The Inherent Problems with Using LLMs to Code

Introduction

Software companies are pushing their developers more and more to use LLMs in order to increase productivity. The argument is that LLMs allow developers to write more code in a shorter period of time. Whether it’s automating processes like requirement gathering or pull request reviews, or creating code like prototypes, unit tests and new features, or refactoring old code into newer more efficient code – there’s a point to be made. Yes AI can help with that. There’s been a joke for a long time that Google searches are a developer’s partner when it comes to coding. LLMs take it one step further and actually write the code you need based on the code it’s been trained on from the internet. But while LLMs can help developers produce more code, faster; what are the downsides? In this article I’m not talking about the technical downsides, or using AI agents to automate processes, but more about the psychological downsides to the workforce, and how that effects the code being generated.

Motivation

I’ve heard from some of my colleagues that they really like using LLMs because they “hate coding.” I find that quite interesting since being a developer means you’re going to be coding a lot. Why did you go into a career where the majority of its purpose is something you “hate”? If you asked a career counselor in high school what careers pay a lot of money, one of the top answers would be “software engineer.” Many of those kids went that route for that answer alone. They had no real interest in computers, or coding. It was just the default job to get into if you wanted to make a lot of money. These engineers will hail AI as their savior, and if you truly don’t like coding, that makes sense.

Then there’s the other people who got into software development. Nerds. There’s a group of people that were into computers as a hobby, and the natural transition was to continue that obsession into a career. These are the people that actually enjoy coding. Making someone who enjoys coding use an LLM to code for them is like asking a mechanic who enjoys working on cars to control a robot to do it. What motivation does a software engineer who enjoys coding have now that you’ve taken it away from them? Why continue to learn new languages? Why stay in the industry at all?

Ownership/Pride

If you’re a software engineer that enjoys coding, you have a sense of pride in what you create. You took on a challenge, designed a solution, and executed that solution. That’s something to be proud of. It’s also something you take responsibility for. You designed it. You wrote the code. It’s yours. If the code is good, you take credit for that. If the code ends up having problems, you have to take credit for that too and take it upon yourself to fix it. In a nutshell, because you have pride in something, you claim ownership of it, and if you own it, you want it to be good.

In the case of using AI, you tell the LLM what you want, how you want it implemented, and ask it to generate the code. What if the code created by the LLM works but is optimized badly? What if there’s security risks? What if the data model that was created is bad? Would a human have the same amount of pride in something they didn’t actually create? I would argue no; It’s just human nature. A developer in this situation SHOULD take as much care as one that wrote the code from beginning to end, but I don’t think that’s realistic. The feeling of ownership isn’t there. If the level of pride is less, then the level of care is also less. In the end, less pride ends up with lesser code.

Skill

What happens to a software engineer’s coding skills the more and more they use AI? Unlike riding a bike, coding is a “use it or lose it” skill. Much like learning a new language, if you don’t speak it often, you get out of practice and start forgetting what you’ve learned. Using AI to generate code lets that “brain muscle” atrophy over time. In fact Anthropic, one of the leaders in AI development, recently published a study showing that a collection of software engineers using AI over a period of time performed worse on a skills test than the engineers that didn’t. The article about the study can be found HERE. Use it, or lose it. What does this predict for the future of software engineering? Will there be a mass of AI generated code that after years of changes no one knows how to troubleshoot? Will companies be locked into paying for AI forever because there’s no one around who can write decent code from scratch anymore? These are probably exaggerated consequences, but they are something corporate leadership with have to deal with. This also leads us into the next topic, the workforce.

Workforce

Early in the push to use AI for coding there was the concern that AI would replace junior engineers. If the executives of software firms were to think logically, and see the big picture, they’d realize that if they’re not training and mentoring their junior engineers, they’re not investing in their future. Junior engineers become senior engineers and you’ll need that expertise to maintain your code in the future. However, that assumes that executives are thinking logically, and that they’re looking past their upcoming quarterly market goals. While using AI as the scapegoat for layoffs is currently rampant in the industry, there’s still reason to be concerned that the more you rely on AI means relying more and more on AI in the future. If using AI does negatively affect the skills of software engineers, if allowed to reach its inevitable conclusion, you’ll end up with code beholden to AI, maintained by engineers beholden to AI. Is that what these corporations want? Their intellectual property tied to the services of an AI company?

Mental Paradigm

I’ve come to realize that for me, using an LLM to write my code is more mentally tiring than writing the code yourself. When you’re working with an LLM to write code, you’re essentially “chatting” with “someone” all day. We interact with LLM‘s using somewhat normal speech. You’re convincing an entity to write the code you want. The thought process is totally different. When you’re writing the code yourself, there’s no conversation, it’s just logic. That’s a different part of your brain. After a day of coding with AI, I feel like I’ve been in a day long meeting because at it’s most basic form – I was.

