Recommended Free Tools
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Retrieve whole notes when your notes are long and a rule only makes sense alongside its exception or rationale. Retrieve chunks or short passages when documents are already brief, because a chunk then loses little. Rules that must always apply should not depend on similarity search at all; load them unconditionally.
That answer comes from a write-up by Tom Jones, “Whole notes, not fragments: the retrieval half,” published 18 September 2026. The article describes a personal-notes workflow and reports its own benchmark results. The figures below are the author’s, measured under the conditions he describes, and they have not been independently reproduced.
What “whole-note retrieval” means in this design
In the described system, notes are stored as plain Markdown files, each with a short header. Each note is embedded as a single unit, and retrieval returns whole notes ranked by cosine similarity to the embedded query. A chunked pipeline would split the same files into passages and return the best-matching passage instead. The difference is the unit that comes back: a whole note keeps its internal context, while a chunk may keep only the sentence that matched.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
The workflow the author describes has three parts:
- Dense path. The note text is embedded with
nomic-embed-textthrough a local Ollama instance, and the query is embedded the same way. - Storage. Vectors sit in a SQLite table using the
vec0extension. - Keyword path. A separate full-text index handles exact tokens such as function names, command-line flags and error strings, which embeddings often blur.
The two result lists are fused. The author’s reason for keeping both is practical: dense retrieval handles paraphrase, while keyword search catches literal strings. Component names and versions are as the article reports them. Treat them as a description of one implementation, not a compatibility guarantee for any other setup.
#1 Best Overall
- ADJUSTABLE HEIGHT DESIGN: The mobile standing desk promotes a healthier workstyle by allowing quick transitions between sitting and standing. The gas spring lift smoothly adjusts the height from 28.3in to 44in, supporting better posture and reducing neck and back strain during long working hours. This portable desk improves daily comfort and productivity across different environments.
- SUPERIOR STABILITY AND DURABILITY: The rolling desk adjustable height model stands out with its sturdy H shaped steel base and reinforced structure, providing stability even at maximum extension. The waterproof and scratch resistant MDF desktop ensures long lasting use, while the retractable keyboard tray and hook create organized storage for accessories. This unique design differentiates the desk from standard folding table or rolling podium options on the market.
- ERGONOMIC AND FUNCTIONAL DESIGN: The portable standing desk offers a spacious 25.6 x 17.7in surface to accommodate a laptop, monitor, or books. A dedicated slot holds phones and tablets, while the 23.6 x 11.8in keyboard tray supports a full size keyboard and mouse. The thoughtful structure allows the small standing desk to serve as a side table, study cart, or computer desk with keyboard tray in living rooms, bedrooms, and offices.
- EASY MOBILITY WITH LOCKABLE WHEELS: The adjustable rolling desk includes four caster wheels that allow smooth movement between rooms. The lockable function secures the desk in place when needed, creating flexibility for use as a rolling laptop desk, classroom furniture, or teacher standing desk. The compact rolling table design makes the desk on wheels easy to move, while maintaining stability during presentations or study sessions.
- EASY OPERATION AND LOW MAINTENANCE: The sit stand desk is operated with a simple hand lever that activates the gas spring for smooth upward adjustment, while gentle pressure lowers the surface. The mobile desk workstation requires minimal maintenance, as the MDF board is waterproof, scratch resistant, and easy to clean with a damp cloth. This reliable raising desk minimizes user effort and ensures long term durability without complex upkeep.
What the public benchmarks show
The article reports three comparisons. They answer different questions, so they should not be read as one scoreboard.
SciFact: a tie with chunked retrieval
SciFact is a scientific-claim retrieval dataset, and the article scores results with nDCG@10, a ranking measure that rewards placing relevant documents near the top of the first ten results. Higher is better. Whole-note retrieval scored 0.7014 and chunked retrieval scored 0.7016, which the author calls a tie. He attributes this to the fact that SciFact abstracts are short enough that a chunk already covers most of its document.
