Prompting the imagination (for example, writing prompts) was a thing before generative AI even emerged in a meaningful sense. And it does work.
This is one of the reasons why I do literate programming using org-mode. It's easy to lose track of what I was thinking when I wrote something, and what the original structure and goal was as I continue to write it. Org-mode helps me keep my thoughts in order in English and interlineate code in with them, then mash a key to spit out compilable source. I don't use it for everything, but it sure comes in handy when I do use it.
I think it was 600 W 7 St. A quick Google of NSA tap plus that address turns up serious press reports from 7 or 8 years ago. It was clearly put up in a hurry. But I can't imagine I was the only one who noticed. I was just some dumbass serving a music site and it was totally obvious something was up.
Someone "invented" a previously unpublished military design.
Century means there is a factor of a hundred between lower and upper frequency that the antenna efficiently works on.
Decade means a factor of 10.
Bandwidth is typical the range of frequencies where the antenna is efficient.
Typically we categorise this by comparing it's gain at the centre frequency with how much you have to deviate before the gain drops some set amount.
Let's say it's resonant at 10GHz with a gain of 6dB. Typically we care about the range it's within 3dB of that.
Gain in this setting is less about amplification and more about concentrating the energy into a region of space rather than out in all directions.
If you go down to 3Ghz (probably 3.16GHz actually)and up to 30GHz( probably 31.6ghz)and it's still withing 3dB of 6dB, you have a decade bandwidth.
If you can do a century you're looking at 1 to 100GHz
An Octave is 1/sqrt(2) to sqrt(2) a decade is 1/sqrt(10) to sqrt(10) and a century is 1/sqrt(100) to sqrt(100) range.
So if you have an A4 note at 440Hz and you want a filter that captures an Octave with that A at its centre you would need a filter from 440/sqrt(2) =311 and 440×sqrt(2) = 622 which are d#4 and d#5. 6 semitones up from D#4 you get A4, another 6 and you have D#5. Musically it sounds like A4 is exactly between D#4 and D#5.
The fact that Century-bandwidth has a dash but decade bandwidth has not is weird, I didn't make the connection due to that (one sounded like a brand name, the other not)
The term "century-bandwidth" is always used in the article as a pre-noun adjective. The term "decade bandwidth" is not used that way. It's the difference between "a well-known author" and "an author who is well known".
Here is a basic explanation and a comment on the patent issue the author is addressing:
1. Antennas are designed to either transmit or receive RF.
2. The antenna is like water spigot for RF energy. If you design the antenna right, you can maximize the amount of RF energy you can get through it. This is equivalent to minimizing the impedance.
3. But, antennas are resonant. In other words the amount of RF that gets through can be maximum at one frequency (or a small range or bandwidth of frequencies), but very bad outside of that range.
4. Antenna designers try all kinds of tricks and techniques--including shapes, elements, delays, etc.--to try to get the antenna to have a broader resonance. An analogy might be the design of a musical instrument like the saxophone.
5. The VSWR the author mentions is a way of charting the resonance of the antenna with respect to frequency, ie. the bandwidth of the antenna.
6. Typical antennas (like a rod or something) have a bandwidth of maybe 5-10% of the center frequency.
7. Fancier antennas like a discone have a fractional bandwidth of up to 10:1.
8. An antenna like the one described by the author claimed to be 100:1.
9. But...even though the antenna may have broad bandwidth...the other factor antenna designers care about is gain (or loss). And this just complicates the design process even more.
Now, what I find most interesting about the author's comments is their suggestion that people continue to reinvent the wheel and then patent it. I too have seen this happen (not necessarily patents, but with other "technology inventions") over the years. I used to work in radio direction finding and every five or ten years someone would claim to have a new way of locating signals--but it's always just the same ideas over and over again. In other words, physics is physics.
"Antennas are designed to either transmit or receive RF". This isn't even close to realistic. Outside of amplified or active antennas, they have to be bidirectional devices. An active antenna is just a normal passive one with a non-linear junction attached.
Sometimes "to either x or y" doesn't mean "to do one of x or y", sometimes it means "to be able to do both x and y (but not necessarily at the same time purposefully)".
The thing about the article is that the new patent is basically the same as the almost 20yo design that the author got when a lab closed shop and he asked nicely to have it.
He also says that the "original" design was also not so original as there are previous publications with similar designs.
The antenna itself is just an antenna that works well over a very broad RF spectrum.
European civil servants are also usually banned from using AI — perhaps with the exception of Microsoft copilot. They live in a bubble where they just don’t know. This goes for most academics as well.
Ah yes. "It's so bad that people in government agencies cannot give sensitive info to US companies or blindly rely on LLMs for their decisions since nothing has ever happened when people in governments blindly trusted black boxes"
The European Commission actually has an internal AI chat platform with a selection of different models, and recommendations on which to use based on the classification level of the information that will be shared with the AI model.
Some are hosted internally (LLAMA models), other are sourced from commercial providers (Mistral, OpenAI).
I don't know of any software or services that would be banned at my university. People use all sorts of LLMs extensively.
At least in Finland also civil servants are free to use what AI services they want, given they don't put in sensitive information. Just like they can use any search engine they want.
I love em dashes. They are so much less pretentious than colons or semicolons — and they help with flow of speech. I learned that key command a couple years ago and it made me feel so smart. I’ve had my comeuppance but I’m not stopping — just a better way to write
But sunlight is essential for the cutaneous conversion of 7-dehydrocholesterol to vitamin D3, whereas ethanol serves no essential purpose, irrespective of whether one enjoys it or not.
Personally I don’t consume ethanol; but I don’t care if others do or not so long as they stay off the roads and are not piloting my flight.
I will say that when I did consume ethanol even in small quantities, my sleep was much worse than it is at baseline; and that effect only worsened as I got older.
Ethanol is food— you also don't need carbs, but they do keep you alive. More importantly, it fairly central to cultural vitality. Not essential but it plays a highly functional role. Maybe could be replaced by religion or other drugs, but short of that, the world is less vibrant without it.
"Hey Claude, can you help me create a strategy to optimize my token use so I don't run into limits so often?" --> worked for me! I had two $200 plans before and now I am cool despite all day use
I've only had to do a major token optimization once. It reduced my memory, claude.md, mcps, etc... that's usually the big issue. but of course it gets dumber without the context of the tools but smarter with the cleaner window. so you have to find your balance.
But like most challenges with claude, if you can just express them clearly, there are usually ways of optimizing further
reply