fix ASON -> aeson in podcast 47 transcript

podcast/47/transcript.markdown

@@ -200,11 +200,11 @@ And so, I think that ultimately, we need to just do more to onboard more people
 
 *AL (0:33:06)*: So, I guess probably it is not that everybody agrees that it is a goal to grow the community. But I would imagine that it is actually something that many people would agree to, and that it is not something that people are actively fighting against.
 
-*AP (0:33:22)*: Right. I don’t know to what extent this is the case in the Haskell world, but some communities in other parts of the open source world, not chiming in on something is consensus for something. So, if people are not present on something and the people who are showing up to do something, they’re not going to be blocked by the people who are not chipping in on it. And I don’t know if that’s the case here. Or maybe it’s partly true in some cases and not in others, but I think it shouldn’t be the case that pushing forward on the academic research sides of Haskell should get in the way of people wanting to push the industrial use cases more forward. But where those two things really collide is when we break backwards compatibility of things. That causes a lot of pain, I think, for a lot of the people that rely on Haskell. And when it’s really late at night and you’re upgrading ASON in your code base and you just want to go to sleep, that’s the time where this impacts our lives as Haskell developers. 
+*AP (0:33:22)*: Right. I don’t know to what extent this is the case in the Haskell world, but some communities in other parts of the open source world, not chiming in on something is consensus for something. So, if people are not present on something and the people who are showing up to do something, they’re not going to be blocked by the people who are not chipping in on it. And I don’t know if that’s the case here. Or maybe it’s partly true in some cases and not in others, but I think it shouldn’t be the case that pushing forward on the academic research sides of Haskell should get in the way of people wanting to push the industrial use cases more forward. But where those two things really collide is when we break backwards compatibility of things. That causes a lot of pain, I think, for a lot of the people that rely on Haskell. And when it’s really late at night and you’re upgrading *aeson* in your code base and you just want to go to sleep, that’s the time where this impacts our lives as Haskell developers. 
 
 *JB (0:34:25)*: Yeah. So are you reporting actual experience right now, or is it something that doesn’t actually bother you personally too much, but you keep hearing from others?
 
-*AP (0:34:34)*: We dealt with the ASON thing very much so at Scarf. That one was not one that I was doing myself. By then, I was already not coding as much at Scarf, but I definitely had times where, yeah, GHC upgrades really took a lot of time, or needing to patch the Haskell Stripe library took a lot of time, and it was using type families. This was a while ago as well. The library took 20 minutes to compile every time I wanted to make a small change. These kinds of things, they’re paper cuts, but they’re really deep paper cuts. And in an environment where you are trying to focus on your business, I don’t want to think about Haskell very much. I’m trying to just get my work done and focus on other things. Yeah, it’s hard. It’s sometimes hard to write Haskell and have it like not have to be really top of mind about Haskell stuff, if that makes sense.
+*AP (0:34:34)*: We dealt with the *aeson* thing very much so at Scarf. That one was not one that I was doing myself. By then, I was already not coding as much at Scarf, but I definitely had times where, yeah, GHC upgrades really took a lot of time, or needing to patch the Haskell Stripe library took a lot of time, and it was using type families. This was a while ago as well. The library took 20 minutes to compile every time I wanted to make a small change. These kinds of things, they’re paper cuts, but they’re really deep paper cuts. And in an environment where you are trying to focus on your business, I don’t want to think about Haskell very much. I’m trying to just get my work done and focus on other things. Yeah, it’s hard. It’s sometimes hard to write Haskell and have it like not have to be really top of mind about Haskell stuff, if that makes sense.
 
 *JB (0:35:30)*: So, from your experience, it is an actual concrete issue that Haskell compiler and libraries tend to break compatibility team more often than you would wish?