SloMoMonday's recent activity
-
Comment on What games have you been playing, and what's your opinion on them? in ~games
-
Comment on What games have you been playing, and what's your opinion on them? in ~games
SloMoMonday Link ParentIt felt like the difficulty was pretty flat for most of the game and it started to become a slog to get through. Then there's a major spike right at the end when they just start throwing enemies...It felt like the difficulty was pretty flat for most of the game and it started to become a slog to get through. Then there's a major spike right at the end when they just start throwing enemies and red zones at you with large spans of no floor/grapple points. It's not full Kaizo insanity and using the time-slowing system to aim your hook helps a ton. There's also a late-game movement tech which that can have very inconsistent interactions with platforms and rails so keep that in mind.
-
Comment on What games have you been playing, and what's your opinion on them? in ~games
SloMoMonday LinkI've had a lot of time to jump around games while traveling with the Steam Deck. Was excited to finally dig into Factorio Space Age and... nope. Can not play this on the deck. I did get up to Blue...I've had a lot of time to jump around games while traveling with the Steam Deck.
Was excited to finally dig into Factorio Space Age and... nope. Can not play this on the deck. I did get up to Blue Science hooking it up to a monitor and keyboard but the way I play is not conducive to controllers and small screens (and to play in planes since I tend to talk to myself when planning). Will get into it properly now that I'm home.
Another game is Sanabi. This is a 2D gimmick platformer by a Korean studio and the narrative plays hard into a sort of sci-fi K-Thriller/Action vibe and takes a lot of inspiration from Metal Gear. While the overall quality is above average, there are occasionally some amazing levels, art and character moments. However, like Metal Gear, there are times when narratives and cut-scenes can become a bit... much. To the point where I sometimes could get control after a scene, walk to the next room and immediately get another scene. The main gimmick is a grappling hook that is pretty intuitive to use but the main game is held back by forgettable combat and levels that can overstay their welcome. There is also a free DLC mission that puts you in the shoes of one of the more standout Bosses and makes for a fitting send-off.
Also played Super Meat Boy for the first time. Since I'm 15 years late to this particular party, all I can say is that it's a well made precession platformer that has plenty of little secrets and is a fun throwback to Newgrounds art style. But it's not nearly as refined as Celeste or interesting like Will You Snail.
Luna Abyss. This is a fun one and I put it down since it technically "runs" on the deck the same way a car can technically run on Nutella. It's British Hollow Knight. And if endearing bugs says something about the Australian psyche, then it looks like Brits have thing for charismatic gender-fluid people because so many of the characters here are just a joy interact with. The game is a first person platformer/arena-shooter in a post-post-collapse lunar colony where you are sent to serve out a 25 year prison sentence. And the game just throws you right in with an FPS that does some genuinely interesting things. Like they are not afraid to get crazy with the levels, color and contrast. Spaces are vast, dark and dead. It scratches the itch I got from Metal Garden for dauntingly large and moody level design. (Also please play Metal Garden. It's like $2 and 2hrs long and you can feel like this dev is onto something special). They also just give you an aimbot. None of the bullet hell parts left me thinking that it was strictly needed, but it has me excited for how hectic combat could get if they don't want you even aiming. The part I'm not to keen about is the main story. Like Silksong, there's a lot of religious (I think its Catholic) undertones, savior/Christlike figures and false prophets. I think it could have come together better if there was the option of Downtime and experiencing the wider world, outside of the oppressive underground. It feels like that must be here but I've done 2 levels and the game forces you to go to sleep and jump into the next assignment. Looking forward to finish this on a big screen.
Bounced between a few tower defense game but none really stuck. Idun feels like a Warcraft 3 TD mod with hero characters, but there's some weird F2P mobile game vibes in the staggering amount of resources/currencies, "filler" missions, and the way some things have pointless limitations.
Defense Grid - The Awakening, it's been well over a decade since last playing it, I have 20hrs in it and cant remember a thing about it. The game really shows it's age and feels way too rigid and streamlined.
Gem Craft - I have no idea what the hell is going on.
