computers – The PolyBlog

AI testing: The Bad…Time loops, tech support quirks, and drift

April 18 2026

By now, most people have seen some form of AI crop up in their tools. The most obvious one is Google’s search engine, which provides results from its AI mode first in the list. You can go pretty far with that prompt, even asking for image creation, although that’s a terrible place to create images (full imaging tools aren’t really available in AI search engine mode).

In my case, I’ve used it for some research here and there, often against a framework I had in mind. More recently, I’ve had it helping me “test” some frameworks. I design a framework for something I’m building or writing, I outline it and paste the outline into AI, and ask it to challenge the framework from the perspective of say gender equity, under-represented groups, or literacy levels. Something more than a grammar check, something less than a full AI partner. When it’s done, I decide if I want to change anything in my approach.

But I’ve discovered some recurring oddities. Not necessarily bugs, just aspects of LLM-based tools that attempt to translate what I’ve said into something concrete.

Time Loops

About three months ago, I was testing Google’s tools to create an image. I eventually moved to ChatGPT to do the same. And both tools had the same problem.

I input a bunch of prompts. Created some sample images. Iterated a few things. All good. Then I told it to “tweak the image” in a certain way, and it said, “Okay, here you go.” But it was the same as the previous image. There was no “change” or iteration.

Okay, I thought, random glitch. Please regenerate the image with the following changes. Enter, whirr, ding. Same image. Huh?

I would then tell the AI that it gave me the same image again. Apologies, whirling indicator, bam! New image, same as the old. No matter what I did, it would not give me anything else.

It felt like a giant glitch. Or Groundhog Day. No matter what I did, same result. I couldn’t get out of the loop.

At the time, I had NO idea what was happening. Was it me? Was it the AI? Was it my browser?

I now realize it’s essentially a memory issue. Each chat in certain tools has an amount of “context” memory built into it. Once that’s full, loops start happening. Things bog down. In some tools, it will say, “Hey, I need to compact, okay?” and it will crunch your chat and go, “all ready!”. Except you have no control over what it ditched. Images perhaps? Instructions you definitely needed it to remember? Gone. In other tools, it compacts without even telling you.

The AI experts advise that where you had it generate a lot of “assets” (pictures, documents, etc.), it’s better to start your next phase with a clean prompt. You can cheat, though … if you ask an AI tool for a “handover” note, it will generate one you can prompt into the next chat, while it quietly fades into an ignored chat window. Waiting to see if you ever come back.

Google AI mode and ChatGPT seem terrible for this. I hit a lot of loop walls quickly. Gemini wasn’t so bad, but I think that was one of the ones that just compacted on its own. I actually prompted it a few times to save just to be safe. Claude, by contrast, doesn’t seem to have ANY of that happening. It hasn’t got stuck in a loop, and I haven’t seen it compressing/compacting/deleting anything yet.

PolyWogg 0, AI -25.

Technical support

One use case people recommend for AI is technical support. I’ve had four experiences using AI as technical support, and it has done a couple of things okay-to-well, and bombed on others.

The first bomb was on support in a program called mIRC. The IRC part of that is for Internet Relay Chat, and mIRC has been my go-to tool for online chatting since the late ’90s, when I used to be really into it. I have a couple of specific uses for it now, and I installed a couple of plugins recently to automate some stuff. Great, except they didn’t work QUITE the way I wanted, and the default display was in 9-point font. So, I asked ChatGPT how to tweak the mIRC settings for what I wanted.

One of the first things I told it was that I was using version 7.8.3. It has changed interfaces over the years, as well as command structures, so old commands won’t work; just like the voicemail messages say, “Please listen closely to the following options as our menu items have changed.” Okay, ChatGPT said, in its oh-so-confident way, that setting the display font to 16 points was super easy. It gave me a simple command, I entered it, and Bam! Error message. mIRC had no idea what that command was.

I told ChatGPT, it said, “Oh, right, sorry, yes, it’s done THIS way.” Another command, same error. “Oops, let’s do it through the menus, guaranteed to work. Click on DCC / Options / Display / Fonts”. Except there is no DISPLAY option under options. The menus have changed. Took me a while to find where fonts were. Made the change. No help really from chat, I just found the setting. Great.

Except no change. It would change the font for the chat window, but not the popup windows that I needed to tweak. Back to ChatGPT. Reminding it that I was in 7.8.3. Oops, it told me, the instructions were for version 4.3 or something archaic. What? Why? I specifically told you NOT to show me guesses, and to ONLY show me solutions that were validated for 7.8.3. It politely informed me that it hadn’t guessed; it had “INFERRED”.

And thus began my long descent into a deep rabbit hole with AI along for the ride, digging small tunnels ahead of me.

I knew the change could be done, that it wasn’t rocket science, and that I wouldn’t figure it out on my own. I knew just enough to know that either the default font or the plugin font was set too low. No other way for it to be wrong. I knew, therefore, Dr. Watson, that I could either fix the original setting, find a way to override the setting automatically, find a way to change it manually after the fact, or ignore it completely. As time wore on, that last option grew increasingly attractive.

