chuckcode

Would like to see the latency and cost of parsing entire 10M context before throwing out the RAG stack which is relatively cheap and fast.

Thanks for the very helpful and detailed reply!

I agree, but as aleksiy123 suggests there is an additional complexity burden and it is a long journey to teach users to make use of a new technology. I think a lot of "advanced" features get de-prioritized as not many people use them and it seems like resources could be better spent helping the masses. I think that the importance of "advanced" features is often under rated by traditional engagement models. Wikipedia is a great example of where less than 1% of users click on the edit button, but that 1% adds all the value for the other 99%.

Do you think part of this is that Netflix has assumed zero effort from user model? My experience has been that Netflix does an ok job of recommendations, but fails at overall discovery experience. There is no way for me to drive or view content from different angles easily. I end up googling for expert opinions or hitting up rotten tomatoes to get better reviews. Netflix knows a ton about me and their content, but seems to do a poor job of making their content browseable/discoverable overall. I do like their "more like this" feature where I can see similar titles.

Author is making an argument against any blockchain or distributed ledger. To quote from article. "Any application that could be done on a blockchain could be better done on a centralized database. Except crime."

I'd like to see people look past the noise of cryptocurrencies to see how important digital trust and new applications of cryptography will be as we try to scale the ability of humans to work together effectively at scale.

I'm a little surprised that the author is missing the critical innovation of crypto which is digital trust and observability. Sure it is easy to make an argument that one cryptocurrency or another is a bubble, but don't underestimate the importance of being able to distribute work and verify trust at scale.

Just look at how git has transformed software development by mapping code to a hash. Or how DNS + SSL has transformed how people trust and transact online. Is the scalable future one where people and organizations trust their data to the cloud or other 3rd parties with no way to verify integrity?

Do people think that the future of human agreements is signatures on little pieces of paper managed by courts and lawyers? Personally I think it will be digital. Given it is digital, do you think there will be some "centralized database" run by government or a commercial entity that can be trusted as single point of failure? Personally I sure hope that there is some way to distribute and verify data integrity even if it isn't full blown proof of work. I'd sure prefer something more like git where I can see if two branches are the same even from different sources.

Thanks for this comment, sums up exactly my experience with python. Do you have a pointer to a larger write up of these issues? I really enjoy python, but agree with you that it isn't well suited for performance or scale necessarily. I find that people really get attached to it though and I'd like to have a good reference for them to help explain why issues like interpreter performance, GIL and typing make a difference on large projects even if not their pet ones.

Great point. I'm not one to over optimize, but seems like parsing messages for internet sized apps is worth spending a little effort to save energy and environment.

Don't get distracted by the click bait title. Effect size should be captured by statistical significance (larger effects are less likely to happen by chance). Author is really complaining that the original study didn't report enough data to check their analysis or do alternative analysis methods. Better title for article would be "Hard to peer review when you don't share the data"

I see a lot of suspicion in thread below, which I very much understand.

I'd like to take a minute though to express my frustration with the banks that refuse to supply any sort of limited APIs. How is it 2021 and I still can't give my tax person read only access to a specific year of transactions? Plaid and others trust issue would be so much easier if the banks had any sort of control over sharing aside from none or authorized to do anything.

+1 to the reality that most productivity tips come off as hopelessly naive to the reality of life with kids, sick parents/spouses, customer demands, etc.

Few things that have helped me:

- Have something like google calendar that reminds you so don't forget something important. Avoid turning small tasks into huge ones by missing a critical deadline

- Change the game when it is stacked against you. Most productivity advice is written about what can you do alone in the current situation (game) to improve, when changing situations could make things much better. I see too many people get stuck in bad relationship, bad job, etc. trying to make it work. Find places where the tide is rising and raising all the boats rather than working against you.

- Be kind and realistic to yourself. Look around, life is challenging and nobody wins everthing. Coach yourself the same way you would coach your kids with kindness and empathy and setting them up for success rather than something impossible.

(edit formatting)