Tower Domination, Repel the Rifts, Mineral Defense - Three roguelikes where the only strategy at high difficulty is to just roll good paths that merge into overtune kill-funnels. Tower Domination probably tries the hardest to balance it out, but overall, there's just way too much variation and even the best runs can be killed with the wrong enemy spawning on the wrong path. -
Comment on No, artificial intelligence is not conscious in ~tech
SloMoMonday LinkAt first I was a little disappointed with this piece, especially since I love his stories Absence of God, Liking What you See and Anxiety Is the Dizziness of Freedom. But to me, the conclusion...At first I was a little disappointed with this piece, especially since I love his stories Absence of God, Liking What you See and Anxiety Is the Dizziness of Freedom. But to me, the conclusion here points to this being something of a of a reluctant exercise for him.
It’s fortunate that LLMs are not conscious, or else the actions of the big AI firms would be even more scandalous than they already are. So why are Anthropic’s employees suggesting that Claude might be conscious? Perhaps it’s just another form of hype; perhaps they have fallen prey to the same spell that they have been casting on their customers. But when they publish a document about Claude’s moral education and have their in-house philosopher do a press tour, we should understand them as asking the rest of us to indulge them in their fantasies. We don’t have to play along. In writing this essay, I have spent more time indulging them than they deserve, in the hopes that it will keep you from spending your time indulging them. If you want to think about LLMs, there are scores of other questions more worthy of your contemplation; you can safely ignore the question of their being conscious.
I can sort of empathize with this sentiment. Ever since the discussion of LLMs/AI really started to get steam through 2024, the topic of machine sentience and consciousness seemed almost a moot point given the many glaring business/costs/RoI/safety issues I needed to resolve. I suspect someone like Chiang would have had his own areas of concern, particularly in psychology and philosophy of humans relationships with tech/corporate in general. I might be projecting, but it feels like the entire second half of the piece is an exercise in self restraint by constantly skirting around the things he wants to talk about. (Parenthood, accountability at scale, infatuation with slavery, IP theft...)
But the idea of machine consciousness has been so thoroughly staged in the public discourse that it was for a long time, one of the only discussions most people could hope to have. "What happens if it breaks containment." "AI is showing signs of introspection." "We need to be prepared for super-intelligence that definitely going to arrive any day now." "We have a model in the lab that makes us afraid for the future." Even now that the industry has finally reached the cart that was miles in front of the horse and costs vs. RoI is unexpectedly the most important thing in the world, I still see the topic of machine sentience come up in non-technical circles and it is just as big a point of an anxiety as it ever was.
I also suspect there's a level of frustration when sci-fi writers are expected to comment on this industry when it's the original trope of the genre: Frankenstein defiled the dead to create life simply because he could. I think Chiang was explored this idea in a story call Life of Software Entities (haven't read it in a while but it's basically Her with some weirdness).
On topic of consciousness itself, that just seems like a philosophical black-hole that doesn't even have a fun thought exercise to make it approachable (trolly problem, ship of Theseus, teleporter proplem). And a person can't exactly share or remove themselves from their own conscious experience to gain insight through perspective. I recently learned that "internal monologues" are literal things for some people and it has completely changed how I approach other people and try to communicate because my own head doesn't seem to work like that.
So how do I describe making a conscious decision when words are not exactly part of that process? Is either conscious experience "correct" or "anomalous"? How does this overlap with brain chemistry, upbringing and life experiences? Is any of this even practically quantifiable and can it be effectively categorized? -
Comment on Nvidia announces liquid cooling system that promises to reduce electricity consumption and cut water use by up to 100% in ~tech
SloMoMonday LinkWhile such developments are objectively good, it does sound like a desperate attempt to save face. At least at this point. Data center roll out is being stonewalled by communities across the board...While such developments are objectively good, it does sound like a desperate attempt to save face. At least at this point.
Data center roll out is being stonewalled by communities across the board and the power draw/noise pollution/environmental impacts are real. Combine that with the climate crisis being particularly impactful recently, SpaceX stumbling, global fuel reserves being strained and the very fresh tech price hikes; it's not looking good for the AI industry.
Its one thing to stand up in a town hall with a press release with some loose technical specs saying "we solved all your issues". Doesn't mean that we shouldn't still do an impact analysis and an environmental review, social equity plans and community consultation. It also means that if this tech exists, it should be retrofitted in existing centers and should demonstrate tangible results. That would support its case when proposing additional roll out. But that's all too slow for how hot NVIDIA has been burning for years.