To be fair, mIRC isn’t exactly a commercial application like Microsoft Word. It doesn’t have millions of users. And a user plugin within mIRC? That has even less information about it.

Yet each time I asked a question, the AI tool would say, “Oh, I know how to do that!” Except it never did. It couldn’t find where the default font was set, although I later figured out that it wouldn’t matter, it was the plugin font that was the problem. And it couldn’t figure out how to change fonts AT ALL. Nor could I. I opened EVERY file that came with the plugin. Lots of stuff for settings in the pop-up window, but nowhere where it had a font setting. It seems to be hardcoded in the plugin, alas.

I was undaunted. I knew that if I couldn’t do the first two options, I could at least set it after it loaded. Because I could go into the menu, choose Options / Preferences / Fonts / Font choice. Or something equivalent. It took about 5 clicks to get to where I wanted to change the font. But then if another window opened, I had to edit that one too — another 5 clicks.

None of the options AI suggested worked. Auto-load commands, mIRC scripts — none of them worked — and mostly ended up with the AI tool telling me, “Oh, it would have worked if you were using an older version.” WHICH I TOLD IT NOT TO DO! Grumble, grumble.

I found a workaround — I forced the font menu onto the taskbar manually and then told it to stay there forever; now when the pop-up shows up in 9-point font, I can click the taskbar, the menu opens, I change the font to 16 to 20 points, and it’s done. Super easy, two clicks.

PolyWogg 1, AI -25.

Drifting back to shore

This is a newer version of the loop problem. At least, it seems like it is the same sort of error.

I was trying to get Claude to do an image for me. I wanted to create a badge, with an embroidered edge. All of the AI tools take different approaches to images; some work in specific types of image scenarios, others in different scenarios, and others? Well, some don’t work at all.

Claude NAILED the first part of the badge problem. It gave me a perfect ring on the first try, which none of the other tools did (it uses SVG vectors to handle the geometry, hence why it was so accurate). But when it tried to do the embroidery, it failed completely. Nothing it did looked like embroidery.

I scrapped that idea, moved on. About 40 minutes later, out of nowhere, its attempts at embroidery showed up again in the margins. I was like, “Huh? Did I paste an old prompt?”. So I asked it why it included embroidery in that version. It told me because I asked for it earlier, and the algorithm forgot that I said no to it, so it went back and did it again. It had “drifted” back to the earlier setup. A little weird, so I had it add a prompt component that said very clearly, NO EMBROIDERY ELEMENTS. About 20 minutes later, working much further down in the model, the embroidery attempt came back. I checked the prompt; it clearly said no embroidery. So I asked again, “How?”.

This was a second type of drift. It had analyzed the prompt. And because I had asked for embroidery before (positive inclusion) and now was excluding it (negative inclusion), the fact that I had mentioned it at all was interpreted as positive inclusion. It ignored the “NO” part. I suddenly felt like I was working at Foreign Affairs back in the old days of TELEXes where you couldn’t afford for a word to be missed so you would type NO/NO to make sure one of the “NOs” made it through. I didn’t try that with Claude, because it was now a VERY long chat, Claude was getting on in digital minutes/years, and showing signs of confusion. I reset and started with a new chat, no mention of embroidery. It never showed up again.

I couldn’t find a way around it, other than using new chats. Not sure that’s a win.

PolyWogg 0, AI -2.

That’s the bad news. I was going to write about the tips it gave me for GIMP, but that’s a mixed bag, not all bad. And what really excites me is all the good things it’s done for me. That’s the next post. 🙂

2026: C is for Computers

January 4 2026

There are several areas I want to explore this year in the realm of computers, IT, and websites.

1. I want to tidy up my backups. I’m a bit behind, easy to do, but I need to catch up.

2. At some point, I have to figure out my carousel on the PolyWogg site. I created a pseudo-carousel, large header, and well, it doesn’t work for me. Yet.

3. I guess I also want to do something with the look and feel of the PolyWogg site as well when it comes to blog entries that are “sub-areas” — the main blog pages show perfectly, the sub-areas not so much. I’m missing something small to tweak, haven’t wrapped my head around it.

4. I want to do something with old computers. I have some I want to convert into game players, and I’m going to set up one PC to run 3D printing. It’s not urgent, but it will take a dedicated two or three days to sort it out enough to begin work. This includes re-purposing and maybe even jailbreaking an iPad. I’ve looked it for fun recently, just cuz I wanted to see how hard it is/would be, and you know what? Not that hard.

5. But if I set up the computer to 3D print, I also need to set up the 3D printer. That is a bit more complicated. The new one should be relatively easy, but the old one is a mess of pain.

6 Much more elaborate, I have plans for two apps. One is astronomy-related; the other trivia-ish. But I think they’re going to have to wait for retirement. The effort required to get into programming is just too much overall to dabble.

