Posts tagged with "AI"

Another itch to scratch

As I mentioned yesterday, I've been a happy user of Pelican for a couple or so years now, but every so often there's a little change or tweak I'd like to make that requires diving deeper into the templates and the like and... I go "eh, I'll look at it some time soon". Another thought that often goes through my head at those times is "I should build my own static site generator that works exactly how I want" -- because really any hacker with a blog has to do that at some point.

Meanwhile... I've had free access to GitHub Copilot attached to my GitHub account for some time now, and I've hardly used it. At the same time -- the past few months especially -- I've been watching the rise of agents as coding tools, as well as the rise of advocates for them. Worse still, I've seen people I didn't expect to be advocates for giving up on coding turning to these tools and suddenly writing rationales in favour of them.

So, suddenly, the idea popped into my head: I should write my own static site generator that I'll use for my blog, and I should try and use GitHub Copilot to write 100% of the code, and documentation, and see how far I get. I doing so I might firm up my opinions about where we're all going with this.

The requirements were going to be pretty straightforward:

• It should be a static site generator that turns Markdown files into a website.
• It should be blog-first in its design.
• It should support non-blog-post pages too.
• It should be written in Python.
• It should use Jinja2 for templates.
• It should have a better archive system than I ever got out of my Pelican setup.
• It should have categories, tags, and all the usual metadata stuff you'd expect from a site where you're going to share content from.

Of course, the requirements would drift and expand as I went along and I had some new ideas.

Getting started

To kick things off, I created my repo, and then opened Copilot and typed out a prompt to get things going. Here's what I typed:

Build a blog-oriented static site generation engine. It should be built in Python, the structure of the repository should match that of my preferences for Python projects these days (see https://github.com/davep/oldnews and take clues from the makefile; I like uv and ruff and Mypy, etc).

Important features:

• Everything is written in markdown
• All metadata for a post should come from frontmatter
• It should use Jinja2 for the output templates

As you can see, rather than get very explicit about every single detail, I wanted to start out with a vague description of what I was aiming for. I did want to encourage it to try and build a Python repository how I normally would, so I pointed it at OldNews in the hope that it might go and comprehend how I go about things; I also doubled-down in the importance of using uv and mypy.

The result of this was... actually impressive. As you'll see in that PR, to get to a point where it could be merged, there was some back-and-forth with Copilot to add things I hadn't thought of initially, and to get it to iron out some problems, but for the most part it delivered what I was after. Without question it delivered it faster than I would have.

Some early issues where I had to point out problems to Copilot included:

• The order of posts on the home page wasn't obvious to me, and absolutely wasn't reverse chronological order.
• Footnotes were showing up kinda odd.
• The main index for the blog was showing just posts titles, not the full text of the article as you'd normally expect from a blog.

Nothing terrible, and it did get a lot of the heavy lifting done and done well, but it was worth noting that a lot of dev-testing/QA needed to be done to be confident about its work, and doing this picked up on little details that are important.

An improvement to the Markdown

As an aside: during this first PR, I quickly noticed a problem where I was getting this error when generating the site from the Markdown:

Error generating site: mapping values are not allowed in this context
  in "<unicode string>", line 3, column 15

I just assumed it was some bug in the generated code and left Copilot to work it out. Instead it came back and educated me on something: I actually had bad YAML in the frontmatter of some of my posts!

This, by the way, wouldn't be the last time that Copilot found an issue with my input Markdown and so, having used it, improved my blog.

A major feature from a simple request

Another problem I ran into quickly was that previewing the generated site wasn't working well at all; all I could do was browse the files in the filesystem. So, almost as an offhand comment, in the initial PR, I asked:

Can we get a serve mode please so I can locally test the site?

Just like that, it went off and wrote a whole server for the project. While the server did need a lot of extra work to really work well¹, the initial version was good enough to get me going and to iterate on the project as a whole.

The main workflow

Having kicked off the project and having had some success with getting Copilot to deliver what I was asking for, I settled into a new but also familiar workflow. Whereas normally, when working on a personal project, I'll write an issue for myself, at some point pick it up and create a PR, review and test the PR myself then merge, now the workflow turned into:

• Write an issue but do so in a way that when I assign it to Copilot it has enough information to go off and do the work.
• Wait for Copilot to get done.
• Review the PR, making change requests etc.
• Make any fixes that are easier for me to fix by hand that describe to Copilot.
• Merge.

