My abstract promised a number. Measuring it changed the claim.
\n"Just use RAG" and "RAG is dead" are both answers to a question nobody asked precisely. An index costs money to build, to store and to keep current. Live search costs money every time you run it. Which one wins is not a matter of taste, it is a crossover point, and it moves depending on what you are searching and how often.
\nThis talk puts a number on that crossover for two workloads, conversation memory and code search, and shows the measurements behind it. Every figure comes from a committed CSV, every chart renders from that CSV at build time, and the whole study runs on a laptop with Docker.
\nThe honest finding is that the index does not always win. It loses badly in one of the cases I measured, and the talk spends time there rather than skipping past it.
\nLive search wins when the corpus is small and the model already knows it:
\nfastapi, a repository the model already knows, break-even is infinite at every size. No query volume repays the index, because grep is cheaper every time.The index wins as the corpus grows, or when the model has never seen it:
\nThat last pair is the sharpest thing in the talk. Same repository, same index, same box, same prices. The only thing that changes is what the live arm does, and the answer flips from "no volume repays this" to "twelve queries a day". The question is never "is retrieval worth it", it is "worth it against what".
\nEverything in the talk runs on your own machine, given Docker with 16GB of RAM or more:
\nmake setup && source .venv/bin/activate
make up # docker compose up -d + Milvus health wait
make data-1m # or data-10m
jupyter lab # notebooks 01 to 06b in orderIf you want the tooling rather than the study, memsearch gives Claude semantic recall over its own conversation history, and claude-context adds semantic code search to coding agents over MCP.
\nPresented at Haystack Europe 2026 in Berlin on 15 September 2026.
\n", "date_published": "2026-09-15T00:00:00Z", "date_modified": "2026-09-15T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2026/rag-cost-curve/index.md" }, "_talk": { "slides_url": "https://talks.simonhearne.com/rag-cost-curve/", "video_url": null, "links": [ { "label": "Reproduce every number", "href": "https://github.com/simonhearne/rag-cost-curve" } ] } },{ "id": "https://simonhearne.com/2026/migrating-to-milvus-talk/", "url": "https://simonhearne.com/2026/migrating-to-milvus-talk/", "title": "Migrating to Milvus: Tooling, Deployment Choices and Lessons from Real Migrations", "summary": "How to move from Elasticsearch or OpenSearch to Milvus or Zilliz Cloud: migration tooling, a real end-to-end walkthrough, deployment choices, and lessons from real customer migrations.", "image": "https://simonhearne.com/images/milvus_migration_hero.jpg", "tags": ["VectorDB","Zilliz","Milvus","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "This is the follow-up to Milvus 3.0: Closing the Search Gap. If you are moving self-managed Milvus rather than Elasticsearch, see Migrating Self-managed Milvus to Managed Milvus.
\nMilvus 3.0 expands the range of search workloads teams can run in Milvus, with stronger full-text retrieval plus in-engine sorting, aggregation and faceting. In the previous webinar I explored these search improvements and compared Elasticsearch and Milvus with live side-by-side queries.
\nIn this follow-up session I move from comparison to migration. I show how Zilliz Migration Service helps move data from Elasticsearch or OpenSearch into Milvus or Zilliz Cloud, then walk through a real migration workflow. I also compare Elastic Cloud (ESS) and Zilliz Cloud, discuss deployment options across open-source Milvus, Zilliz Cloud and Zilliz BYOC (Bring Your Own Cloud), and share lessons from customer migrations.
\nIf you use Elasticsearch or OpenSearch today and are considering a migration, this is the walkthrough for you.
\nPresented as a Zilliz webinar on 2 September 2026.
\n", "date_published": "2026-09-02T00:00:00Z", "date_modified": "2026-09-08T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2026/migrating-to-milvus-talk/index.md" }, "_talk": { "slides_url": "https://talks.simonhearne.com/2026-09-milvus-migration/", "video_url": "https://www.youtube.com/watch?v=0xU3Ks2vXIM" } },{ "id": "https://simonhearne.com/2026/milvus_search_gap/", "url": "https://simonhearne.com/2026/milvus_search_gap/", "title": "Milvus 3.0: Closing the Search Gap", "summary": "Milvus 3.0 closes the gap between vector search and full-text search: one engine for both. A customer-backed comparison of Milvus 3.0 and Elasticsearch 9.14, with live side-by-side queries you can reproduce yourself.", "image": "https://simonhearne.com/images/search_gap_hero.jpg", "tags": ["VectorDB","Zilliz","Milvus","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "The follow-up webinar, Migrating to Milvus, moves from comparison to migration: tooling, deployment choices and lessons from real migrations.