Limiting it to new chips means that there is still a lot of stock in the world that is an environmental hazard and there is no incentive to better the infrastructure that is already causing real harm. It just feels like a desperate ploy to sell more chips. Even if all the existing stock still isn't being fully utilized. (Which is a wild idea on its own. It's not like you can just strip down $100k in unused boards for $100k of parts to make new ones)
-
Comment on What do you think about Destiny 2’s imminent death and games as a service? in ~games
SloMoMonday Link ParentIf you'd like a bit of a lighter story about the game. This was around Deep Stone and I was running it with my sibling and some friends. Warm up with the strikes and the whole way through, my bro...If you'd like a bit of a lighter story about the game. This was around Deep Stone and I was running it with my sibling and some friends. Warm up with the strikes and the whole way through, my bro is going on about how hes so hungry. We tell him to get some chow but he says he's good... then proceeds to wine about it all the way through the twin Oger fight.
So I get out my phone and order him delivery. In the form of random sides and extras from as many places that would allow it (add that it's a prank in the notes and people are surprisingly game). We had to wipe every few minutes because he had to go get 2 chicken wings and a handful of serviettes and a cup of ice and a can of bagle sprinkles and waffle toppings without the waffle and a platter of BBQ rib sauce over grilled peppers.
After the half dozen wasabi tubs, my brother threatens to dip and one guy does a flawless Zavala impression: "Guardian, are you abandoning the mission because you find your rations... unsatisfactory. And am I to understand that you find humble gifts from your brother in arms to be beneath you. Is that truly the conduct you wish citizens should expect from a Titan of the traveller."
We didn't even make it to the sorrow fight that night. Was it worth the money? No. But I was playing Destiny so that should explain where my head was at.
-
Comment on What do you think about Destiny 2’s imminent death and games as a service? in ~games
SloMoMonday LinkGoing to be a bit more emotional since played Destiny on and off since Peter Dinklage listlessly woke us up in the Cosmodrome. We had a regular raid group that did Vault of Glass, Crota's End,...- Exemplary
Going to be a bit more emotional since played Destiny on and off since Peter Dinklage listlessly woke us up in the Cosmodrome. We had a regular raid group that did Vault of Glass, Crota's End, Kingsfall and Machines Wrath all blind. Did Trails and Nightfalls religiously. Fought for god-roll gear. Sherpa for newer players. Got both my bothers playstations, copies of the game and PS+ when I left home so we could still play. And finished Rise of Iron thinking it would only get better.
Then, in spite of Destiny 2 being on the PC, it instantly felt like so much more of a chore. Forsaken and Last Wish were the last high points and Shadowkeep looked like it could carry that momentum. But Beyond Light and Witch Queen is where things started going pretty bad. I really should have drawn the line at Sunsetting content I paid for, or the over emphasis on Eververse, or the god aweful transmog, or the way PvP was left to rot. But my wallet kept being dragged back in.
There were plenty of bright spots here and there. But on the whole, it was simply a case where I could count on being let down with increasing frequency. And it was finally Lightfall that broke the camels back. Just deleted the game, sunk cost be damned. I even stopped watching Byf videos to keep up. It just wasn't worth it. Don't even know how Final Shape ended.
So yeah. After that, I just have zero patience for live service and MMOs. Even general PvP and co-op "friend slop". Destiny showed good enough numbers to supercharge the enshitification of premium online gaming. The idea, combined with Fortnight, CoD and Apex effectively killed countless other studios and promising projects, including Arcane Austin, Overwatch, Halo and a big chunk of Sony's unreleased First Party ambitions.
Just like Destiny, anything live service feels like waiting for a scrap of positivity in a deluge of bad business or poorly aging gameplay.
I would not be nearly as negative if I could boot up the game right now and have instant access to hundreds of hours of increble gameplay from over a decade of creative and community efforts. 95% of this work can exist as a single player, linear experience. I can't see any reason why this can't exist as peer-to-peer or through dedicated server setups that allow for people to curate their own experience. We've already had sensible looter style RPGs as far back as Diablo 1 and with the game going offline, there's no need to maintain the balance.
Last Christmas, I played halo splitscreen with my cousin and our kids. Got the AlphaRing mod, hooked up the pc to the big TV and spent nights going through Reach, ODST and multiplayer on 3. The same way I used to play with that cousin and our parents over a decade ago. It was just magical to have the whole family there and to see my kid go a little feral when we got that grav hammer.