In fact, the first step had some sub-steps to it too, I was finding. What I was doing, more than ever, was writing issues like I'd write sticky notes: with simple descriptions of a bug or a new feature. I'd then come back to them later and flesh them out into something that would act as a prompt for Copilot. I found myself doing this so often I ended up adding a "Needs prompt" label to my usual set of issue labels.

All of this made for an efficient workflow, and one where I could often get on with something else as Copilot worked on the latest job (I wasn't just working on other things on my computer; sometimes I'd be going off and doing things around the house while this happened), but... it wasn't fun. It was the opposite of what I've always enjoyed when it comes to building software. I got to dream up the ideas, I got to do the testing, I got to review the quality of the work, but I didn't get to actually lose myself in the flow state of coding.

One thing I've really come to understand during those 5 days of working on BlogMore was I really missed getting lost in the flow state. Perhaps it's the issue to PR to review to merge cycle I used that amplified this, perhaps those who converse with an agent in their IDE or in some client application keep a sense of that (I might have to try that approach out), but this feels like a serious loss to me when it comes to writing code for personal enjoyment.

The main problems

I think it's fair to say that I've been surprised at just how well Copilot understood my (sometimes deliberately vague) requests, at how it generally managed to take some simple plain English and turn it into actual code that actually did what I wanted and, mostly, actually worked.

But my experiences over the past few days haven't been without their problems.

The confidently wrong problem

Hopefully we all recognise that, with time and experience, we learn where the mistakes are likely to turn up. Once you've written enough code you've also written plenty of bugs and been caught out by plenty of edge-cases that you get a spidey-sense for trouble as you write code. I feel that this kind of approach can be called cautiously confident.

Working with Copilot², however, I often ran into the confidently wrong issue. On occasion I found it would proudly³ request review for some minor bit of work, proclaiming that it had done the thing or solved the problem, and I'd test it and nothing had materially changed. On a couple of occasions, when I pushed back, I found it actually doubting my review before finally digging in harder and eventually solving the issue.

I found that this took time and was rather tiring.

There were also times where it would do the same but not directly in respect to code. One example I can think of is when it was confident that Python 3.14 was still a pre-release Python as of February 2026 (it isn't).

This problem alone concerns me; this is the sort of thing where people without a good sense for when the agent is probably bullshitting will get into serious trouble.

The tries-too-hard problem

A variation on the above problem works the other way: on at least one occasion I found that Copilot tried too hard to fix a problem that wasn't really its to fix.

In this case I was asking it to tidy up some validation issues in the RSS feed data. One of the main problems was root-relative URLs being in the content of the feed; for that they needed to be made absolute URLs. Copilot did an excellent job of fixing the problem, but one (and from what I could see only one) relative URL remained.

I asked it to take a look and it took a real age to work over the issue. To its credit, it dug hard and it dug deep and it got to the bottom of the problem. The issue here though was it tried too hard because, having found the cause of the problem (a typo in my original Markdown, which had always existed) it went right ahead and built a workaround for this one specific broken link.

Now, while I'm a fan of Postel's law, this is taking things a bit too far. If this was a real person I'd tasked with the job I would have expected and encouraged them to come back to me with their finding and say "dude, the problem is in your input data" and I'd have fixed my original Markdown.

Here though it just went right ahead and added this one weird edge case as something to handle.

I think this is something to be concerned about and to keep an eye on too. I feel there's a danger in having the agent rabbit-hole a fix for a problem that it should simply have reported back to me for further discussion.

The never-pushes-back problem

Something I did find unsurprising but disconcerting was Copilot's unwillingness to push back, or at least defend its choices. Sometimes it would make a decision or a change and I'd simply ask it why it had done it that way, why it had made that choice. Rather than reply with its reasoning it would pretty much go "yeah, my bad, let me do it a way you're probably going to find more pleasing".

A simple example of this is one time when I saw some code like this:

@property
def some_property(self) -> SomeValue:
    from blogmore.utils import some_utility_function
    ...

