It comes as a surprise to me that Github stars can be purchased for money. Personally, I don’t bother looking at the number of stars because even if they were accurate, it wouldn’t really tell me anything that I couldn’t learn by looking at other aspects of the repository instead.
I have a habit of checking the age difference between the earliest and most recent commits because doing so allows me to ensure that the project is not one that someone has spent a couple of weeks coding up, dropped on github, and then forgotten about. I’ll look into those problems on there as well. I’m looking for more closed issues than open ones, but before I do that, I’ll give each of them a cursory look to get a ballpark estimate of how many of them are actually significant problems. In addition to that, I receive signals from the readme and the docs. If there are problems with those, it is not necessarily a deal breaker, but it would be very helpful to my opinion if they existed and were both clear and detailed.
When I need to select a specific tool for a project from among several different repositories, I find that using stars is helpful.
When deciding between two repositories, I give a lot of weight to which one has a significantly higher star rating. The recency of the commits is unquestionably significant; however, the fact that many other people have starred the repository is evidence that there is activity and attention being paid to it.
Github stars were mentioned in the job description and during recruitment pitches at the company where I previously worked. They did this on a regular basis to encourage employees to go to the company’s repos in Github and star them. One of the things that was reported in all-hands meetings was the progress that had been made on Github: “we’ve surpassed X in Github stars” (applause).
(However, the name of the company X is more well-known than that of my former employer was.)
Quite some time ago, I listened to an episode of Freakonomics in which the host discussed how companies use proxies to both improve their image and cover up any incompetence that may exist within the company. To illustrate this point, I will use the fact that many companies have chosen ritzy names beginning with A (such as AAA Plumbers) in order to ensure that they are listed first in business directories. It was later discovered that these companies were either extremely incompetent or even fraudulent.
When evaluating an open-source software (OSS) project, the activity of the community is an important indicator. The number of Github stars a user has is not a reliable indicator of community activity. To begin, as the aforementioned article has demonstrated, it is susceptible to gaming. Additionally, since selecting stars is a very low-barrier action, it does not necessarily indicate that the individual who starred the project will actually use it.
Github issues and comments on slack, Discord, or Discourse are, in my opinion, two excellent indicators of the level of community activity. One of the most important aspects of github issues, in my opinion, is that it is not a good sign if the core team is responsible for the majority of the issues on github. You want to see a wide variety of problems reported by customers or end users, and not by the team. This is a useful indicator that can be used to determine whether or not the project is solving a real problem. The entry requirement for playing Stars is extremely low. The same principle applies to the comments in the slack channel; they need to have both volume and freshness.