[GH-ISSUE #1103] My opinion of the online text first two chapters. #5654

Closed
opened 2026-04-21 21:38:11 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @Huebr on GitHub (Jan 7, 2026).
Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/1103

I’m really enjoying the reading and finding it very helpful as I work toward transitioning into machine learning. I’ve just finished the first two chapters and genuinely like the content overall. That said, I wanted to share a few suggestions that I think could further improve the reading experience.

One thing I noticed is that some ideas are repeated multiple times, which makes the text feel longer than necessary and can be a bit discouraging while reading. For example, the discussion about physical latency limits between California and Virginia appears in slightly different forms in multiple places. While the point itself is important, reiterating the same numbers and conclusions several times gives the impression that the text is circling back to the same idea rather than moving forward. Condensing these repetitions could make the chapters feel more concise and easier to progress through. This becomes less of an issue later on, where there is a lot of genuinely engaging and fresh information.

Another point concerns the side notes. There are quite a few of them, and while many are useful, some contain information that feels important or especially interesting. In those cases, it might be worth integrating them into the main text, so readers don’t miss valuable context or insights.

I also had some difficulty with the structure in a few sections, where certain concepts are introduced before they are fully explained. For instance, in the passage introducing the AI Triangle, the text refers to it as a foundational framework, but its meaning isn’t immediately clear until later, when an image shows the relationship between data, algorithms, and infrastructure. Providing a brief explanation up front could help readers better follow the discussion from the start.

Finally, some figures appear to be AI-generated and can be unclear conceptually. For example, Figure 13’s flowchart feels ambiguous and doesn’t clearly define a procedure, which makes it harder to understand the intended message. Clearer or more explicitly structured diagrams might help reinforce the ideas being presented.

Overall, I’m really enjoying the material and find it very informative. These are just a few suggestions that I think could make the text even clearer and more engaging for readers.

Originally created by @Huebr on GitHub (Jan 7, 2026). Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/1103 I’m really enjoying the reading and finding it very helpful as I work toward transitioning into machine learning. I’ve just finished the first two chapters and genuinely like the content overall. That said, I wanted to share a few suggestions that I think could further improve the reading experience. One thing I noticed is that some ideas are repeated multiple times, which makes the text feel longer than necessary and can be a bit discouraging while reading. For example, the discussion about physical latency limits between California and Virginia appears in slightly different forms in multiple places. While the point itself is important, reiterating the same numbers and conclusions several times gives the impression that the text is circling back to the same idea rather than moving forward. Condensing these repetitions could make the chapters feel more concise and easier to progress through. This becomes less of an issue later on, where there is a lot of genuinely engaging and fresh information. Another point concerns the side notes. There are quite a few of them, and while many are useful, some contain information that feels important or especially interesting. In those cases, it might be worth integrating them into the main text, so readers don’t miss valuable context or insights. I also had some difficulty with the structure in a few sections, where certain concepts are introduced before they are fully explained. For instance, in the passage introducing the AI Triangle, the text refers to it as a foundational framework, but its meaning isn’t immediately clear until later, when an image shows the relationship between data, algorithms, and infrastructure. Providing a brief explanation up front could help readers better follow the discussion from the start. Finally, some figures appear to be AI-generated and can be unclear conceptually. For example, Figure 13’s flowchart feels ambiguous and doesn’t clearly define a procedure, which makes it harder to understand the intended message. Clearer or more explicitly structured diagrams might help reinforce the ideas being presented. Overall, I’m really enjoying the material and find it very informative. These are just a few suggestions that I think could make the text even clearer and more engaging for readers.
Author
Owner

@profvjreddi commented on GitHub (Jan 7, 2026):

Hi @Huebr,

Thanks so much for taking the time to share this, and I'm glad to hear the book is helpful as you transition into ML.

You've touched on a challenge we've been grappling with: the repetition issue. Some of it is intentional (we know people jump between chapters), and at other times it's because I write sections at different times and tend to revisit points repeatedly. It’s worth reviewing to tighten that up. So, thank you for pointing those out. I will work on fixing it.

The side notes point is well taken too. We probably got a bit too enthusiastic about using them, and some genuinely interesting material ended up in the margins where it's easy to overlook. Good catch. I am working on those right now as we "speak".

Regarding the AI Triangle: fair criticism. We should clarify what we mean before presenting it as a "foundational framework." An easy fix.

The figures are an ongoing project. They aren't AI-generated; they are actually Tikz code, so they look a bit rigid. Some were created early on and haven't aged well, I suppose, cause I have been improving material. If you notice specific ones that aren't effective, feel free to flag them. It's always helpful to know which visuals are causing confusion rather than clarifying it.

Appreciate the feedback, and I hope the rest of the chapters continue to be useful!

<!-- gh-comment-id:3716840804 --> @profvjreddi commented on GitHub (Jan 7, 2026): Hi @Huebr, Thanks so much for taking the time to share this, and I'm glad to hear the book is helpful as you transition into ML. You've touched on a challenge we've been grappling with: the repetition issue. Some of it is intentional (we know people jump between chapters), and at other times it's because I write sections at different times and tend to revisit points repeatedly. It’s worth reviewing to tighten that up. So, thank you for pointing those out. I will work on fixing it. The side notes point is well taken too. We probably got a bit too enthusiastic about using them, and some genuinely interesting material ended up in the margins where it's easy to overlook. Good catch. I am working on those right now as we "speak". Regarding the AI Triangle: fair criticism. We should clarify what we mean before presenting it as a "foundational framework." An easy fix. The figures are an ongoing project. They aren't AI-generated; they are actually Tikz code, so they look a bit rigid. Some were created early on and haven't aged well, I suppose, cause I have been improving material. If you notice specific ones that aren't effective, feel free to flag them. It's always helpful to know which visuals are causing confusion rather than clarifying it. Appreciate the feedback, and I hope the rest of the chapters continue to be useful!
Author
Owner

@profvjreddi commented on GitHub (Jan 7, 2026):

Fixed in 6af21ccf8f

<!-- gh-comment-id:3716878733 --> @profvjreddi commented on GitHub (Jan 7, 2026): Fixed in 6af21ccf8f1f0ea8dcae6722d6fa5679dc8a8e7c
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/cs249r_book#5654