I got pretty burned out by this project and haven't been active for many months. This project is called Awesome and lists are expected to be of (awesomely) high-quality. Unfortunately, many submissions were not and required a lot of effort on my part to just comment on things that were already listed in the pull request guidelines. This doesn't scale to one person when there are many new submissions a day.
I'm ready to put some work into this project again, but I need your help.
There's a new requirement for submissions now. When you submit a new list, you're required to review at least 2 other open pull requests. This is my experiment to try to make the Awesome project self-sustaining. Read more here.
This requirement also applies to existing submissions, so if you like to see your submission merged, review at least 2 other pull requests and I'll prioritize getting your submission merged.
Please also go through the list guidelines again on your own submissions. You have probably missed something.
Originally created by @sindresorhus on GitHub (Jul 28, 2018).
Original GitHub issue: https://github.com/sindresorhus/awesome/issues/1363
I got pretty burned out by this project and haven't been active for many months. This project is called Awesome and lists are expected to be of (awesomely) high-quality. Unfortunately, many submissions were not and required a lot of effort on my part to just comment on things that were already listed in the pull request guidelines. This doesn't scale to one person when there are many new submissions a day.
I'm ready to put some work into this project again, but I need your help.
There's a new requirement for submissions now. When you submit a new list, you're required to review at least 2 other open pull requests. This is my experiment to try to make the Awesome project self-sustaining. [Read more here.](https://github.com/sindresorhus/awesome/blob/master/pull_request_template.md#requirements-for-your-pull-request)
**This requirement also applies to existing submissions, so if you like to see your submission merged, review at least 2 other pull requests and I'll prioritize getting your submission merged.**
Please also go through the [list guidelines](https://github.com/sindresorhus/awesome/blob/master/pull_request_template.md) again on your own submissions. You have probably missed something.
Glad to see you back active on this! How about trying to automate part of the review process? At least some of the requirements for the pull request could be validated by a script I will put down some ideas here: https://github.com/sellisd/awesomeReview to see if it can be of use
<!-- gh-comment-id:408637069 -->
@sellisd commented on GitHub (Jul 28, 2018):
Glad to see you back active on this! How about trying to automate part of the review process? At least some of the requirements for the pull request could be validated by a script I will put down some ideas here: https://github.com/sellisd/awesomeReview to see if it can be of use
@transitive-bullshit commented on GitHub (Jul 29, 2018):
@sellisd see awesome-lint which seems like a solid starting point, though it's in need of a lot of love.
<!-- gh-comment-id:408650197 -->
@transitive-bullshit commented on GitHub (Jul 29, 2018):
@sellisd see [awesome-lint](https://github.com/sindresorhus/awesome-lint) which seems like a solid starting point, though it's in need of a lot of love.
Making the project self-sustaining and avoiding burnout is a good idea. It would be helpful to provide a bit more guidance regarding the review process.
A list of PRs requiring review. Unless I'm missing something, now reviewed and unreviewed PRs all appear together.
A way to mark PRs whose owners have fulfilled the "review at least two PRs" requirement or to notify you that the requirement has been fulfilled.
<!-- gh-comment-id:408667749 -->
@dspinellis commented on GitHub (Jul 29, 2018):
Making the project self-sustaining and avoiding burnout is a good idea. It would be helpful to provide a bit more guidance regarding the review process.
* A list of PRs requiring review. Unless I'm missing something, now reviewed and unreviewed PRs all appear together.
* A way to mark PRs whose owners have fulfilled the "review at least two PRs" requirement or to notify you that the requirement has been fulfilled.
<!-- gh-comment-id:409053873 -->
@TheKitty commented on GitHub (Jul 31, 2018):
@sindresorhus - we'll be reviewing
https://github.com/sindresorhus/awesome/pull/1160 and
https://github.com/sindresorhus/awesome/pull/1127
We'll be noting issues that are not in line with the guidelines to include syntax, comments, etc. Is there's anything else you require?
It seems obvious that the current approach of merging pull requests does not scale. Currently, there are four pull requests that have been open for more than one year and there is a total of 102 open pull requests. At least some of them seem to have met all requirements including the authors having done the reviews of other pull requests. Judging from the list of commits, @sindresorhus is the only (active) user with merge privilege.
