Blog

Unless you have been living under a rock over the last couple of years you have probably run into something that’s been touched by generative AI. These tools are interesting in concept and they aren’t going to be going away any time soon but I’m definitely getting sick of the people who are taking these tools and using them like their existence makes them a domain expert in something that they have no business messing with.

Some of these tools are able to do some crazy stuff, and the people that I’m seeing using these tools and getting success with them are all folks that already know what they are doing in a specific domain. So a developer who writes code can definitely take these things and use them to generate code quicker, sanity check their work for security issues or common bugs, or even layout UI elements for quick prototyping.

Now take that same tool and put it into the hands of someone who doesn’t have that same background and now you have someone who’s building a application but doesn’t have the background to understand what’s getting produced by the tool. Sure you might be able to get a app working but without knowing how it’s working do you know if your inputs are all being validated to make sure that you don’t have security issues with someone tying to inject SQL commands into the application. Do you know how your database is configured and how the data that it contains is being protected?

And while these are all things that I’m sure the the AI tool can address if you don’t have a background in application development and you don’t know that input validation can cause problems how will you know to address that problem in the first place?

I’m not an application developer myself, but I do write a number of scripts for application deployments and configuration of systems, and I’ve played with using these tools to build some of those scripts. Starting with ones that I have already built that are working properly to see what the AI tool would come up with when prompted.

Generally I’m not hugely impressed.

This might be just me not knowing how to prompt the things to do what I want but in general the results from the AI tools don’t work any better than my scripts and in most cases take longer to build out since I have to keep going back and telling the tool what I want it to do and iterate over and over with them to get things working properly.

Where I do see better results is when I take an existing script that’s working and tell it to clean up my syntax and check for typos. Even then the results so far haven’t run better or faster than what I do normally, occasionally a loop is a few lines shorter but in the scripts that I’m working with those few extra lines don’t make a difference in the long run.

At this point I have been hearing stories about people building app without any experience and starting businesses based on these apps. I haven’t actually seen anybody do this successfully where they didn’t already have the ability to do so without the AI tools. Even playing around with some of the larger applications that claim to build apps like that I still don’t see them building things that are useful without spending a large amount of time telling the tool exactly what you want done. With the tool that I was playing with I was building a very basic app to try to replace a excel sheet that was used to track device inventory and it was so tedious that I just gave up after having to tell the AI Agent that we not only needed a way to check out a device but that we would need a way to check the device back in. Something that any competent developer that I have ever worked with would have built from the start.

At some point these tools will likely get to where the advertisements are trying to make us believe that they are at, but It’s still feeling like a ways off.

So. I’ve kind of run out of content for this blog and I’m struggling to come up with topics to post on.

It’s not like there’s nothing to post about, there’s plenty of things that are stirring the pot all around the world, however I generally try to avoid topics like politics and such that can be super divisive when discussed since I really don’t like getting into arguments over the internet with random people and I sure as hell don’t need a target painted on my back because I didn’t happen to agree with someones specific opinion on a controversial topic.

And it’s not like there’s a lot interesting in the tech space right now that’s not connected with AI in some way. And personally I’m not a huge fan of the AI generated stuff since it’s got a lot of people running around generating things that are functionally useless.

That might be a good rant now that I’m thinking about it…

Earlier I had mentioned that I’m somewhat picky about the tools that I use, and pens are one of the tools that I make use of a lot and because of that they are one of the tools that spend more on than someone who’s not looked into higher end writing tools would expect.

So for background I’m left handed and a lot of cheaper pens don’t dry fast enough when I’m taking notes and that results in a bunch of smeared writing and ink on the side of my hand as I drag my hand across what I’m writing. At some point back many years ago I found myself walking past a pen shop that was going out of business while I was out of town for work and wound up picking up a Lamy All Star and a Lamy Safari for extremely cheap as an impulse purchase.

After using these pens for a couple of days on the trip I noticed that they didn’t wind up leaving ink all over my hand when I was using them. A combination of how I was holding the pen and how quickly the ink dries out resolved that issue and from that time forward I’ve stopped using the really cheap pens and have exclusively been using fountain pens for note taking.

Here we are going to talk about the first two pens that I bought. The Lamy Safari and the All Star.

Generally speaking the Safari and the All Star are the same design of pen. The Safari is a plastic pen that is available in a number of colours with special editions being released periodically that have different colour combinations between the feeds and bodies. The All Star is the same general design, however the feeds are generally transparent plastic with the cap and body being made out of aluminum not plastic and with similar special editions being released periodically.