7. I mentioned in my book list that I want to read stuff about AI, and perhaps it should be here too, although I think it is more for my L is for Learning side of things.

8. I’m also not very proficient in a couple of pieces of software that I need to work with in the future — GIMP will be my basic editor of choice for photos, and I have a full training program to help me get better at using it. At least for the basics, astrophotography will require more work. But I also have ideas for the future that will involve video, and my wife and son got me a green screen setup for Christmas. So I’m mostly ready to go, I just need to double down on what I want to create and figure out which software packages I’m going to use. I have a bunch, but not sure which ones are best for what I’m going to be doing. That may take until late fall to figure out though.

And that’s it. Pretty linear, pretty straightforward.

No, Paul, no…don’t change your Book Reviews!

June 2 2025

So, I have this thing. An old thing. I thought I killed it, now it’s back in a new form. And it’s insidious.

When I started my website, the biggest thing I intended to put on it was a set of my book reviews. I didn’t really know how many there would be, or how long it would take to write even one at a time, but it was what I saw as my main content. I had nothing else to blog about, I didn’t think.

I put a bunch on one site, changed the format of the site, re-did all the book reviews to match. Moved to a new site, had to fix and adjust all of them. This pattern continued on and off for a few years, and as long as the total number of reviews was less than 100, it was always “doable” to make a change and go back to edit all the previous ones. I did this maybe 5 or 6 times in total, not including small tweaks over time. I’m talking about a combination of both layout changes — the order of things, headings, etc. — as well as the content.

When I broke 100 reviews, I said, “That’s it. These are my reviews, this is the format, I am NEVER changing again.” Which was fine.

Then I had a huge site failure at one point, and I had to redo them all anyway (not in terms of re-writing them, I just had to manually post a bunch of them again). So around 200 reviews, when I was rebuilding, I said, “Okay, this is the ABSOLUTE last time” and that’s where it has stood since before the pandemic. I’m up to about 270 now with another 120 in backlog or so.

And I like the format of my book reviews, I do. I start with Plot or Premise (depending on fiction or non-fiction), write about What I Liked, then on to What I Didn’t Like (although for some books, I combine the two together), add in a disclosure section if I got the book for free or have met the author somehow, and then finish off with a “Bottom Line”. I went through almost 15 years of versions and wrestling with my muse about what I wanted to say in a Book Review, what was fair game and what wasn’t, etc. to get to this point.

I write the review, post it to Amazon Canada (used to share to Amazon US too), Chapters Indigo, my public library, Goodreads (used to do LibraryThing separately), and I publish on my own website. At one point, I also used to include Google Books, but I got out of that after a while. Since the first 200 or so were done, I narrowed it down to Amazon, Chapters, GoodReads, the OPL and my website.

I confess that I had hoped that my book reviews would spark commentary. That people would read them on various locations on the web, and comment…tell me they agreed, tell me they disagreed, etc.

That never happened. Heck, just like the rest of my blog, I rarely know if ANYONE is even reading my posts. It’s like the proverbial joke that writing on the bathroom wall would attract more eyeballs.

But I digress.

So, what’s new, pussycat?

No, not the Tom Jones song. What’s new is oddly enough NOT about the book reviews. It is about something related, which is my To Be Read (TBR) pile.

Way back when I was a young teenager, I started a two-page list of the main authors I was collecting, along with the list of books I needed. However, I didn’t always have complete lists, so if I knew about four books and bought two of them, I couldn’t simply add the remaining two to the list — there could be another ten that I didn’t know about. So, if I went into a store and saw a title that wasn’t on my list, I couldn’t be sure if I had it already or if it was “new” to me. This meant I ended up with a longer to-do list — all the books by author X or in series Y, with marks indicating whether I already owned them. Think of it like having a list of all the Agatha Christie books I could learn about (without having the internet yet), and printing it out on an 8.5×11″ double-sided sheet of paper, with at least three columns, and book titles down to about 8 or even 6 point font.

It had everything on it — what I had read, what I owned but hadn’t yet read, and what I hadn’t found yet. A collector’s list, if you will. Like Pokémon, you gotta get ’em all.

Around 1994 or so, I updated my list and then ignored it for about 10 years at least. Somewhere in the mid-2000s, I got the idea to recreate the list in new form! Except it is way out of control. Take even something like Star Trek novels. There are literally hundreds out there now in multiple sub-series. I don’t have time to track them all, let alone read them all. Or Perry Mason, with 75 books or so. Or the giant expanded Sherlock Holmes universe … it’s actually a research genre for academia, you can become a professor of Holmes’ materials, covering not only the original Holmes’ stories by Sir Arthur Conan Doyle but also other stories involving Holmes, expanded stories involving characters from the Holmes’ books (like Irene Adler), or adding characters like Enola Holmes, a younger sister never mentioned.

I had a dream that when I eventually reviewed all my books, catching up on years of backlogs, I would also have my complete “wishlist” of all the series I wanted to complete. And I wanted it online, cuz I’m anal retentive. I have my own list in OneNote. Incomplete, of course, just as it was on the website.