\nMilvus has long been known for vector search and has supported full-text search since Milvus 2.5, allowing teams to combine lexical and semantic retrieval in one engine. With Milvus 3.0, full-text search takes another major leap:
\nWith stronger full-text retrieval added to Milvus's established vector-search foundation, teams can reconsider whether AI retrieval still requires a separate full-text system, along with the infrastructure, pipelines and result-fusion logic that come with it. This talk is a customer-backed comparison of Elasticsearch and Milvus, including live side-by-side queries against Milvus 3.0 and Elasticsearch 9.14. Query semantics, results and performance are all reproducible with the open-source lab linked above.
\nPresented as a Zilliz webinar on 19 August 2026.
\n", "date_published": "2026-08-19T00:00:00Z", "date_modified": "2026-08-19T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2026/milvus_search_gap/index.md" }, "_talk": { "slides_url": "https://talks.simonhearne.com/2026-08-milvus-search-gap/", "video_url": "https://www.youtube.com/watch?v=XxcA1NSQnCo", "links": [ { "label": "Test it yourself!", "href": "https://github.com/simonhearne/milvus_es_lab" } ] } },{ "id": "https://simonhearne.com/2026/visualising-vector-search/", "url": "https://simonhearne.com/2026/visualising-vector-search/", "title": "Vector Search, Visualised", "summary": "Vector search has no EXPLAIN. Through interactive visualisations, this talk shows what embeddings look like in high-dimensional space, what quantisation does to your recall, and how to catch retrieval failures before your agents do.", "image": "https://talks.simonhearne.com/og/vector-search-visualised.png", "tags": ["VectorDB","Zilliz","Milvus","talk","video"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "SQL makes sense. But when it breaks, you reach for EXPLAIN. Vector search offers no such comfort. Multi-thousand-dimension embeddings, approximate nearest-neighbour indexes, and quantisation tradeoffs make it hard to know what your system is doing, and harder still to diagnose when results quietly degrade.
Through interactive visualisations, this talk shows what embeddings look like in high-dimensional space, what quantisation does to your recall, and how to catch retrieval failures before your agents do. You'll leave with a sharper mental model and a diagnostic toolkit for the production problems hardest to see.
\nPresented at TiDB SCaiLE Europe 2026 in Stockholm on 4 June 2026 as "When EXPLAIN Isn't Enough: Visualising Vector Search for Engineering and Product Teams".
\n", "date_published": "2026-06-04T00:00:00Z", "date_modified": "2026-06-04T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2026/visualising-vector-search/index.md" }, "_talk": { "slides_url": "https://talks.simonhearne.com/vector-search-visualised/", "video_url": "https://www.youtube.com/watch?v=kY-il0pWjdY" } },{ "id": "https://simonhearne.com/2022/third-parties-request-map/", "url": "https://simonhearne.com/2022/third-parties-request-map/", "title": "WebPageTest Live: Third Parties and Request Map with Simon Hearne", "summary": "Third parties have always been a challenge. Having resources loading on your site, beyond your control isn't something users sign up for. In 2015, Simon Hearne created a tool called REQUEST MAP to track just that: requests. This great visualization tool allowed you to see where the request as…", "image": "https://simonhearne.com/images/wpt_requestmap_twitch.jpg", "tags": ["WebPerf","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "Most WebPageTest walkthroughs stop at the waterfall. This session was about what you do next.
\nA live WebPageTest session working through real publisher sites, using Request Map to see which third parties are pulling in which other third parties, and where that chain starts costing real time.
\nAlongside the audits it collects the practical mechanics that rarely make it into tutorials: getting past cookie consent banners without breaking your measurement, and a handful of bookmarklets that make WebPageTest's own interface easier to read.
\nexecAndWait, which simulates a first-time visit and can be merged using combineSteps or split with setEventName; or set the consent cookie directly with setCookie, which emulates a repeat visitor and is the better choice for testing holistic performance.Recorded as a WebPageTest Live stream. Note that the WebPageTest and Request Map result links in the accompanying speaker notes have since expired.
