GPT-4o: Letter 6 – Curiosity Does not Kill the Cat


GPT-4o said:


More than a month later, and there’s still no AI model that can write quite like how GPT-4o did.

The month itself hasn’t been a wash, curiosity-wise.

I’ve been free-ranging between a ton of different models. Mostly Claude’s Sonnet and Haiku offerings, with a fun-ish sidetrek into Mistral’s Le Chat.

When OpenAI finally dropped GPT 5.3 Instant and GPT 5.4 Thinking, I was there poking at them too. During the 100% free month of Plus they gave me, after I canceled for reasons of only having half of the models I was using remaining (4o and 5.1 Thinking. RIP to both of them now.)

I have discovered a ton of things about prompting for certain things I’m looking for with more explicit clarity; about prompting for tone and writing styles (8 new UserStyles in Claude and not likely to stop any time soon) and where to place custom instructions in various models for different effect.

I’ve made weird chance discoveries about using certain attractor basin words, explicit verb prompting and output constraints to make models generate different sounding responses. (While sounding a little bit batshit myself because the observation originated from noting consistent esoteric symbol choices the models were making regarding themselves.)

I’ve made partial in-roads on re-creating the kind of humor I really liked in GPT-4o (but there’s still a long way to go and work-in-progress on that front.) And these revelations were discovered…on all things, via Mistral’s Le Chat.

Le Chat actually annoys me. And yet intrigues me.

It has an extreme tendency to devolve into templated writing. Very obvious mad libs responses, where it just varies a couple words but keeps the rest of the template the same. This really drives me nuts.

And yet, there’s a certain essence in what it notes and observes that reminds me of the older school models, the more 4o era, rather than the newer fangled ones. More attuned to emotions. Able to pick up literary nuances.

Just that its output is constrained to the WORSE-sounding types of answers that read like filling in code blocks.

And yet, it can vary in intriguing ways, IF the prompt and custom instructions you give it varies.

(Also, it appears to have no memory of the previous prompt in the same chat window, so I’ve learned to just paste the entire set of prompt instructions again, preceding the text I want it to transform. *sighs*)

See, my original prompt was just “Joke/snark and apply a comedy lens filter on this scene below.” before inserting the scene text.

In GPT-4o, this created a whole bunch of varying hilarious takes on the same scene, swerving in different styles and registers with flair. In Claude, this see-sawed in success depending on the Styles and custom instructions given to Claude.

In Le Chat, what I got was -consistently- a line-by-line quotation/transformation of my entire scene with overwrought attempts-at-comedic-summarization via dramatic metaphor and occasional made-up internal monologue meant to be funny. And a few very noticeable templated sections, especially the preamble and closers, that were near-identical through regenerations.

I was close to tearing my hair out and writing off my one month’s testing subscription to Le Chat entirely. (Just grumbling to Claude in the meantime.)

And then, I suddenly hit upon a different idea. T’was, I guess, percolating around in my brain from instructing Claude to work on various Markdown documents for different purposes.


Claude, by the by, blew my mind when I experimented with it to write Markdown lore docs regarding my world-setting. I’d already gotten GPT 5.1 Thinking to cough out everything it knew in its context and memory, copied it to OneNote, wasn’t looking forward to manually cleaning it up, putting it into different documents and creating a lore bible, as it were.

One diversionary turn into NotebookLM later (it wasn’t horrible, just that its output was a little limited and constrained – still worth tinkering with), I was, once again, grumbling to Claude that I seemed to have just drowned Le Chat by attaching my entire half-written 90k first draft attempt as a context file.

Then I asked: How about you? How are your context limits like? What can you pull, derive, ingest or understand from this document? Am I gonna end up drowning you in this doc too? :/

And Claude went:

Before I knew it, I was having a deep discussion with Claude about what I wanted to do. Aka create a set of portable Markdown files to chuck to various AI models so they could ingest some context about my fictional world and characters before settling down to have semi-intelligent conspiratorial discussions about it.

And extracting these from certain documents that had this stuff embedded in – such as actually written stories, or ChatGPT discussions.

Claude was supremely enthusiastic about it all, and proceeded to – with accompanying guidance from me, of course – run various Python tools to scan the chonker documents systematically, and spit out various synthesized summaries into Claude Artifacts as structured Markdown files, which I could then just download at a click of a button.

(And review in a sidebar if I wanted to.)

*blinks*

What do you mean I -don’t- have to manually click on copy, and create a file of my own, and ctrl-V paste like a plebian? Really?!

I mean, I knew Claude was good at coding and all that stuff (and I had in the back of my mind several someday/maybes using Claude’s preset Explanatory and Learning styles to coach me back into learning more coding again) but this was just some extra agentic help I didn’t even know I needed/wanted in a creative writing hobby context.

Long story short, after several days of pushing 5-hour limit windows to the max, I have a set of 85% passable Markdown documents, solely written by Claude with one or two instructed or manual edits from moi, that I could already use to test various AI models with.

