Ruben Taelman on LinkedIn: KNoWS Query Research - Progress Update (2024-09-24) (2025)

A useful insight https://lnkd.in/dfKvBpQf

3

I want to compare the "performance" of different vector embedding providers & dimensions using pgVector within a webapp. What would be the best strategy:a) Database per providerb) Table per dimension & providerc) Table column per dimension & providerd) One big column to hold all dimensionse) Use a different vector database :) Target is to have a similar table as below but per embedding provider & dimension. Suggestions welcome!

2

1 Comment

Is your website too slow?Learn how to implement caching when fetching data using React and RTK Query!

9

1 Comment

🔗 𝐔𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝𝐢𝐧𝐠 𝐔𝐑𝐋 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 🌐Ever wondered how URLs are structured? Here’s a quick breakdown:𝐒𝐜𝐡𝐞𝐦𝐞/𝐏𝐫𝐨𝐭𝐨𝐜𝐨𝐥 : Specifies how to access the resource (e.g., https://).𝐇𝐨𝐬𝐭/𝐃𝐨𝐦𝐚𝐢𝐧 : The server's address (e.g., www.example.com).𝐏𝐨𝐫𝐭 (Optional): The port number on the server (e.g., :8080).𝐏𝐚𝐭𝐡 : The specific resource or page (e.g., /folder/file.html).𝐐𝐮𝐞𝐫𝐲 𝐒𝐭𝐫𝐢𝐧𝐠 (Optional): Additional data for the resource, starting with a ? (e.g., ?id=123).𝐅𝐫𝐚𝐠𝐦𝐞𝐧𝐭/𝐀𝐧𝐜𝐡𝐨𝐫 (Optional): Points to a specific section, starting with # (e.g., #section2).💡 Example: https://www.linked.com#TechTips #WebDevelopment #URLStructure

5

P.S.: Understanding the meaning of specific HTTP response status code is important. Image Source: MDN Web Docs: https://lnkd.in/gge_D975

2

New Post: Compare the Numbers of Different Types

4

New Post: Convert a Map to a Spring MultiValueMap

19

𝟴 𝗽𝗼𝗶𝗻𝘁𝘀 𝘁𝗼 𝘂𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱 𝘁𝗵𝗲 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗯𝗲𝘁𝘄𝗲𝗲𝗻 𝗛𝗧𝗧𝗣/𝟭.𝟭 𝗮𝗻𝗱 𝗛𝗧𝗧𝗣𝟮[1] HTTP or Hypertext Transfer Protocol is used to transfer data between client and server over the web.[2] It started as a single request-and-response protocol. For each cycle, a TCP connection was opened and closed.[3] HTTP/1.1 made Persistent Connections the default and eliminated the overhead of closing the connections for every request.[4] HTTP/1.1 also introduced the concept of Request Pipelining.[5] Despite these improvements, HTTP/1.1 had fundamental performance problems such as Head-of-Line blocking and limited connections per domain.[6] Also, TCP needs a 3-way handshake and follows the slow-start algorithm resulting in more problems.[7] As the web grew, HTTP/2 was launched to address the problems with HTTP/1.1[8] HTTP/2 brought some cool improvements such as:✅ Binary Framing Layer✅ Request and Response Multiplexing via Streams✅ Stream Prioritization✅ Server Push✅ Header Compression with HPACKBut all of these points are just a glimpse into the amazing technical details.I'm sure you'd want to know more about the HTTP/1.1 and HTTP/2 features.𝗜'𝘃𝗲 𝘄𝗿𝗶𝘁𝘁𝗲𝗻 𝗮 𝗱𝗲𝘁𝗮𝗶𝗹𝗲𝗱 𝗮𝗿𝘁𝗶𝗰𝗹𝗲 𝗼𝗻 𝗛𝗧𝗧𝗣/𝟭.𝟭 𝗮𝗻𝗱 𝗛𝗧𝗧𝗣𝟮 𝗮𝗹𝗼𝗻𝗴 𝘄𝗶𝘁𝗵 𝘀𝗼𝗺𝗲 𝗴𝗿𝗲𝗮𝘁 𝘃𝗶𝘀𝘂𝗮𝗹𝘀 𝘁𝗼 𝗲𝘅𝗽𝗹𝗮𝗶𝗻 𝘁𝗵𝗶𝗻𝗴𝘀.Here's the link:https://lnkd.in/gkyCc5XA

46

3 Comments

You sometimes need to step into the dark and grab a leak or two from the Tor Onion sites for analysis. The problem is it is slow as hell and the files are usually gigabytes of data long, so... I built a tool to help you just with that. Make sure to read the warning. This approach has some drawbacks.https://lnkd.in/ewJ9j3Yr#tor #utility #download

3

New Post: How to Intercept a Request and Add Headers in WebFlux

7