The whole-note arm was run three times and returned 0.7014, 0.7019 and 0.7014. A keyword-search control scored 0.6644. This is the clearest counterexample to an unconditional whole-note advantage: on short documents, the choice of unit made no measurable difference.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11NFCorpus: whole notes against keyword search
On NFCorpus, across 323 queries, whole-note retrieval scored 0.3417 and keyword search scored 0.3098. The article presents this as a win for whole notes over the keyword baseline. It does not present the result as equivalent to the SciFact comparison. It tests a different baseline and a different question, so it should not be combined with the SciFact tie into a single conclusion.
Rank #2
- 【32” x 19” Perfect for Small Spaces & Corner】 Specially designed with a compact 32" x 19" desktop, this small electric standing desk seamlessly fits into limited areas like apartments, bedrooms, and cozy home office corners without crowding your room. It is the ultimate space-saving, height-adjustable solution to pair with under-desk treadmills and walking pads for remote workers, freelancers, and students
- 【4 Memory Presets & DIY Wheel Ready】 This adjustable desk features a smart control panel with 4 programmable memory presets for effortless one-touch height adjustment (28.3" to 46.5"). Plus, built-in universal M8 screw holes on the desk feet allow you to easily install your own casters/wheels to DIY it into a mobile rolling desk.
- 【176 lbs Max Load & Rounded Safety Corners】 Constructed with heavy-duty steel rails and a solid desktop, this small stand up desk supports up to 176 lbs with exceptional stability while transitioning. The tabletop features smooth rounded corners to protect you, your family, or pets from accidental bumps in tight, compact spaces.
- 【Rigorously Tested for Long-Lasting Use】 Engineered for daily reliability, our motor and lifting system have been rigorously tested to withstand up to 50,000 lift cycles under full capacity. Enjoy a whisper-quiet, smooth sit-to-stand transition that keeps you focused and productive all day.
- 【Easy Assembly & Budget-Friendly Choice】 Comes with detailed instructions and all hardware included for a hassle-free, quick setup. Get premium electric sit-stand functionality at an unbeatable, budget-friendly price. Risk-free purchase with dedicated customer support ready to help.
The internal 14-task evaluation
The most striking figure is an internal result: 52% for whole-note retrieval against 27% for standard snippet retrieval, across 14 tasks the author wrote for his own notes. Responses were scored by a model. The author states that this evaluation cannot be rerun externally, so the tasks, scoring and exact prompts are not available for others to check. Read it as the author’s report of his own corpus, not as an estimate of how much whole notes would improve results elsewhere.
| Comparison | Dataset or setting | Whole-note result | Comparison result | What it does and does not show |
|---|---|---|---|---|
| Whole note vs chunked | SciFact, nDCG@10 (public) | 0.7014 (three runs: 0.7014, 0.7019, 0.7014) | Chunked 0.7016 | A tie on short abstracts. Shows no advantage where documents are short. |
| Whole note vs keyword search | NFCorpus, 323 queries, nDCG@10 (public) | 0.3417 | Keyword 0.3098 | A gain over a keyword baseline on this dataset. Not a chunking comparison. |
| Whole note vs snippet retrieval | Author’s 14 internal tasks, model-scored | 52% | 27% | A large gap on one personal corpus. Cannot be rerun externally. |
When whole notes are the better unit
The author’s argument is not that whole notes are better in general. It is that they help under a specific condition: the document is long, and it contains a rule whose exception or rationale matters. A snippet that states the rule without the caveat can lead a model to apply it wrongly. Returning the full note keeps the caveat attached to the rule.
Short documents have little room to benefit, because a chunk already contains most of what a note contains. The SciFact tie is consistent with that boundary.
