@ves93 replying just so I'll get notification if someone posts something interesting. I haven't found anything truly satisfying myself, most of the highly functional apps seem to be subscription models.

@sam @tk yeah the bottom has definitely dropped out of the used and vintage motorcycle market. I'm not entirely surprised. They might actually have a hard time getting rid of it at this price right now, even though it's value should be way higher. Nobody is buying though.

Cool find.

@dalias I think you have to be willing to back that one up with actions should it come to it, but other than that, the messaging is spot on.

Yes, it's good to have a bot to offload the work onto when you're short-handed. But who left you short-handed in the first place?

And how sure can you be that you're not going to be just as short-handed AGAIN, as soon as corporate figures out exactly how finely they can pare the flesh down to the bone WITH AI assistance?

I'm glad those folks have a tool that abstracts away the most odious parts of their job. But I could wish we didn't make it so odious in the FIRST place.

I just want to leave this photo as a reminder of the almost inevitable outcome of using armed forces to manage a protest. I live about twenty minutes from KSU. We haven't ever forgotten.

Soldiers are not trained for this situation. They are trained to fight and kill enemy soldiers. That's not what is needed right now.

@dalias @codonell what I really don't love is that rather than answering on being called out for terrible coding practice some if the comments amount to, "You were mean talking about us on the internet. We're not listening to you."

Seriously, be a f%#$ing adult. Thicken up your skin.

@arnowi apparently someone contributing to Glib used it for a union field, as I mentioned in my top post. Makes sense in that context and compiled just fine until gcc15, when c23 became the default.

@arnowi the problem arises because 'bool' is now a reserved word, so you can't use that name for a variable.

Took the zip-out line out of my riding jacket this morning in what I'm guessing is once again an overabundance of optimism. This weather has to stay warm sometime, right?

One of the things that I think has always been a benefit of C is that old code has generally continued to work with new code due to how conservative language changes have been. I think C23 is going to cause some pain though, and since GCC now defaults to the C23 standard that's kind of important.

For 50+ years everyone has been defining their own boolean types in C due to the language not having that facility built in. Now we get 'bool', 'true' and 'false' as keywords. That's going to break a lot of code. I've seen breakage in the wild just this morning. The pkg-config project (the one hosted at freedesktop.org, not the other one) includes an old version of glib to satisfy circular dependency issues which uses 'bool' as a union member.

While I don't write a lot of C, some of what I have written gets broken here, too.

I'm not saying this was a bad idea, just that seeing things break due to such a small and innocuous seeming change in the language was enlightening. Made me question that particular value of C that I always assumed to be true - that old code will generally continue to work unchanged.

@jik all of that said I want to impart to you the importance of checking certain things regularly enough that you'll catch them and doing regular maintenance. I was taking a long camping trip earlier this year and just into Kentucky my clutch cable snapped just as I was pulling into a gas station. I knew I was lucky that it happened where it did, while pulling off the road and going slow. But a few weeks later something happened that gave it new significance for me.

My brother in Florida lost a good friend just a few weeks later. His friend, who runs a bar, was following a young lady home on his bike to make sure that she made it there all right and crashed on the way home. Fatal. The cause? He had a frayed clutch cable, abs they rubble that it gave way just as he was inching up to a dangerous intersection to look stood the corner. The engine (a big V-Twin cruiser) had enough torque to pull him right out in front of a car before he could stop it. Gave me chills considering I'd just had the same mechanical failure myself, and made me re-commit to doing more regular maintenance.

Not telling you this to scare you. You just should make it a habit to be checking those common wear items and failure points. Things like tires, brake lines and cables need to be changed every few years regardless of the mileage.

@jik before every ride is probably excessive. But I like to take a little time each weekend to clean and check things over. Now, I commute to work on the bike so summer and a far into the cold season as I can stand, but it's a short commute (ten to fifteen minutes across town). I take much longer rides when I have the chance, and if I'm doing something that's going to take me out of town I'll do a better check.