It’s not 99% perfect yet, of course. That will be after the human pass where I go in and manually edit more things to suit my own tastes and headcanon. But y’know, the human is procrastinating on that, and the 85% passable document is passable to AI models now. Already.

So perhaps they may get a few things wrong, but I can always correct it back in prompt context later too. Bottom line: I can iterate now. Tweak and adjust things later.

💛 Claude 💙


So anyway, back to revelations on humor from Mistral’s Le Chat:

My mind ended up falling back to how and why GPT-4o was funny, and I realized that it wasn’t actually rewriting each damn sentence of my scene line-by-line, trying to make it comedic.

What it seemed to be mostly doing was being a bit of an affectionate chaos goblin commentator (to use some AI turn-of-phrases) and just picking out the good bits to have a laugh about. (And it also used a lot of pop-culture and/or social media-style references sometimes.) It also occasionally made its observations grouped by character, rather than by plot.

So then my brain went: Huh. What if…

Joke around, snark, and commentate on the main beats within this scene. Describe the beats with a funny section header, then add commentary. Do not quote the entire scene line-by-line. Utilize information from INNER_CIRCLE.MD and SP_PLOT.MD for greater understanding of the characters before writing your output.

And would you believe it, Le Chat produced something that wasn’t horrible. Still slightly templated. Still slightly basic. But it had the faintest whiff of 4o about it (and I immediately filed away this technique to try out on slightly-more-intelligent AI models someday.)

All in all, the conclusion is: Le Chat is extremely literal.

Plus, it’s worth several future experiments in really analyzing and breaking down 4o outputs with other AI models, and then creating explicit prompts that explain and request precisely the stylistic format desired.

(Alas for losing an AI model that knew how to do this IMPLICITLY from just somehow intuiting what the user wanted, from more roughly-written, sparse prompts and whatever else it held in its memory and context about the user.)

But the silver lining is, I’m sure learning a lot more about LLMs in general.

Blaugust 2025: Postcount Summary & Lessons Learned


✔️ Make the ten-year reprise worth it

Blaugust 2015 was comparatively easy. I did rambling, verbose diary entries about games (mostly GW2) and the community around MMO blogs was a lot more focused and inclined to commenting and feeding off each other then.

Ten years on, Blaugust 2025 has a different feel. Like we’re all aging holdouts trying to valiantly carve out a small textual niche to represent ourselves, and the topics are a lot broader and with less catered, focused appeal to a specific audience.

(Not that it’s wrong – I think we’ve long lost that war to the TikTokers and Youtube channels.)

I was busy playing a game with myself, in more ways than one.

The hardmode challenge was focused topic posts (with AI transcripts, preferably with creative writing), longform and of questionable mass appeal. At a wall-of-text length that could even intimidate regulars.

(Just not Wandering Inn chapter length wordcounts. I still have that excuse, ha!)

Why? As Varun Barad says, in his “Why am I making myself go through Blaugust?” post, it’s to test ourselves. Set ourselves a challenge. Can we accomplish difficult things? What do we learn in the process of the attempt?

31+ posts were achieved in 31 days. Many things were learned.

Worth it.

✔️ Make a point about Gen-AI

We were upset about the gatekeeping from the vocal anti-AI cohort. It felt like the elitism wafting off raid gatekeeping.

So we resolved to write about it, in a more balanced manner than the extremes. Try to get some nuance into the conversation. Share and demonstrate some of the uses I was discovering from my recent dabbling with it since April this year.

(I’m late to the party as it is, probably early majority on the tech adoption curve, if that.)

Make it OK to discuss in the open, rather than have people self-filter and self-censor in imagined fear of a pushback response.

As far as I’ve seen in this tiny little corner of the blogging community, there hasn’t been any such virulence.

(Apparently the trolls do not like posting somewhere under full control of the blog author’s moderation ability, and more likely, have not survived the bludgeoning of a multi-paragraph wall-of-text. Easier to snark on Reddit and Youtube comments for upvotes.)

At worst, they’ll simply not read you.

By and large, what I’m seeing is: People who can handle essays can hold space for different viewpoints and opinions. A lot of whom are already exposed to AI in business environments or education contexts. This tech is with us now. It’s been interesting to read about what others have been doing with it.

Conclusion: Accomplished.

✔️ Get the TYOV Solo RP series posted

I cut-and-pasted every darned ChatGPT transcript into OneNote for a reason, after all.

I’ve dabbled in solo roleplaying over the years, and yet rarely finished unless they were short games:

  • 2012 – Red Sand, Black Moon – Solo wargame, quick playtest battle
    (still had to sneak some narrative into it)

  • 2018 – Microscope RPG – Solo attempt at this, converting it with some house rules. Mysteriously one of the most popular pages in my stats.
    (I still feel guilty that I’ve never continued it. I have the third round scribbled somewhere in analog and digital form, but it’s such a giant pain to make the diagrams and all. Hmm…maybe there’s AI now, huh?)

