I am really impressed by the practical approach to handling retries and ordering in data pipelines. The notes on schema drift are particularly useful for catching breaking changes early. It is exciting to see such clear explanations of idempotent consumers and how to backfill partitioned tables without taking down production. This resource is a fantastic addition to my toolkit for managing the unglamorous plumbing of data engineering.
Questions about Dead Letter Notes
What is Dead Letter Notes?
Dead Letter Notes is a personal blog written by E. Marlow that publishes articles on data pipelines, message queues, and related failure modes. The site features posts on topics such as the small files problem, schema drift, backfilling partitioned tables, idempotent consumers, and dead letter queues. It is built with plain HTML and offers an RSS feed for following new entries.
Who is the target audience for this blog?
The blog is designed for people working with the unglamorous plumbing that moves rows from one place to another, such as batch and streaming data pipeline builders. E. Marlow writes about the specific failure modes that occur in between these systems, including retries, ordering, partitioning, and the eternal argument about exactly-once processing. The author notes that opinions on these topics tend to change as new things are discovered.
How can I stay updated with new content?
The website explicitly states that it has no comments section and no newsletter, but it does provide an RSS feed for users to follow new entries. The site also features a footer that confirms it was built with plain HTML and includes a copyright notice for 2026.
What specific technical topics are covered in the articles?
The blog covers specific technical challenges including the small files problem and four ways the author has dealt with it. Other detailed topics include catching breaking changes before consumers do, backfilling a partitioned table without taking down production, and stopping the fear of retries through idempotent consumers. The site also provides practical guidance on dead letter queues.
What are the main sections of the website?
The site operates with a simple structure, featuring a home page, an about page, and a footer that includes the RSS feed link. The about page provides background on the author E. Marlow and their experience building and babysitting data pipelines.
What are the key features of the website's design?
The blog emphasizes a no-tracking approach, stating that there is no newsletter and no comments section, focusing instead on the RSS feed for updates. This minimalistic design aligns with the plain HTML construction of the site.
How is Dead Letter Notes rated on Trustburn?
On Trustburn, Dead Letter Notes has a rating of 4.2 out of 5 based on 5 reviews.
AI-generated from reviews on this page. Verify anything important with the company directly.
The notes on dead letter queues and backfilling partitions provide some useful context for understanding data pipeline failures. However, the information is somewhat scattered across different posts without a clear structure. I would have liked to see more detailed explanations of the four ways to deal with the small files problem. The reliance on plain HTML is nice for speed, but it limits the ability to present complex diagrams or interactive examples that could help clarify the technical concepts being discussed.
I have been reviewing the available notes on Dead Letter Notes to see if they might serve as a useful reference for our team's ongoing work with data pipelines and message queues. The topics covered, such as schema drift and idempotent consumers, are certainly relevant to the challenges we face daily. The author mentions using tools like Postgres, Kafka, and Spark, which are part of our own stack, suggesting that the advice might be applicable to similar environments. However, it is important to remember that this is a personal notebook rather than a comprehensive manual. The opinions are stated to change as the author gets burned by new things, which implies that some content might become dated or require verification against current best practices. I appreciate the transparency about the author's experience, but I would prefer to see more explicit connections between the notes and the specific tools mentioned. The lack of comments and tracking is refreshing, but it also means there is no community discussion to help clarify ambiguities. For someone considering using this site as a primary resource, I think it is a good starting point for understanding the basic concepts of retries and partitioning, but it should be supplemented with official documentation from the vendors themselves. The technical depth seems adequate for intermediate practitioners, though beginners might find some of the terminology challenging without additional context. Overall, it is a useful, albeit limited, resource that offers a practical perspective on the unglamorous plumbing of data engineering.
The plain HTML design makes the content load instantly without any clutter. It is a reliable resource for technical insights on data pipelines and message queues.
Reading through the available posts on data pipelines and message queues feels like having a reliable reference library at my fingertips. The fact that the site is built with plain HTML makes it incredibly fast and easy to navigate without any distractions. It is a convenient resource for anyone looking to understand the failure modes between different systems.
News
No news yet. This is where the business shares its own updates and press.
Photos
No photos yet. Photos shared by reviewers will show up here.
Widgets
Get the widget for your site →About Dead Letter Notes
Dead Letter Notes is a personal blog that publishes articles on data pipelines, message queues, and related failure modes. It features posts on topics such as small files problems, schema drift, backfilling partitioned tables, idempotent consumers, and dead letter queues. The site is built with plain HTML and offers an RSS feed for following new entries.
- Website
- deadletternotes.com