The ideas at the bottom of the list might be just as good as those at the top but haven’t been voted up simply because they don’t get the EXPOSURE they need to rise in popularity. 4. 1 TylerRick Possible solutions: 1. reward RECENT votes/trends/activity/comments: if a suggestion has gained some votes recently, list it ABOVE a suggestion that ALREADY has lots of votes but has been fairly static/hasn’t attracted any new votes recently (order by “what’s hot”). I’m not saying its RANK should go down simply for not attracting recent votes, just that it doesn’t need to be DISPLAYED at top. 5. 1 TylerRick (The original suggester shouldn’t be overly rewarded for voting on their own suggestion, so only start rewarding for recent vote activity when OTHER people have voted for it…) 6. 1 TylerRick 2. Allow bad suggestions to be voted DOWN (negative votes ). So when a new user comes to the list and sees that the #1 suggestion has 25 +’s and 3 -‘s, he MIGHT stop long enough to read the comments left by the negative voter before deciding to waste his votes on “me tooing” an already popular suggestion… 7. 1 TylerRick 3. Order the list view RANDOMLY by default and only order by popularity if they click “most popular”…? 4. Order by date and show the newest (with >= 2 fans) at the top…? 5. Order by number of distinct voters each suggestion has…?

Not used:

In other words, like many services that rely on rankings (Google search, Digg, etc.), it suffers from popularity amplification/perpetuation: those things are popular already become more popular still, while those things that are relatively obscure become obscurer still, through no fault of their own.

It feels like it’s biased towards the status quo (the most popular suggestions REMAINING the most popular) and it’s too hard for new ideas to gain popularity.

Digg, etc.

…

Tags (del.icio.us, …)

…

Case study: In context of [search engines (category)]

So I search for keyword “foo”. Google dutifully gives me a list of the top 10 pages that match that keyword, after ranking the hits on the basis (at least partially) of how many other sites out there link to that page. In other words, Google essentially ranks pages by how popular they are — based on the reasoning that 1000s of people wouldn’t be linking to this site if it weren’t relevant, credible, trustworthy, etc. I check out the top 10 hits and find that about 5 of them are actually good/interesting. I don’t bother going on to page 2 (results 11-20), because what would the point be? — I’ve already found what I’m looking for.

So then I, myself a site author, come along and build a page about “foo”. Naturally, I want to include some links to other sites about “foo”, because some of those pages may have content that is interesting but beyond the scope of my article. So I add some of the top links from my previous search to my “links” section.

And in a way I’ve just perpetuated the popularity of those sites…

Even if there were better ones out there… They will never get discovered, and hence never linked to, because they aren’t popular enough to appear on page 1 of the search results.

Obviously that’s an over-simplification, but I think it’s largely true.

Languages

Mandarin Chinese is the most commonly spoken language in the world.

Spanish is the second most commonly spoken language in the world. That and the fact that there are a growing number of Spanish speakers in America probably explain why Spanish is one of the main languages taught as a second language in America.

English is also one of the most spoken languages in the world. And therefore people learn it.

Programming languages

One of the main reasons PHP continues to be such a dominant language used on the web is because of its popularity. Because “everyone” uses/wants it, hosting providers naturally have it installed on their systems, because they know most users will demand/require it. And so the ubiquitous “LAMP” platform was born everyone just naturally assumes it will be available.

Other languages, on the other hand (Smalltalk, Python, Ruby), don’t enjoy that kind of pre-established popularity and their users must therefore struggle even to find a hosting provider that has those languages/modules available.

This lack of availability is then used against the languages — “we don’t want to go with language X because it’s hard to find support for that language” — thus unnecessarily turning people away from it and perpetuating their lack of popularity. … when it’s not even the languages’ fault that they’re not as popular — it’s the fault of those who refuse to adopt it: they’re the ones keeping it from becoming popular. But it’s a vicious, interconnected cycle; I’m not blaming any individuals for it.