Conclusion

AI and using LLMs to assist with software development is here to stay. It IS useful. It does help developers create more code at greater speed. It also allows developers to create things they might not have been able to create without it. But like most things in life, AI is a double-edged sword. The good needs to be balanced with the bad, and I hope the challenges I discussed here are things that we as an industry can avoid or find solutions to.

G&L Neck Profiles

This a copy of the post by “Gordon” at Upfront Guitars, a premier G&L dealer before G&L went out of business. I’m copying it here for preservation purposes.

This is an update to an earlier blog post to cover the changes G&L made to their neck offerings during their 2015 mid-year model changes.

One thing that may not be clear to buyers is that neck profile and fretboard radius are independent. Meaning that while the Modern Classic neck that comes standard on most guitars has a 9.5? radius, it’s available in 7.5? or 12? radius too. Any profile is available with any radius, with the exception that you can’t get the 7.5? radius with a Bigsby. So by decoupling the profile and radius specification, G&L has actually expanded what’s available in terms of options. 

As we cover the various neck profiles, we’ll discuss them as much as possible independent of fretboard radius, and we’ll also assume the neck has the standard Jescar 57110 fret. In parenthesis will be the old neck designation where applicable. 

Lastly, all neck profile tolerances are +/- .015 (1/64th) relative to their stated dimension (it is wood after all not titanium). From the upper to lower end of the tolerance, this is a difference many people can feel. If you are on the fence about a certain profile but are super sensitive to thickness for example, opt for the thinner/narrower selection.

Modern Classic – The Modern Classic is now the standard neck on all popular Legacy, S-500, Comanche and ASAT models (.820? at the first fret and .870? at the 12th fret). On most guitars they use a 9.5? radius, with the exception of the Invader Models which are 12?.  Think of the Modern Classic as a Slim C with a 1-11/16? nut width, and the same string spacing. The “MCNK” addresses the two issues of: 1) Occasional string falloff with the 1-5/8 nut width, and 2) “What feels most like a Fender?” With the thinner profile and 9.5? radius, the additional width is hardly noticeable, whereas on C Plus (Wide C) it can start to feel a little too manly. The MCNK is a good all-things-to-all-people neck, although I find it a little lacking in palm support in the upper frets (which my son really likes, and his hands are a little smaller).

Classic C (#1) – Up until July 2015, the G&L 12? radius #1 neck was the standard. It is the only neck available on the F-100 and SC-2.  Measuring .830? at the first fret and .960? at the 12th fret, it’s mildly beefy and fits most people well. Versus the MCNK, I like the extra thickness in the higher frets to anchor my palm. Every once and a while some necks do exhibit some string falloff on the high E, but it’s rare. We’ve also tried this neck with the Dunlop 6230 vintage fret option, but the combination of flat neck with skinny fret makes the frets seem undersized, and they look a little lost on the wide flat fretboard. (1-5/8? nut width)

Slim C (#1a) – This is your MCNK neck without the nut width. This is the onlyneck profile available on the Fallout (yes, really). The G&L #1a is about the same size as the #1 at the first fret, but only .870? at the 12th. There is very little taper to the neck, so it feels quite slim as you move up the frets. Good for shredders and people who like to be able to reach around the neck and hammer the notes. (1-5/8th nut width)

Heritage ’86 (#1b) – Before the MCNK, when we intentionally wanted a slim neck, this is the one we’d order. The “Heritage C” profile is great for women, people with smaller hands, or folks who like a thinner profile. Feels instantly comfortable and tapers nicely up the frets to .910?. I’ve always liked the feel of this profile more than the #1a, and it’s still a nice option. (1-5/8th nut width)

Modern U (#1c) – This is G&L’s Modern “U” shape, which at .850? is pretty beefy in the lower frets, but tapers less than most to .910? at the 12th fret. We’ve only had one of these, and it feels very much like a “C” neck. If you like a more generous neck proportion down low — like a 50’s Gibson — this neck will do the trick. But if you have no strong opinions on neck shape, a Classic C is probably a safer choice. (1-5/8? nut width)

 Deep V (#1d) – This is G&L’s Modern “V” neck, which is a shape that not a lot of players have experience with, but I’m a big fan. At .890? at the first fret, it is G&L’s thickest neck, and it tapers to .930 at the 12 fret. The V shape provides a lot of “beef” but since it tapers more rapidly to the sides than a “C” profile, it does not wind up feeling bulky. So you get depth without the drawbacks of a large “C” profile. It’s a pretty neat feel, and personally I find it very comfortable and use it on my pine ASAT Classic. Players such as Eric Clapton have been proponents of this shape, and it’s featured on some of his signature guitars. (1-5/8? nut width)

Classic C Plus (#3) – This is the Classic C neck with a 1-11/16? nut width. A good neck for players with larger hands, as the combination of wider nut and classic profile is something you can really feel. Conventional wisdom is that “thinner is faster” but if you like a neck that provides generous palm support, the C Plus is the ticket.

