techdragon

That’s it in a nutshell. The current cost of a Synchrotron is more then the cost of the equivalent capability using non-synchrotron technologies.

It might seem crazy but it turns out Synchrotrons are pretty rare machines too… and they are less “operationalised” in an industrial sense than the crazy tech inside the EUV machines… between the vacuum levels needed and the various linear or circular accelerator technologies available, at the end of the day it’s just way more of a pain in the ass to maintain a large industrially pure vacuum machine operating at continuous round the clock fashion. which might seem nuts but talk to anyone who’s works with high vacuum systems about vacuum purity, out gassing and contamination and yeah… it will make sense pretty quickly.

As a long term python dev I was starting to warm up to c# and considering it for some upcoming projects… even to the point of evaluating and considering paying for commercial licenses for third part c# libraries to fit the projects…

But this rug pull deeply shakes my faith in .Net as a cross platform development ecosystem…

I’ve done a complete 180 on my entire enthusiasm for .Net here… I know the underlying tools will continue… but if they don’t give enough of a shit to maintain their IDE (which for the .Net world is absolutely crucial… I cannot emphasise this enough… .Net development requires an IDE to maintain your sanity…) well…

This is basically step one in a slow abandonment of non Windows platforms… and everyone who lives on those non Windows platforms… be they Linux of MacOS… we know it spells doom.

The marketing message around the suburban lifestyle has been sold as as the “American dream” or the “Australian dream” for more than a half century… its a profitable social manipulation that benefits politicians, land owners and all the businesses involved in house construction… it’s not that the end result is evil, it’s just pushed as an aspirational ideal that everyone should strive for when there are downsides for society at large when it comes to things like infrastructure maintenance and other costs shared by state/local governments…

The effectiveness of public transport is significantly affected by the population density.

Without sufficient people per unit area, you won’t have enough passengers to make the public transport work, depending on all the obvious factors like cost/quality of the public transport, proximity to housing without parking, the existence of housing without parking, etc.

I would view the existence of housing that doesn’t come with parking to be a forcing function, it on average only gets built where public transportation is viable, and once built it ensures ongoing continued demand for that transportation which keeps public transport more viable around that location.

Not a lot of housing development (even removing suburban homes where driving is basically a necessity) in Australia or New Zealand gets built without at least one parking space per dwelling be it a town house an unit or even a studio apartment… it does happen but it’s kind of rare. In large part due to the conflict between inner city real estate where public transport density is high enough to make car free living practical for normal people, is sufficiently expensive that the market selects for people who are rich and thus you’ll see apartments built where the entire first few floors are a multi story carpark squashed in under the actual living space because no one that can afford the apartments would even consider not having room to park their luxury car.

It’s sort of a vicious cycle, the cost also helps drive realestate speculation and the ongoing rise in prices and the continued cyclical rise in real estate prices year on year.

This is one of the useful tactics that unions will use when they aren’t legally allowed to strike for one reason or another, they will engage in large scale “work to rule”, which is pretty much what it sounds like, you do your work following the rules to the letter and no further, stubbornly by the book no matter how inconvenient or unproductive or disruptive the rules are when you suddenly start following them exactly.

It’s particularly effective if the rules are a large part of their collective grievances with management, as it highlights to all involved that the current rules need changing, and can sometimes result in fairly quick and positive changes since it demonstrates to management that the current “rules as written” are not the most fiscally effective set of rules for running the business.

Well… now in addition to my PyCon AU conference talk, I’ve got myself a lightning talk … fun times. I was using this with some ASP.NET 6 based experiments and didn’t realise it had an RFC behind it, I was just enjoying the nice error messages.

Got 89% overall, was a couple that I had to think about but mainly because I forgot to treat them as news headlines not simple statements of "fact"... But overall not bad... this was interesting. It did seem to have a bit of a "stuff you might have seen this year" sort of a vibe, which would have been interesting to see countered with an anachronistic old time headline version...