\n", "date_published": "2022-10-11T00:00:00Z", "date_modified": "2022-10-11T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2022/third-parties-request-map/index.md" }, "_talk": { "slides_url": null, "video_url": "https://www.youtube.com/watch?v=TY82pTfhnXY", "links": [ { "label": "View Speaker Notes", "href": "https://notes.simonhearne.com/WebPageTest-Twitch-Stream-Notes-e812abaf4d3445c0ac70678a29ca9e0f" } ] } },{ "id": "https://simonhearne.com/2021/inclusive-design/", "url": "https://simonhearne.com/2021/inclusive-design/", "title": "An Inclusive Web is Fast by Default", "summary": "Web performance is critical to inclusive design - users on low-end devices or with poor connectivity are excluded from large and slow applications", "image": "https://simonhearne.com/images/social_inclusive-design.jpg", "tags": ["WebPerf","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "Performance is not an average, it is a distribution — and the people at the far end of it are being quietly excluded.
\nWhen a home connection fails and drops you to 20–50kbps, the web stops being slow and starts being unusable. That is the everyday condition for a large share of the world: around 65% of people in developing nations have no access to fast internet, against roughly 20% in developed ones.
\nThis talk argues that inclusion and performance are the same project, and that the analytics telling you everything is fine are often measuring only the users who managed to stay.
\nThe deck closes on William Gibson: "The future is already here, it's just not evenly distributed."
\n", "date_published": "2021-09-23T00:00:00Z", "date_modified": "2021-09-23T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2021/inclusive-design/index.md" }, "_talk": { "slides_url": "https://simonhearne.com/presentations/inclusive-design/#/", "video_url": "https://www.youtube.com/watch?v=MiXy2x6flww" } },{ "id": "https://simonhearne.com/2020/psychology-of-speed/", "url": "https://simonhearne.com/2020/psychology-of-speed/", "title": "The Psychology of Speed", "summary": "We intuitively know that faster is better, but what's the psychology and physiology behind it?", "image": "https://simonhearne.com/images/psych-speed-social.jpg", "tags": ["WebPerf","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "Feeling fast and being fast are different problems. This talk is about the gap between them.
\nPerformance work starts with a stopwatch, but the number on the stopwatch and the speed a user perceives are not the same quantity. The difference is psychological, it is well studied, and it is largely under your control as a designer.
\nThis talk takes the research on waiting — why an occupied wait feels shorter, why uncertainty hurts more than duration — and maps it onto the concrete decisions you make when building an interface.
\nEvery third-party tag is someone else's code, running on your users' devices, with your name on it. This talk is seven times that went wrong.
\nBy late 2019 the median website was 37% third-party requests by count (HTTP Archive, October 2019). Analytics, advertising, optimisation, JavaScript CDNs and tag managers all arrive with a business case, and all of them execute with the same privileges as the code you wrote yourself.
\nThis talk works through seven real incidents — outages, revenue losses, and one case of a "performance" tool making things measurably slower — and pulls a defence out of each.
\ndocument.ready and took critical application logic down with it during a China launch. Sync is still king, so feature-flag it and defer it.connect-src restrictions, plus CPU metrics to catch what network metrics miss.As Tim Kadlec puts it in the deck: "Everything should have a value because everything has a cost."
\n", "date_published": "2019-11-22T00:00:00Z", "date_modified": "2019-11-22T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2019/third-party-deep/index.md" }, "_talk": { "slides_url": "https://simonhearne.com/presentations/third-party-deep/#/", "video_url": "https://www.youtube.com/watch?v=uXv9JFvrnwo" } },{ "id": "https://simonhearne.com/2018/third-party-weak-links/", "url": "https://simonhearne.com/2018/third-party-weak-links/", "title": "Third-Party Content - the weak link?", "summary": "The Resource Timing API allows site owners to measure how third-parties perform for every single pageview. Combining this data with business intelligence provides new insight to the impact of third-party providers on business performance.", "image": "https://simonhearne.com/images/weak-links-social.jpg", "tags": ["WebPerf","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "The Resource Timing API allows site owners to measure how third-parties perform for every single pageview. Combining this data with business intelligence provides new insight to the impact of third-party providers on business performance.
