# Meeting 7/12/2020

Feedback Dr. Naithani

- Break up abstract in two paragraphs (content is okay)
- Introduction should has some weird jumps in structure
- give overview of upcoming paper structure throughout paper
- reference to section numbers every time something new is introduced
- Figure 1 is too small
- Acronyms have to be defined as their full version, but the meaning does not have to be explained
- MLP must be cited/explained
- "pseudo sequential execution": ajeya was confused by this but I don't know if it was okay or not after explaining it
- Reaction Raf: I think he was not and suggested us to use (almost) sequential

Reordering:
- Balance should be between spatial and temporal localicty **AND overhead**, make this very clear!

Partitioning:
- Too many reuses of "et al."
- Some figure captions are too short:
- caption should explain the figure thoroughly, or at least be a complete sentence
- Table 1 should be positioned earlier in the text

Prefetchers:
- Page 11 example:
- citation is wrong (of figure?)?
- it is not paraphrased enough, almost copied exactly
- DROPLET is too short
- describes what they did, but not why and what the results are
- look at the observations 1-6 in the original paper
- 5-6 lines before explaining the architecture
- Hybrid prefetchers
- explain first why
- then say we will discuss two implementations, and name them
- then first subsection follows

Related work:
- Maybe add one more subsection to reference 3-4 more works at least

Other:
- Provide separate document with labor division