I'm not going to show my kid how to Solo the Abyss or rag on my brother because he's never been able to outrun the flaming Servitor (it doesn't stop being hilarious to see him bonk the same cliff every bloody time). The infinite loot cave is gone. The Mythoclass, Bad Juju and Souros Regime that I dragged into every, even when it was off meta. No one will have a chance to reimagine the city factions or the criminally under used Last City (and even the Traveller itself). It's all just gone. And needlessly so.
After people paid for it. Thats probably the part I'm most salty about.
-
Comment on Who’s buying SpaceX and Anthropic? in ~finance
SloMoMonday LinkI've actually just set a meeting with a financial advisor to make sure we have managed exposure to SpaceX and the AI public offerings. In general, the instant rules are being broken to get people...I've actually just set a meeting with a financial advisor to make sure we have managed exposure to SpaceX and the AI public offerings.
In general, the instant rules are being broken to get people "in on the action", it's the biggest red flag to me and it seems like the exact case here.
In terms of spaceX, there a lot of weirdness going on:
Most notably, the company is putting up a laughably small percentage of shares. Musk will still maintain full control of his company and have no incentive to act in investors best interest. If thr company goes public at the requested price, he then folds it into Tesla and is practically entitled to his insane "performance bonus" based on a damn technicality. And if SpaceX is rammed into every index fund and passive portfolio, governments will be forced to bail it out to protect pension funds. And all that money just goes to Musks slush fund.Similarly, with AI, I do not see any sustainable path to recoup the costing. 2 Trillion dollars is the cost-to-date on the AI roll out. And that's loose change compared to the $15 trillion valuation across the 6 already existing AI companies.
Can EVERYONE make a significant RoI to justify these types of costs and valuations. This is with an escalating energy crisis. And an inevitable everything-shortage. And an increasingly likely Chinese invasion of Tiwan. And the fact that Data Centers are not coming online anywhere fast enough for nvidia to install Blackwell's and recoup costs to buy the latest high end tech. And to still perform in the face of cheaper Chinese models.
-
Comment on Outsourcing plus local AI will soon become more economical vs frontier labs in ~tech
SloMoMonday Link ParentApologies. Think I just misread the tone of your response. I'm used to "we'll figure it out" being a thought-terminating argument used to justify bad planning.Apologies. Think I just misread the tone of your response.
I'm used to "we'll figure it out" being a thought-terminating argument used to justify bad planning. -
Comment on Outsourcing plus local AI will soon become more economical vs frontier labs in ~tech
SloMoMonday Link ParentI really don't think that's fair on AI/LLM customers and not how you run an industry aiming to IPO in the multiple trillions. Simply because from the very start, it was all sold on the idea that...I really don't think that's fair on AI/LLM customers and not how you run an industry aiming to IPO in the multiple trillions. Simply because from the very start, it was all sold on the idea that AI will: get cheaper, get better, become self improving, improve outcome quality, end global warming, cure cancer...
The enterprise AI cost scaling only really started this year and became the norm in the last two months. And I'm sure Cluade Code constantly "glitching" to poorer output quality at peak usage times had absolutely nothing to do with it. Also, I'm sure retail costs are going to stay flat forever with "Tokenmaxing" being the new craze.
At no point do I recall customers or the public being primed to shoulder escalating OpEx when using this tool. At the same time, this pricing change was made well after most AI adopters have already reflected their "AI Savings"... which was in the form of firing a lot of technical and white collar staff. That means that the AI tools NEED to start generating value, in excess of the price increase. And I don't think that's going to be by the tech massively increasing product offerings and revenue centers. My hunch is that it will simply be more price increases, job losses, cut corners and all the other misery we've come to expect.
And I don't see any of the fabled new AI powered competitors coming over the horizon to stabilize the industry. The only people that can afford this tech is incumbent companies. Any new entrants will either be bought out by those incumbents or just have any AI/LLM solution replicated by the AI Labs (see: Cursor - Suppliers sell at a loss to identify where they can extract profit and then become a competitor to their customer).
Yes, people should know better. But people should also trust that their service providers and suppliers are honest. I did not offer any sort of AI solutions/strategy at my last job because I could not develop solutions I still am confident enough to stand by that the long run. There was a massive muck up on the e-commerce integration we made a few years back and I took the hit to my bonus over anyone below me. And after that, I put systems in place that kept that service running without issue until just recently (after an AI integration). I could not do the same with AI.
