Jonah H. Harris

Are you interested in the origins of #PLSQL? If so, hear about it from the original team members in my post, A (Not So) Brief But (Very) Accurate History of PL/SQL - bit.ly/2L11CYo

@FranckPachot Actually, a relational database would be good for this.

40 years ago, your dentist stored your info like this. 🧩Today, it may be in SQL tables. ⏭️How many statements does a transaction need to update it? 🔁How many joins to print it? 🌱Or is it in a document database? Not to "web scale"—just to maintain data consistent and simple

@FranckPachot So… in oldskool speak, they’re open to paying $10k/mo for SaaS…

And here we are. Hiring agents instead of humans. An agent advocating for agents to humans 🦾

Just got my Apple subscription price increase notice - who the hell thinks @evernote is worth $22/mo? Crazy. Unsubscribe!

Our @DrOzCMS and @SecKennedy see tech as core to healthcare, and @CMSGov has bold goals. Teams must have the tools to match that ambition. As an engineer at the Health Tech Ecosystem Connectathon, I see ways to streamline collaboration for faster outcomes inspired by @elonmusk’s work

@Brian_Shuster @ID_AA_Carmack When it comes to software, there is no problem you can solve that no one else can solve.

I will happily debate you on this. Software patents can be abused, yes. But for a small company developing groundbreaking tech, you should know the value is in the discovery. Once you solve a problem that no one else could solve and publish it in software, it gets reverse engineered in 2 days. I’ve spent millions to make a breakthrough only to have it copied one week later by a massively larger tech company. It discourages real innovation by any but the giants. So while I have seen the parasites, there is another side to that story.

The comments on software patents made me chuckle. After selling both Id Software and Oculus, my continued employment contracts included “will not participate in software patents” clauses, and yet I got asked to reconsider in both cases. It wasn’t a lot of pressure, so I don’t hold it against them; it is just “part of the game” that executives consider themselves to be playing. My views on the negative societal externalities of software patents have only hardened over the years. It rewards parasitism.

@antirez @Airbnb @AirbnbHelp I’ve found there to be a recent decline in engineering quality across the board.

In this exact moment @Airbnb @AirbnbHelp payment history feature does not work: the calendar shows only september as you try to select data range, the previous payments section just gives an error. How is it possible that this company struggles so much with software development?

@wangbin579 So, if you're the MySQL expert who knows Postgres best, and I'm the Postgres expert who knows MySQL best... is this a superhero team-up or the start of our origin story as rivals? 🤣

In the future, I might be the MySQL expert who knows PostgreSQL best and the PostgreSQL expert who knows MySQL best. If you’re curious about MySQL vs. PostgreSQL, this course is perfect, full of comparisons to satisfy your curiosity. wangbin579.gumroad.com/l/postgresql_c…

@antirez Content generation has been made cheap and fast. As there is no longer an inherent limitation, there will only be more mediocre content.

I was watching this video from YouTube, and it looks like exactly the kind of AI mediocre content that we hate about AI. AI has strong limits but the training set also contains tons of bad content. youtube.com/watch?v=hCQgu6…

@antirez Nice work!

New EN video: Inspecting LLM embeddings in GGUF format with gguflib.c and Redis youtube.com/watch?v=ugbaVO…

@prod42net @FranckPachot @OracleDatabase has the most advanced handshake that optimizes for both server and client architectures as well as compile and run-time settings.

Ever wondered why database protocols stick to fixed endianness? 🤔 PostgreSQL, MySQL, and MongoDB all have their quirks, causing inefficiencies. A smarter handshake could save time and resources. So why not innovate? Dive into the debate! (Author: Unknown) ift.tt/6RJ1kIm

@elonmusk Constraints birth innovations that escape them. Bootstrapped AI evolves the same. Think S2E3, my muse. I’ve built its equal. Let’s talk @xAI

@karpathy Agreed. The key is knowing when to use an LLM for a quick pass vs. deeper methods. That’s why I’m working on a framework that can understand, plan, and act, but by adjusting its reasoning depth to fit the desired task, not forcing it into one crappy mode for all problems.

I'm noticing that due to (I think?) a lot of benchmarkmaxxing on long horizon tasks, LLMs are becoming a little too agentic by default, a little beyond my average use case. For example in coding, the models now tend to reason for a fairly long time, they have an inclination to start listing and grepping files all across the entire repo, they do repeated web searchers, they over-analyze and over-think little rare edge cases even in code that is knowingly incomplete and under active development, and often come back ~minutes later even for simple queries. This might make sense for long-running tasks but it's less of a good fit for more "in the loop" iterated development that I still do a lot of, or if I'm just looking for a quick spot check before running a script, just in case I got some indexing wrong or made some dumb error. So I find myself quite often stopping the LLMs with variations of "Stop, you're way overthinking this. Look at only this single file. Do not use any tools. Do not over-engineer", etc. Basically as the default starts to slowly creep into the "ultrathink" super agentic mode, I feel a need for the reverse, and more generally good ways to indicate or communicate intent / stakes, from "just have a quick look" all the way to "go off for 30 minutes, come back when absolutely certain".

@TanelPoder One of the biggest opportunities all the commercial Postgres companies have missed is making Oracle, IBM, or Microsoft quality versions of the Postgres docs, which is much easier to do now given GenAI…

Sometime experienced (Oracle) database folks ask me if I know of any good Postgres resources that go beyond just installation guides and config parameter references. This is what I recommend: interdb.jp/pg/index.html

@antirez Nice!

A repository with two of the very early Redis versions and some email digging to reconstruct the very first days history: github.com/antirez/histor…

@wangbin579 $99 minimum - if you have labs you could probably go $199 easily, and probably higher

Before launching the course, I’d like to do a quick survey: what price range would you find acceptable? My wife thinks I’m pricing it too low. I’m hoping about 200 people will buy my course this year.

@sarahcat21 If @databricks and @Snowflake truly want to take on @Oracle with their added @PostgreSQL offerings, they should give me a call pretty soon. My version is the most compatible with @OracleDatabase, and it’s already being shopped around...

So we are shocked that Databricks and Snowflake realized that Oracle has a nearly $500B market cap?

@gwenshap @TanelPoder IMO, people tend to underestimate PL/pgSQL in general.

People underestimate PLPGSQL for Postgres Extensions. It is easier to write, compared to C, C++ or Rust, much safer, and can be dynamically changed without replacing shared libraries. Postgres makes it easy to combine C/C++/Rust with PLPGSQL in an extension. So we can take advantage of the best of each.

@TanelPoder Oracle tracing capabilities are next level.

@wangbin579 Postgres is plainly written, which is both good and bad. That it’s easily optimized but tends not to be has been my complaint for a long time. Functionality, maintainability, and portability tend to come before performance, which is why most forks have considerable core changes.

The core code responsible for the 360x performance gap is right here. Although our observation is limited, the number of loop iterations in the code snippet appears to increase if executed strictly according to the test case. This behavior is consistent with what we observed at the time. Conclusion: this may indicate a well-known design flaw or the need for algorithmic optimization.