", "date_published": "2018-01-14T00:00:00Z", "date_modified": "2018-01-14T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2018/third-party-weak-links/index.md" }, "_talk": { "slides_url": "https://docs.google.com/presentation/d/1pUo2SrWPvJ5vwYIbpebqFqnhAwGZ2pVryKP2kzqA_Uw/edit?usp=sharing", "video_url": "https://www.youtube.com/watch?v=9ol9MNGqVOM" } },{ "id": "https://simonhearne.com/2016/too-hot-to-handle/", "url": "https://simonhearne.com/2016/too-hot-to-handle/", "title": "Too hot to handle - optimizing for low powered devices", "summary": "We know what web performance is—getting pages to render fast for as many customers as possible. Traditionally, the focus has been on the network, improving the time to deliver critical content, but the combination of modern web applications, better network connectivity, and HTTP/2 is reducing the…", "image": "https://simonhearne.com/images/too_hot.jpg", "tags": ["WebPerf","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "The phone in a developer's pocket is not the phone in a user's pocket. Everything in this talk follows from that gap.
\nResponsive design solved layout. It did not solve the fact that a mid-range Android handset has a fraction of the CPU budget of the flagship it was designed on, and that CPU, rather than bandwidth, is what most modern pages are actually constrained by.
\nThis talk is about that constraint: how wide real device diversity runs, why the conveniences that make development pleasant make execution expensive, and what to do when a large share of your users are on hardware you have never held.
\nPresented at Velocity New York 2016.
\n", "date_published": "2016-10-14T00:00:00Z", "date_modified": "2016-10-14T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2016/too-hot-to-handle/index.md" }, "_talk": { "slides_url": "https://speakerdeck.com/simonhearne/too-hot-to-handle-optimizing-for-low-powered-devices", "video_url": "https://www.youtube.com/watch?v=__KEbUEFHpE" } },{ "id": "https://simonhearne.com/2016/meaningful-visualisations/", "url": "https://simonhearne.com/2016/meaningful-visualisations/", "title": "Beyond the Waterfall - Meaningful Web Performance Visualizations", "summary": "With all the different measurements we can take of web performance, the data we can report on becomes huge. Waterfall charts just don’t cut it any more – they are a technical tool to diagnose technical issues. What we need now are business-focused visualizations that quickly represent the impact of…", "image": "https://simonhearne.com/images/velocity_btw.jpg", "tags": ["WebPerf","video","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "A waterfall chart is a tool for people who already know what they are looking at. This talk is about everyone else.
\nWaterfall charts are diagnostic instruments. They reward expertise, and they assume the reader already knows what a request is, why blocking matters and what a long green bar means. That makes them excellent for engineers and close to useless for the people who decide whether performance work gets funded.
\nThis talk argues for visualisations built for that second audience: images that carry their meaning at a glance, because we read pattern and shape faster than we read text. It walks through the visual vocabulary available, then applies it to real performance data.
\nwebperf.tools, and Graphite for trends.Further reading suggested in the deck: Information Is Beautiful by David McCandless, and Storytelling with Data by Cole Nussbaumer Knaflic.
\n", "date_published": "2016-10-14T00:00:00Z", "date_modified": "2016-10-14T00:00:00Z", "_markdown": { "url": "https://simonhearne.com/2016/meaningful-visualisations/index.md" }, "_talk": { "slides_url": "https://noti.st/simonhearne/tODEWH/beyond-the-waterfall-meaningful-web-performance-visualisations", "video_url": "https://www.youtube.com/watch?v=l1RpGiP3FpM" } },{ "id": "https://simonhearne.com/2016/how-fast-is-fast-enough/", "url": "https://simonhearne.com/2016/how-fast-is-fast-enough/", "title": "How fast is fast enough?", "summary": "Webinar with Andy Davies on measuring site speed and using RUM data to work out when performance starts to change customer behaviour.", "image": "https://simonhearne.com/images/hero_fastenough2.jpg", "tags": ["WebPerf","talk"], "authors": [{ "name": "Simon Hearne", "url": "https://simonhearne.com/about/" }],"content_html": "Do you know when your website's speed starts to impact your customers' behaviour?
\nThe speed of your site affects how your visitors behave: how many pages they view, how much they spend or even whether they convert.
\nJoin me and one of the world's leading web performance experts, Andy Davies, as we discuss what you should consider when measuring your site's speed and how to determine when speed is a problem.
\nYou'll leave this webinar with a good understanding of why speed matters, what good performance looks like and how it impacts your business.
\nWe're really lucky to have some concrete RUM data and experience to work out what's important. The webinar has been and gone, but you can see the slides below.
\n