What's the recourse when the LLM Sales agent and the new AI powered dev team and the AI executive assistants and the AI analytics suite all start incurring true cost and my mess up looks like pocket change? Hell, we already saw just how flagrant and chaotic model upgrades and feature/service changes can be, on top of effectively meaningless Service Level Expectations (on that note, I'm curious if there is there an enterprise AI Services SLA floating around and how the terms and metrics are measured and enforced).
"People will figure it out" is not good enough when there's 300 low-income operations employees and dozens of downstream clients and multiple people who already lost their jobs in just my personal example. Multiply that over every medium/major company in every industry across the world. Sam Altmans and Dario Amodei are not on the line, even after convincing everyone and their grandparents that their marvelous machines are the future of business.
-
Comment on Outsourcing plus local AI will soon become more economical vs frontier labs in ~tech
SloMoMonday Link ParentThe "cost-per-token" is an interesting idea because that raises the question of return-per-token. The value of LLMs is that it was a flat cost seemingly unlimited compute and your efficiency gains...The "cost-per-token" is an interesting idea because that raises the question of return-per-token.
The value of LLMs is that it was a flat cost seemingly unlimited compute and your efficiency gains came in the form of tasks/ticket completion vs. the cost/time of a human developer.
And people couldn't exactly compete against that. The fact that so many people I know were incentivized to burn maximum tokens as possible without any performance/quality metric never sat well with me. Pay-as-you-go for an infrastructure heavy service seemed like the logical outcome. (At least based on how uber, airBnB, AdobeCS, MSOffice, AWS, Netflix, Spotify and general online advertising all turned out).As soon as we moved to prorata billing, the discussion flipped to RoI very, very quickly. And if were scoping out project, question 1 would be: "what are we burning tokens on?" and question 2 is: "how can I tell that money is well spent?" Especially if it's standard practice to burn several hundred thousand tokens across concurrent instances.
Is it economical to just use LLMs on the easy stuff like setting up a dev environment, spinning up and connecting all the services and configuration oauth/ldap and maintaining documentation? Or are we just using it for QA and bug hunting? Or are we going to save it for the most complex tasks and have it chip away at it in the background?
Also, can we really trust any AI lab not to "nudge" models to maximize their own RoI. Especially after teams let go of a chunk of their developers and are in a pretty captive position (ie. See adobe, Netflix, AWS...)
-
Comment on Who else is as excited as I am for the Backrooms movie tomorrow? in ~movies
SloMoMonday Link ParentIts sort of a meme and in a weird cultural/creative space. There's the Backrooms as Kane Parsons project that the movie is based on. It's his personal vision on the idea with it's own story,...Its sort of a meme and in a weird cultural/creative space.
There's the Backrooms as Kane Parsons project that the movie is based on. It's his personal vision on the idea with it's own story, characters and world rules. I think The Oldest View and People Still Live here are better concepts on his YouTube channel so I hope those get some traction after this movie.Then there's the Backrooms Wiki which is an SCP style collaborative fiction. It's a real hodgepodge of liminal space fiction where people are tossing around ideas and concepts with a very flimsy sense of cohesion. There's not any real standout ideas for me like SCP (I love everything about the Antimemetics Division and cognitio hazards) and it looks like they are really trying to uplift older posts with a rewrite drive. But looser rules and more outlandish universe lets people get more creative, at the cost of being a cohesive concept ( Red Knight is a good example ).
Then there's Backrooms/SCP crossovers. I've only ever seen it in Backroom-themed Games and those tend to be little more than long walking sims/interactive blender art projects. On the whole, I don't think there's a lot of overlap between the two creative initiatives.
-
Comment on Life Below | Launch trailer in ~games
SloMoMonday LinkI'm a sucker an interesting city builder and itching for something to tide me over till Frostpunk 2 DLC. Will give it a try. Over the last few months I've tried Airborne Empire, New Cycle,...I'm a sucker an interesting city builder and itching for something to tide me over till Frostpunk 2 DLC. Will give it a try.
Over the last few months I've tried Airborne Empire, New Cycle, Captains of Industry, Stranded Alien Dawn and Memoria Polis. They're all pretty good and have their own charms. But they lack the spark to keep me coming back or to get me really invested.I tend to revert to playing my Timberborn megaproject or Jurassic World with the kid.