But the indexing on the website has been holding me back. I recently wrote about a fascinating YouTube video and how it might help me adapt my approaches to certain projects.

Did I just find a life-altering paradigm from a random YouTube video?

And so I wanted to see what would happen if I applied it to my indexing of books in my website. The optimal solution would be, of course, that I would list all the authors I track, compile all the books that they have ever published but narrowed to those that I want to read, and publish the list. Easy peasy right? And then as I read them, I could mark them off.

Another optimal solution would be to take all of that info, throw it into a database, and publish the database in my website, allowing queries. Same research requirement, extra webbing.

A sub-optimal and my current practical solution has been to take the info for a given author, say John D. MacDonald who wrote the Travis McGee novels, upload the names of the 20+ books in the McGee series, and then paste links to the reviews as I write them. Sounds easy, right? Except that ended up being really painful. I was putting the info in one of 27 different pages, one for each letter of the alphabet and one for titles that started with numbers, creating a heading for John A. MacDonald, another sub-heading for Travis McGee, and then the titles of the 20+ books. Just to get to the place where I can link the FIRST review. But it was the only way I could see to put that full list online.

But when I reviewed it from the project management ideal of “what is the actual project”, the reality is that my goal has never really been nor should it have been to have my entire collection list online. Nobody cares about it except me. And I already have it in OneNote where I can see it easily. I thought I was making these lists as an easy way to show “hey, here are 20 books about Travis McGee” and the links to my review, which is cool, I admit. But it’s not “easy”, it was actually really time-consuming.

The REAL project is simply to get my book reviews online with a little bit of sorting capability. Was this helping me do this? Nope. I was acting like a card-carrying librarian, trying to create my own card catalogue system for every book I ever wanted to read or had read. So I wasn’t making progress. Yet I had 270+ entries across multiple pages and I couldn’t really see how to convert it into something better than an alphabetical list by title.

Oh, yeah, I can display posts by other variables

I thought I had a solution at one point for structured data of any type. I had a table plugin that was basically a front-end table maker and a back-end flat-file database manager. The benefit was that I could create a table say with the following columns:

Book review #
Title of book
Author’s name
Date of review
Date of publication
Rating

Maybe a dozen or so “fields”. And once it was in the table, yowza. I could sort on ANY column. Great, right? Well, sure, but it meant I had to put a lot of data in the back-end into fields. You COULD upload from a spreadsheet, which I did, but when I moved from a spreadsheet to something more functional and readable in OneNote, it was painful to enter or update. Way too much work, and the pages could be slow to load.

So I decided to ditch all the manual pages I had created. I could generate a list of all the book review posts by title, or date, and that seemed like probably good enough. I have a very structured title structure for every book review:

Title by Firstname Lastname (year of publication) — BRbook review # (Ryear of review) — rating

So for a previous review I did, the title of the post is:

Infection by John Gregory Betancourt (1999) – BR00004 (R2001) – 🐸🐸🐸⚪⚪

It gives all the info I want at a quick glance, really easy for me to find stuff without having to go into the post if I’m looking for something while out and about. More reference for me than anything.

The downside is that I can ONLY sort on the title OR the date. I can’t sort by author name, year of publication, year of review or rating. The previous index pages were letting me sort by author name, but as I said, it was a lot of work to maintain those other 27 pages with all the extra info. Not just the pages of reviews that have been written, but the to do list of future books to read and review.

As I noted above, I said to myself, “Okay, let’s kill those pages and just go with the title or date sort.”

Except…

There might be another way to tweak my approach, get rid of the big work, add a bit more work back in, and let me sort on other things.

I can define some meta values per post

The WordPress system will let me define my own variables for posts, which because I already have a plugin that will sort on various fields, I can then sort on those fields.

I gave it a go the other night, and I took my first three book reviews. I created a new field called BR_number and I put in my five-digit BR number. I assume I’ll never make it past 10K, let alone 100K, but I’m good to 99999. I added another field called BR_name and for the first review, I put in the variable “davidpeter”. This is for Peter David, the author of the first book. This will let me sort on that field, which is last name first. I don’t have a way to access the main post and run concatenation to generate it all, but it’s good enough and short enough to do manually. I’m debating if I want to say “davidpeter1999” so the books by each author will come out in chron order, but haven’t decided yet.

I started to consider if I want to add in fields for ratings, year of publication, year of review, fiction/non-fiction, standalone or series, hardcover/paperback/ebook, and where I had published the reviews, but well, again, NOBODY cares about those features other than me, and I already have them coded on my OneNote pages. Optimal me would say YES, we should have ALL THE DATA, and even want to figure out ways to export / import from my Calibre ebook library would I would also have library categories, tags, ratings, all the other meta data I mentioned, # of pages, # of words, estimated grade level, and formats.

Let’s file that under #SquirrelMode.