I have seen some pull requests that were closed by their respective authors because they lost patience or interest. The process of being listed is really cumbersome and I understand everybody who is not willing to create a pull request just to have it open for months (or even over a year) without any reasonable prospect of being listed at all.
I understand that maintaining this project is time-consuming and often even annoying. But letting this project die a slow death is a pity. This only works when more people with proper privileges are involved. Sindre, I think it is time to add more contributers to this repository. What do you think?
<!-- gh-comment-id:425671250 -->
@troxler commented on GitHub (Sep 29, 2018):
It seems obvious that the current approach of merging pull requests does not scale. Currently, there are [four pull requests that have been open for more than one year](https://github.com/sindresorhus/awesome/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc) and there is a total of 102 open pull requests. At least some of them seem to have met all requirements including the authors having done the reviews of other pull requests. Judging from [the list of commits](https://github.com/sindresorhus/awesome/commits/master), @sindresorhus is the only (active) user with merge privilege.
I have seen some pull requests that were closed by their respective authors because they lost patience or interest. The process of being listed is really cumbersome and I understand everybody who is not willing to create a pull request just to have it open for months (or even over a year) without any reasonable prospect of being listed at all.
I understand that maintaining this project is time-consuming and often even annoying. But letting this project die a slow death is a pity. This only works when more people with proper privileges are involved. Sindre, I think it is time to add more contributers to this repository. What do you think?
@troxler Just because there are lots of open PRs doesn't mean this project is dying. This repo is just a list of what I consider good. The beauty of open source is that anyone can create a list. It doesn't have to be listed here. There are currently 1500 repos tagged with the awesome-list tag and probably lots more that are untagged. I try to prioritize reviewing PRs here, but most PR reviews are still just me copy-pasting items from the contribution guidelines, which is very demotivating. I also have a lot of other OSS responsibilities.
I have tried adding more collaborators multiple times before without success. You don't even need to be a collaborator to help out. Merging the PRs are easy. Thoroughly reviewing them is the time consuming part, and anyone can help out with that, but few do.
I'm also working with community members on a awesome linter which will hopefully lessen my PR review work.
<!-- gh-comment-id:425672403 -->
@sindresorhus commented on GitHub (Sep 29, 2018):
@troxler Just because there are lots of open PRs doesn't mean this project is dying. This repo is just a list of what I consider good. The beauty of open source is that anyone can create a list. It doesn't have to be listed here. There are currently 1500 repos tagged with the [`awesome-list`](https://github.com/topics/awesome-list) tag and probably lots more that are untagged. I try to prioritize reviewing PRs here, but most PR reviews are still just me copy-pasting items from the contribution guidelines, which is very demotivating. I also have a lot of other OSS responsibilities.
I have tried adding more collaborators multiple times before without success. You don't even need to be a collaborator to help out. Merging the PRs are easy. Thoroughly reviewing them is the time consuming part, and anyone can help out with that, but few do.
I'm also working with community members on a [awesome linter](https://github.com/sindresorhus/awesome-lint) which will hopefully lessen my PR review work.
Just because there are lots of open PRs doesn't mean this project is dying.
No it does not indeed. I've been following this project for a while and I think during this time, the number of open pull requests even decreased. But many of them were closed because their authors apparently lost interest and gave up or just did not answer anymore. I strongly think that is also has to do with the amount of time it takes from the request creation to actually being merged (or closed).
I try to prioritize reviewing PRs here, but most PR reviews are still just me copy-pasting items from the contribution guidelines, which is very demotivating. I also have a lot of other OSS responsibilities.
I do realize that (kudos and thanks btw!). And I especially get the "demotivating" part as I've seen many examples of authors that clearly did not read the guidelines at all.
You don't even need to be a collaborator to help out. Merging the PRs are easy. Thoroughly reviewing them is the time consuming part, and anyone can help out with that, but few do.
That may be true. But don't you need tools to see at a glance which pull requests could potentially be merged (i.e. someone reviewed it and thinks it is OK) and which need more work? Do you go through all 100+ requests regularly just to check that (well, just the once that were updated recently but you get the point)? Wouldn't it be helpful to have appropriate labels or a being able to filter by approvals? Contributors without privileges can't use either. It would also make sense for the reviewers to quickly find requests that actually need a review without skimming through a long list first. And personally, I also find it demotivating to do reviews without proper tools.