Both of the pens can be operated with ink cartridges or using bottled ink if you have a converter handy.

Neither of the pens is very large or bulky, however the All Star is a touch larger in diameter than the Safari, and both have a grip that has a triangular profile. This makes it really easy to pick up the pen and get it into the correct position to be written with quickly.

The nibs on the pens come in a number of different sizes including some italic ones. I find that the fine or medium nibs are what I like to run with. They are not anything crazy, just plain steel either in black or silver depending on what Lamy decided to put on that specific pen when it was mad the Fine nibs that I have write more like a medium, and some of the medium nibs that I have write more like fines. So if you are picking one up and you can try the pen out in the store to get a idea of how it writes that’s likely a pretty good idea.

Writing with them, even given the variation in the nibs is generally pretty good, and any that I’ve gotten that were scratchy or dry were generally adjustable enough to make them work the way that I like without too much hassle.

Over the years I’ve picked up a few of these.

More than 60 of them.

They are fun to collect, relatively inexpensive as far as these things go, and good enough quality that I don’t think that I have had one break on me in over 20 years of collecting them.

All in these things are great daily drviers and while I have some other pens that I do enjoy writing with (yes there’s some Twisbi’s in that case) the Safari’s and All Stars are still my go-to pens with at least two inked in my backpack at any given time.

So I’ve had the chance to play around with my Nikon Z50 II for a bit and get out to take shots in a couple of different places. And while the kit lenses work well enough in most situations that I’ve come across there are a couple of things that I’m thinking that I might want to pick up something a bit more specialized for some of the shooting that I’ve been doing.

The one thing that I don’t really care for on the 50-250 that I have is how far the lens extends out from the front of the camera when you have it locked into a shooting position. When you have it locked for storage it’s a little over 5 inches long, but when you open it up for shooting it’s immediately extended out to between 7 and almost 9 inches in length depending on what you have it zoomed out for.

While I would prefer a lens with internal zoom the only one that I can find is the NIKKOR Z 24-70mm f/2.8 S II, a lens that’s going to cost me over 4k CAD by the time that taxes and such are applied to it.

And while I wouldn’t mind picking one up I have other things that I think that that 4k should be spent on.

So, what other options are out there?

Looking at the shots taken at the IBJJF Worlds most of those where shot at 50mm, there were a few where I couldn’t get close enough and had to push the lens out to the 85 – 100mm range, and a couple where I shot candid shots across the mat and those where all around the 250mm level. So one option might be to pick up a couple of prime lenses that do what I want and just rock a second body. A second Z50 II and a 50 to 85mm lens would probably be less than the cost of the one superzoom. It would also give the benefit of having some redundancy in case something happens to one of the bodies.

So time to snoop around and see what’s avlaible in prime lenses around the 50mm + range.

Occasionally I get to play with some neat hardware for work. Servers, laptops, occasionally some other strange things. Recently I was asked to purchase a rather powerful piece of hardware on order for a project at our office.

Yep, that’s 18K CAD for a single computer.

It’s got to be the most expensive single computer that I’ve been told to source while working at my office and I’m somewhat interested in seeing what it’s capable of once we get our hands on it. The 16-18 week lead time is something of a bummer but given the number of these things that are getting gobbled up for LLM use it’s not really shocking to see.

This post might be a bit of a downer. If you aren’t in the headspace to handle that just keep on scrolling and move on.

So, a couple of years ago we lost my Mom. Skimming over a lot of it we have been trying to deal with all the stuff that she collected over the course of her life. This includes the stuff that Dad had collected over the course of his life and that mom was just unable to deal with when he passed.

My family are collectors. When we get into a hobby we dive straight into the deep fucking end of things. In my Dad’s case that came down to piles of Coca-Cola collectables and stamps. Now you are probably thinking that Stamps don’t sound like a huge issue. However Dad collected them for a long time, and he had a lot of them. Something like 20 index card boxes and 14 banker boxes full of the things. Now stamps aren’t that big, so you can get a lot of them into a bankers box.

So. Yea. There are a lot of stamps here.

Recently however the focus has been on dealing with a very large rolltop desk.

Now, rolltops were a thing back in the day, you know when people would use a desk for writing and not just holding your computer off the ground.

My understanding about the desk is that it’s something that my Grandmother picked up “back in the day” and that it was passed down through her kids until the thing wound up in my mothers hands, where she used it for many years. Now that she’s gone however me and my brother are faced with what to do with it.