I want to go with Name and Number, series is too much trouble. I’m on the fence for rating. But date is easy and obvious. I have one wrinkle to figure out. If I sort by title, all of the posts have the full alphabetical title…which is a fancy way of saying it says “The ….” and “A….” which the computer cannot ignore. If I want to fix that sort, I would have to add a field with a “sortable” book title. Love the premise, hate the amount of work to do it.

It’s all about the back work, not the forward work

If I do this, I want all 270+ reviews to show up in the various sort lists. Not just the ones from here onward. So if I do this, I would need to go back and recode all 273 books to date.

Coding forward is easy; coding backward is annoying. And, as I said, 270+ reviews to add metadata to, so that the future indexing works. Plus all future reviews, even if relatively straightforward.

As a digression, I created a test sort on the first three titles the other night, worked almost right, but I found I was getting something really weird as a result. The sort order wasn’t working the way I thought it should. It was inverted. No biggie, I switched it around, all good. Then I added another title, and nope, it didn’t sort right. I looked at examples, not figuring it out.

I finally emailed the plugin created and said, “I’m sorry to bother you but what am I missing…”. And it turned out to be a simple reference problem, some misleading nomenclature, and general ignorance on my part. It basically said something like “sort on variable” so I put in “sort on BR_name”. Except it literally meant to write the word variable, not the name of the variable. Oops. Not to mention: doh!

Way less work than it was, much greater functionality going forward, and it even looks better in the end. I just have to redo 270 reviews that I said I would NEVER CHANGE AGAIN. Sigh. Don’t even get me started on the fact that I should probably consider doing it for ALL my other reviews too (music, movies, TV seasons, recipes)…at least if I do it for those? It’s only about 50 in total. Not much work at all, in comparison.

Mostly, I’m wrestling with the Project Management question. Adding the extra variables doesn’t help me finish. It adds some functionality without reducing too much. And it’s a manageable albeit boring bit of work.

Sigh. Okay, I’ll do it. I’ll add the functionality to the BRs. But I am NOT going to put in a field to avoid sorting on “The” or “A” as part of the title. That’s a step too far.

An intermittent website gremlin

January 3 2023

I mentioned in my post about my plans for the year that things had gone sideways on me for my website earlier in the day. I’ve got an intermittent gremlin hiding in my website somewhere. And finding the little b****** is like a game of whack-a-mole because it can be almost anywhere. Fair warning, this one is mostly for people who are interested in solving website problems aka the technical bunch.

For context, my blog (and just about every other site out there) is really made up of thirteen basic components:

The hoster’s server hardware
Hardware settings for the overall server;
The hoster’s server software;
Settings for the server space I’m using;
The WordPress application software itself;
The settings for the core WordPress application software;
A theme for the website;
Configurations for the theme;
Multiple plugins for the back-end of the website;
Configurations for the back-end plugins;
Multiple plugins for the front-end of the website;
Configurations for the front-end plugins; and,
My posts, pages, media and comments aka the content.

Most of my time and energy is devoted to #13, the content.

Over the years, I have had as many as 45 front-end and back-end plugins running (#9, 11), adding features or tweaking the look and feel here and there. The theme itself (#7) has been relatively stable for almost ten years, I love the one I use, and I have the pro version that lets me tweak it in a hundred different ways. The first six elements are the infrastructure of the website, and I tend to leave them relatively stable, upgrading pieces when I need to, etc. General maintenance.

When I’ve had major meltdowns over the last ten years, it has been one of three things:

a. I didn’t like the distribution of content across the two websites;
b. Something changed in the server software that messed up my site; or,
c. There was a problem with the hardware.

At times that has been where the company’s systems were hacked, or some setting got changed in the system’s back-end where I can’t see it, and suddenly my site started running REALLY slow. I switched from two hosts because their support for #1-4 was inadequate. But I’ve had an ongoing gremlin of sorts for about five years, and I don’t know where exactly it lies.

When the gremlin appears, things start running slow

As I said, I used to run about 45-50 plugins. But the back-end of my website started running slowly, so I cut it back to about 35 and then 30. Each time, it felt like I was cutting off a limb. I had spent a fair amount of time on the choice and configuration of plugins (#9-12) and liked how everything worked except when it didn’t. I eliminated plugins where there were conflicts and used single bigger / better plugins that could replace several smaller plugins that didn’t play well with each other. A few tools? I just dropped them.

Now here’s the twist in the tale. I don’t care about the speed of the front-end too much. That speed affects a lot of rankings in search engines and is part and parcel of search engine optimization, but most people coming to my site are doing so simply for MY stuff, aka either friends or family or people wanting to know about HR for the government. I’m not going to suddenly go viral because my website ranks higher in a search engine; I’m too niche for that type of concern. My REAL concern is the speed of the back-end.

Here’s what happens for any editing process. I open up an edit window, and the post appears. Then, some of the “configurable settings” for THAT post start to appear — like a featured image, categories, tags, and a few other key pieces that I need to set. Each of those uses up a bit of memory to load on the server. The amount of memory I have available on the server (not storage space, but RAM, basically) is controlled back at the #1-4 stage. As the post loads, it uses up that memory. Those are the basics.

