Decline of easy-to-fix tags in issues

[Anand Bansal]

I have intermediate knowledge of python and I am currently studing mathematics. I want to contribute to sympy but easy-to-fix issues are very less and old. Can someone suggest where I should start? I am currently reading the codebase and want to get my feets wet into these easy to fix issues. Then I will start working on harder issues

Thank you

[Oscar Benjamin]

Hi Anand,

Yes, it is true that the issues labelled as "easy to fix" are not such
a good source of things to work on. The problem is that whenever an
issue is labelled as easy to fix if it really is easy then it quickly
gets fixed. What is left then is typically the issues that are not
actually as easy as they seem which is not very helpful.

My suggestion is to pick some topic or area of the codebase where you
are familiar with the mathematics or that you are interested in
learning about and then focus on that. Issues are generally labelled
by topic and reading through issues for a particular topic label is
probably the most helpful way to find something to work on.

There are other ways to contribute besides just opening pull requests.
For example many of the open issues are in fact already fixed and just
checking if the issue still applies and commenting that it could be
closed is very helpful. Also what many issues really need is for
someone to investigate the problem and after that investigation the
code changes might be very simple so the more useful part is the
investigation rather than opening a PR.

Many hopeful contributors right now just seem to be asking an AI to
generate pull requests which is the least useful thing that an AI can
do. I would be much more interested in seeing people use AI to figure
out which issues are already fixed, which PRs are duplicates, or to
actually debug something and identify the cause of a given problem.
The fact that this isn't happening makes me think that AI is not
really capable of it so it is good that you are actually reading the
code yourself rather than just using an AI to spew out code.

Oscar

On Tue, 16 Dec 2025 at 11:37, Anand Bansal <anand4...@gmail.com> wrote:
>
> I have intermediate knowledge of python and I am currently studing mathematics. I want to contribute to sympy but easy-to-fix issues are very less and old. Can someone suggest where I should start? I am currently reading the codebase and want to get my feets wet into these easy to fix issues. Then I will start working on harder issues
>
> Thank you
>

> --
> You received this message because you are subscribed to the Google Groups "sympy" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sympy+un...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/sympy/b7da9c93-d5bb-44f2-bafd-cfe49540c989n%40googlegroups.com.

[Anand Bansal]

I completely understand about the easy-to-fix issues.I will start by reviewing some pr from my field of interest .Also, even if AI actually is able to map pr and issues solved, It can't be trusted and that leads to code bloating.

Thank you

Anand (AnInsaneJimJam)