It’s most places, due to the international pressure from multinational organisations like the WIPO (World Intellectual Property Organisation) and WTO (World Trade Organisation) … they have longstanding campaigns to harmonies members national legislations to facilitate easier international trade and respect of intellectual property rights between countries…

As a result most countries (but, yes, not all countries) have the Bern convention “life plus 70”

Or slightly less charitably… 3k of News Corps local news stories were some degree of made up and are even less trustworthy than they used to be (they don’t have a great reputation for quality already)

We’re trying to revolutionise most of it… because at the moment everything we use in space has to be built on earth and launched in one piece on a rocket big enough to launch it one piece… and made tough enough to be launched on a violently shaking rocket… and everything has to be made bulletproof because once it’s up there it’s impossible to repair… all of which makes basically any space construction project extremely expensive to engineer… the only shortcuts are mass production of identical disposable units like planet labs doves or spacex starlink… which are kind of wasteful but work if you have a cheap enough ride into space…

But building anything in space is a massive set of challenges of its own … a space station module requires skilled welding of a pressure vessel, fitting a satellite requires careful cable management and manipulation of sensitive components… low earth orbit is practically a low density oxygen plasma compared to higher orbits (and oxygen plasma will oxidise just about anything given enough time) …

We analysed a lot of options and ended up designing (and are slowly building due to limited budget) a modular system of robotic spacecraft designed to work together in teams like a construction or assembly team, able to work on larger projects like space station modules or simple things like assembling a cubesat from a bin of parts… and designed to handle both construction and maintenance so that we can provide value to the massive amount of existing assets already in space, and because it’s a multi element system capable of maintenance and repair work, each robot can be more low cost and straightforward, like a Bobcat you can fit a bucket or dozer blade or jackhammer or claw to… they can swap tools on each other just like when you use one bobcat to lift a heavy tool up off the ground for attachment to another bobcat, and when one breaks the others can open it up and replace what’s broken…

It seems a lot more complicated when it’s described out in terms of capabilities, but fundamentally there’s a small number of modular components and a handful of support systems that are much simpler than the primary robot platform units are.

The website doesn’t have a lot of PR material at the moment since most of the effort is still on engineering design and prototyping work, but the website https://www.tetraorbital.com will have more information when we have stuff to show, it also has contact info for more specific enquires, and the mailing list is a zero fluff no spam future announcement only list (we’ve literally not sent a single message to it yet since there’s no PR or announcement material to share yet)

Wow that’s disheartening, almost makes me think it’s not worth applying… I’m outside the USA doing a hard tech problem involving a difficult industry with high capital costs… I think my chances are basically zero, why would YCombinator think some random company from Australia can revolutionise space construction…

Really disheartening when you add the fact they added a special early AI around to deal with the thousands of AI startups ( most of which are probably just built on ChatGPT and don’t have a lot of long term value proposition as the real value is coming from OpenAI’s APIs and then adding a thin layer of prompt engineering, a dash of industry specific knowledge, and a bunch of marketing, doesn’t really seem like something that companies won’t be able to easily replicate )

I cannot fathom a single faster way to annihilate all the brand goodwill they have earned than to begin inserting advertisements into ChatGPT output… short of ChatGPT spewing Nazi slogans in German like a pop up advertisement of WW2 era black and white Hitler rally… that’s about the level of “how fast this will destroy your customers good will”… people using ChatGPT on GPT4 are paying customers… if you inject advertising into paying customers who are already rate limited and unable to use VPNs and unable to fully use the model due to risk of getting banned if they red team and try pushing boundaries, and have a CEO who is rapidly trying to get a seat on the US Olympic bullshit artist team…

OpenAI are either blind, hypnotised by their own hype, utterly mismanaged, or fucking stupid… and I’d feel pretty safe putting my money on mismanaged base of the current nonsense coming from Sam Altman…

People used to sell this sort of information, it was the classic digital multi level marketing scam without the overarching marketing company. There were ads on YouTube google and Amazon all the time for get my book about how to successfully make money self publishing books on Amazon… people bought it tried to do the harder part of writing stuff then eventually some decided they could crib the book that convinced them to do it in the first place and plagiarised/rehashed the material and wrote the next generation of how to make money writing books to self publish on Amazon and the cycle would continue, slowly growing as more customers fell for the cheap garbage books and the promise of easy income…

Then large language models utterly annihilated this as it becomes obvious that you just write spam books and hope people buy them, job done…

Kind of amazing to see thevalue collapse reaching a new phase, where the only reliable way to drive sales for garbage books is to try and parasitically leach reputation by pretending a real author who has a real reputation is the author of the garbage book.