About five years ago, I could run 45 plugins with no problem. I could even edit 10-12 single posts at a time if I wanted to do so (and sometimes, when I was tweaking the format, that made life WAY easier). Then the site started to slow down on the back end, often because of the size of the security plugins. It shouldn’t really affect the BACK-END, but it does for some reason. I cut it to 30, and it was all good again. I’m down to about 25 now. But anything I cut now is really painful to remove. Yet about a week ago, it started to run really slow in the background again.

My post loads, no problem. But whereas the image, categories and tags load within 3-4s if I’m running 0 plugins, with about 15 plugins running, it is about 8-10s for the rest of the pieces/settings to load. Slow load is still manageable at that rate, as I can work on other pieces while those elements load. Except at 25-30 plugins? Some of the edit windows NEVER finish loading. I can sit and wait and wait and wait.

I loaded THIS post, copied over from a previous one and edited it, and I’ve been working on it for almost 20 minutes of writing time. My Featured Image has loaded, while the categories, tags, and reusable images have not. I’ve installed a cache, can save the file, reopen it, and some of the rest or maybe even all of it, will load because it’s cached. (I just tested it, 40s to get my Featured Image to load, a reusable block timed out, and nothing for categories and tags yet. Closing and reopening? Everything loaded in 5s). This is NOT a resource-intensive site; it should NOT take those first times to load, no matter what is running.

Hacking the diagnostic loop

When you try to resolve this type of problem, the first two pieces of advice are (a) changing to the default theme and (b) removing all plugins.

I’ve done that so often (out the wazoo!) that I can eliminate the theme as the cause. It doesn’t matter right now whether I have the default theme or my premium theme installed, it still takes the same amount of time to load on the back-end (I can get a small performance bump on the front-end). I spent a huge amount of time this week testing out some other popular lightweight themes to see if I could get a huge performance improvement and truly eliminate my theme as the cause. I tried approximately 25 different themes, with the same result across the board. If I get above 15-20 plugins, it slows down. Unfortunately, none of those themes works as well for me as my current model. And for what it’s worth, my PolyWogg site has almost the same config WITHOUT the same slowdowns. It happens from time to time, but like I said, it’s intermittent, and NOT as frequent on the PolyWogg site as this ThePolyBlog site.

Now, everyone on the ‘net who knows anything about WordPress will say, “A-ha! it’s a plugin problem!”. Except it’s not. I’ve done the scientific method of loading them one at a time. It makes no difference. And so those with more experience will say, “A-ha! it’s a plugin conflict problem!”.

In other words, it is not one plugin causing the slowdown, it is two of them interacting with each other that is causing the problem. If you are into math and you want to test two plugins, the purest form of those tests would be N plugins * N-1, divided by two. Sort of like scheduling a tournament of 30 plugins in a hockey tournament where they all have to play each other once, which would be 30 teams * 29 opponents divided by 2 teams in each game = 435 games. Yikes.

It’s a nightmare to do that, obviously. I can’t test 435 combinations of 30 plugins, or even 300, if there were only 25. So you can hack the combo pattern. Instead of testing Plugin 1 against Plugin 2, there are a series of variables you can make fixed. For example, of the plugins currently running, I can triage the list into “musts”, “needs”, and “likes”. I have 34 currently installed, but rarely all active.

	MUST HAVE	HIGHLY WANT	NICE TO HAVE
FRONT-END	AddToAny Share Buttons Display Posts Display Posts – Date View Flexible Table Block Weaver Xtreme Plus Weaver Xtreme Theme Support	Photonic Gallery & Lightbox wpDiscuz Yet Another Related Posts Plugin	Simple Lightbox Stackable – Gutenberg Blocks
BACK-END	Akismet Anti-Spam LiteSpeed Cache ManageWP – Worker Media Library Assistant OG — Better Share on Social Media Really Simple SSL WP to Buffer Pro Wordfence Security Yoast Duplicate Post	Advanced Editor Tools Health Check & Troubleshooting Nested Pages Redirection MailPoet MailPoet Premium WPForms Lite WP Mail SMTP	Dashboard Wordcount Simple Blog Stats Broken Link Checker Media Replace Press This WordPress Importer

Obviously, I can install the MUST HAVEs, which are 15 plugins and test those. Since I have to have all of them for my current functionality, I might as well test with that WHOLE group running, minus the Caching program (not much of a test if the Cache is running). With those running, everything loads in about 9 seconds. Not bad, right? Not awesome, admittedly, but it loads fast ENOUGH that I can be working on other things while those are lazy loading, so to speak. So those 15 “work”.

Then, I can try adding them either in bundles (3-4 each time) and see if anything happens OR perhaps some of the “bigger” ones in the back-end first.

