Franck Pachot
@franckpachot.bsky.social
https://dev.to/franckpachot
🥑 Developer Advocate at 🍃 MongoDB
🔸 AWS Data Hero
🅾️ Oracle Certified Master
🐘 PostgreSQL and▝▞ YugabyteDB
🥑 Developer Advocate at 🍃 MongoDB
🔸 AWS Data Hero
🅾️ Oracle Certified Master
🐘 PostgreSQL and▝▞ YugabyteDB
Meeting many friends at @doagev.bsky.social
November 18, 2025 at 3:28 PM
Meeting many friends at @doagev.bsky.social
I've never realized before that SQL*Plus can tell you when it disconnects "with complications" 🤓
November 7, 2025 at 7:15 PM
I've never realized before that SQL*Plus can tell you when it disconnects "with complications" 🤓
I enjoy staying connected with multiple database communities. Even if I share more about one today 🌱🐅, I'm happy to have feedback from others 🅾️🔸️🐘🐬 on what I share
October 30, 2025 at 9:51 AM
I enjoy staying connected with multiple database communities. Even if I share more about one today 🌱🐅, I'm happy to have feedback from others 🅾️🔸️🐘🐬 on what I share
Nice to see a public presentation about SQL Assertions by @toonkoppelaars.bsky.social at @doagev.bsky.social
my.doag.org/events/userc...
my.doag.org/events/userc...
August 2, 2025 at 7:48 AM
Nice to see a public presentation about SQL Assertions by @toonkoppelaars.bsky.social at @doagev.bsky.social
my.doag.org/events/userc...
my.doag.org/events/userc...
The level of 💩 on Medium is high. This article references non-existent versions and features, as well as "early access internal builds" of open-source software 🤣. AI is using this garbage to build the knowledge base for the next decade. Long live human experts who rely on factual information!
July 29, 2025 at 6:10 AM
The level of 💩 on Medium is high. This article references non-existent versions and features, as well as "early access internal builds" of open-source software 🤣. AI is using this garbage to build the knowledge base for the next decade. Long live human experts who rely on factual information!
With literals, what is in the array is interpreted as the arguments for the $ operator, so 5 and [5] are the same. Cf. doc 👇🏼
It's different if you pass a field (= one argument) that is an array
(I removed my previous post as it's misleading🫤)
Example with both:
mongoplayground.net/p/TfuvZsKJHW2
It's different if you pass a field (= one argument) that is an array
(I removed my previous post as it's misleading🫤)
Example with both:
mongoplayground.net/p/TfuvZsKJHW2
July 20, 2025 at 7:53 AM
With literals, what is in the array is interpreted as the arguments for the $ operator, so 5 and [5] are the same. Cf. doc 👇🏼
It's different if you pass a field (= one argument) that is an array
(I removed my previous post as it's misleading🫤)
Example with both:
mongoplayground.net/p/TfuvZsKJHW2
It's different if you pass a field (= one argument) that is an array
(I removed my previous post as it's misleading🫤)
Example with both:
mongoplayground.net/p/TfuvZsKJHW2
I'm reading E.F. Codd's paper "Further Normalization of the Data Base Relational Model" and spot the name of my database teacher at university 😀
July 3, 2025 at 6:33 AM
I'm reading E.F. Codd's paper "Further Normalization of the Data Base Relational Model" and spot the name of my database teacher at university 😀
Question for Bsky followers, (or @toonkoppelaars.bsky.social )
For the six reasons for normalizations, what does "neutral to the query statistics" mean? Is it about data distribution or access patterns? Or other? (I'll explain my guesses in🧵 )
For the six reasons for normalizations, what does "neutral to the query statistics" mean? Is it about data distribution or access patterns? Or other? (I'll explain my guesses in🧵 )
July 2, 2025 at 8:08 AM
Question for Bsky followers, (or @toonkoppelaars.bsky.social )
For the six reasons for normalizations, what does "neutral to the query statistics" mean? Is it about data distribution or access patterns? Or other? (I'll explain my guesses in🧵 )
For the six reasons for normalizations, what does "neutral to the query statistics" mean? Is it about data distribution or access patterns? Or other? (I'll explain my guesses in🧵 )
No RDBMS has implemented all of EF Codd's rules. Do you believe that AI-generated applications could benefit from them (more data independence when iterating on application code) or must be more flexible (iterating on persistence at the same pace)?
June 12, 2025 at 8:32 AM
No RDBMS has implemented all of EF Codd's rules. Do you believe that AI-generated applications could benefit from them (more data independence when iterating on application code) or must be more flexible (iterating on persistence at the same pace)?
Searchable Encryption: encrypted from the application, searchable but not decrypted in the database, even for range queries. Which database can do that?
May 13, 2025 at 9:26 AM
Searchable Encryption: encrypted from the application, searchable but not decrypted in the database, even for range queries. Which database can do that?
I got the answer (amazing colleagues). This was documented in 2015, but since there were no builds or testing on OEL, it was wiser to refer solely to the RHEL kernel.
But I'm sure you can run MongoDB on this unbreakable penguin 🐧
But I'm sure you can run MongoDB on this unbreakable penguin 🐧
April 10, 2025 at 5:49 PM
I got the answer (amazing colleagues). This was documented in 2015, but since there were no builds or testing on OEL, it was wiser to refer solely to the RHEL kernel.