-
Comment on US military launches strikes on southern Iran amid talks in Qatar in ~society
SloMoMonday Link ParentI think a lot of the global fuel price-cuts is due to releasing the strategic oil reserves to stabilize prices. All the talk of a deal was to calm markets to bring down the speculative price....I think a lot of the global fuel price-cuts is due to releasing the strategic oil reserves to stabilize prices. All the talk of a deal was to calm markets to bring down the speculative price. Materially, production is most likely at a cold stop. Even if things went back to pre-war arrangement right now it doesn't look that will even be enough to push through the slow restarting of production and then supply lines.
And unless the US government steps in and effectively nationalizes the oil companies and considers local people, they will keep exporting the strategic reserve for $100+/barrel. (odd choice given that you're at war). They know KNOC, EBN and the CNPC have an incentive to buy supply wherever they can get it. There is a world where the strategic reserve is used for it's purpose and carry people through a supply shock. But where's the money in that?
The faintest glimmer of a silver lining is that I can see this being a massive blow to gas cars and force air/cruise travel to evolve, at least outside the States. Hybrids or full EVs have become the norm in under a year where I am and we're well behind Brazil, Mexico, India and SE Asia. Even some of the recovering parts of the middle east are plastering their buildings with panels to reduce grid dependency. Pretty sure Europe, Australia and Canada will start catching up as a necessity too. Beyond that in the 15+ years it takes to start bringing more nuclear plants online, most countries could probably get the bulk of domestic users onto supplemented renewable VPP grids. It's either that or keep the risk of insane people taking the entire world hostage for a quick buck.
-
Comment on A random sci-fi question for you in ~talk
SloMoMonday LinkSoooooo... what do I need to do to get to Terra Nov... I mean New Mongolia. (I really liked the idea of Terra Nova. Unfortunately the show did not get the space to find its groove and audience.)...Soooooo... what do I need to do to get to Terra Nov... I mean New Mongolia. (I really liked the idea of Terra Nova. Unfortunately the show did not get the space to find its groove and audience.)
But seriously, even if I do suffer some horrible frontier-related death; the thought of a clean slate planet is intoxicating. I'd happily take being mauled by some mega-fauna predator over slowly being killed in a windowless office by my neighbors second hand Vape smoke.With a colony like Mars, I can easily see it becoming a distilled corporate hellscape. Food, water, data and even air being at a premium and provided through monopoly services. Not hard for those seven years to be dragged out with every excess charge and no real recourse, even with a united workforce. (If this sounds like your thing, can I recommend the game Hardspace Shipbreaker. The plastic free food option is not bad).
-
Comment on From neat lawns to wild havens: how No Mow May is transforming England’s gardens in ~enviro
SloMoMonday Link ParentI'll see if I can find my notes, but from what I recall from way back when: a lot of nutrients and minerals in the soil tend to sink because they're heavy. And if the vegetation is sparse, its not...I'll see if I can find my notes, but from what I recall from way back when: a lot of nutrients and minerals in the soil tend to sink because they're heavy. And if the vegetation is sparse, its not exactly going to be used anywhere else. What weeds tend to do is capture those nutrients in their roots and since they don't exactly need much to grow, they sit on the bulk of it.
Weeding pulls the roots and all their captured nutrients out of the soil and then tilling jumbles up any roots, mycelium and insect biome made that space home and was effectively circulating all that energy. Throw on all your fertilizer and weed killer and all those nutrients are sinking about 15cm-20cm with nothing to catch it. It's where you start transitioning into B-Horizon density and it's not doing anyone any good there. So naturally a lot of industrial farmers resort to just saturating spaces with fertilizer, which is a way of getting nutrients to your crops but its hell on the soil ecosystem.
Now this is going to be a tangent, but remember "its not much but it's honest work" meme. That's the late David Brandt, and that gentlemen is a pioneer soil health in the US. Here's a lot of the discoveries and strategies he developed and I'm sure there's plenty of literature from him is you know where to look. There's also Charles Dowding with his own good resources but I'm not the biggest fan of his style. I have a thing for chaos gardens.
