I wish Google had spent their efforts building stuff like this into Google Sheets instead of launching a Gemini tab that follows you everywhere and says "sorry I can't help with that" if you try to use it
When you look into the data structure and organization that is underneath Google Sheets, and if you have any experience with LLMs and structured outputs, Google shit in their sheets as far as having an overly complex internal representation for a spreadsheet. Their data representation needs to be translated into something far easier to work with for machine learning models to work with that data, without a massive amount of filtering just to reach the spreadsheet data itself.
> When you look into the data structure and organization that is underneath Google Sheets
I don't use Google Sheets very often, so maybe this is obvious to someone that does. But, how do you look at the data structure and organization underneath Google Sheets? Are you an employee or is this available to the general public? Through an api?
Look at their API and the data structures it uses, the relationships between them. It's more complex than necessary, ridiculously so, it feels like a student's learning project they abandoned out of boredom. API consistency, what's that?
Believe it or not, we already do, it is just not obvious and so far, I've only exchanged info with very few that realize this, and how to access that capability.
I think if you trained a MLM to replace certain tokens in a CSV with formulas, you can accomplish that without worrying about the internal representation of a spreadsheet.
My point is that no training is necessary, none at all. The major LLMs already know the internal data structures and APIs of open source spreadsheets, because that information is in their training data, all open source software has their source code in major LLM training data. The trick is accessing that knowledge.
Not so much of a "trick" as it is understanding how to create a context the retrieves/generates from the more/most accurate training data.
In the case of open source software, a specific open source app or library or API, often all it takes is to reference a library or API by name, use a few terms specific to that library or service, and you'll get an LLM context that retrieves/generates from that area of knowledge. That context knows information about the library/API that is useful for integration and will tell you all about them just by explaining what you're trying to do. Note this is essentially conversational R&D, which is then backed up by verifying the data structures, library calls and/or APIs are indeed as they are being discussed. I include verifying checks during the conversation, rather than build a strategy and then check if what was discussed is even possible.
If an integration is possible, if the above conversation yields useful data and function/API calls (which they tend to do), then it's just a matter of a well structured prompt that's probably generating structured output. The end result is a process where a useful library/API call retrieves essential data of some sort that the user can ask questions about, modify, and if modified the prompt generates the data that is then fed back into the library/API.
That would have been the future had OpenAI and the other company not ventured to offer AI search power in a chat window. small incremental AI implementations would have been the norm. Like auto reply in email. Or email summary at end of day.
I think it's hard to advise a process as it's quite personal. Some people write down bullet points and expand, others write and rewrite until it's what they want. I think we're far from having enough research to know if one way is 'better' than another or not, so for now we mainly just advise writing as much as possible (https://styleguide.ritza.co/improving-your-writing/how-do-I-...) and figuring out what works for you.
Structure is really hard so I don't have any easy answers for you there either, but https://styleguide.ritza.co/ritza%27s-writing-rules/Style/#k... talks a bit about how the writers job is to go through exactly what you're describing so that the reader doesn't have to.
I'd really recommend Matt Levine's various columns on this. Here's one [0] - it's paywalled, but if you sub to the newsletter you get the full texts for free. Not sure if you can get historical ones, but I'm guessing he'll talk about it again in the next issue this or next week. I added the excerpt here [1] too. Very entertaining and informative take on most things finance.
I encountered this weird issue yesterday[0] (forum post is not mine but I also could not log in with Safari). It made me think I had somehow forgotten my master password for a while.
If you'd told people 50 years ago that "literacy" would decline to the level of the reddit poster's style of writing, with poor sentence structure and words in ALL CAPS, they'd be pretty shocked. Now it's just normal communication.
Older people are shocked that young people can't do long division, or easily add a column of numbers mentally.
I'm not saying LLMs won't negatively influence the next generation, but the next generation has always been the first to use and abuse new technology and we are all still alive and (most) bridges and airplanes stay up and maybe we will even put people on the moon again some day.
7:05 shows the lever for a Brother typewriter in this video and a bit before that shows a different lever for heavyness (although I don't know if that would do bold): https://youtube.com/watch?v=d_dTu6T87Cg&t=6m50s
And AFAIR you could somewhat do bold by overtyping a few times (backspace, retype, repeat) or by hitting keys harder.