<!-- gh-comment-id:425674400 -->
@troxler commented on GitHub (Sep 29, 2018):
> Just because there are lots of open PRs doesn't mean this project is dying.
No it does not indeed. I've been following this project for a while and I think during this time, the number of open pull requests even decreased. But many of them were closed because their authors apparently lost interest and gave up or just did not answer anymore. I strongly think that is also has to do with the amount of time it takes from the request creation to actually being merged (or closed).
> I try to prioritize reviewing PRs here, but most PR reviews are still just me copy-pasting items from the contribution guidelines, which is very demotivating. I also have a lot of other OSS responsibilities.
I do realize that (kudos and thanks btw!). And I especially get the "demotivating" part as I've seen many examples of authors that clearly did not read the guidelines at all.
> You don't even need to be a collaborator to help out. Merging the PRs are easy. Thoroughly reviewing them is the time consuming part, and anyone can help out with that, but few do.
That may be true. But don't you need tools to see at a glance which pull requests could potentially be merged (i.e. someone reviewed it and thinks it is OK) and which need more work? Do you go through all 100+ requests regularly just to check that (well, just the once that were updated recently but you get the point)? Wouldn't it be helpful to have appropriate labels or a being able to [filter by approvals](https://github.com/sindresorhus/awesome/pulls?q=is%3Apr+is%3Aopen+review%3Aapproved)? Contributors without privileges can't use either. It would also make sense for the reviewers to quickly find requests that actually need a review without skimming through a long list first. And personally, I also find it demotivating to do reviews without proper tools.
I strongly think that is also has to do with the amount of time it takes from the request creation to actually being merged (or closed).
Yup, but in most cases it takes a while because people keep opening PRs without having put much effort into their list or PR, so they require a lot of review time, which means I have less time for other PRs, and suddenly there are 100+ open PRs...
Do you go through all 100+ requests regularly just to check that
I go through my Awesome repo notifications once in awhile to see if any new PRs are ready for another review.
Wouldn't it be helpful to have appropriate labels or a being able to filter by approvals? Contributors without privileges can't use either.
Sure, this all sounds nice, but you don't just add random people to a project and hope they do a good job. You add people that already regularly do a good job.
It would also make sense for the reviewers to quickly find requests that actually need a review without skimming through a long list first.
All PRs need review, including already reviewed ones. Even when I get external review help, they miss a lot of things, so the more reviews of the same PR, the better.
<!-- gh-comment-id:425675577 -->
@sindresorhus commented on GitHub (Sep 29, 2018):
> I strongly think that is also has to do with the amount of time it takes from the request creation to actually being merged (or closed).
Yup, but in most cases it takes a while because people keep opening PRs without having put much effort into their list or PR, so they require a lot of review time, which means I have less time for other PRs, and suddenly there are 100+ open PRs...
> Do you go through all 100+ requests regularly just to check that
I go through my Awesome repo notifications once in awhile to see if any new PRs are ready for another review.
> Wouldn't it be helpful to have appropriate labels or a being able to filter by approvals? Contributors without privileges can't use either.
Sure, this all sounds nice, but you don't just add random people to a project and hope they do a good job. You add people that already regularly do a good job.
> It would also make sense for the reviewers to quickly find requests that actually need a review without skimming through a long list first.
All PRs need review, including already reviewed ones. Even when I get external review help, they miss a lot of things, so the more reviews of the same PR, the better.
Yup, but in most cases it takes a while because people keep opening PRs without having put much effort into their list or PR, so they require a lot of review time, which means I have less time for other PRs, and suddenly there are 100+ open PRs...
Yes I can imagine. First thing I would do is to close the PRs of authors that did not respond to any comments in months. It seems useless to let people do yet another review on those anyway.
All PRs need review, including already reviewed ones. Even when I get external review help, they miss a lot of things, so the more reviews of the same PR, the better.