If you want a practical overview for a home garden/veg patch, this is my method
You start with 3 rules: 1. Don't ever weed or pesticide you garden. 2. Don't till the soil. 3. Please don't ever weed or pesticide you garden. If you want perfectly controlled environments and clinical precision, hydroponics or micro-greens might be your speed (and they're fun projects anyway so give them a try as well if you can). To "tame" your plot for planting, start by collecting as much compostables/mulch as possible and ensure you have a good stockpile until a few weeks before planting season. Also collect clean brown cardboard. Minimal black ink is fine but avoid the color/plastic stuff or ones that come from food places (grease and meat smell).Mark out your plot and you can cordon it off with some wood beams if you like. Then tile the cardboard over that space. Unless there is a considerable shrub there, you want to preserve as much plant matter as possible. Soak the cardboard with water, put a bit of manure to level it out and then a thick layer of compost and top it off with mulch (some people do plastic but I find that a magnet for slugs and snails. Periodically top up the compost and throw your bokashi broth and other soil supplements if you like. And when its time to plant, just shift the soil for seedlings or poke it for seeds. Then you just need to be very generous with the mulch. If you want a bit of order or irrigation flow, gently push the soil to form two mound beside your planting line.
If weeds are encroaching on your crop, don't pull them. Just smother it out with a small piece of wet cardboard. But the much should help prevent that and retain moisture to bring down the watering costs.
As for pests: there are super targeted and minimum impact ways to deal with most of them. But a general use killer like roundup is indiscriminate and goes for pollinators and generally valuable fauna. There's also companion plants that tend to be very fragrant and natural pesticides. Mint, lavender, marigolds, basil. Plenty to choose and it has the added benefit of brightening up the color and smell of your space. (Just try to keep mint and lavender in a pot, that stuff spreads like crazy). Can also encourage natural predators, birds, bats, spiders, ladybugs. Even if they take their share of produce, they can make a big difference with even household pests.
I don't know if its past planting season up north, but if you missed it, you can even start now and just litter your space with a mixed bag of green fertalizer seeds. Then you can cover over that growth next season for a bit of a head start.
-
Comment on How I feel about LLM (AI) writing in ~tech
SloMoMonday LinkLLMs supercharge so many issues with the internet at large so it's hard to have opinions on one specific part of it without being contradictory on the whole. So I'm going to be pedantically...LLMs supercharge so many issues with the internet at large so it's hard to have opinions on one specific part of it without being contradictory on the whole. So I'm going to be pedantically specific:
I think LLM writing is fine (or even fun) for entertainment and creative purposes, provided the models are fairly trained on works by consenting and compensated artists. Those models should be run on sustainable hardware, and in a well considered sandbox that protects the users privacy and with the proper guardrails and deterministic measures taken to prevent abuse and model drift.And I need to be so granular with that view because things outside that parameters goes into territory I don't agree with and consider harmful.
I'm only going to rant on first part of that condition: "Entertainment and creative" reasons. It drives me insane that people use these tools for information, correspondance and decision making. Common sense would say that you should always verify whatever you find, especially regarding important matters. But it doesn't help that 9 out of 10 times, when any web page is published after 2024, its just more LLM generated content. Same thing with emails. I have a tag specifically for people that I know communicate via LLM. Because if I receive anything that contains technical or precise details; I will call to double check. If I send them something with important details; I will call to make sure they got the info in case their AI summary of their entire mailbox missed something. And the fact that these tools are becoming increasingly accurate is more of a problem because that means when the model does messes up, it's harder or less likely to be caught.
And even if they are more accurate than a human, the issue is a matter of accountability. My professional liability cover does not include AI related errors. I specifically asked for it over the last two years and they don't even entertain the idea. So I fave a report and an LLM filler line leads to a bad decision, that's on me. Hell, if I look up a formula online and get something slightly off from an LLM generated page, I'm still on the hook for that.
And that carries over into LLM written content thats flooding every corner of the internet. Especially when it's an LLM centipede as far as the eye can see. I looked up the recipe for a caramel sauce and AI auto-result spit out a method I knew was wrong. No butter, the sugar boils, too little cream no instruction to take it off the heat. It's going to burn or become a grainy fudge block. There was a reference to an LLM written recipe on a site created by a "home baker" in 2024. The site has several hundred recipes, each with a touching emotional story and photo that made the dishes look magical. It's obvious whats happening and there is no incentive to stop it. Especially when the metrics for online success is attention, scale and advertising space. Not authenticity and factuality. Yes it's been like that for a while, but it's near unmanageable.