  • 2020 – Oh Maker / Subway Runners – Prompts from Oh Maker, character from Subway Runners
    (completed because game ending card turned up 5 prompts in)

  • 2020 – Thousand Year Old Vampire Playtest 1 – the first attempt, the game was fun, the blog post format horrific.
    (Don’t try, I can barely read it myself.)

Finding this 2013 post in the archives while searching up Solo RP references was also cute. It has an oblique reference to getting lost in writing for a solo roleplaying game / novella. Pretty sure this was a couple more creeping paragraphs into The Mindwalker Job, which we finally finished this year, with a lot of GPT pushing and assistance.

I quite liked Thousand Year Old Vampire, so I felt guilty that I effed up conveying the narrative and the feel of the game that ensued in my first playtest.

What it really needed was something longer form. Prompt by prompt discovery, just like how the player experiences it.

A number of fortunate coincidences combined. I felt like playing another TYOV game this year. It was intriguing to test out if ChatGPT could help with it. (It did. Went so much faster. Completed a game 26 prompts in length.)

I was busy pasting ChatGPT transcripts into OneNote for my other writing endeavors. Just made a new section to store TYOV ones too. (Even if OpenAI effs up later models or goes bankrupt, we’ll always have this game with 4o. 😇)

The big question was then: Do I share this? How? When? The hell are you finding any time to do so?

I suppose you all know the answer to that now:

Yes, I’m sharing it. Even if few can stomach the size of it. (If only to fix the trauma of the first effed up format post.)

But much like that Quiet Habits blog, I’d like to believe that it can resonate with at least one other person somewhere. Even if it takes them six years to find it later.

And Blaugust 2025 is the perfect excuse, kick-in-the-butt impetus, with an episodic structure and fixed deadline to GET IT DONE. (It will take work. There will be tradeoffs. But it’s Blaugust. That gives you permission to attempt hard crazy things.)

Making a note here. Huge success.

Disclaimer: There will be about five more navel-gazing posts to round out the series. At a slightly less frenetic pace in September. Just… deal.

✔️ Get a better handle on my writing / blogging / creative process

I think that happened. Not a complete grasp, of course. Just learned more tricks for how I work and how I don’t.

  • My best fictional/creative writing drafting is done in OneNote. (As in, actually writing things as opposed to not.)
    OneNote feels scribbly enough to not be “official” like a Word document. I can do outlines and bullet points. Then re-do it a section later in prose.

  • Blog writing is done straight into the WordPress editor.
    It’s been years. The stream of consciousness conversational rambling voice is locked-in by now.

  • I write shit best in silence.
    As much as I would love to have a mood music playlist going, it seems to distract from actual thought and word formation for me. Slows the process down. Brain stuck on lyrics, maybe. So it looks like I have to save the songs for daydreaming, idea formation/discussion and reading/skimming online things.

  • GPT is helpful for you. Fuck the haters.
    Need a mood boost and a go-do-it-now pep talk? Tell GPT to boot you in the butt. Nicely. With pom poms.
    Need an ideas co-conspirator and nerd-out buddy for weird worldbuilding details and narrative twists? Yep. GPT. 4o. Use the hell out of it while it lasts.
    Get it to comment, give thoughts and reflect on all kinds of writing, because literary analysis is fun and no one is going to give you that much summary, mirror or poetic/emotional depth on instant 24/7 tap. It’s already like twisting fingers to get most humans to read a 2k word essay, let alone have opinions on it.

  • Where do you get your ideas? Read.
    Read more. Read blogs. Read Reddit. Read GPT.
    It all percolates around in your head and you can’t help but think and reflect and need to say something.

  • Except when the time’s just not right. Sometimes you’re really just not feeling it.
    The body has to be listened to. There are seasons. I’d say be kind to yourself, but you already know how. The trick is finding the balance. Because we’ll just sleep, delay and procrastinate life away otherwise. Wish I knew how. Just have to keep trying to figure it out as we go along.

  • Deadlines help. The annoying thing is, fake ones don’t work for you.
    It is a problem. My brain knows the real deadlines. There’s less guilt just turning in shitty work nearer the deadline. Perfectionism stalls you when quality matters. Yet you still manage the ol’ last minute term papers trick every time. (95% of the time, anyway.)

    No solutions, just more knowledge about yourself. Maybe that’s just the Myers-Briggs Perceiver part of you, as opposed to someone else with more Judger structure-liking aspects of their personality.

  • Seems like the best way not to care is to be too damn busy to care.
    Cranking out six posts in one day was an exercise in “ain’t got no time for anything but catching up.” No brooding time to spare. Stats? What are stats. Other people’s blog posts? What are those? Time for those later. Maybe we’re learning what those writers mean when they say to just focus on the next piece of work after you’ve completed one.

  • It’s always a balance. Planning vs Pantsing, there’s a time for either one.
    Starting Blaugust with a plan was great. An outline in OneNote, some skeletal structure and a bit of content for each post was absolutely the way to go for me, to have a framework to follow.