A practical way to apply the test is to ask two questions of your own notes:
Rank #3
- [INTEL POWERED CONTENT] - Built with a 8th Generation Hexa-Core Intel i5 and 32GB of DDR4 RAM; Modern, Windows 11 ready, with 4K support, Executive multitasking, media streaming and smooth, multi-tab web browsing; Perfect as an all-purpose multimedia computer; built for content creators; Plenty of RAM and Mass storage for photo and video editing powered by Intel HD 630
- [LATEST WIRELESS TECH] - This Dell Desktop Computer easily connects to the internet through the Built In WiFi / Bluetooth
- [SOLID STATE STORAGE] - This Dell Computer setup comes with an ultra-fast 1TB Solid State Drive (SSD); Setup as the primary boot device; Boot and load programs with lightning speed ; Additional expansion available
- [BUY & OWN WITH CONFIDENCE] - From the world's largest Microsoft Authorized Refurbisher; Quality Guarantee and Free Tech Support; Award-winning Customer Service; | Support Sustainable Business
- [MODERN HI-SPEED PORTS] - USB 3.0 (x4) | USB 2.0 (x4) | DisplayPort (x1) | HDMI Port (x1) | Audio Combo Jack (x1) | Audio Out (x1) | RJ-45 Ethernet (x1) | Internal SATA (x3)
- Does a typical note contain a rule followed by a condition, exception, or reason that changes how the rule should be applied?
- Would a passage that matches the query, read without the rest of its note, be misleading?
If both answers are yes, whole-note retrieval is worth testing. If the notes are short and self-contained, chunking is the cheaper choice and the SciFact result suggests you will lose little.
The trade-offs: context cost and reader capability
Whole notes preserve neighbouring context, but they cost more. Each returned note consumes more of the model’s context window, which means fewer distinct results fit in the same budget and each token carries less relevant material. The author treats this as the price of keeping the condition attached to the rule.
The reader matters as much as the retrieval unit. The article reports that a large-context model did better when given whole notes, while a small local model preferred compact records. The author presents this as an observation from his own setup rather than a general law, so if your model has a small window, test both shapes before committing.
Rules that must always apply
The author draws a firm line here: rules that must always apply should not be retrieved by similarity. Similarity search returns what looks relevant to the query, so a safety rule that the query never mentions can simply not appear. He states the position as “Safety rules are never retrieval-gated.” That is a design position of the article, not a formal safety standard.
Rank #4
- Create Instant Active Standing - VIVO’s desk riser provides on-demand standing throughout the day for the freedom to get out of your chair and relieve muscle tension, reduce stress, and increase productivity. --Patented--
- Space Efficient 31.5" Surface - The top surface measures 31.5” x 15.7”, which maximizes space while still providing room for dual monitors. The 31.3" x 11.8" (10.5" in center) keyboard tray raises in sync with the top surface to create a comfortable workstation.
- Strong 33 lbs Lift Assist - Go from sitting to standing in one smooth motion using the innovative simple touch height locking mechanism (Adjustment Range: 4.5" to 20"). Lift design elevates straight upwards.
- Very Minimal Assembly - This riser is almost ready to go right out of the box! Place on your existing desk, attach the keyboard tray, and start organizing your workstation.
- We've Got You Covered - Sturdy, high-grade steel design is backed with a 3-Year Manufacturer Warranty and friendly tech support to help with any questions or concerns.
In practice, this means splitting your context into two layers:
- Always-loaded rules. Short, mandatory instructions placed in every prompt, regardless of query.
- Retrieved knowledge. Notes and passages selected by the hybrid search described above, where a missed result is an inconvenience rather than a failure.
Limits of the evidence
Three limits should shape how much weight you give these results.
- The public comparisons use SciFact and NFCorpus, both of which are scientific or medical literature, not personal notes. Their relevance to a personal knowledge base is an inference.
- The 52% versus 27% result is the author’s own, scored by a model, and cannot be rerun by others. It shows what happened on one corpus under one setup.
- The benchmarks compare unit choice under specific baselines. They do not measure every alternative, such as hierarchical chunking or re-ranking, so they do not rank whole notes against all retrieval designs.
What the evidence supports is narrower than a headline. Whole notes can preserve a rule with its reason and exception, that benefit depends on document length and structure, and the cost is context. Whether it nearly doubles retrieval quality is a claim the article’s internal result does not establish for your corpus.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →The Bottom Line
Use whole notes when long notes carry rules whose exceptions matter, and test the choice against your own queries and reader. Use chunks when documents are short. Load must-always-apply rules directly, outside retrieval.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