I'm not a fan of imports in the body of methods unless there's a demonstrable performance reason. I asked Copilot why it had made this choice here and its reply was simply to say it had gone ahead and changed the code, moving the import to the top of the module.

I see plenty of people talk about how working with an agent is like pair-programming, but I think it misses out on what's got to be the biggest positive of that approach: the debate and exchange of ideas. This again feels like a concern to be mindful of, especially if someone less experienced is bringing code to you where they've used an agent as their pair buddy.

The overall impression

Now I'm at the end of the process, and using the result of this experiment to write this post⁴, I feel better informed about what these tools offer, and the pitfalls I need to be mindful of. Sometimes it wasn't a terrible way of working. For example, on the first day I started with this, at one point on a chilly but sunny Sunday afternoon, I was sat on the sofa, MacBook on lap, guiding an AI to write code, while petting the cat, watching the birds in the garden enjoy the content of the feeder, all while chatting with my partner.

That's not a terrible way to write code.

On the other hand, as I said earlier, I missed the flow state. I love getting lost in code for a few hours and this is not that. I also found the constant loop of prompt, wait, review, test, repeat, really quite exhausting.

As best as I can describe it: it feels like the fast food of software development. It gets the job done, it gets it done fast, but it's really not fulfilling.

At the end of the process I have a really useful tool, 100% "built with AI", under my guidance, which lets me actually be creative and build things I do create by hand. That's not a bad thing, I can see why this is appealing to people. On the other hand the process of building that tool was pretty boring and, for want of a better word... soulless.

Conclusion

As I write this I have about 24 hours of access to GitHub Copilot Pro left. It seems this experiment used up my preview time and triggered a "looks like you're having fun, now you need to decide if you want to buy it" response. That's fair.

So now I'm left trying to decide if I want to pay to keep it going. At the level I've been using it at for building BlogMore it looks like it costs $10/mth. That actually isn't terrible. I spend more than that on other hobbies and other forms of entertainment. So, if I can work within the bounds of that tier, it's affordable and probably worth it.

What I'm not sure about yet is if I want to. It's been educational, I can 100% see how and where I'd use this for work (and would of course expect an employer to foot the bill for it or a similar tool), and I can also see how and where I might use it to quickly build a personal-use tool to enable something more human-creative.

Ultimately though I think I'm a little better informed thanks to this process, and better aware of some of the wins people claim, and also better informed so that I can be rightly incredulous when faced with some of the wilder claims.

Also, it'll help put some of my reading into perspective.

For about 2 and a half years now this blog has been built with Pelican. For the most part I've enjoyed using it, it's been easy enough to work with, although not exciting to work with (which I think is a positive thing to say about a static site generator).

There were, however, a couple or so things I didn't like about the layout I was getting out of it. One issue was the archive, which was a pretty boring list of titles of all the posts on the site. It would have been nice to have them broken down by date or something, at least.

Of course, there are lots of themes, and it also uses templates, so I could probably have tweaked it "just so"; but every time I started to look into it I found myself wanting to "fix" the issue by building my own engine from scratch.

Thankfully, every time that happened, I'd come to my senses and go off and work on some other fun personal project. Until earlier this week, that was.

The thing is... I've been looking for a project where I could dive into the world of "AI coding" and "Agents" and all that nonsense. Not because I want to abandon the absolute thrill and joy I still get from writing actual code as a human, but because I want to understand things from the point of view of people who champion these tools.

The only way I'm going to have an informed opinion is to get informed; the only way to get informed is to try this stuff out.

So, here I am, with my blog now migrated over to BlogMore; a project that gives me a blog-focused static site generator that I 100% drove the development of, but for which I wrote almost none of the code.

At the moment it's working out well, as a generator. I'm happy with how it works, I'm happy with what it generates. I also think it's 100% backwards-compatible when it comes to URLs and feeds and so on. If you do see anything odd happening, if you do see anything that looks broken, I'd love to hear about it.

As for this being a "100% AI" project, and how I found that process and how I feel about the implications and the results... that's a blog post to come.

I took lots of notes.

After having knocked up obs2nlm the other day I realised there was another source of sources available to me that might be fun to throw into NotebookLM: one or more of the Norton Guide files in my collection!