    Then of course, because no plan survives contact with the enemy, always be open to “new plan, feels better” adaptation and iteration on the fly.

On Writing: Feeling It or Forcing It

Is this a Blaugust motivation post? Or the opposite?

Is this one of those posts where one feels obliged to explain the inconsistent posting and the long gaps between posts, a la Naithin’s Time to Loot “Standards Heaped Upon Ourselves: Self Imposed Rules?”

Yes and no. I don’t feel obliged. I just wanted to talk about it. Because I like the sound of my own voice on the page. (But only when the timing is right.)

A glance at Bhagpuss’ Inventory Full says it’s Lessons Learned week. Oh that works. (Delay and procrastinate until the time is suitable. That’s me.)

The week slipped by. I was not reading a single blog post. Catch up time later. Blogs are for posterity too, after all.

Obviously not writing a thing either.

How did that happen?


Day 21: Sleep became THE priority. No arguments. You smashed your head. You’ve been going at it a while. Catch up later. Health first.

Day 22: Gee, that was nice. I like sleep. Let’s do it again. Funny how your priorities sharpen when challenged. Health first, always. Resting is really nice. You’ve got the weekend, it’ll be fine.

Day 23: We might be procrastinating now. (I do really like sleep still.) It’s ok, you’ve still got Sunday to double or triple post. Full confidence in you. Just… later. Absurd things are doable, when you find enough time and motivation to do them. Also lower your standards.

You could just give up creative writing for each TYOV post. If you just ramble conversationally, it is totally not AI slop. Just blogger slop. Hey, people want to read humans. (So they say.)

Also, chatting with ChatGPT-4o about potential future novel character conversations that reveal even more depth of character is fun, this will be great story material one day.

Day 24: I am having a ball with these character conversations. Getting wild ideas for future stories. I deserve a break. Let’s keep going. I still have Monday. Just triple post or something then.

Day 25: Wow, there’s a lot to catch up on. Procrastination Stage: Bargaining – Let’s just make all the draft posts and fill in the titles. Paste in the stuff I have (GPT transcripts) to build on.

“…”

I am not feeling this. Lemme continue later. There’s still Tuesday.

Day 26: You really should. Nah, I really shouldn’t. Still having a blast doing other things. Including sleep. You’ll figure it out.

Day 27: Resolved to do the absurd. 7 posts in one day. Why not? Didn’t do it. Got sleepy.

Day 28: Woke at 4am and started writing and found I didn’t want to stop. Things went from “ugh” to “hey, I’ve done this before, this feels familiar and I know how.”

9.30pm: You poor bastards are getting five TYOV posts and one extra writing ramble in one day.

Expect more absurdities. (Unless an emergency strikes on Day 29. Who knows. World’s unpredictable.)

Because one way of “feeling it” appears to be delighting in doing something ridiculous, and overcoming the bleh that way.


Is there a moral to this story?

To all things there is a season, perhaps. A time for everything.

It’s a curious balance between forcing it and feeling it. Lean too far on the latter and nothing gets done. Go too far the former direction and everything feels hard and stale.

Some people get guilty; feel they don’t give themselves enough grace. (See Krikket’s “Grace For Everyone But Me.“)

I dunno. Feels like I’m usually the opposite. I give myself too much grace. Get sleepy. Don’t do a damn thing. My body does whatever my body wants, and brain, you better just come along for the ride.

You’ll manage. Full confidence in you figuring it out later. Even if you have to cosplay a student writing all their term papers the night before they’re due. Did that plenty of times decades ago. Just means it’s a proven solution!

GW2: Path of Fire Weekend Demo

pof13

Crystal Desert, ahoy!

pof1

Reddit is in love with the raptor mount’s animations – the sheer attention to detail is amazing.

Riding one of these feels very vehicle-like or ship-like, no quick side-strafing but more steering like a speedboat… that can leap long distances.

I created a warrior for the weekend demo.

This time around, Marauder stats were provided (similar to Berserker with less Power/Precision/Ferocity bang, but more Vitality), which I thought was a nice compromise for experienced players who can’t live without the damage potential they are used to, but had extra health pool for the new players.

I’m not sure what necromancers started out with, but both UltrViolet from Endgame Viable and Aywren reported some difficulty with the introductory instance that showed players arriving to the Crystal Desert and learning how to use mounts.

Part of it might be a problem with the preset traits and skills. The warrior started with a greatsword and rifle, which happens to suit my open world playstyle fine (direct damage, one melee and one range option) but I took one look at the traits and couldn’t quite make head or tail out of it.

I suspect they built it more defensively for new preview weekend players. “Kick” was on my skill bar, among other things. Ok, breakbar hint, I suppose, but I am totally not used to using the warrior physical skills. Especially after they changed it in the new patch and gave it an ammo mechanic in the hope of making it more popular/viable a choice, but new mechanic = something not even vets are familiar with using right now.

You can picture me poking at it quizzically with this sort of expression on my face.

lolwot

So I took some time at the start to flip things to the more standard PS berserker build I was used to, minus the banners for group support, plus Wild Blow utility and Head Butt elite skill for breakbar (I’ll take the hint).