While I would love to have kept it at my house I just don’t have the room for another desk in my house unless I’m swapping out one of the existing ones, and those are all piled with existing stuff that doesn’t really go on a rolltop desk. My main computer desk has too many computers and monitors on it, and the workbench that is on the other side of the room that holds my 3d printer and tools is deeper than the working space on the rolltop.

My brother is kind of in the same place. No space, doesn’t work for what he’s doing.

So, we reached out to the family, however most of the remaining aunts and uncles are downsizing and don’t have the room and none of my cousins either have space, or interest in the desk.

So in the end we put the thing up on a local marketplace and sold it to someone who really wanted the thing.

Do I feel bad about selling the desk?

Yes, and No.

The desk itself didn’t really mean anything special to me. And frankly that’s the case for 95% of what was sitting in my parent’s home when they left us to deal with their belongings. There were a couple of things that me and my brother found sentimental, a camera, a candy dish, small things like that, however nothing large was really something either of us had interest in.

China? Stamps? Other collectables?

Meh.

They are just a pain in the ass to get rid of at this point. At least at the scale that they are sitting around here at now.

And looking at it I’m sure that I’m going to leave similar pains in the ass for my kids. I think I’ve mentioned it in the past but I collect pens, specifically fountain pens. While I enjoy them I don’t know that my kids are really going to care about them or the other things that I’ve been collecting over the years. I’ve made it clear to them that I don’t care what they do with the stuff once I’m gone so that they don’t have a huge struggle about what to do with all the stuff.

I don’t see myself becoming a minimalist, but I think that I can definitely slow down the accumulation of stuff if nothing more than as a favor to my kids in the far flung future.

So one thing that I miss having on my computers at home is proper remote management of the device. Yes, I have my servers setup so that I can ssh into them and get access to things but out of the box you can’t access things like the BIOS or do a reload of the OS without having to be physically present in front of the device.

I’ve been keeping eyes on a couple of devices that should help with this type of problem, and a opportunity came up where I was able to grab a Sipeed device for a decent price and I decided to give the thing a try and see if it scratched my need for physical level remote access.

The specific device that I picked up was the NanoKVM-PCIe device.

A lot of these devices come in a external package, this one caught my eye because it’s something that I can tuck into the case and not have to worry about having it floating around. Yes this is a little less flexible since I can’t move it around like a external device could, but it also means that I don’t need a daughterboard to be able to control the hosts power.

Wiring it up wasn’t anything specifically hard, internally you put it in line with the front panel header for your power, reset, and indicator LED’s and then run a HDMI cable between the monitor and the NanoKVM and you are almost ready to go. You still have to get a USB connection to emulate a keyboard and mouse, and that can be done using internal connectors or a external jumper. In my system there are no more internal USB connectors so I’ve got a USB A to C cable connecting the card to the rest of the system.

The card can pull power from the computer through the PCIe connector, and you can wire up a external power source through one of the USB-C ports on the back so that the device keeps running even if you turn off the host machine. There is a option that you can get that will power these over POE, but since I don’t have a POE switch I didn’t bother paying the extra for that.

The last part is the network connection. Mine’s just got the option to do wired ethernet, but there are some versions that have WIFI as a option, but again I don’t think that’s something that’s a huge bonus for me.

So after getting the thing wired up and installed I let it boot up and grab a IP address so that I could see what’s up with it. The interesting thing about this is that it’s got a little OLED on the back that shows you what the current IP address and connection status is. Handy since that means I’m not having to dig into my router to look for the device and find what address it picked up.

And once I got logged in and changed my password and user name it was all fairly straightforward. None of the configuration options are specifically esoteric and to be honest there’s not a lot of them. Strangely enough there isn’t even a option in the UI to set a static IP address, if you want one you have to either reserve it on the router side of things or you have to configure it in the terminal on the device itself.

The device does have a option however to tie into TailScale directly so that nice. Although it’s apparently a version that’s out of date.

The one thing that I will note is that these devices were initially flagged as having some security concerns. Some of those concerns are things that don’t really give me much pause – specifically the hardcoding of Google’s public DNS servers (I use those anyhow). However there were some things noted that were a concern, specifically how some of them had a microphone embedded on the device. Turns out that’s because the board they were using is a repurposed development board that happened to have a microphone on it and that fact just seemed to be missed by a lot of people.