And so, after a wee bit of hacking on a wet Sunday afternoon, ng2nlm was born.

In terms of what it does, it's pretty much the same as obs2nlm, only it uses ngdb to read the contents of a Norton Guide and turn it into a single Markdown file which can then be used as a source in NotebookLM.

So far it seems to be working a treat.

Here I grabbed the guide to assembly language Norton Guide, and the DOS interrupts Norton Guide, turned them into sources, created a notebook and let it do its thing.

I can't vouch for how valid the answer to my question is -- it's a long time since I wrote any 8086 assembler code and it's a moment since I last had to dig into DOS interrupts and the like -- but I like the result.

Probably the biggest issue I see at the moment is with the code examples: it seems to have a habit of adding some sort of citation marker that links to nowhere when it's emitting code. I think this is down to the instructions I've given it in the source I output, at least that's what it's suggesting when I ask it:

Me: What are all the numbers in square brackets for?

NotebookLM: The numbers in square brackets are passage indices used to cite which specific part of the sources supports each statement [Instruction]. Within the source files themselves, these markers identify distinct entries or sections within the Norton Guide databases.

Given I am trying to encourage it to cite its sources, I think this is where the confusion comes from. In "normal" paragraphs it does do a good job of citing its source and linking to it:

so presumably, when it "thinks" it's outputting code it doesn't do the markup to actually do the linking, and so things end up looking a little confused.

If I ask that citations aren't included in the code, this does seem to have the desired effect:

So, given this, perhaps I can use the --additional-instructions switch for the particular source to encourage it to not add citation links to code? Or of course I could tweak ng2nlm itself to include that instruction to start with. While it's a little specific to one use case, Norton Guide files do tend to be coding-related so it could make sense.

Anyway, in the very unlikely event that you have a need to turn one or more Norton Guide files into sources to throw at NotebookLM or similar tools: ng2nlm exists.

I'm sure I've mentioned a couple of times before that I've become quite the fan of Obsidian. For the past few years, at any given point, I've had a couple of vaults on the go. Generally I find such vaults a really useful place to record things I'd otherwise forget, and of course as a place to go back and look things up.

But... even then, it's easy enough to forget what you might have recorded and know that you can even go back and look things up. Also I tend to find that I can't quite figure out a good way of getting a good overview of what I've recorded, over time.

Meanwhile: I've been playing around with Google's NotebookLM as a tool to help research and understand various topics. After doing this with my recent winter break coding project (more on that in the future) I realised I really should get serious about taking this approach with my Obsidian Vaults.

I'm sure this is far from novel, I'm sure lots of people have done similar things already; in fact I'd quickly dabbled with the idea a few months ago, had a bit of a laugh at some of the things the "studio" made of a vault, and promptly forgot about it.

This time though I got to thinking that I should try and take it a little more seriously.

And so obs2nlm was born.

The idea is simple enough: enumerate all the Markdown files in the vault, wrap them in boundary markers, add some instructions to the start of the file to help NotebookLM "comprehend" the content better, throw in a table of contents to give clues to the structure of the vault, and see what happens when you use the resulting file as a source.

So far it's actually turning out to be really helpful. I've been using it to get summaries regarding my work over the past 18 months or so and it's helped me to consolidate my thoughts on all sorts of issues and subjects.

It's not perfect, however. I've had it "hallucinate" some stuff when making things in the studio (most notably in the slide deck facility); for me though I find this an acceptable use of an LLM. I know the subject it's talking to me about and I know when it's making stuff up. This, in turn, makes for a useful lesson in how and when to not trust the output of a tool like this.

Having tested it out with a currently-active vault, I'm now interested to find out what it'll make of some of the archived vaults I have. Back in 2024 I wrote a couple or so tools for making vaults from other things and so I have a vault of a journal I kept in Journey for a number of years, a vault of a journal I'd kept before then in Evernote, and I also have a vault of all the tweets I wrote before I abandoned Twitter. I also have a vault that was dedicated to recording the daily events and thoughts of my time working at Textualize. It's going to be fun seeing what NotebookLM makes of each of those; especially the last one.

Anyway, if Obsidian is your thing, and if you are dabbling with or fancy dabbling with NotebookLM, perhaps obs2nlm will be handy for you.