Month: August 2025

more nonsense about sports and streaming

This is the third post I’ve written about sports and streaming this month. (See here and here for the previous ones.) Sorry. I guess I’ve been retreating into watching (and reading about) sports because the “serious” news is too stressful right now.

I mentioned the new ESPN streaming thing in one of my previous posts. I gave in and signed up for it yesterday. My previous Disney+ bundle was the “legacy” one that included ad-free Disney+ and ad-supported Hulu. That was $22/month. There’s no equivalent bundle with the new ESPN service. The bundle with ad-supported Disney+ and Hulu is $30/month. The bundle with ad-free Disney+ and Hulu is $39/month. So I went ahead and signed up for that. It’s probably too expensive, and I don’t know if I’ll stick with it. And that $39/month rate is a promo; after a year, it goes to $45/month.

It just seems like all of this stuff is getting too expensive and too complicated. I’m still pretty happy with my MLB.TV subscription, which lets me watch most of the Phillies games. (I wish I could watch all of them, but I guess it’s good enough.)

I’m still not sure if I’m going to be happy with my NFL situation this season. I’m still planning on watching RedZone on Sundays (via NFL+) and hoping that’ll scratch my football itch. And I can watch Thursday Night Football on Amazon, Sunday Night Football on Peacock, and Monday Night Football via my new ESPN subscription. So that should be enough.

I’m also trying to see if I can get into Premier League Soccer. I was randomly clicking around in the Peacock app last Saturday, and happened to notice that it was week one of their season. And Peacock has a whip-around show (similar to RedZone I guess) called Goal Rush, so I can watch that for a couple of hours on Saturday and see bits and pieces of a few different games. So now it’s week two, and I’ve watched a bit of this stuff, and it’s kind of fun, though I’m mostly treating it as background noise. I haven’t developed any real interest in it yet.

And I briefly thought about giving up on the NFL and following the CFL instead. That thought came from the storyline that ran in the Crankshaft comic strip recently, where the main character went to Winnipeg to see a Blue Bombers game. It turns out that streaming rights for the CFL in the US aren’t terribly complicated, but they’re not terribly convenient for me either. Some games run on CBS Sports Network, which I’d probably have if I still had YouTube TV, and which might be included with Paramount+, though I’m not sure. The rest of the games can be watched on CFL+, which is a free streaming service. So that sounds good! It’s free! But there’s no Apple TV app. I’d have to pull up the website on my laptop and then AirPlay it to the Apple TV. That’s too much work.

So overall, my plan for distracting myself with sports for the next few months is approximately as follows:

  • Phillies games on weeknights, via MLB.TV.
  • Some other random baseball games on ESPN and Apple TV+.
  • NFL games on Sunday via RedZone on NFL+.
  • Thursday, Sunday, and Monday night football via Amazon, Peacock, and ESPN.
  • Maybe Premier League soccer on Peacock on Saturdays.

This will probably all fall apart at some point. If the Giants are as bad as they were last year, I may lose interest in the NFL. Or, conversely, if they do well, I might get frustrated about not being able to watch all of the games.

And I don’t even want to think about how complicated watching the MLB and NFL playoffs will be. Well, that’s a problem for another day.

playing with AI and enjoying sports

OK, so that’s two completely different topics, but I’m going to throw them together in one post. I’ll start with AI.

At work, I’m still very limited in the AI tools I can use: basically, just Copilot and our own internal chatbot. (And it’s just the plain Copilot chat that I can use, not GitHub Copilot.) I’ve been wanting to experiment with some of the coding tools that are out there, so I decided to try some stuff out on my (personal) laptop. I was going to try Claude Code first, but I decided that GitHub Copilot would be easier for me to set up.

I started by enabling it in my personal GitHub account, and setting up the extension in Visual Studio Code. Then I tried something simple: asking it to write a “hello world” console app in C#. It did that successfully, but of course that’s a pretty low bar. I then tried to get it to help me create a git repo for the project. That worked, but it didn’t create a .gitignore file , so all of the binary output files got added to the repo. When I pointed that out, it apologized and created a .gitignore file, but it didn’t really offer me any help with cleaning up the mess it created. And, while the .gitignore file it created was “good enough”, it was a little questionable.