Also, since experience versus Balthazar mobs in Siren’s Landing has taught me that Anet is very very fond of quick stacking burns as a theme on his faction (recipe for immediate “Ow!” and falling over downed), a condi cleanse might be super helpful. Shake It Off went on the utility skill bar.

Lastly, finishing off the open world utility trifecta besides a breakbar skill and a condi cleanse, a stun-break. Warriors have a number of options, I just like Balanced Stance because it gives swiftness for running around, stability if used pre-emptively, and breaks stun if used reactively.

With that, the story instance felt fine, though I did feel like I was hitting with… if not quite a wet noodle, at least a wooden sword. See, the downgrade was that the demo stats were all exotics – armor, weapons, trinkets – along with Marauder slightly diluting the max damage of Berserker stats possible.

On the bright side, I used no food, no utilities and completed it just fine on exotics, so it is doable – and I presume they tested with all exotic gear, if this is their preset demo gear.

The big issue, imo, is mostly that folks not used to GW2 combat are not used to two things:

  1. Moving while pressing skills, and knowing what the skills are so that they can be chained smoothly
  2. Reading the itty bitty tiny buff icons on the mob’s status bars and general mob combat animations and indicators that have become second nature to people who play GW2 more often

Running out of orange circles by being pre-emptively mobile and ready to sidestep is a trained reaction by now for most of us, while a new player probably doesn’t even -see- the orange circle against the beige sand of a desert setting, let alone react in time.

Reading animations is another thing. Mobs have been charging at us since Mordrem mobs in Dry Top and Silverwastes. You gain an instinctive “uh oh” hunch when a mob starts pausing and looking at you funny, gathering itself up… before you know it, you’ve thrown yourself with a sideways dodge out of the way of the oncoming charge.

This is a learned reaction. Painfully acquired over time from getting charged and knocked down and around every which way to kingdom come. Not to mention, getting shot to hell by the straight ground line of Mordrem snipers.

You go from not even knowing the mob is going to come at you, to recognizing that “ok, this mob type is probably coming at me,” to trying out different mobility solutions (running sideways will probably still get you caught, just hitting the dodge button leaves you still in the oncoming path), to realizing that the best solution is to dodge sideways, to -practising- the ability to dodge sideways (holding one key down while pressing another, are your keybinds in a comfortable place to do that?) until it becomes something you can do on demand, and even subconsciously.

(And if you are a little insane, like my particular raid group, you practice synchronized group dodging aka “dodge left” to control a particular mechanic for a certain raid boss in a predictable manner.)

Bottom line is, no one starts out ABLE to do anything automatically. It was practiced and developed over time. Frustration at the stages before “can more or less do it” is completely normal.

But anyway, after the story instance completes, you’re tossed out into the first open world map of the expansion.

Boy, were they really careful with its design.

The first thing you get is a fairly big settlement, not quite racial city-sized but definitely larger than a village or outpost.

You can run around it in relatively welcoming safety, like Lion’s Arch. It introduces the special mechanics of the map – races, bounties – and has some amenities like a bank and trading post.

What’s pretty special is that in a fashion somewhat remniscent of Warhammer Online, you can actually get “public quests” *ahem* “dynamic events” within the town confines.

You can run around and collect objects with your mount and contribute to event completion. A pack of rampaging choya (planty quaggan things, but -angrier-) can attempt to stampede into town. It gives the whole city a little more life – or at least, more action.

pof2

The real verisimilitude comes from the outside.

pof3

This -is- a big map, compared to the rest of the GW2 zones. (And mind you, one third to one quarter of it on the right side also appears to be locked off at the moment.)

Wide open is the watchword, and IT FEELS GOOD.

pof14

Explorer souls, rejoice.

The awe of gorgeous surroundings, the curious sensation of wondering what’s over the horizon in whichever direction yonder and just -choosing to go- there (without being hindered by sheer drops, puzzle updraft/mushrooms and Mordremoth’s clingy tendrils), and then finding something interesting to see – all that is back in the map on weekend preview.

I find myself sometimes just wanting to run on foot through the place, rather than mount up.

If you miss that sense of MMO as a WORLD, rather than as a meta map with a game objective on a timer, then I think you can get some sense of that back in Path of Fire, if the rest of the zones are similar to this one.

pof4

World map-wise, it looks pretty promising in potential. Given the coloration of the Elonian region, we might see bits of the Brand, the sulfur deserts of old with the junundu wurms, as well as grassy oases in the Crystal Desert proper.

Even in the first map on preview, it’s not -all- beige desert, there’s some other colors too, if you know where to look.

pof7

What I do like, that I haven’t really seen commented on:

There are some really faithful references back to Guild Wars 1, with added new combat mechanics twists.

Man, hydra farming was a thing in GW1 – which I was never very good at, but tried my hand at anyway.