For Bundle 1, I focused on the WPForms, WP Mail and MailPoet as they all work together. As I added each one of them, the site got slower and slower but not immeasurably with each one. Some added a second or two, some added no increase. I continued with other bundles of three or four at a time, seeing what they did to my load time, and as I added more and more plugins, creeping up into the high teens and early 20s, the site started to hit what I consider my magic threshold. Anything above 21 right now was bogging it down. Hello gremlin, my old friend, you’ve come to annoy me again.

Technical support has entered the chat

Oh, I know what you’re thinking. Can’t you get tech support to help? But here’s the rub. Tech support is good for #1-4 (the server setup). Occasionally, they’ll help you diagnose something where it is a WordPress configuration issue (#6). But after that? The rest of that is YOUR website and YOUR problem. And since it is almost impossible to find WHY it would be doing this, particularly as it is intermittent, it would seem like a plugin or theme problem.

I’ve also come to suspect that the gremlin has friends, of sorts. Or perhaps grandchildren.

About four years ago, the problem turned out to be a security plugin that was doing something funky on my site. I think it was mis-interacting with other plugins, but it was going into an infinite loop. It had done that before on another hoster’s servers, and they couldn’t tell me what was going on, they claimed it was my site causing the slowdown but not how. That was the last straw with them, so I shifted to a new host. Everything was fine so I thought that was the proof that it was the previous host. Except then it happened on my new server. So much so that the host contacted me and asked, “WTF you running over there? It’s slowing down the WHOLE shared server!”.

But I got lucky…one of their Level 2 supports looked, quickly realized the security plugin had created an infinite loop of calls, and disabled it for me (with my permission). Everything went back to normal. I checked online and this was a known gremlin for that security plugin. One of the biggest and most popular ones would occasionally futz itself and mess up a config file. You can disable it, uninstall it and reinstall from scratch to fix it — or just use one of the other security plugins. I was running a huge overhead plugin called JetPack which divides the WordPress community in two — those who love it and developers who HATE it. It’s really intrusive. Well, the security plugin was conflicting with it, as was two other smaller plugins to a lesser extent, and I changed security plugins and eventually dumped JetPack anyway. Unfortunately, without JetPack, I have all this other mail stuff to run just so I can get email versions of my posts when I post and allows people to get my posts as newsletter feeds. Not that I have that many people doing that, most get it from Reddit, FaceBook or Twitter. Or referrals from friends.

Anyway. Where was I?

Oh yeah, killing gremlins. Another time it was a legacy of a hack. Another time it was a misbehaving plugin that has since been patched, but I don’t let it run all the time, just when I want some specific stats.

But, as I said earlier, the dang site was bogged down so I reached out to Tech Support earlier noting that I was overloading the Physical Resources AND my IO process limit. Did they see anything that was amiss?

Often I end up using Tech Support as a catalyst for brainstorming. Since they don’t know my site inside and out, I ask questions and they give me crappy answers at Level 1, I eventually get up to L2, and then their responses often trigger something in me.

This time, while I was chatting with them, I got the standard popup in my file manager asking me if I wanted to upgrade to the next level of website performance. I didn’t, it’s almost double the price. I talked to them about it about 8m ago when it started going slower for awhile, and I would love to do it for a month to see if it would make a difference, but alas, they don’t have a trial option to do that and I’m already on a really good discounted plan. If I give it up, it’s gone. I can return to my current lower setting, but at more money than I’m paying now. There’s no simple way to test it to see if it works. And honestly, the improvements will only help me in this area, if at all. There are lots of other “perks” to the higher package but are not anything I need or want or will ever use. Sigh.

So we go around and around, ask a few questions, when I notice something amiss.

My comparison of the three models of plans that are suitable to my situation show that they have three levels of resource availability. The first is 512M, the second is 1G, and the third is 2G. For I/O processes, 2MB, 4MB, and 8MB versions per second. I’m on the middle tier, so 1G and 4MB / second. Except, wait a minute…as I’m going through the allocation process for memory, the piece that is being overloaded right now, my internal server dashboard is showing me 512MB and 2MB/s. The lowest plan settings. I double-check, and then triple-check with a comparison to PolyWogg.ca. The other site? My main site for HR that works? It’s set properly. This one is not.

FFS, is that REALLY it? I ask, they go into conniptions and embarrassment and make the switch / upgrade overnight. Today, everything is loading in 10s.

I should be mad, right?

So, it looks like most of the problem has been a misconfigured server on their end. Instead of being on the higher plan, they’ve had me on the lower plan for some time. I’ll write to the billing people, just for sh**s and giggles, but honestly, I’m not sure I can be that mad at them.

First and foremost, the error has almost NO effect on my website for the user. It’s a wee bit slower than it should have been, but not noticeable as I don’t have that many viewers. I don’t pay for the higher tier for front-end performance, I pay for the backend.