There are at least two sides of the argument which have a recipe they believe is correct… it won’t be held onto forever since both of them saw the original recipe guarded so closely it went nowhere… one will cave and the other will follow… classic prisoners dilemma stuff.

This waste of effort by way of duplicating unpublished negative results is a big factor in why replicated results deserve to be rated more highly than results that have not been replicated regardless of the prestige of the researchers or the institutions involved… if no one can prove your work work was correct… how much can anyone trust your work…

I have gone down the rabbit hole of engineering research before and 90% of the time I’ve managed to find an anecdote or subsequent research footnotes or actual subsequent research publications, that substantially invalidated the lofty claims of the engineers in the 70s or 80s (which is amazing still despite this, a genuine treasure trove of research unused and sometimes useful aerospace engineering research and development) and unfortunately outside the few proper publications, a lot of the invalidations are not properly reverse cited research material and I could have spent a week cross referencing before I spot the link and realise the unnamed work they are saying they are proving wrong is actually some footnotes containing the only published data (before their new paper) on some old work that has a bad scan copy on the NASA NTRS server under some obscure title and no related keywords to the topic the research is notionally about…

Academic research can genuinely suck sometimes… particularly when you want to actually apply it.

This feels monumental… can we get the black bar for the entire weekend.

This is the passing of someone that has fundamentally touched the lives of so many programmers… his work is now part of our very folklore, we make vim jokes, we emulate it in other tools, even the “enemy” (eMacs) eventually added evil mode (mind my humour it has sharp edges)… vim is mentioned in science fiction and has survived the journey from Unix to a plethora of other operating systems…

He will be missed.

I think this might be a bit exceptional as far as public engagement goes. So I wouldn’t necessarily judge public engagement based on this case.

“Rocks float” vs “rocks not float” is a very easy success criteria for the average person to judge by, lowering the bar for the average person to feel like they can add something to the conversation… so when we add in the potential revolutionary aspects of a room temperature superconductor we have a recipe for significant engagement… it’s even engaging the gawker reflex and people are picking up on it be LK99 is a weird trending topic and people will check to see if it’s an airplane that crashed or something …

In essence it was, by sheer coincidence, bound to go viral… and only because of a number of properties that others won’t have…

It took a few minutes sleuthing to validate that my gut reaction of "but there's way more than 50000 movies" was right ... But thanks to UNESCO statistics (http://data.uis.unesco.org/) I've got the sad truth.

Your 50,000 movies is a lot but its not even half of the movies made since 1995 that UNESCO were able to reliably cite data for. I can't deep link to the exact spreadsheet of data but it's under the culture data section and its feature film statistics. The total number of feature films (which will exclude some things like short films and other stuff based on various data processing considerations) produced around the world between 1995 and 2017 is at least (because there are likely more movies shot than produced) 107,432.

Assuming your "over 50,000" is the usually marketing line and being generous and assuming its somewhere between 50,000 and 55,000 movies, then if true that's approximately half the "feature films" produced between 1995 and 2017...

A a sensible lower bound extrapolation based on the data is in the range of half a million feature films (again recognising this data is likely excluding things we would on average collectively call a movie), making your service closer to 10%... or possibly even lower...

Over 50,000 movies feels like a lot until you dig into how much media we make as a species. That's just feature films, likely only ones with a theatrical release (I don't think I can reliably translate a lot of the source documents even if I wanted to validate the source data criteria myself, hence I'm using the weasel word "likely"), meaning its excluding a LOT of film/movies, and its only for 1995 to 2017, covering an era where the "amateur film" scene was rapidly exploding due to falling costs of the technology behind recording, editing, and distribution a "movie"...

None of this is to tear down the effort of hard working preservationists... both legal and illegal. I agree with the archivists I've had conversations with, whose collective opinion can be summed up quite simply. "Any copy is better than no copy."

Freon memory the issue they faced was the FDA not really having a path/category for them to be licensed under. It was a live bacteria “vaccine” with a mechanism of action that basically amounts to “out compete the bad bacteria and stick around forever in the human mouth because the other bacteria would have anyway” … which was a brilliant idea!