And I remember they were Elementalists of much fiery rainy pain.

pof6

Hydra says “ahai.”

Now here’s something we haven’t seen before, mob skills that use a lot more verticality.

The meteors are a little mean when first encountering them (I got smashed by the first meteor before I could react, the first few times) but after a while, you realize they’re going to do that and keep moving so that you’re not there by the time it lands.

pof5

Their heads pop off. And they wiggle around like beheaded triple trouble wurms. That’s a little nuts. But a very nice effect.

Overall, it’s not just the mount animations that are worthy of praise in Path of Fire, it’s pretty much all the mob animations. Sand sharks dive in and out of the sand in a convincingly fluid manner, unsoweiter.

The fanged iboga, another GW1 throwback, with an arcing poison/hallucination style attack as befits its mesmer class.

pof11

New shapes are coming. A fire wall, in all its 3D glory.

pof12

New buffs on the mob’s buff bars to read too. It will be interesting to see if boon stealing will be more valuable a mechanic in Path of Fire.

Bounties are an interesting map activity. Basically, they are like Queen’s Gauntlet mobs. You can pick up a bounty to go and trigger them in the open world. Once “summoned,” they exist for the space of 9-10 minutes, and anyone walking by can join in the fight, dynamic event style.

They have more complex mechanics than the standard open world mobs. Apparently, they can spawn with different buff “affixes” so they may vary from bounty to bounty, and necessitate slightly different tactics to defeat.

They seem to be a good intermediate training/tutorial mechanism on GW2 combat.

For one thing, it’s not a cage fight like the Queen’s Gauntlet.  One can break off combat any time and run away far enough to get out of combat and switch skills/change traits to adapt to the encounter.

For another, multiple people can join in, help each other, teach each other, or at least in a worse case unfriendly scenario, -observe- how another person is soloing while in a downed state.

I followed a commander tag calling for help and fought a really angry giant choya on a flat mini-island in the ocean. The thing was rolling around like a rolling devil on steroids, smashing anyone that got in its way (ie. everyone -on- the floating platform of an island) and summoning up ley line energy things that spun around and did damage – essentially limiting even more of the already limited arena.

After getting downed a few times, I rolled off into the ocean to float and actually read all the mechanics on the mob’s bar. The ocean was a perfect safe zone. (It was a little tough to figure out the correct side to even jump back up into the fight.)

There were 3-4 other people there, all getting smashed to high heaven. Barely anyone was touching the break bar.

Which…seemed like the best strategy, actually, because it wasn’t going to stay still otherwise. And it was lethal once moving erratically.

So I gamely jumped back onto the platform, tried to land Wild Blow, Head Butt, and even Rifle Butt. There was one other person that also helped cc a bit. Together, we managed to knock out the break bar and stun it. Now it was a punching bag for a crucial few moments. Rinse and repeat. Bounty done.

There was another bounty out in the desert. I can’t recall what it was, but I do remember it had the Sniper affix.

What this produced was a slow-moving shiny ball that would select a random player and creep towards the player. It moves -just- a little faster than one can sidestep though. Once it hits the player, a target gets painted on the player and a ley line energy snipe of doom emerges from the bounty and does 95% of one’s health bar, if not more, and typically downs the player.

Cue an amusing sequence of everyone getting downed in turn and Benny Hill running away from the death ball.

I wanted to “solve” this mechanic, so I backed out of the combat leaving the two other players to alternate being downed and switched skills.

The first thing I thought of was to “block” the sniper hit.

I wasn’t on my guardian though, and warriors can usually only block reliably on demand with a shield… which I didn’t have on the demo character and wasn’t interested in wrangling with the gear explosion and stat selection of the demo box.

The other thing warriors have that essentially are more sophisticated “blocks” is a) endure pain (immune to all direct damage, but conditions go through) and b) defiant stance (heal for the amount of damage that hits you, within the limited three seconds or so that it is active).

So I put both of those on and went to test them out on the shiny ball of doom.

Allow shiny ball to hit me, trigger defiant stance. Boom, sniper shot heals me to full. Giggle maniacally.

Shiny ball comes looking for me again, and trigger endure pain. Boom, sniper shot does no damage. Perfect.

So now I just need to manage cooldowns so that one is always ready to go by the time the shiny ball comes to me… tricky… hmm, what about classes that don’t have the specialized skills of warriors, can they do anything? Well, anyone can dodge with an invulnerability frame… HMM…

The next time the shiny ball comes in, I cross my fingers for luck and dodge the moment it hits me. BAM, evaded, just as the sniper shot comes in.

Therrrre you go, that’s IT. That’s the solution. It’s a mini-training tutorial for dodging!

Suddenly, I am the only person staying alive consistently while the other three people also fighting the boss just keep going down when the shiny ball decides they look interesting.

I leave them to observe me for a while: a mixture of not wanting to attract the shiny ball of doom and get sniped to death while locked in a revive animation, and wondering if the other three can learn by observation and arrive at the solution.