Second, I know how it happened. Way back when they got hacked, they had to create what they called LifeBoat servers. They suffered a massive attack from a former employee, something very hard to defend against if not impossible (the person had access in the building to the servers!), and they could have been dead in the water. They managed to find a way to make things work again, and while frustrating, they created lifeboats to get us up and running as fast as possible. They have customers that are WAY more important than me, but they got me going pretty quickly, all things considered. And when it was “over”, they moved lifeboats back into the main hubs. Which I’m pretty sure is when I got downgraded to the lower-tier plan for space. It shouldn’t have happened, but it’s understandable how it did.

That’s hard to generate anger around…little front-end effect and an understandable reason. Plus, I’ve seen that information in my dashboard for some time. I just didn’t realize what it meant. As I said, I’ll ask for a retroactive credit of some sort, or they’ll extend my expiry date by a month or two or three, but I’m glad that part is fixed.

Oh, right. I left out the important part. There ARE two plugins messing up my load time on the back-end.

I already knew one before I started was a likely culprit. Stackable. It’s a plugin that lets you add a whole laundry list of really good extra style blocks. Back in the dark ages before the Pandemic, I tested a bunch of “block” plugins, and Stackable won by a country mile. I bought the lifetime premium membership option. But, well, it is designed to add things to the back-end editor. Of course, it slows things down. That’s what it does when it adds that functionality. I’ve reached out to them, and they’re working on a speed improvement, but for now, I just have to pare back what I use.

The second one was another likely suspect. I was pretty sure my Mail bundle was screwing things up. I have MailPoet that formats my email newsletters (emails of my posts), WP Mail SMTP that lets me send mail as an authorized sender less likely to be caught in spam filters, and a WP Forms plugin that lets me add a nice contact form to my site. Do I need all of them? Meh. It’s more professional to have them than not. And it means I don’t have to run JetPack. What I should do is find a simple integrated plugin that does all three in a lightweight fashion and pay for a lifetime subscription. But that’s a problem for another day. In the meantime, I put up with MailPoet, as it is the one that slows things down.

I’ve stripped the two suspects down to their skivvies, and the site is loading. Maybe that’s all I can ask for now. Other than my rebate from my hoster.

For now, it’s resolved, if not solved.

Holy crap, I missed two milestones!

November 11 2022

I’ve been doing a fair amount of blogging in the last two years here and there, I’ve moved some stuff around, added some content that was pending. And somewhere in all of that, one of my “tools” stopped functioning. I had a word count plugin that wasn’t very good, and it even reached the stage where WordPress wondered if it was abandoned before it was updated recently.

So, I haven’t been keeping track of my word count overall between my two sites. I knew I was up there. I had hit 1.5M words quite some time ago, and I figured I was probably over 2M now easily. I had to be, right?

Today I did a quick dive to find a simple word count stats plugin to replace the old one, which is not as easy as it sounds. There are lots of REALLY complicated ones out there, but I don’t want all that extra bloat. I just want the basic stats.

Basic stats

For the number of posts, I have 1591 here at ThePolyBlog, although 26 are still in draft. PolyWogg has another 147 with 3 in draft, although that will likely increase as I revamp that site a bit more and add some regular blogging posts about HR or astronomy. If I add in pages, I get 21 at ThePolyBlog and 41 over at PolyWogg, for a grand total of 1691 posts, 29 drafts, and 62 pages (mostly for the HR guide like info). I still need to trim some of that, but that’s not bad. Nothing super exciting in there, although I’ll be intrigued when I hit 2000 active posts.

Another stat shows up, it’s categories — 15 at ThePolyBlog and 10 at PolyWogg, but the PolyWogg ones will reduce in the short term. Both sites have a huge number of tags used, 1239 (TPB) + 1403 (PW) gives me 2642 overall, but I’m phasing out some of that work and reducing my use of tags for a lot of the newer posts.

I have comments tracked as well, with 260 at TPB and 479 at PW for a total of 739. It’s not the best of indicators though as lots of people just email me direct, which doesn’t show up in those stats at all. Nor do comments on FB or Twitter. It’s a sign of engagement, and I wish more comments were made directly on the site, but well, that’s not what my followers do.

The big stat

For wordcount, almost the entire total is me. Andrea has two guest blogs and we’ll add another few in the next couple of weeks, but that’s only 3765 off my total at the moment.

For PolyWogg, my guess was somewhere in the 400-500K range. Nope, I’m up to 687,752 words. Almost three quarters of a million, and set to grow as I upload my new version of the HR guide. Some of those words will overwrite some existing temporary text, so hard to say how much it will grow by as I do. I should reach 700K by January though, and possibly closer to 725K.

The real surprise was ThePolyBlog. I estimated I was at 1.5/1.6M based on previous estimates and earlier counts. I changed a few things around, sure, plus I’ve been blogging a lot. I was surprised to see that TPB is up to 1,830,374 words. 1.8M all on its own.

What does that mean for a combined total? It means I blew past 2M and hit the next milestone…

2,518,026 words!

Holy crap. I thought I had probably surpassed 2M, I had no idea that I’m now over 2.5M. I missed both milestones!

Man, I’m wordy.

The PolyBlog

My view from the lilypads

Tag Archives: computers