Plus many early home computers only had UPPERCASE (e.g. initial Apple ][) and were monochrome (so inverse was often a highlight thing instead of using colour).
Yes, agreed. Actually the very existence of the problem described in the original post is an indication that the work being assigned is ultimately useless. If the teacher can't reliably distinguish between a student's answer and an AI answer then what's the point of giving that assignment? Kids with street smarts or a hacker mentality are always going to figure out the easiest path from A to B, and it's arguable if we should even be discouraging that—it's quite an important skill in real life.
We're at a major transition point and I don't envy teachers who need to figure out ways to adjust to it, but giving the same exact assignments clearly isn't going to cut it. They should instead try to come up with assignments where an AI may be a helpful assistant but isn't able to complete 100% of the task to a high standard. Then evaluate the result regardless of how the student got there. Presentations, discussion-based evaluation, and projects with a physical component are all examples.
I remember when I was a kid, some teachers would assign research papers and then tell us we weren't allowed to use the internet; we needed to go to the school library and use physical books instead. This holy war of teachers vs. chatgpt is likely going to seem similar in hindsight.
>If you'd told people 50 years ago that "literacy" would decline to the level of the reddit poster's style of writing, with poor sentence structure and words in ALL CAPS, they'd be pretty shocked. Now it's just normal communication.
I would argue that the people who would have written like 50 years ago just never had the opportunity or audience that people do today. Except maybe in a Letter to the Editor of a newspaper or something.
So many more people are expressing themselves in public than ever were able to 50 years ago so I think it just seems that people have become less literate than they were. So many YouTube comments and Facebook posts that never would have occurred 50 years ago.
It's a democratization of the broadcasting of expression more than literacy decline.
yeah big +1 to this. People are obsessed with automation, but doing a bit of manual work for stuff like budgeting and project management actually really helps cement it mentally I think (a bit like hand writing lecture notes, etc).
Spreadsheet + Pivot table is all you need in many cases.
There's a bunch of good advice and resources here but in the end sales and marketing is not that different from coding.
To do it well, you have to do it a lot. And you're going to suck at it in the beginning.
The good news is that if you're good at coding then you already know how to be persistent and self-critical and disciplined. You'll see what is working and what isn't and do more of the former and less of the latter until you get good.
I know that's not particularly useful in terms of "how do I speed run this" but I think it's something that most devs don't want to fully grok.
Pixel phones run song identification constantly now. They have a local database of the top 1000 (?) most popular songs. It has negligible impact on battery life.
Not saying I agree that 'phones are listening to show us ads', but technically we have the capability for that to happen (sampling audio every X intervals and matching against a local database of keywords)
Add at least two zeros to your number. Pixel phones can detect the top 11k songs while being offline (it used to be more). The fingerprint database for this is around 500 MB in size.
I think it is very easy to sneak a few (thousand) extra fingerprints in this database and do all kinds of tracking with it. All while the green microphone icon is disabled.
For argument’s sake, let’s be generous and stipulate your phone is listening for 11k keywords to serve you ads.
Why would “pool fencing” take up one of those valuable keyword slots on everyone’s phone?
And you’re going to see way less than 11k ads per day. Why would the ad server prioritize serving an ad for pool fencing (a phrase said once) over all the far more common topics a person talks about in a typical day, like movies, TV shows, food and drink, clothes, cars, consumer electronics, music, etc?
"look into" is a much more likely trigger, then send the 30 seconds before and after to a server for more analysis. "buying" could be another. It's not like it would be that hard. Especially with some of the pretty good vocal compression for audio. It would be a small blip on a modern connection, even wireless.
I'm not saying it is or isn't happening but it wouldn't be hard.
Your argument plays with the idea that the phone listening stuff is the only source of information for the ad networks. But it would be much more complex. It would be only one of many signals, that are used to serve the consumer the right advertisement in the right moment. So it doesn't need to have the exact phrase "pool fencing" in the database. It just need to detect that something about pools, or swimming, etc. was talked about. Since Google has thousands of signals and statistics (like browsing history, current location, the other smartphones that are near, and those histories etc.) about this person, it can sell the ad space to "pool fencing" and expect a high click through rate.
Selling ads is a bit like the current LLMs. It's just a stochastic parrot, that hallucinates stuff. But the stuff is often that advertisement that brings in the most money.