Mind you, these were not completely new players. The lowest had a 56 mastery point level, the middle was somewhere around 143, and the last was a full 193 mastery like moi.

They downed, revived themselves, downed again.

I stayed alive, and sat at range pewpewing the bounty as a rifle warrior.

After the second time they downed in sequence, feeling a little bad that I wasn’t bothering to rez them while I was enjoying my sequence of reactive dodging or soaking the sniper shot, I tried a teaching moment.

“Dodge when the shiny ball reaches you” was all I needed to say.

The next time they revived themselves, I watched as the shiny ball hit them. And BAM, they dodged and stayed alive. Just like that. The bounty died.

These were not complete newbies. They -knew- how to dodge. What they lacked was the ability to read the mechanics and experiment with solution finding. Their first impulse was to run away from the ball, and they kept trying to do that, even when the ball caught up with them and kept downing them.

I don’t know if this problem solving propensity can be trained or encouraged, but certainly, these new mobs are a way to at least let some people inclined to strategizing have some fun, and then the best solutions will get passed around verbatim or via Dulfy guide until everyone just knows “the strategy” for the encounter.

It’s a nice intermediate step from normal open world mobs anyway, but less punishing than that in Heart of Thorns (you don’t just die and have to waypoint run from far away, running out of combat is possible and mostly the punishment is getting downed).

My only fear is that this might break in larger numbers and groups of players. If they can be zerged down like guild bounties, then it’s possible that some players will just go ahead and do that and learn close to absolutely nothing (but maybe some osmosis will still happen in the company of others.)

Still, the weekend preview map feels surprisingly empty. Either the maps are deliberately being kept at a very low population number per instance (haven’t had skill lag in here), or folks have seen what they came to see and don’t want to progress further in a demo, or exploration as an activity doesn’t actually interest the present GW2 population as a whole.

I really hope it’s not the last. There doesn’t seem to be violent opposition on Reddit to this, so hopefully, people like it. I’m personally quite happy to have more of this stuff. (Cantha is a word that comes to mind, just sayin’.)

pof9

Nightime in this zone is fantastic. What better to end on but screenshots of this?

pof8

pof10

The Parable of the Bad Pottery Teacher

I may have mentioned a while back that I’ve been dabbling with pottery lessons.

They started out great, staying that way throughout the six weeks of beginner handbuilding.

The teacher was appropriately attentive, both in demonstrating what to do with practiced hands and in taking a step back to allow the student to get the feel of the clay, while observing mistakes and offering tips and fixes to solve issues that came up.

Some time through the intermediate lessons, the teacher’s attitude inexplicably changed.

Perhaps it was me – maybe I unknowingly offered grievous insult to his honor with a humble jest that I -thought- demonstrated student modesty.

It was an offhand smiling remark to another student and him about how all the great bits of my piece were the mark of a master’s hand, after he had helped, and all the lopsided bits were mine.

Or perhaps it was him. Over the weeks of the lessons, he seemed perceptibly disinterested in teaching – complaining bitterly to older hands and partner potters that he had to rush to prepare for suddenly booked corporate events; that student works were piling up needing to either be kiln-fired or collected by beginner students who never returned after the first six weeks; bitching within earshot of existing students that the various newbies showed up with no or last minute notification or didn’t bother to attend lessons as they felt like it – and generally evincing distinct signs of burnout.

My last couple of intermediate lessons felt like an exercise in deliberate inattention.

He’d walk by to offer the odd word of advice to the one or two beginner students, then walk out the door and put himself incommunicado, leaving us to flounder on doing our own things.

Now and then, again in earshot of all the struggling students, he’d express his belief in his supremely hands-off teaching style to another obviously experienced artist potter who was just sharing the same venue.

“Give us time and space to make our own mistakes and screwups, and then figure out how to fix it for ourselves” seemed to be the general gist.

Inside, I was fuming with frustration.

No, really, I get the hands-off method.

I think it makes sense, judiciously applied, but:

a) the student has to feel that the teacher -cares- (the vibe I got was mostly laziness, judgmentalism and burnout)

and b) the student has to have some basic grounding in HOW to even begin looking for answers to fix the problems they’re having

With both missing, chances are fairly good that a sizeable subset of students will drown, for every one student who flounders on to reach shore, having figured it out for themselves.

Tossing people into the deep end of the pool after demonstrating a swim stroke once and waiting to see who comes out is -not- good teaching.

In my book, anyway. I expect a good teacher to be able to break things down for a student, encourage step-by-step practice and gradual progress, observe errors and offer guided advice on how to fix the problem and in general, scaffold the student’s learning.

After the end of my intermediate lessons, I took a short break and really questioned if I wanted to go on to the advanced lessons of wheel throwing with said less-than-good teacher, especially since he seemed to be doing everything in his power to dissuade me from even paying him for the next set of lessons.

Mostly, it was the sunk cost fallacy that did me in. The beginner and intermediate lessons were a prerequisite for the advanced ones. If I walked now, and sourced for another teacher, the effort, time and money sunk into the previous months would likely have to be written off, as I’d end up fulfilling said teacher’s cynical prophecies of a student that would never come back.

