Votes or Popularity rating in AUR?

When you download a package from the AUR, do you generally pay attention to the votes or the popularity rating of the package?

1 Like

I virtually never go to the AUR itself for anything - yay whatever is good enough for me :laughing:

1 Like

If I find an app I’m interested in and I’m not familiar with it and it’s available in the AUR, I do go and check whether or not it’s flagged out of date. That’s about it, though.

Similar to using alternativeto.com, occasionally those stats help pinpoint what’s at least popular, when reviewing a range of alternatives.

I wouldn’t make that the final word on the matter though, it’s just a start.

Agreed. The reason I even bothered to ask this question is I see an inconsistency between the two ratings. Was just curious how users here generally perceive it.

It can help you to know how widely used it is.

It may be useful, for example, when comparing different options for the same software.

yay also shows you votes and popularity. That is what the numbers in parens are:

1 aur/librewolf 130.0.1-1 (+136 4.76) 
6 Likes

Huh, TIL. I have never even questioned what those numbers mean :sweat_smile:

3 Likes

great question except I always thought the +number was # of people with downloaded/installed package.…not popularity etc… silly rabbit.

@dalto
so with my ignorance it really mean nothing at all. those numbers are meaningless.
EX:
package A performs well but small following +22
package B performs well and large following +100

per capita what’s the difference? i.e. no correlation between popularity and performance.
the only number per capita (I know I am using that loosely) that matters now is how many people think it’s a sucky app. then you might have a rounded picture…

1 Like

It isn’t intended to tell you if the software/package is good or bad. If you were thinking it was similar to a like/dislike system, that isn’t what it is at all.

the notion of ‘popularity’ sent me on that good/bad tangent, of course. these numbers still murky to me but its ok

I tend to look at a range of things, whether or not it’s out of date (and by how much and when the last update was/update consistantinty), the comments and if there are multiple packages for a similar thing is when I look at the popularity and votes. Using the combined info is a pretty good way of determining whether to install or not)

2 Likes

“Popularity is the sum of all votes with each vote
being weighted with a factor of 0.98 per day since its creation.”

It only shows votes cast by people - so it’s not necessarily an accurate reading.

It sometimes helps when comparing similar items though, and votes are weighted down over time which stops a popular package from 1995 dominating the field:

audacious-vortigo-themes, for example, got 19 votes…

if no more are cast, after a month that’ll go down to 10 votes, making it a ‘live’ rating.

2 Likes

Yeah, I pay attention to it and make sure to upvote it if I think it is good. Over the many years that I have been using an Arch based distro. Nothing makes me happier than seeing AUR packages become official packages.

3 Likes

Don’t tell the wife :rofl:

Today you get the :star::star::star::star::star: Geek award.

3 Likes

No.

Most software is a “no choice, no problem” scenario for me. I’m willing to take the minor risk of it breaking my system (nothing has, yet), and if it doesn’t work, well, at least I tried it.

1 Like

For me I take in to account both but I also check the update history to see if it’s the same person under the changes area for a while. Comments are good to check too in case there is a current issue that is flagged that needs fixing. The other is the date, I haven’t come across any but there’s bound to be an outdated package at some point.

And I also have a quick look, especially if it’s a new package I am downloading, the PKGBUILD so see what it does to see if there are any obvious issues, if an updated package you can see the changes made to it. It doesn’t necessarily mean the issue is malicious, sometimes mistakes can potentially happen when making them. Even Steam wasn’t immune to a mistake, such as wiping out the whole OS.

And if someone wants to then I also check the upstream URL to double check.

It seems like a lot, but quite quick for me to do now, mostly for new packages. If it’s any relief I haven’t come across a single dodgy program yet. But it’s not like this isn’t an issue with any other package, I just like checking this stuff since I can and it’s transparent while on windows I couldn’t.

If I’m mistaken someone will let me know, I don’t exactly do this often as I don’t download packages all the time.

2 Likes

Yes, once you get used to doing this, it is very fast to do some basic checks.

I have never found anything malicious, but I do occasionally find packages using unofficial upstream sources which I avoid.

4 Likes

I’m a bit late for this reply, but good you mentioned this. I mentioned I checked the upstream URL source but not what I checked for.

I usually take into account both metrics, like pikaur does by default when it shows package list.