So I stepped back a bit, deleted the git repo and the .gitignore file, and tried again in Visual Studio 2022. I went through the usual steps in VS 2022 to initialize a git repo, and that created the usual VS .gitignore, so that was a lot cleaner than what Copilot did. Then, I set up Copilot in VS 2022. It probably works a bit better in VS 2022 over VS Code, at least for me.

I had Copilot in VS 2022 make a few little changes to the program to see how it worked. I had it add a line to output the current date and time, and that was simple enough. I then had it add some code to show the operating system and version, and that was a bit dicier. I had to coach it through making changes so that the program would actually output “Windows 11 Pro” instead of Windows NT or Windows 10.

So my opinion on this stuff hasn’t changed much. It’s interesting, and can be useful, but it’s wrong about half the time, and it’s often no more efficient than just doing an internet search and copying something from Stack Overflow. I kind of want to try it on something bigger, but I’m not sure what I should try.

On the sports topic: as per my last post, I signed up for NFL+ Premium recently, and have been watching some preseason football. The annual Jets/Giants preseason game is tomorrow night. That’s usually fun.

I’m still not sure what I’m going to do about the new ESPN service. I’ll have to look at the price for switching my Disney+/Hulu/ESPN+ bundle to include the full ESPN.

And I’m still planning on skipping the new Fox service, on general principle. Which is a bit tough, since the first two regular season Giants games are on Fox, so I won’t be able to watch them. My plan, for now, is to watch NFL RedZone (via NFL+) on Sundays and see if that’s good enough to keep me entertained and show me enough of the Giants game.

I’ve been looking into RedZone a bit; it’s interesting. It seems to be quite popular. I’d been aware of it previously, but have never watched it or really learned anything about it. There’s been some concern about what will happen to it after the NFL/ESPN deal goes through (assuming it does). That shouldn’t affect anything this season; it should be at least a year before that’s finalized.

Meanwhile, I’ve watched a lot of Phillies baseball this week. I’ve just been in the mood for baseball instead of, really, anything else. They’ve lost their last three games, so that’s disappointing, but I still like watching them.

Streaming Chaos

Preseason football has started up, so that’s got me thinking about how I might be watching (or not watching) football this year. Last year, I had YouTube TV, and the year before that, I still had the broadcast basic package from my cable company. So this is the first time I won’t have any way to watch regular old broadcast TV during football season. (I could still break down and re-subscribe to YouTube TV, but I’m not planning on doing that.)

Streaming is getting even more complicated this year, it seems. Here’s a fun article about how to watch football this year. ESPN has a new streaming service starting up soon. If I understand things correctly, my ESPN+ sub will become an ESPN Select sub, so no big change there. I guess I could switch to the “full” ESPN service, and then I’d be able to watch… whatever games are on ESPN/ABC but not ESPN+. I’m honestly not sure what those will be. Monday Night Football might be ESPN/ABC only, or might also be on ESPN+. Who knows?

For Sunday night games, I should be OK. I still have a Peacock subscription. I don’t have any way to watch Sunday day games right now though. I canceled my Paramount+ sub, so no CBS games. Fox has a new streaming service starting up, which should carry their Sunday games, but that service also includes Fox News, and I’m dead set against supporting that stuff, so I guess I’m going to go without.

I did just pay for a year of NFL+ Premium. That gives me NFL Network and NFL RedZone, so that’s something. I’ve never actually had RedZone before; maybe I can just watch that on Sunday? The NFL+ sub should let me watch all the Giants games on my iPhone or iPad, I think, just not on TV. So maybe I’ll just watch football on my iPad from now on.

Here’s an interesting article on the new ESPN/NFL “mega-deal.” I’m not sure I even understand what’s going on with football right now.

And, of course, I could just give up on football. I’ve done it before. I’ve gone back and forth on it, over the last decade. Some years, I’m enthusiastic about it, and some years I’m not. (Often, that has to do with how well the Giants are doing that year…)

Meanwhile, I’m still enjoying baseball. I watch the Phillies on MLB.TV a few times a week. I’m always a little annoyed if a game is blacked out because they’re playing the Mets (or Yankees), or because it’s on ESPN or wherever. But there are so many games in a baseball season, it doesn’t bother me that much if I miss a few.

Of course, if the Phillies get into the post-season, that becomes a problem, as those games aren’t on MLB.TV. Sigh.