The manufacturer did post back a response on this shortly after the concerns came to light;

https://github.com/sipeed/NanoKVM/issues/301

And looking through the list of issues it seems like the vendor has been addressing them, and at this point I’m not terribly worried about the device in my network. Granted it’s on a VLAN that has no access to the outside world, but that’s something that I would be doing regardless of who made the card in the first place.

At this point I think that the thing works well enough for what I’m doing, and some more may wind up finding their way into some of the other machines that I’m working with at home since getting servers with “proper” IPMI cards installed is well outside of my price range.

So it’s been a couple of weeks since I was at the IBJJF worlds and now that I’m back into my normal flow of life I guess that it’s time to think about what’s next.

First question, am I going to compete again? I’m thinking that I will so I’m going to proceed under the assumption that I’m making the trip again next year. So now I’m looking at what I need to do to get on the podium next year. Thankfully the answer is simple. I just need to get better at JuJitsu.

And that answer is oh so helpful.

But it’s also completely true. There was a huge gap in skills between the two guys that I fought. The first one I was able to get control of, play my game, and eventually get a submission. The second guy I fought, I was the one getting controlled. Both were apparently 4 stripe brown belts but the second guy that I fought felt more like rolling with a black belt than any other brown belt that I’ve rolled with.

So there’s obviously a huge range in skill when you talk about people in a given belt range, and that means that I need to be getting myself to the top of that range if I want to be competitive. That means that I need to basically start over and work everything I know from the ground up. Every position, movement, and technique needs to be re-evaluated and polished. Some things I think that I know are things that I haven’t revisited since I was a purple belt and touching them again may open up new pathways to other positions that I’m just missing right now.

Either case, it’s a lot of work ahead of me.

Now I just need to shake of this head cold that I picked up and I’ll be back on the mats.

I’m in somewhat of a grumpy mood so I’m going to complain here. Feel free to bail if you don’t like that type of stuff. If you are hanging around you have been warned.

So. The internet. I’ve been online for a long, long, while. I was online before webpages were really a thing and most of the interesting content was sitting in newsgroups and BBB’s forums. During that time there was little that really tied your activity online to your actual physical identity and that anonymity could let you express opinions freely without a lot of worry about backlash.

Obviously there are things that a lot of people do that wind up linking you back to your original identity. Even here I’m sure that someone could figure out who the hell I was with little issue since I’m not really hiding who I am here. However for some reason people still seem to think that just being online shields them from consequences.

And social media has not helped the situation any.

Now generally I’m really a live and let live type of person. However there were a few things that do get under my skin. And recently I’m seeing more and more people doing things that just cross a line into an area where I’m not comfortable with the whole live and let live approach.

Yes. Vauge bullshit. I know, but I’ll be right fucked if I’m going to draw attention to the crap I’ve been seeing.

My biggest gripe is the crap that’s flooding discussion about political matters. The ability to hide your identity on the internet and Generative AI’s ability to fake almost anything means that suddenly I’m getting flooded with political posts that are designed to enrage me and push me towards a specific outcome. The saving grace on a lot of those posts is that I tend to pay attention to what’s going on and many of the people using these tools don’t realize how similar all the generated content seems when you look at it.

So now I’m sorting through posts that are obviously AI gene I can tell that the information is being heavily manipulated, if not just misrepresented. However many people don’t pay attention to things, see the posts, and then get enraged about what’s going on and suddenly you can’t actually discuss anything about whatever the topic is without people painting you as some type of jackass who’s trying to do horrible things.

And it’s not like there’s specific bastions of truth that you can trust any more, if there ever were.

Even if it’s not something state run media there’s generally still some bias involved. In some cases it’s obvious what the bait is because the folks involved scream it from the top of their lungs like it’s a badge of honour. In other cases it’s not really apparent until something kind of pulls back the mask a bit and you start digging and realize what’s going on. Suddenly you realize that the news source that you were looking at is something that’s bene trying to manipulate you from day one.

Is it tempting to crawl under a rock, assume everything is bullshit, and try to hide.

Yes.

Hells. Yes.

However I still prefer having easy access to cat videos and the ability to get a warm pizza delivered to my hose within 15min.

I was poking around some WordPress themes, kind of toying with the idea of putting something up that looked a little fancier than what’s currently on here.

And after poking around for a bit I decided that it really wasn’t worth the time to spend screwing around with new templates and such. At the end of the day I would rather have the template look plain but have content that’s worth reading rather than having something that’s super fancy looking but has nothing of value on it.

I might look into playing with the fonts a bit and seeing if I can find one that’s a bit more to my liking than the set that’s currently in place. That’s however something for another day, since it’s getting late.