Classic C Wide (#4) – The “Shrek Neck” which is 1-3/4? wide with additional string spacing. We have not sampled this neck yet, but if you think you need it, you probably do.

Soft V (#2a) – Another neck that we have not tried, but it sounds intriguing given our infatuation with the Deep V profile. In theory it should be the solution for players who find the Classic C a little too chunky down low but want less of a taper than a Slim C. We might have to get one of these….. (1-5/8? nut width)

Quartersawn versus Flatsawn Necks– All standard G&L necks are flatsawn. If you were to take off the neck and look at the end of the neck you would see that the grain of the wood is parallel to the fingerboard. This makes a stable neck, and also they also get more Flatsawn necks per piece of maple stock, which keeps the cost down. Quartersawn necks have the grain perpendicular to the fingerboard, so the wood is much stiffer in the direction that the neck typically bends. But cutting this way yields fewer necks per piece of maple stock, which increases cost.  Sonically, the stiffer quartersawn neck is felt to be more percussive, with a quicker attack and less note compression than a flatsawn neck (theoretically it makes sense). While it’s not practical to A-B two necks on the same guitar, personal opinion is that guitars I’ve played with a quartersawn necks have a little more attack, but maybe a touch less complexity. Some players swear by them, and if you live in an area that varies widely in temperature and humidity, a quartersawn neck may require less tweaking. Not a bad idea either on bass guitars with their longer necks and string tension. And if you are worried about “dead spots” on a bass, my recommendation would be to opt for quartersawn.

Amp Frustrations and the Plexi Paradox

In *theory* I love the “Plexi” sound. To be more precise, I mean the Marshall Super Lead over-driven sound. Every time I try a profiled one of these amps on my Tone X profiler, it’s my go-to sound. Everytime I hear someone demoing one online – it’s the sound I’m looking for. I also quite like the sound of a Marshall JTM 45 as a pedal platform (again based on online demos), so much so that I have a JTM 45 pedal I use in front of my Fender amp, and I really dig it.

So, I figured it was time to dip my toes into the “plexi” world by buying some amps based on the plethora of demos I’ve seen online. I’ve tried the PRS HDX, the Tone King Royalist, and the Marshall Vintage 20 Studio. I’ve even tried the Plexi’s granddaddy, the Bassman, and its russian clone, the MIG. Here’s what I’ve decided. I don’t like a “real Plexi”.

Now before you immediately jump to the comment section to spout off about how these amps need to cranked to sound good, and that the cabinet really matters I already know that. But even cranked these amps just don’t sound good to me. There’s a harmonic to the gain that I find especially displeasing, an overtone that doesn’t sound musical. That overtone can be tamed a bit by dialing back the middle frequencies, but it’s always going to be there. I have cabs with Celestion Greenbacks, Celestion V30s, and Creambacks – all speakers that I love, but they don’t “fix” the tone on all the “plexis” I’ve tried.

The real frustration is this. Trying out amps before you buy them these days is a real chore. If you’re anything like me, you’re main music shop is Guitar Center. If you don’t want a Boss Katana, a Fender combo, a Vox combo, or an Orange Crush, you’re kind of out of luck. The independent mom and pop stores tend to carry a small selection of amp brands they’re particularly fond of, and if you don’t want one of those you’re again up a smelly creek without a paddle. What happens is that you end up buying an amp online and then returning it if you don’t like it. Many stores accept returns, but they don’t pay for the shipping back. So, you basically end up spending $65 or more each time you want to try an amplifier. This has been my life the last few months trying to find an amp I like.

So why do I love the Super Leads on my profiler but not the real thing? And if I like them in my profiler so much, why do I care about getting a real one? Obviously, the amp sound from the profiler is idealized, and is coming out of my studio monitors. It sounds like a recorded amp should sound. I know I wouldn’t get that out of a real amp being played loud in the room, but I at least thought I could get it close. As far as the second question goes, the answer is simple. I like tube amps. I like how they look, how they glow, how the hot tubes smell. I like the switches and knobs. I wish I could be happy with just a profiler with a bunch of built in amp sounds, but I just love amps too much.

What’s the lesson here? I guess what I’ve learned (and I don’t like it) is that I need to buy amps from Guitar Center online. That way if I don’t like it, I can just take it back to the local store without any shipping fees. That will unfortunately cause me to lose out on some deals however, but I guess it really isn’t a deal if I have to pay shipping to return the amp if I don’t like it. I just can’t deal with the gamble anymore.