Frustration

I have a bunch of stuff on my mind, and a few topics for blog posts I want to write. But I’m going to start with a little decompression about my work week. (I should first say that I need to be careful about this, since our policy on blogging about work has been tightened up recently. But none of this will come anywhere near to violating it, I think.)

We’re working on an upgrade to our AX environment right now, and we started some testing in a new test environment this sprint. This includes (for me) getting some web services working on the new test server, and getting them to point at the new AX test env. So a lot of my time this week has been spent whacking away at web services with a hammer, trying to get them to work. That’s always frustrating. It always seems like I need to try six different things until I finally hit on the one that actually fixes the issue. And I always feel like an idiot for not being able to get there faster.

I’ve been thinking about why this kind of work is frustrating, and I think a big part of it is having no clue how long it’s going to take, or how close you are to solving it. With certain kinds of work, you can break a problem down into parts, and work through them, so you know when you’re 10% done, or 30% done, or 80% done. But with this kind of troubleshooting, you don’t know if you’re going to need to try 5 things or 10 things or 20 things before you find the right thing. And you don’t always know if there’s one thing wrong, or two things, or ten things.

I won’t get into details on the actual issues I hit, but I will say that one of them was basically a new permutation of this issue I hit back in 2011, so it was fun to see that again! And another one was related to the topic of this post from 2020.

I’ve been using AI tools to help me with this work, and I’m not sure if they’re any more useful than just plain old internet searching. Sometimes it seems like they are; I can get a lot of seemingly-useful ideas from a chat with Copilot. But then the real answer comes from a simple search that leads to a Stack Overflow question/answer, or a GitHub issue. (For instance, one big problem I had this week was solved by reading through this GitHub issue from Microsoft’s WCF repo.)

My big project for this afternoon was working on a side-project for a little competition we’re having at work. (It’s one of those things where small project teams propose something, do a little work on it, then present it to the CIO. And then I guess the winning project actually gets green-lit.) That project is turning out to be a bit frustrating too. (It was supposed to be a nice break from the frustrating stuff!) First, I needed to get myself set up under our corporate developer account with one of the big package carriers, which we only managed to get done yesterday. And I’m still not completely set up, to where I can do what I actually need to do. But I can start testing, so I started work on that.

The project involves using the carrier’s webhook API for package tracking. My idea was to start out with a simple Azure function that I could point the webhook at. Then, to start, I’d just log the info I got from the carrier. The next step would be to hook that Azure function up to some kind of queue in Azure. Then, after that, writing another Azure function that would pick stuff out of the queue and send out notification emails. Basically, I was going to put my AZ-204 skills to work on a real project! (Well, not entirely real, but a POC at least.)

It took me the whole afternoon to get a simple Azure Function set up. I started out trying to use Copilot chat with GPT-5, which they rolled out yesterday, to help me out. That looked really impressive, showing me status info about what it was doing, and generating a lot of text with pretty specific instructions. But it seemed to be giving me quantity over quality. I turned GPT-5 off, and then managed to get a lot farther.

My initial idea was to just enter the C# code for the function directly in the Azure portal, but I set it up in isolated mode, which apparently doesn’t let you edit code directly in the portal. So then I tried using Visual Studio Code, which is supposed to be an easy way to quickly develop and deploy Azure Functions. That led me down a path of installing a bunch of stuff locally. I was initially doing this on my laptop, so I had to install .NET SDK 8, plus a bunch of VS Code extensions. At some point, I got nervous about how much stuff I was installing on my laptop, and switched over to my regular dev VM, which is where I’m supposed to do development. So then I had to install all that stuff there too, since I’d never installed the .NET 8 SDK there, or set up VS Code for Azure dev. So I probably killed an hour just getting my environment set up.

And then, once I had everything set up, I started hitting a bunch of weird little snags with editing, compiling, and running the function. I eventually gave up, and with some help from Copilot, set up the simplest Azure Function project I could, entirely from the command line, then got it running and uploaded to Azure. So that probably took another hour, and the end result was just something that would print “Welcome to Azure Functions!” when you hit the URL for the function.

After that, I got the thing working in VS Code, and managed to change the function from a GET to a POST function. And I thought I had gotten far enough with it that I could hit it from the carrier’s test page, but all I can get back from that is an error, with no details giving me any clue what’s wrong. And that’s pretty much where I ended my week.

I don’t really drink anymore, but I think I kinda want a beer right now. Or maybe just a bowl of ice cream.