Between being early on the applied genetic manipulation tech wave, having a product that reproduces itself and sticks around forever (cue the FDAs concern meter rising) and at least in the eyes of the FDA not really fitting into any existing licensable food/drug categories they have established procedures for… the company stalled … and pivoted to selling some kind of water product I think.

Important correction… the catalog of movies is likely significantly smaller than all the movies ever made… the ongoing removal of new movies for business reasons and the market fragmentation mean that reduced profits are leading to corner cutting with respect to the size of stream catalogs, licenses cost money… pirating via torrents while larger also has diminishing returns as you step outside the mainstream movies, unless your lucky and find someone diligently sharing bandwidth to keep content online happens to share an interest in that niche… we’re rapidly losing access to an entire generation of analog only media as people throw source material into the trash and corporate stewardship of original recordings slowly fails one papercut at a time for the decades it will take for work to enter the public domain and be properly archived by the people trying to hold onto all of this.

We could have maglev trains and superconducting supercomputers and copyright will still be deleting our culture to “save money” for corporate copyright owners to increase their profit margins.

The only “secret “ feature I want is a reliable way to permanently disable the bullshit. I could even live with the new taskbar that’s a bad dock ripoff (and I say that as a Mac Windows and Linux user who has tried alternative windows shells and multiple Linux Desktop environments for the last 20 years)

But I am currently holding back the upgrade because i keep hearing about adverts in notifications, about absurd spyware levels of telemetry, of massive steps back for users who are more advanced and more experienced because Microsoft didn’t just “move the cheese” they decided to replace the cheese with a tomato while moving to to another continent…

Oh absolutely, there’s a lot of potential issues with the simple suggestion of requiring replication. I was tempted to write a bunch more to outline strategies and counter strategies from a game theory perspective but realised it was basically unnecessary since it will never happen.

Effectively you could take a watch glass or a petri dish, either not compress to a pellet in the first place, or just break up your sample gently between your fingers into the onto the watch glass or into the Petri dish, then just place that on top of the biggest magnet you can find and pluck away any floating fragments with plastic tweezers… seems simple… assuming you appreciate the sorts of thing’s about solid powder chemistry (which can seem a lot more like applied geology than the kind of chemistry most people are used to)…

if you want to see some interesting things and like a good chemistry video I thoroughly recommend searching for dry powder on powder chemical reactions… most of them are fire related demonstrations but it can be interesting to see them gently spoon a powder on top on another powder and nothing happens, until they poke it with a stirring rod hard enough to press the powder firmly against the other powder and get the reaction surface area to increase and then it’s off to the races… physical contacts between materials can sometimes not be in as much contact as you think… even when it’s literally pilled on top of the other stuff.

This further confirms the existing discourse around sample makeup and quality, the video, to me at least, looks like a large flake of inert material attached to a smaller whiteish looking bit of reactive material at one end of the flake, each time it’s demonstrated that end of the sample is the end that moves and the rest of it looks like it’s hanging down… the same sort of thing we saw with the original demonstration with the larger lopsided sample in the very first publication on by the QCenter original Korean researchers…

As I said in an earlier post on another article, smash the samples… let’s see some clear unambiguous levitation of small grains with little dead weight, this process is clearly not yet refined enough to give large samples of high enough quality to visibly demonstrate levitation… several of the existing failures to replicate could actually just be failures to have a high enough ratio of superconducting to normal mass and thus no levitation… these powered based bulk material processes are tricky as hell, even when you aren’t trying to modify the chemical properties and do reactions it can be hard to get structural cohesion of samples and a lot of dry powder based ceramic products have had notoriously bad quality control issues… we “mastered” wet clay ceramics thousands of years ago, yet depending on what your trying to do, something as “simple” as dry ceramic sintering can be challenging!

I don’t think a lot of the researchers rushing to perform replication are being sufficiently thoughtful about the nature of the samples they are producing. They are smart capable people who potentially have very little powder process knowledge or may have it from a background where this sort of thing isn’t as important… heck I only have my knowledge of this stuff from research into ultra high temperature ceramics for use in rocket engines, where poor powder process control can affect the structural and thermal properties and lead to all sorts of erosions cracks and other kinds of material degradation and sample destruction… so yeah I don’t expect all these teams to fully understand the challenges of powder based solid chemistry and I think we are seeing it brought out in the visible evidence we’ve seen posted online.