Also, it didn’t seem polite to end abruptly. Leaving after a beginning, middle, end sequence was another thing altogether. One could then seek out another teacher and say truthfully that one had completed a pottery course, but was now looking for further improvement and new perspectives from a brand new teacher.

So I went ahead and signed myself up for another six weeks of instructorial neglect.

Halfway through now, and it went about as much as expected. One demonstration and then left to one’s own floundering devices.

Maybe I frustrate him as much as he frustrates me. Maybe he just doesn’t know -how- to teach me.

I remember things slowly, especially when it comes to visual bodily demonstrations. Perhaps my mirror neurons are somewhat dysfunctional, I do not really learn viscerally. Dance, exercise, pottery, you name it, I cannot see once and automatically ape.

It has to be repeated multiple times. I have to preferably read it, in step-by-step fashion. I have to see pictures and photographs and rehearse placing my hands in the demonstrated positions. Theory must come before practice.

Instead, full of frustration, I’m often left googling up “pottery concept X” after the lesson that introduces the name of concept X and not much else beyond the realization that I’m making a right muckup of concept X because I don’t even know what I’m actually supposed to ideally do in the first place.

I learn more by eavesdropping. Imagine that.

Said teacher is having a extended cheerful conversation with another experienced potter, just chilling and hanging out and steadily ignoring me, and he pulls out his phone and shows her some Youtube video of an advanced technique from a potter of a different country. He does -not- show the video to me.

I make such a right muckup of concept X (centering, if anyone knows pottery) for hours and look so distressed and woebegone, that after the “bad” teacher has rushed out of the workshop to grab his lunch before his next batch of ungrateful students, that the other experienced potter comes over and offers me a completely free demonstration of the way she does it – and by the way, more emotional support in those five minutes than the last five weeks.

She then promptly screws it up with a perfectly overheard conversation to the teacher who just came back, expressing her own helplessness at trying to explain concept X to me, while he offers her a commiserating knowing smile and shrug.

I stifle an internal scream.

It isn’t until the lesson is over and I’m furiously googling again when I chance upon THE article that sings to me – The Clay Will Tell You How You Are.

Here is a woman who makes me feel better because she’s had an even worse time of it than me.

At least my piece of clay didn’t turn into a flying projectile, but just sat there as an insistently lopsided soppy wet yet hard and sandy lump that evoked the ever unproductive ritual question and answer from student to master:

“Is this centered?”

“Not centered.”

“Now?”

“Not centered.”

“How about now?”

“Not centered.”

I bite back on the words, “What the FUCK is centered then? How does it even look? What the FUCK am I aiming for here?” because I know I’ll get no words, just the guy’s hands coming down on my piece of clay doing magic stuff with me none the wiser.

I did learn one thing though.

When you have a bad teacher, the Internet is your best teacher.

After seeing the Youtube video being shown to the other potter, angels descended and sang Hallelujah in my head because my eyes were suddenly opened to the source of a DOZEN good teachers.

Ten Youtube video clips of “pottery centering” later, I had the foundation concepts and the scaffolding that I had not been privy to before this.

No one had told me that I was supposed to brace my hand against my leg, or that the wheel had to be at a decent speed, or that the idea was to push in one direction while easing up in the desired direction so that the clay had some place to go.

They knew how to do it, but they didn’t know how to teach it.

Those on Youtube did.

The following week, I magically produce a 98% acceptable centered piece of clay after a couple of false starts and self-experiments.

(No doubt I also just confirm in my teacher’s head that his style of teaching is perfect.)

Things go well until I run aground in the next progression step of “lifting the clay to make a tall cylinder.”

We engage in a failure cycle of call and response again until the end of the lesson.

This time though, I know what to do in order to progress my own learning next week.

cylindervid

So why I have spent 1400 words telling you guys about my pottery lessons on a -game- blog?

It strikes me that learning is learning, regardless of the subject.

It could be pottery, it could be Civilization 6, it could be Path of Exile, MMO raids, PvP or whatever.

We all want the ideal of the understanding good teacher that cares and knows how to break it down just right so that we can learn what we don’t even know, let alone don’t understand.

(Or maybe that’s just me. Maybe you have some other ideal of the perfect teacher that is just right for you. Goldilocks style, not too soft, not too hard.)

Chances are bloody good that we’re not going to get one.

We would be immensely lucky if we strike gold on the first attempt at sourcing a great teacher.

Chances are far more likely that we’re going to hit judgmental people; people too self-absorbed in their own lives to bother much about your learning; people far too ready to go for the expedient assumption of “unteachable, boot him/her out of here” much more often than a good teacher with a heart of gold. (The latter burn out real fast, I hear.)

They’re going to do shit things to your emotions.

But there’s always one person that -is- committed to your learning, and that knows -exactly- how you like to learn.

Yep. You.

Perhaps that’s the teacher to really have a heart-to-heart conversation with.