That would be ideal yes. What currently is happening is the following: Some PRs (mostly newer ones) get multiple repetitive reviews while other PRs (mostly older ones) don't get any. My own PR has been open for exactly one year today and did not get a review. I tried pinging you, I updated the list several times to comply with updated rules, I fixed the merge conflict caused by #1343, I ran the linter and did some more updates, I did reviews on other PRs and pinged you again. Well, a year has passed since I created the PR and it is still open. How would anyone be interested to go through all that hassle just to be ignored?
I highly appreciate what you do (here and on the many other projects you have) and I definitely understand that it all takes a lot of time. All I'm saying is that it does not seem to scale. Having that many open PRs of which many are so old does not attract new contributions in my opinion.
That is all I wanted to say :-)
<!-- gh-comment-id:425962975 -->
@troxler commented on GitHub (Oct 1, 2018):
> Yup, but in most cases it takes a while because people keep opening PRs without having put much effort into their list or PR, so they require a lot of review time, which means I have less time for other PRs, and suddenly there are 100+ open PRs...
Yes I can imagine. First thing I would do is to close the PRs of authors that did not respond to any comments in months. It seems useless to let people do yet another review on those anyway.
> All PRs need review, including already reviewed ones. Even when I get external review help, they miss a lot of things, so the more reviews of the same PR, the better.
That would be ideal yes. What currently is happening is the following: Some PRs (mostly newer ones) get multiple repetitive reviews while other PRs (mostly older ones) don't get any. My own PR has been open for exactly one year today and did not get a review. I tried pinging you, I updated the list several times to comply with updated rules, I fixed the merge conflict caused by #1343, I ran the linter and did some more updates, I did reviews on other PRs and pinged you again. Well, a year has passed since I created the PR and it is still open. How would anyone be interested to go through all that hassle just to be ignored?
I highly appreciate what you do (here and on the many other projects you have) and I definitely understand that it all takes a lot of time. All I'm saying is that it does not seem to scale. Having that many open PRs of which many are so old does not attract new contributions in my opinion.
That is all I wanted to say :-)
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @sindresorhus on GitHub (Jul 28, 2018).
Original GitHub issue: https://github.com/sindresorhus/awesome/issues/1363
I got pretty burned out by this project and haven't been active for many months. This project is called Awesome and lists are expected to be of (awesomely) high-quality. Unfortunately, many submissions were not and required a lot of effort on my part to just comment on things that were already listed in the pull request guidelines. This doesn't scale to one person when there are many new submissions a day.
I'm ready to put some work into this project again, but I need your help.
There's a new requirement for submissions now. When you submit a new list, you're required to review at least 2 other open pull requests. This is my experiment to try to make the Awesome project self-sustaining. Read more here.
This requirement also applies to existing submissions, so if you like to see your submission merged, review at least 2 other pull requests and I'll prioritize getting your submission merged.
Please also go through the list guidelines again on your own submissions. You have probably missed something.
@sellisd commented on GitHub (Jul 28, 2018):
Glad to see you back active on this! How about trying to automate part of the review process? At least some of the requirements for the pull request could be validated by a script I will put down some ideas here: https://github.com/sellisd/awesomeReview to see if it can be of use
@transitive-bullshit commented on GitHub (Jul 29, 2018):
@sellisd see awesome-lint which seems like a solid starting point, though it's in need of a lot of love.
@dspinellis commented on GitHub (Jul 29, 2018):
Making the project self-sustaining and avoiding burnout is a good idea. It would be helpful to provide a bit more guidance regarding the review process.
@TheKitty commented on GitHub (Jul 31, 2018):
@sindresorhus - we'll be reviewing
https://github.com/sindresorhus/awesome/pull/1160 and
https://github.com/sindresorhus/awesome/pull/1127
We'll be noting issues that are not in line with the guidelines to include syntax, comments, etc. Is there's anything else you require?
@TheKitty commented on GitHub (Aug 1, 2018):
Reviewed
https://github.com/TheKitty/awesome-microbit/pull/1
https://github.com/TheKitty/awesome-jupyter/pull/1
for 1160 and 1127
@myshov commented on GitHub (Aug 4, 2018):
@sindresorhus hi!