Since things should be daily though. It only takes a minute, assuming you have a center stand, to check the oil level through the sight glass. At the very least do this when you fill up with gas. Bikes can sometimes catch you by surprise with just how quickly the engine oil can disappear, particularly vintage bikes like mine.

A lot of things you just look at when you walk around the bike before getting on. Things like tire tread depth.

@antolius kinda makes you yearn for the days when we all had a curated bookmarks list.

@jk I went on a rant about exactly this in one of my college courses while we were discussing disability, pointing out where the ebook reader webapp we were forced to use to access certain library materials hijacked select, cut, copy and paste actions.

@Ailantd calling it an industry implies that it creates a tangible product or service. That's always a stretch.

Anyway, good. Let's kill it.

This morning I null routed another dozen IP addresses for scraping my personal git server using repeated http requests. As per usual, a quick inspection reveals that at least some of them are scraping for LLM data. As always, I have not consented to this use of my non-maintained code, experiments, college coursework, and miscellaneous crap that I for whatever reason decided to self host rather than pushing it to Codeberg.

I mean, if you really want to feed your LLM on a diet that includes gems like a homegrown RSA encryption program written in C and limited to 64 bits that I did to understand the scheme, or an unmaintained vanity Linux distro that builds from source using Makefiles, or my Discrete Mathematics notes written in Markdown and LaTex, maybe I should let you? It's not really a good idea. But what really pisses me off is when I log into my little Raspberry Pi that's running the Forgejo instance only to notice the lag between key presses because all four CPU cores are maxed out. At some point this shit just needs to stop. Your robots are basically DOSing my private infra.

What's really stupid about all of this is that instead of all of these individual http requests to get the changes that took place to some file between such and such dates these idiots could just clone the damn repo and examine it using git. I wouldn't be happy about it, but the resource usage would be orders of magnitude lower on both ends of the transaction. Then, if they wanted the changes for my code that I'm not actively working on anyway a simple pull would tell them that nothing has changed, rather than hitting every file individually. It's fucking insanity.

My understanding is that anyone who is running a public facing code forge is dealing with these same issues. Anyone feel like getting together for a class action lawsuit? I mean, the time alone that I spend null routing new IP addresses should be compensated.

@hosford42 @todymotmot @zakalwe exactly what I got tired of after years of being the accommodating one.

@hosford42 I agree, and yet...

The only portrayal of an Autistic character in a major film when I was growing up was Rain Man. Don't get me wrong, I think that Dustin Hoffman gave a great performance, but it really reinforced the public perception that Autism means you are unable to take care of yourself or have meaningful human contact. Now, there are those on the spectrum who have severe disabilities as comorbidities, but with The Accountant it's nice to see a character who, while he faces challenges, ultimately is a fully realized and capable adult.

Could it be better? Obviously. Is it problematic in its portrayal of Autism in completely different ways? Again, yes. But on the whole it's a win for me.

Another thing to note in this conversation, after the absurdity of RFK's recent characterization of Autism, I think it's vital to get the message out that we're not "defective" humans on as many fronts as possible. I'll pick my battles and give The Accountant a bit of a pass for its flaws, because I think it is overall more helpful than harmful.

@tk surprisingly common in early overhead cam engines. The Rolls Royce Merlin V12 used a bevel gear drive for the camshafts. Simplest and most reliable way to transfer the motion up to the top of the heads.

@narthur to each his own obviously, I'm not going to criticize anyone for their taste. I do wish the manufacturers would offer their bikes without the extra tech though. Let those who want it order the options. For me, I have a phone holder on my bars and a Cardo in my helmet. That totally meets my needs for music and navigation and I don't see the point in integrating it into the bike itself.

I would be interested in trying a BMW someday. The dealership I stopped at for parts on my last trip had a nice '75 R75-6 sitting in their showroom. Beautiful bike. Whenever I hear about a demo day for any brand I try to go. Not sure if BMW does those kinda of events but if they do maybe I'll get a chance to ride a couple sometime.