Also… It’s interesting how often people are using the same new AI powered video translation service … which I’d literally never heard of before the initial release and some of the interested parties using it to translate the original Korean information. Everyone following along has heard of this service now and it seems to be a happy accident that they are getting publicity and interest from being associated with the community gathering around the internet to follow LK-99 related work!

The simple truth is we don’t have a culture of replication because no one is paying and no one is paying because no one is demanding it as part of the incentive system that the money is part of…

If a prerequisite of publication in a top tier journal was someone having replicated the work… anyone at all… then the quality of write up would improve dramatically because all the stuff that never gets written up would be written up so that the second lab costs them less money to do the replication work, so the second lab doesn’t spent a month failing because they didn’t bake the powder in a vacuum for a week before performing the entire synthesis under dry nitrogen purge when the methodology just said the powder had been dried and the synthesis was performed at ambient pressure in a zero humidity atmosphere…

Sad follow up for anyone coming across this later… but alas it has indeed now changed to X for me too…

I think about it from a “what would a hypothetical random person say”… it’s not intrinsically holding a view or set of facts, instead it’s holding a weighted average sum of what people have said and written, including all the conflicting viewpoints inherent in that.

Despite some initial fun, I quickly got hesitant to do any more amateur “red team” experiments with ChatGPT because I was finding it useful enough that getting banned would be too big of a risk…

It was surprisingly easy to do once you internalised the word prediction foundation and thought about the corpus of writing you have ever read in a “tropes” fashion. For instance it’s really easy to get it to give you some “mills and Boone” romance novel type sentences or dialog, but when you can keep it on track it’s terrible at romantic “escalation” (the stuff where the relationship or interaction between characters is undergoing the implicit consent negotiation and furtive suggestions as they work out things will “move to the next base”) … And it’s terrible because the bulk of the available writing corpus of romance and interpersonal/intimacy is padded for page count and long sentence descriptions of small details…

It’s actually quite remarkable to see how it can be good at the bulk of the text, but simultaneously so terrible at a specific subset of that kind of writing.

Unequivocally yes… which is why I wasn’t surprised when the first publicised reproduction involved a tiny grain, why the big pressed pellets seem to be staying put, why people are suggesting breaking the samples up, and why the original pictures/video showing the levitation that kicked all this off, is an oddly shaped sample that sort of looks like it flaked off a larger piece.

Smash the samples… see if any of the grains float… (it’s becoming a bit of a refrain on Twitter whenever a big chunky pressed pellet sample is published as a failed replication) … it won’t be hard to improve the process and try and get larger contiguous samples that show the effect once we know for sure that it’s actually happening even if it’s happening in smaller more obviously superconducting samples, be it the size of a sand grain or a grain of rice, as long as it’s clearly a room temperature superconductor… once we have more samples… the analysis experts can get stuck in, there’s a global army of analytical experts… crystallographers, spectrographers, and all the sub disciplines of analytical chemistry… with clearly superconducting samples these people can swing into action and work out what the magic is and what we have to do to make more and bigger floating rocks.

This is fundamentally the major problem.

It’s obvious there’s a limited number of samples and they potentially don’t feel like they have a complete grasp on how to reliably produce more using their current level of knowledge regarding the process they used to create the first one… sometimes science involves luck. They have been working on this for years, I can completely believe a non-trivial portion of that time was spent trying to go from a fluke to having at least some ability to produce samples that demonstrate superconducting properties.

They may have only a half dozen samples of varying purity and quality…

The samples may not be particularly “strong” and given the powder based process involved may be too brittle to safely “send” and thus require careful hand courier delivery like a fresh organ for transplant, just without the urgency…

It’s entirely possible they may be mistaken and it’s not a superconductor, but these people are not fools, so I’m inclined to “give them enough rope” and to let this all play out and see where the either falter and the falsehoods exposed by the absence of any further evidence, they are are shown their mistakes once others get their hands on samples, other’s reproduction proves them correct, or their samples are studied by others and the evidence from their paper proves to be correct regardless of novel theories regarding the mechanism of superconductivity involved.

Science is a process, and to be honest it’s fun watching how fast people are pouring on the gas and trying to find out if this is real or not … stark contrast to the abject scepticism and shoestring budgets given to many innovative potential ideas.