2026-04-28 Tuesday - Thoughts on Date Handling

 

[image credit: Orca on pixabay dot com]

(this post was prompted by a question posted on LinkedIn by Robin Moffat  

A placeholder for organizing my thoughts on date handling...

Benefits of Using Native Date Types:

• Data Integrity: Native types like DATE, DATETIME, or TIMESTAMP enforce correct formatting and valid dates (e.g., preventing a date like February 30th).
• Performance: Native date types are typically stored as compact numeric values. This makes sorting and indexing significantly faster compared to string comparisons.
• Date Arithmetic: Using native types may allows you to easily perform operations like adding days, finding the difference between two moments, or extracting components (year, month, day) without complex string parsing.
• Standardization: Most relational databases and programming languages provide robust libraries for handling native dates, ensuring consistent behavior across different systems.

 

Best Practices: 

• Always use UTC: Store all timestamps in Coordinated Universal Time (UTC) to avoid complexities with time zones and daylight saving changes.
• Convert at the Edge: Only convert dates to strings (formatted for a specific locale) when they need to be displayed to a human in the presentation tier.

 

When Storing/Transmitting Dates as Strings May Be Necessary:  

• External API Constraints: When an external service only accepts or provides data in a specific string format (e.g., JSON which lacks a native date type).
• Historical Edge Cases: Some database DATE types do not support dates in the distant past. For example: 
• SQL Server's DATETIME limit of 1753-01-01
• MySQL TIMESTAMP restrictions starting at 1970
• Lack of BCE (Before Common Era) support

 

Standard-based Date Format:

•  Use ISO 8601 (i.e., YYYY-MM-DD)
• https://en.wikipedia.org/wiki/ISO_8601 
• ISO 8601-2:2019 Standard

 

Native Date Storage Considerations:  

• C#
• ... TO-DO, add entries
• C/C++ 
• ... TO-DO, add entries 
• Java
• ... TO-DO, add entries 
• Python  
• ... TO-DO, add entries 
• Rust 
• ... TO-DO, add entries 

 

Other misc. articles/posts that may be of possible interest: 

• Software Engineering Stackexchange 
• Why not return dates as a string from the database?  
• dev.to 
• Time in Software, Done Right
• Noah Sussman (LinkedIn):
• infiniteundo.com
• http://falsehoodsabouttime.com 
• Falsehoods programmers believe about time 
• Also see: 
• DOI 10.5281/zenodo.17070517 
• https://www.linkedin.com/feed/update/urn:li:activity:7453872223800430593/ 
• https://www.linkedin.com/feed/update/urn:li:activity:7391815745963261952/ 
• https://infinite-undo.medium.com/falsehoods-programmers-believe-about-time-0033203f0410 
• More falsehoods programmers believe about time; “wisdom of the crowd” edition 
• Iain MacDonald (LinkedIn,  Github)
• RFC 3339 vs ISO 8601

 

2026-04-22 Wednesday - AI Governance and Compliance Resources

[image credit: geralt on pixabay dot com]

 

About two weeks ago (~ 2026-04-06, or so), I responded with a series of comments on a LinkedIn post by Alvin Reyes (Co-Founder and Chief Technology Officer, Koneksi), citing some resources that might be of interest, based on the question he raised:

"Anyone here working on AI governance? Be good to connect and share notes." 

This blog post is a placeholder to organize and assemble interesting resources that I find, from time-to-time.

 

Academic Research and Governance-Related Organizations

• Berkeley Center for Responsible Decentralized Intelligence

 

Possible Thought Leaders to follow on LinkedIn

• Jesper Lowgren (Agentic Enterprise Architecture Lead - Agentic System Design, Agentic Governance, DXC Technology)

• Kin Lane (re: his posts and articles related to AI usage in the implementation of governance for APIs), for example:
•  API Governance Rules Are Not a Checkbox. They're a Conversation.

• Luiza Jarovsky, PhD (Co-Founder, AI, Tech & Privacy Academy), for example:
• "Important new article about the negative implications of sycophantic AI and the URGENT need for better AI governance mechanisms: ..."

• Songyee Yoon, PhD (Founder & Managing Partner, Principal Venture Partners), for example:
• AI’s Red Lines Cannot Be Drawn by a Handful of People

 

Interesting Articles, Related to AI Governance

• An important component of AI Governance must consider security. You may find some of the posts/graphics graphic by Aakash Abhay Yadav (Senior Security Risk Intelligence Engineer, Roblox) of possible interest, for example:
• Your AI agent just went rogue. Not because it was hacked - because no one built the guardrails.
• Staircase to Responsible AI Deployment 

 

Possible Vendor Solutions to consider exploring

(Caveat Emptor applies, always. An entry in this list does not constitute an endorsement, or recommendation)

• Collibra
• LinkedIn Post, by George Trujillo Jr. (Senior Business Value Consultant, Collibra)
• "We’re not just building tools, we’re building the future of governed data and AI", (see PDF, re: "Next Gen AI governance is here! Spring '26 product feature drop")
• https://www.collibra.com/products/ai-governance  

 

Book Suggestions

There is a new book under development, through Manning Publications, that is available via their Manning Early Access Program (MEAP): 

AI Governance, Secure, privacy-preserving, ethical systems
by Engin Bozdag, PhD and Stefano Bennati, PhD

2026-04-19 Sunday - Architects are the Guardians of the Enterprise Keep

 

[image source: bojanboyss on pixabay dot com]

We stand, as Guardians of the Enterprise Keep, against the vagaries of chance and the arbitrary whims of fate. No matter how many times the wheel spins - and we may win...eventually, the accumulated balance of anti-chance will come due. 

  

2026-04-14 Tuesday - On The Value of Architecture Diagrams

[image source: Vilkasss on pixabay.com]

 

There is a subspecies of humans (let's call them Tech-Troglodytes of the "Rabid Agile™" clan) - that insist: 

a. You don't need diagrams 
b. You don't need documentation 
c. The source code is all that matters. 

Requiring each person (many of whom may not be technical) to solely develop their understanding of an area by reading [M * N] lines of source code (where [N] may be 10^P, and P may often be > 4)] - encompassing a single application, suites of applications, or the context and behavior distributed across multiple enterprise application domains (some of which may be owned by external partners, or commercial products/services) - would be wasteful, inefficient, and egregiously demonstrates a mind altogether too content with its first and poorest notion. 

A good architecture diagram is a work of art - that provides some of the finest levels of abstraction for reasoning about the design, capabilities, and limitations of a system.