Skip to content

Commit 3cc5f09

Browse files
committed
Addressing PR feedback on SWG post
1 parent 4092476 commit 3cc5f09

File tree

1 file changed

+7
-7
lines changed

1 file changed

+7
-7
lines changed

content/stability-working-group/index.md

+7-7
Original file line numberDiff line numberDiff line change
@@ -13,9 +13,9 @@ What is stability? What has the Haskell Foundation Stability Working Group been
1313

1414
## Who am I?
1515

16-
The health of the Haskell ecosystem is very important to me. Most of my professional career to this point has involved using Haskell. As such it is in my own best interest to help the community to be healthy and grow. That way I can continue to work in the language and ecosystem! Even prior to my professional career, I used Haskell as an undergraduate. Despite it not being a part of the curriculum and being met with skepticism from several advisors that I could ever find work in Haskell.
16+
I am Trevis, chair of the Stability Working Group. The health of the Haskell ecosystem is very important to me. Most of my professional career to this point has involved using Haskell. As such it is in my own best interest to help the community to be healthy and grow. That way I can continue to work in the language and ecosystem! Even prior to my professional career, I used Haskell as an undergraduate. Despite it not being a part of the curriculum and being met with skepticism from several advisors that I could ever find work in Haskell.
1717

18-
Working as first a member and now chair of the Haskell Foundation Stability Working Group has been enlightening. Over the last few years two things have become very apparent. First, that there is a real appetite for movement from practitioners on some notion of stability. Second, getting things done in this area takes much more time and effort than it would seem.
18+
Working first as a member and now chair of the Haskell Foundation Stability Working Group has been enlightening. Over the last few years two things have become very apparent. First, that there is a real appetite for movement from practitioners on some notion of stability. Second, getting things done in this area takes much more time and effort than it would seem.
1919

2020
With the appetite from practitioners and effort in mind, let me describe my views on stability, the operating goals of the Stability Working Group, who the group is, what we have been working on, and how you can help.
2121

@@ -29,15 +29,15 @@ Recognizing this diversity means realizing where there are gaps in representatio
2929

3030
Now with the ecosystem in mind, what exactly is stability?
3131

32-
For many they will read that word and think immutability. We are talking about the Haskell ecosystem after all. However, stability here does not mean a complete lack of change. To forego change entirely would mean to not be able to incorporate new ideas or react to the rest of the world changing. If this was the case Haskell would never have grown to include a huge number of things, including the IO monad! Interaction with he rest of the world will continue to require changes, for example the first version of The Unicode Standard was published after Haskell 1.0!
32+
For many they will read that word and think immutability. We are talking about the Haskell ecosystem after all. However, stability here does not mean a complete lack of change. To forego change entirely would mean to not be able to incorporate new ideas or react to the rest of the world changing. If this was the case Haskell would never have grown to include a huge number of things, including the IO monad! Interaction with the rest of the world will continue to require changes, for example the first version of The Unicode Standard was published after Haskell 1.0!
3333

3434
So we accept that not only will change happen, it must. This leads to how I would describe stability: the ability to react to change in an informed and graceful way. With change as a given, we must be able to react to it. But the other parts are also extremely important. Change can be extremely disruptive. In the ecosystem, change might mean:
3535
- industrial users need to spend time away from their employer's goals
3636
- library and tooling maintainers have to adapt and support multiple paths, taking away from feature development
3737
- academic materials might need to be rewritten
3838
- any other number of issues that are difficult to predict
3939

40-
But we just said that change is going to happen! As such, _ecosystem stability_ is managing the *impact* of change as felt by all members described above . In order to reduce the negative impacts of change to the ecosystem there must be a graceful handling path, not a lack of change entirely. For example, deprecating something before removal. Another important factor here is the timing. Being predictable in change management means that the various groups can schedule work rather than being interrupted. This is why it is critical to be able to handle change as gracefully as possible.
40+
But we just said that change is going to happen! As such, _ecosystem stability_ is managing the *impact* of change as felt by all members described above. In order to reduce the negative impacts of change to the ecosystem there must be a graceful handling path, not a lack of change entirely. For example, deprecating something before removal. Another important factor here is the timing. Being predictable in change management means that the various groups can schedule work rather than being interrupted. This is why it is critical to be able to handle change as gracefully as possible.
4141

4242
## The Stability Working Group
4343

@@ -51,16 +51,16 @@ The SWG has worked on several projects, some of which continue to this day.
5151

5252
We worked on a [review](https://jappie.me/analyzing-haskell-stability.html) of breakages tracked by the GHC team. This was illuminating to highlight and enumerate a high level set of changes. There is still room to do more extensive analysis here as well.
5353

54-
Another project we have been working on is defining a framework for classifying GHC language extensions. This has been a long process. But ultimately is the type of communication enhancing improvement that is needed for predictability.
54+
Another project we have been working on is defining a [framework for classifying GHC language extensions](https://github.com/ghc-proposals/ghc-proposals/blob/master/proposals/0601-extension-lifecycle-framework.md). This has been a long process that is still [ongoing](https://github.com/ghc-proposals/ghc-proposals/pull/669) But ultimately is the type of communication enhancing improvement that is needed for predictability.
5555

56-
More concretely we have defined a set of compiler features to help with stability. These were ideas that already existed in some form. Examples that we were able to bring to attention and have since landed in GHC include warning categories and deprecation of exports. Those both allow for more communication from compiler and library authors to users.
56+
More concretely we have defined a set of compiler features to help with stability. These were ideas that already existed in some form. An example that we were able to bring to attention and have since landed in GHC is [deprecation of exports](https://github.com/ghc-proposals/ghc-proposals/blob/master/proposals/0134-deprecating-exports-proposal.rst). This allows for more communication from compiler and library authors to users.
5757

5858
Finally, we find ourselves with a recurrent item. That is to be generally facilitators of communication. This means bringing groups together, highlighting issues, and generally fostering discussion.
5959

6060

6161
## How you can help
6262

63-
We have been working on many things, but we can still use help! Hopefully I have intrigued, inspired, or just otherwise cajoled you into a desire to contribute. There are several ways to help. The lowest effort is to contribute to Haskell Foundation. Another is to reach out! As I said above we have several projects that could use someone to champion. The SWG often hosts guests at meetings to discuss particular issues or concerns. You do not have to commit ongoing time in order to contribute!
63+
We have been working on many things, but we can still use help! Hopefully I have intrigued, inspired, or just otherwise cajoled you into a desire to contribute. There are several ways to help. The lowest effort is to [contribute](https://haskell.foundation/donations) to Haskell Foundation. Another is to reach out! As I said above we have several projects that could use someone to champion. The SWG often hosts guests at meetings to discuss particular issues or concerns. You do not have to commit ongoing time in order to contribute!
6464

6565
With your help we can continue to strengthen and grow the ecosystem!
6666

0 commit comments

Comments
 (0)