2,585 karma · joined December 5, 2011
Have you dealt with people paying for ads? They want your rapt attention, not some possibly subliminal suggestion. The "good" thing about chat AIs is that they can, with a little work, make it nearly impossible to block. You want to use it for free, you get ads.
But ads can be put in AI.
This is a cross-join. It is the equivalent of a nested loop in programing and almost never what you want to do in a database unless you like watching high-combination data spewing out and CPUs burning. It gives you every combination of items in each table you name.
What you want is to find something in two or more of those tables that can be linked together somehow using (very likely!) "INNER JOINS". Look into examples or tutorials of those on the internet and use what you find to work through the tables on the "investigations". Luckily, the SQL Noir site tells you exactly the steps to take in the first case. You need to find a link somehow between the crime_scene and suspects tables and then another one between the matching suspects from the suspects table and the interviews table.
Note: there is no direct easy link between the crime_scene and suspects tables. You will have to use your investigation skills to trim the suspects list down.
Interesting, because the hero image is a Mac App screenshot.
I will give you a real-world example. My girlfriend works for the government and they have an initiative to do outreach to people in the local community and let them know of all the funded help available to people for different situations.
However, they can't go to all 30k+ households at once and even in targeted areas (which is the way it works now) it takes a very long time to get to everyone.
The thinking about a recommendation is that perhaps they go to a home and let the person know about all the free programs available and leave a list with phone numbers. And this person might not need that information at the moment, but maybe they know someone in their family or from church or elsewhere that could use it and they let them know.
I do understand that if you help people who have lost their home due to fire it is silly to do a survey and ask them if they would or did tell someone else in case their home burns down. However, it isn't across the board silly even or not for profit cases.
>All root servers have a dedicated 1 GBit uplink by default and with it unlimited traffic.
>Inclusive monthly traffic for servers with 10G uplink is 20TB. There is no bandwidth limitation. We will charge € 1/TB for overusage.
So it sounds like it depends. I have used them for (I'm guessing) 20 years and have never had a network problem with them or a surprise charge. Of course I mostly worked in the low double digit terabytes. But have had servers with them that handled millions of requests per day with zero problems.
>3. Hypothetical answer generation from a query using an LLM, and then using that hypothetical answer to query for embeddings works really well.
What sort of performance are you getting in production with this one? The other two are basically solved for performance and RAG in general if it is related to a known and pre-processed corpus but I am having trouble thinking of how you don't get a hit with #3.
Although I create software, my degree is in finance. Calculus is everywhere, and I mean everywhere, in finance after you get past the basic intro courses, for school at least. However, in real life I have only used it not that often and mostly related to machine learning and optimization. Even then, I didn't do it by hand, I just knew what to do and let computers calculate the rest.
Apps that use LLMs or apps made with LLMs? In either case can you share them?
>which tells me you aren’t working with very creative and competent people
> In my professional network beyond where I work now I know at least a dozen people who have successful commercial applications of LLMs.
Apps that use LLMs or apps made with LLMs? In either case can you share them?
No one doubts that you can integrate LLMs into an application workflow and get some benefits in certain cases. That has been what the excitement and promise was about all along. They have a demonstrated ability to wrangle, extract, and transform data (mostly correctly) and generate patterns from data and prompts (hit and miss, usually with a lot of human involvement). All of which can be powerful. But outside of textual or visual chatbots or CRUD apps, no one wants to "put up or shut" a solid example that the top management of an existing company would sign off on. Only stories about awesome examples they and their friends are working on ... which often turn out to be CRUD apps or textual or visual chatbots. One notable standout is generative image apps can be quite good in certain circumstances.
So, since you seem to have a real interest and actual examples of this, I am curious to see some that real companies would gamble that company on. And I don't mean some quixotic startup, I mean a company making real money now with customers that is confident on that app to the point they are willing to risk big. Because that last part is what companies do with other (non LLM) apps. I also know that people aren't perfect and wouldn't expect an LLM to be, just want to make sure I am not missing something.
I think this is still common. People like to think a son or daughter will govern just like a parent. But it sometimes doesn't work that way.
I wouldn't bet on that! I can see legislation to require this for many reasons ... related to intellectual property, cheating, detecting the root of hate-speech or harassment, "stealing" from employers by not performing work or putting them at legal risk, "stealing" from artists by duplicating their style, political speech that can not be traced (it could be from a bad actor!), tracking down generated revenge porn (or much worse!), tracking down people using LLMs to grift the elderly, and on and on. Why, if you are not using a watermarked LLM, it could be an op by Russia, China, or Iran! In fact, part of the legislation could be a requirement of social media or office tools or government tools or political tools or educational tools to check for a watermark and not work if an approved one is not found. Ideally this list will be private, because you want companies to be able to automate away workers, and do the least possible for customers, you just want to make sure you're doing it above-board, you see.
>And there is no good reason for a provider to watermark - they aren't helping the customer.
No one cares about customers, they care about money. And you know what helps make a lot of money? A legally defined moat for yourself and a couple of others that blocks anyone else.
>They'd be helping some other party who isn't paying them.
Yes! That party is themselves!
They are/can be anyway. I had data with 50,000 dimensions, which after applying dimensionality reduction techniques got it "down to" around 300! ANN worked very well on those vectors. This was prior to the glut of vector dbs we have available now, so it was all in-memory and used direct library calls to find neighbors.
If measured by $$$$$ clickhouse certainly has more installations.
Would mind sharing those apps to view. Not the code, just the apps. I have a suspicion about the bi-modal experience.
And frequently from connected and/or rich families. That can be very useful in success, it turns out.
Yeah, I'm not sure I agree with that. Both of these are some of the most eye-popping books I have ever read related to math and computing [0] [1] (note, both pdf files) Both are well-written (by the guy that created both APL and J) in extremely approachable language and the introduction of J into the mix is almost obvious compared to doing some of the common operations in any other non-array language. These two documents and Joe Armstrong's PhD thesis were ones that really made me "think different" about programming.
[0]https://www.cbp.gov/newsroom/stats/southwest-land-border-enc...