But I'm sure you can run MongoDB on this unbreakable penguin 🐧
But I'm sure you can run MongoDB on this unbreakable penguin 🐧
Have you ever dreamed about tracking the scores of rejected execution plans? To identify plans that can flip when data distribution changes. It may also inform the decision to drop an index if another offers similar benefits. Here's what we can get from the #MongoDB plan cache:
March 14, 2025 at 6:49 AM
Have you ever dreamed about tracking the scores of rejected execution plans? To identify plans that can flip when data distribution changes. It may also inform the decision to drop an index if another offers similar benefits. Here's what we can get from the #MongoDB plan cache:
Had a great week visiting the MongoDB offices in Austin, and meeting new colleagues. Awesome discussions with 20 Developer Advocates in a room running through the workshops for the next Developer Days 🤩
February 22, 2025 at 1:26 PM
Had a great week visiting the MongoDB offices in Austin, and meeting new colleagues. Awesome discussions with 20 Developer Advocates in a room running through the workshops for the next Developer Days 🤩
❄️🏃 Morning run in Austin before flying back from a week of workshops with MongoDB colleagues
February 21, 2025 at 4:55 PM
❄️🏃 Morning run in Austin before flying back from a week of workshops with MongoDB colleagues
If you are at FOSDEM, see you this afternoon in the cloud-native database dev room fosdem.org/2025/schedul... (just in front of the #PostgreSQL booth) - schedule: fosdem.org/2025/schedul...
February 1, 2025 at 11:13 AM
If you are at FOSDEM, see you this afternoon in the cloud-native database dev room fosdem.org/2025/schedul... (just in front of the #PostgreSQL booth) - schedule: fosdem.org/2025/schedul...
In 2025, I will embrace change and expand my horizons, such as switching to a Mac after years of using a Windows laptop 👀
January 29, 2025 at 6:19 PM
In 2025, I will embrace change and expand my horizons, such as switching to a Mac after years of using a Windows laptop 👀
marketing vs. community
Be careful about what you read online from vendors blog.
Some companies want to be the first to post about "new features" even before release.
For your information, SPLIT/MERGE PARTITION is *not* a PostgreSQL 17 feature 🤦♂️
Be careful about what you read online from vendors blog.
Some companies want to be the first to post about "new features" even before release.
For your information, SPLIT/MERGE PARTITION is *not* a PostgreSQL 17 feature 🤦♂️
January 22, 2025 at 11:57 AM
marketing vs. community
Be careful about what you read online from vendors blog.
Some companies want to be the first to post about "new features" even before release.
For your information, SPLIT/MERGE PARTITION is *not* a PostgreSQL 17 feature 🤦♂️
Be careful about what you read online from vendors blog.
Some companies want to be the first to post about "new features" even before release.
For your information, SPLIT/MERGE PARTITION is *not* a PostgreSQL 17 feature 🤦♂️
It is intriguing to compare the execution plan of the same query in Oracle Database with MongoDB API with that of the true MongoDB 🤓
December 23, 2024 at 10:06 PM
It is intriguing to compare the execution plan of the same query in Oracle Database with MongoDB API with that of the true MongoDB 🤓
If you still use DTAB you may need to upgrade
December 23, 2024 at 7:49 AM
If you still use DTAB you may need to upgrade
I've seen a remark in the chat about truncate and rollback.
With PostgreSQL transactional DDL, both are fast.
(but an exclusive lock blocks concurrent reads and writes)
With PostgreSQL transactional DDL, both are fast.
(but an exclusive lock blocks concurrent reads and writes)
December 14, 2024 at 5:49 PM
I've seen a remark in the chat about truncate and rollback.
With PostgreSQL transactional DDL, both are fast.
(but an exclusive lock blocks concurrent reads and writes)
With PostgreSQL transactional DDL, both are fast.
(but an exclusive lock blocks concurrent reads and writes)
What do you think? Using a SQL database with those magic design patterns (no foreign keys, no joins, JSON documents). This slide is from Canva at #AWSreInvent keynote
🤔 Why not using a document database then?
🤔 Why not using a document database then?
December 5, 2024 at 5:50 PM
What do you think? Using a SQL database with those magic design patterns (no foreign keys, no joins, JSON documents). This slide is from Canva at #AWSreInvent keynote
🤔 Why not using a document database then?
🤔 Why not using a document database then?
But it was a good question and I should blog about it.
Oracle Database also uses "strong" for this exact same thing (except that it uses SCN sync rather than bounded clocks, so higher latency to be strongly consistent)
docs.oracle.com/en/database/...
Oracle Database also uses "strong" for this exact same thing (except that it uses SCN sync rather than bounded clocks, so higher latency to be strongly consistent)
docs.oracle.com/en/database/...
November 29, 2024 at 11:05 AM
But it was a good question and I should blog about it.
Oracle Database also uses "strong" for this exact same thing (except that it uses SCN sync rather than bounded clocks, so higher latency to be strongly consistent)
docs.oracle.com/en/database/...
Oracle Database also uses "strong" for this exact same thing (except that it uses SCN sync rather than bounded clocks, so higher latency to be strongly consistent)
docs.oracle.com/en/database/...
Let's get some engagement in bsky (/cc @oraesque.bsky.social) 🤓
Full room for @ludovico.bsky.social session at #doag2024:
Is PostgreSQL Catching Up to Oracle Database?
Full room for @ludovico.bsky.social session at #doag2024:
Is PostgreSQL Catching Up to Oracle Database?
November 20, 2024 at 2:32 PM
Let's get some engagement in bsky (/cc @oraesque.bsky.social) 🤓
Full room for @ludovico.bsky.social session at #doag2024:
Is PostgreSQL Catching Up to Oracle Database?
Full room for @ludovico.bsky.social session at #doag2024:
Is PostgreSQL Catching Up to Oracle Database?
The best time precision to run a distributed database with high-performance ACID
🥇 #aws #timesync #ptp #clockbound
🥇 #aws #timesync #ptp #clockbound
November 15, 2024 at 11:44 PM
The best time precision to run a distributed database with high-performance ACID
🥇 #aws #timesync #ptp #clockbound
🥇 #aws #timesync #ptp #clockbound