The self-expressed goal of this kind of test is to pick a phrase or topic that is so random that it escapes that person's existing ad data profile. As the comment above said, "He doesn't own a swimming pool, doesn't want to, and has never expressed any desire to."
So showing that person an ad for pool fencing is a complete waste; they're never going to click it. If that's what an alleged audio targeting system does, it would make the ad network less profitable than just using the data they already have. So why would anyone build it that way?
I dont know if phones listen to us to serve ads, but 11K is a decent vocab. Most adults have a vocab of 20K. Therefore I could imagine it including the words "pool" and "fencing".
Now Playing only has to sample for a few seconds every few minutes when the phone is powered on for other reasons (like to participate in cellular check-ins). This is because a song is typically several minutes long and you only have to fingerprint for a few seconds. It doesn't matter which few seconds. It's not continuously listening, so it's not the same thing at all.
Cashflow problems regularly kill business. Now you need a much higher cash buffer than before (in parent example where is that 800k coming from?). Combine with much tighter VC market and there are definitely many startups closing up shop because of this.
It's interesting that the idea that 'banks should refuse to do business with potential terrorists' is something that most people I know support, or don't care about.
Until it happens to them. Then suddenly they get quite passionate about the state that the world is currently in.
I don't have data on how many people are being debanked, but anecdotally it definitely seems to be some kind of 'line go up' number, and I wonder if cumulatively it will get to some kind of critical mass where we have a large enough percentage of people make enough noise that the laws are reigned in, or if they'll be 'that group of weirdos' forever.
> It's interesting that the idea that 'banks should refuse to do business with potential terrorists' is something that most people I know support, or don't care about.
I tell them about the "terrorists" truckers who went on strike in Canada and who got debanked for going on strike.
> (I almost got debanked once -- I had my account re-instated after many weeks of extreme stress ...
You tell me. I had to fill nearly a PDF with nearly 50 pages justifying where and how I bought groceries and how I put tank in my car because they somehow thought I may not be living in the country I said I was living in.
Same things: weeks of insane stress but, eventually, they said things were OK.
You're not a weirdo. People don't realize we live in a Brave New World, complete with secret ESG score (assigned by banks to customers), etc.
Many simply have blind love for the state and they'll never question any rule made by the state.
So they'll state: "But banks are only following the law!". Which is precisely the problem: such laws shouldn't exist.
I live in a country where it's all about banking finance and most don't realize they don't produce anything at all. It's all about KYC/AML/compliance/SARs/fiscal lawyers/etc. A big, gigantic, void.
While you see SpaceX going from their Raptor 1 design to their Raptor 3 design, using the "idiot index", where things become more efficient and way cheaper, bureaucracy (and their banking following dogs) works the other way: everything becomes less efficient and more costly.
all of the laws meant for terrorists are being applied to normal citizens at an astounding rate, things like border crossings and these unconstitutional laws like the patriot act have really undermined people's rights
A few years ago in Russia, when protesting was still barely legal and possible, prominent opposition leaders were 'fined' for ~1M USD each for damages to businesses what happened due to some protest activity. Banks simply deduced every account balance by -50000000,00 RUB and have a nice day.
Whoever thinks that 'but that was bad, evil Putin's government and our nice, good government would never do such a thing' should think again. The government should not have such power at all, neither directly, nor via bank proxy that would 'independently' refuse you service.
When you happily cheer that in EU you can't pay with 500 EUR bill (because tax evasion, crime, etc!!!), you are basically helping the government create means to financially strangle anyone the government doesn't like. One day it might strangle you.
How could they not have the power? All power the banks have is directly derived from their ability to operate as they do within a currency regime that is directly controlled by the state. Best case is it's your state; plenty operate in a system controlled by a foreign state.
People in countries that are still free should take it away from them and legally protect direct money transfer between people. Cash, anonymous cryptocurrencies,etc, so that no middleman can ever take someone's money.
The problem I have with this debate is that it's targetting the wrong thing. Of course tyrannical states are a problem, but they are a problem because they can ride roughshod over your rights. Having a legal right to money transfers or whatever is only worth the value of the legal right. If the legal right is ignored or revoked, you're in no better a situation than never having the right in the first place.
The problem is a political one - if you want your state to be not shit, then you need to make your state not shit, not campaign for some magic work around to the state.
reply