Where’s the Settings.xml File for Elite Dangerous on Linux?

~/.steam/steam/steamapps/compatdata/359320/pfx/drive_c/users/steamuser/AppData/Local/Frontier Developments/Elite Dangerous/Options/Graphics

Enable Administrator Login in Windows 11

Did you accidentally lock your administrator login by attempting the wrong password too many times? Do you need to re-enable the account, and reset the password? Try the following :

  1. Hold SHIFT and click the power icon on the lock screen. Continue holding SHIFT and click on “Restart”. This will boot the machine into Troubleshooting mode
  2. Select “Advanced Options” on the “Troubleshoot” screen
  3. Select “Command Prompt”. This will open a command prompt window.
  4. In the Command Prompt window, type the following command :
    net user Administrator /active:yes
  5. Now that the Administrator account has been re-enabled, you can set a new password by typing this command :
    net user Administrator <type your password here>
  6. Close the Command Prompt window
  7. Restart the computer
  8. Login with the Administrator account with the password you just set

Fender Mike McCready Stratocaster – A Counter-Review

When searching Google for reviews of the Fender Mike McCready Stratocaster, a made in Mexico replica of Mike’s beloved 60’s Strat, you’ll mostly find positive things being said. Yes, the fretwork is great. The rounded fret board edges are super comfortable. The electronics quality is impressive. If you like relic’ed guitars, then yes – it looks cool. BUT at a cost of $1899 (I bought mine B-stock for slightly less) there are some things about the guitar that has prompted me to return it.

I’m not going to bury the lead here, so I’ll start with the biggest issue, the back of the neck. The McCready model is supposed to have a “worn in” neck back where the gloss of the nitro finish has been worn off. If you look at the photos provided by the Fender website, you can see what I mean :

First off, one of the reasons I bought this guitar was because I was hankering for a 60s Strat but I’m not a fan of gloss necks. This sixties-esque model would have solved that dilemma for me but for whatever reason, the one I received wasn’t very “broken in”. Yes the gloss has been dulled, but not to the extent as shown in the marketing photos. The neck is STICKY. So much so, that sliding up and down the neck is a chore. I don’t think this is intended by design, but an indication of Fender’s recent quality control issues. My guess is that this one made it out of the factory with the “worn in” neck barely worn. (Perhaps my expectations are unrealistic here, but I’ve owned other MiM relic’ed guitars with the neck back worn, and in those cases it was almost a bare wood feel.) For this reason alone, I was highly considering a return.

Secondly, after playing the guitar for a good while, I found the neck pickup especially dark. So much so, I had to EQ my amp differently from all my other Strats. It was dark enough that at first I thought the neck tone knob was turned down. It wasn’t. And even with amp EQ changes, I still wasn’t happy with the neck pickup sounds.

Third, the action upon receipt was HIGH. I’m not one of those guys who needs his action super low to enjoy playing. Many of my guitars have what I’d call “medium” action, but this guitar’s setup exceeded even my tolerance. (Granted this could be an issue with the shop I bought it from, and their setup – not the fault of Fender) I adjusted the action to meet my needs and it was much better, but when paying as much as you do for one of these guitars you shouldn’t have to make such an adjustment.

Finally, we come to the case. It’s cool that it comes in it’s own special case. I dig the brown tweed, and the yellow interior. However, the case is not the highest quality. It’s flimsy. It flexes easily. Setting it down any kind of non-level surface will prevent the locks from lining up. The stitching of the interior fabric is not the result of someone being shall we say, detail oriented. There are sloppy folds, and wads of fabric in the corners. Being used to my G&L cases, this one was disappointing.

In the end, I’ve decided to send the guitar back. This is an expensive Mexican model (if not the most expensive) and for that price point it should really be an awesome experience to own one. For me it wasn’t. At that price point you could buy impeccable Indonesian models from other manufactures with more features and better quality control. In fact, I might just forgo my desire for a 60s-like Strat and go with one of those manufacturers. Or better yet, another G&L might be in my future.

New Tone X Software and Firmware Released!

The new Tone X update is out. Upgrade your software and firmware today!

While I don’t agree with several of this reviewer’s complaints, (especially given the price point), the demo is very good.

IK Multimedia to Release New Effects for Tone X in November

https://www.ikmultimedia.com/news/?item_id=16996&srsltid=AfmBOorbvRzTVqYb1O2zCi2UhbTMByd_67BQGx30OgSN8-_AOEMiAfZt

New effects will be coming to Tone X in November. While the previous release added a tuner, this future release will include 2 delays, chorus, spring reverb, tremolo, phaser, flanger and rotary.