It seems innocuous, but what if one of the millions of instantly generated, untested methods, suggest someone deep fry something frozen in hot oil. Or it misses a warning when some step could be dangerously reactive. Or a person has an important event and it's ruined because of bad guidance. It's not out of the question with how much content is generated in the food space alone. Multiply this issue across every field and every question. And then factor in that there is incentive to use these platforms in advertising. Include the fact that many LLM providers are very cozy with the current US president and their CEO's have, unique beliefs.
We make decisions based on perception of the world. LLM's have flattened that image into a general average and now its reinforcing itself ad nauseam. The tech is fine, I guess. My friend paid me to train his LLM game on some of my old TTRPG notes and the prototype I tested was fun for a few hours. I have a local model that has a good handle on the technical manuals and reference books I use and can point me towards things I might need. A guy I know has a model that is starting to preemptively flag issues on his print-farm. It's got utility, but it just seems like its the only tool being used for every possible thing, just to justify its existence.
-
Comment on US Government UFO document release in ~society
SloMoMonday Link ParentTo be fair, it does feel like the US military is being run by a mid 30's basement dweller that uses his mom social security money to my a bunch of pay-to-win crap and still gets wrecked by a lobby...To be fair, it does feel like the US military is being run by a mid 30's basement dweller that uses his mom social security money to my a bunch of pay-to-win crap and still gets wrecked by a lobby of 13 year olds.
But in all seriousness, it does look like the current us military has a thing for the aesthetics of the military, without any of the tedium or rationale that would make the biggest war machine in history even somewhat effective. Like that AI written speech after that pilot rescue thing. It's like someone is fishing for quotes or narratives that will fit into a game depicting this war. Hell, even the rhetoric they use reeks of insecurity. Insistent in only talking about "warfighters", constant use of religious imagery, constantly reiterating just how completely Iran's capability has been destroyed.
It bleeds over into this web design with the faux military, cyberpunk, conspiracy styling. Like this project had more forethought than the whole "who would have thought they would choke out global oil supply" situation. Its so tragically juvenile.
-
Comment on What games have you been playing, and what's your opinion on them? in ~games
SloMoMonday (edited )Link ParentIts probably one of the best games to show off with if you get the opportunity. You can watch a rapid jump from "Its just one button. How hard can it be?" to "Its just the third level. How is it...Its probably one of the best games to show off with if you get the opportunity. You can watch a rapid jump from "Its just one button. How hard can it be?" to "Its just the third level. How is it so hard?". And then you casually FC 1-X while holding a conversation, because that 7-count is burned on your brain.
Also, why am I 60 hours in and only leaning that there are difficulty levels. I probably pushed it up to hard when I first got the game, and forgot. But I could have saved so much grief on 5-X and 1-XN? -
Comment on 'Blue dot fever' claims Post Malone, Pussycat Dolls concerts. What's really behind it? in ~music
SloMoMonday Link ParentAlso throw in the fact the modern events are pretty miserable. In the last few years, I only managed one music act but I work in conventions, sports and stage shows when I travel. Beyond the...Also throw in the fact the modern events are pretty miserable. In the last few years, I only managed one music act but I work in conventions, sports and stage shows when I travel. Beyond the endless seas of phones and rowdy crowds, really feels like there no more effort from the organizers side. Concession stands are far less and of worse quality. Similar thing with merch which are mostly crappy blanks with even worse prints. Stewards and security beyond the main seating is lacking and often have no clue how to handle a crowd. Constant software bugs, poor signage and miscommunication everywhere from parking to tick check and dispersing afterwards. Ton of places with seats that are essentially a waste of money for the experience you get. Don't know if its just me or a new trend, but the last comedy show, a hospitality convention and a circus act all had horrible AV.
The only show I have planned for this year is the new run of Rocky Horror Show and not keen to waste time or money anywhere else.
For me, Celeste has some very clean controls and visuals that allows you to consistently drill certain screens and movements. I only really appreciated that when moving from Switch to PC and didn't have to deal with the considerable stick drift (if anyone finished Farewell playing on joycons... How?). Maybe I'd say the same with 100+ hours on SMB. But after 6 hours, the controls were just too floaty with the high variability in jump height and the scale of everything felt a bit too small to get consistent results.