Here are my reviewed pull requests:
https://github.com/sindresorhus/awesome/pull/1154#pullrequestreview-143387275
https://github.com/sindresorhus/awesome/pull/1347#issuecomment-410463100
https://github.com/sindresorhus/awesome/pull/1292#issuecomment-410464225
@troxler commented on GitHub (Sep 29, 2018):
It seems obvious that the current approach of merging pull requests does not scale. Currently, there are four pull requests that have been open for more than one year and there is a total of 102 open pull requests. At least some of them seem to have met all requirements including the authors having done the reviews of other pull requests. Judging from the list of commits, @sindresorhus is the only (active) user with merge privilege.
I have seen some pull requests that were closed by their respective authors because they lost patience or interest. The process of being listed is really cumbersome and I understand everybody who is not willing to create a pull request just to have it open for months (or even over a year) without any reasonable prospect of being listed at all.
I understand that maintaining this project is time-consuming and often even annoying. But letting this project die a slow death is a pity. This only works when more people with proper privileges are involved. Sindre, I think it is time to add more contributers to this repository. What do you think?
@sindresorhus commented on GitHub (Sep 29, 2018):
@troxler Just because there are lots of open PRs doesn't mean this project is dying. This repo is just a list of what I consider good. The beauty of open source is that anyone can create a list. It doesn't have to be listed here. There are currently 1500 repos tagged with the
awesome-listtag and probably lots more that are untagged. I try to prioritize reviewing PRs here, but most PR reviews are still just me copy-pasting items from the contribution guidelines, which is very demotivating. I also have a lot of other OSS responsibilities.I have tried adding more collaborators multiple times before without success. You don't even need to be a collaborator to help out. Merging the PRs are easy. Thoroughly reviewing them is the time consuming part, and anyone can help out with that, but few do.
I'm also working with community members on a awesome linter which will hopefully lessen my PR review work.
@troxler commented on GitHub (Sep 29, 2018):
No it does not indeed. I've been following this project for a while and I think during this time, the number of open pull requests even decreased. But many of them were closed because their authors apparently lost interest and gave up or just did not answer anymore. I strongly think that is also has to do with the amount of time it takes from the request creation to actually being merged (or closed).
I do realize that (kudos and thanks btw!). And I especially get the "demotivating" part as I've seen many examples of authors that clearly did not read the guidelines at all.
That may be true. But don't you need tools to see at a glance which pull requests could potentially be merged (i.e. someone reviewed it and thinks it is OK) and which need more work? Do you go through all 100+ requests regularly just to check that (well, just the once that were updated recently but you get the point)? Wouldn't it be helpful to have appropriate labels or a being able to filter by approvals? Contributors without privileges can't use either. It would also make sense for the reviewers to quickly find requests that actually need a review without skimming through a long list first. And personally, I also find it demotivating to do reviews without proper tools.
@sindresorhus commented on GitHub (Sep 29, 2018):
Yup, but in most cases it takes a while because people keep opening PRs without having put much effort into their list or PR, so they require a lot of review time, which means I have less time for other PRs, and suddenly there are 100+ open PRs...
I go through my Awesome repo notifications once in awhile to see if any new PRs are ready for another review.
Sure, this all sounds nice, but you don't just add random people to a project and hope they do a good job. You add people that already regularly do a good job.
All PRs need review, including already reviewed ones. Even when I get external review help, they miss a lot of things, so the more reviews of the same PR, the better.
@troxler commented on GitHub (Oct 1, 2018):
Yes I can imagine. First thing I would do is to close the PRs of authors that did not respond to any comments in months. It seems useless to let people do yet another review on those anyway.
That would be ideal yes. What currently is happening is the following: Some PRs (mostly newer ones) get multiple repetitive reviews while other PRs (mostly older ones) don't get any. My own PR has been open for exactly one year today and did not get a review. I tried pinging you, I updated the list several times to comply with updated rules, I fixed the merge conflict caused by #1343, I ran the linter and did some more updates, I did reviews on other PRs and pinged you again. Well, a year has passed since I created the PR and it is still open. How would anyone be interested to go through all that hassle just to be ignored?
I highly appreciate what you do (here and on the many other projects you have) and I definitely understand that it all takes a lot of time. All I'm saying is that it does not seem to scale. Having that many open PRs of which many are so old does not attract new contributions in my opinion.
That is all I wanted to say :-)
@sindresorhus commented on GitHub (Feb 26, 2019):
Closing this as I'm back to merging PRs again.