Why Maemo Downloads are screwed

The very most basic way to find good apps is by popularity. Both iPhone and Android have the option to find the most popular apps, and in fact it’s the default. This is so brain-dead simple that in the app store review videos they just mention it: you can see popular applications (sure, as you would expect).

Maemo, on the other hand, doesn’t even have a concept of popularity. This ensures both that users would not be able to find cool apps, and that cool app writers would not be rewarded for their efforts.

This has to change.

Evidence

If you go to the Maemo downloads page, you would find a link that says “Popular“, but when you click it, you go to the “Hot” section. Is this really the same?

Well, let’s compare:

msn-pecan
downloads: 200390
download rate: ~700
votes: 29
stars: 4.3
page: 18

rulerjinni
downloads: 38742
download rate: ~450
votes: 1
stars: 3
page 1

givemefive
downloads: 51025
download rate: ~500
votes: 1
stars: 3
page 1

So, no, whatever measure is being used to determine the place in this list, it’s definitely not popularity. msn-pecan is clearly a more popular project, and it appears on page 18, which is like place #450, while both rulerjinni and givemefive share the page of the #1.

Could it be hotness? Nah, msn-pecan has a higher download rate, and it actually has comments and votes. So what is this listing is beyond my comprehension but it’s not useful.

In fact, if you look at the download stats of rulerjinni and givemefive you would see that they are almost identical, which is probably due to the fact that they come from the same author, so perhaps he made releases at the same time, which is the only kind of activity these projects get, and then, they get to the front page, where they get downloaded at the same rate, because they get the same visibility.

Totally unfair.

Fix

Three months ago I filed a bug report. The only response is basically saying that popularity is not measured in popularity.

Please, vote on the bug report so that this has a chance of getting fixed.

My guess is that the only way the community is finding good apps is through word-of-mouth, social media, blog posts, etc. Because clearly, there’s no place to find popular apps.

About these ads

5 thoughts on “Why Maemo Downloads are screwed

  1. Pingback: Why Maemo Downloads are screwed « Felipe Contreras | Maemo Meego

  2. The ‘hot’ downloads list is not meant to be all-time popular, it instead uses a concept called “application karma” to determine the most popular downloafs a the time. There is a separate list for most downloaded apps, you can find it in:

    http://maemo.org/downloads/downloads/Maemo5/25/

    So the correct fix to your bug report would be to link this as “see the most popular apps of all time”

  3. @Henri I know what the ‘hot’ list is meant to be, but it’s not doing it right; msn-pecan has more downloads a day than rulerjinni, so it should be ‘hotter’ shouldn’t it?

    But ‘hot’ != ‘popular’. So they should not put a “popular” link and then go to the ‘hot’ list (which is wrong anyway).

    Also, “see the most popular apps of all time” is redundant and too verbose. What other kind of list popular apps is there? If it’s temporal, there’s already “hot” for that, and if it’s not “popular” can be used just fine.

    Moreover, total number of downloads should not be the only way to determine if some app is popular, there’s also the number of stars, and votes.

  4. Yes, app karma uses ratings and comments too. And agreed, hot vs. popular might be the good wording. As for why rulerjinni is ‘hotter’ than msn-pecan, it is probably because while total downloads per day is lower, the change in them is bigger.

  5. @Henri So the link you provided uses “app karma”? It seems maemo.org is back up, so I could take a look, and that indeed seems to be a good popularity indicator, however, I don’t see any link pointing to that url.

    This “app karma” should also be used when clicking on the categories (right now seem to be sorted by number of downloads).

    And finally, that “hotter” calculation is horrible. If an app has 1 million downloads per day steadily it would be less hot than one that doubled its downloads from 1 to 2. The hotness should be calculated as the velocity of downloads, not the acceleration. If you want acceleration, then a different term should be used, like “trending”.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s