TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

MQL – Client and server to query your db in natural language

59 点作者 akashkahlon大约 1 年前

9 条评论

brudgers大约 1 年前
My experience with this kind of tool is that it is at least as hard to learn the tool as it is to learn the technology it abstracts over.<p>I think that&#x27;s because thinking about the problem I am trying to solve is always the hardest part and I have to learn a syntax and semantics no matter what. And the syntax and semantics of SQL is mathematically linked to the mathematics of relational databases. Natural language isn&#x27;t.<p>Furthermore there&#x27;s decades of good technical documentation for SQL written by diverse authors for diverse levels of technical experience. Natural language projects are one off and writing documentation is usually a lower priority than making code go.
评论 #39972310 未加载
评论 #39987493 未加载
kshitijb大约 1 年前
For the majority of people from non-tech business functions, the ability to ask for insights from data is liberating. Tools like this can unlock their potential to make more informed decisions. Imagine a store manager of a hyperlocal grocery startup managing a dark store. What if they could ask questions like &quot;What is the fulfilment rate of a certain SKU between 12-3 pm in their store for the past 7 days?&quot;
mritchie712大约 1 年前
text-to-sql is a dead end. There&#x27;s no way for a model to correctly interpret the meaning of every column in a real world database using the `information_schema` alone. Most cloud warehouses (e.g. Snowflake) don&#x27;t use foreign keys, so you don&#x27;t even know the joins.<p>Imagine you hire a highly skilled data analyst (e.g. 9 out of 10 proficiency in SQL) and start asking them questions about your database. They won&#x27;t answer them, they&#x27;ll ask you more questions. The conversation would go something like:<p>you: what is our churn rate by channel?<p>new analyst: where do we store &quot;channel&quot;? what do we use to process payments? where is that data stored? do we include discounts in MRR &#x2F; churn? etc.<p>If a human can&#x27;t do it, an LLM can&#x27;t either. An LLM isn&#x27;t able to write the SQL from scratch get the right answers without a ton of additional context. We&#x27;re working on an approach using a semantic layer at <a href="https:&#x2F;&#x2F;www.definite.app&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.definite.app&#x2F;</a> if you&#x27;re interested in this sort of thing.
评论 #39971028 未加载
评论 #39970947 未加载
评论 #39975836 未加载
评论 #39970997 未加载
dragon96大约 1 年前
Genuine question: does anyone here actually want to query their database with natural language?
评论 #39973138 未加载
评论 #39971346 未加载
zainhoda大约 1 年前
Nice job getting something released! How does this compare to the other similar open source solutions like Vanna AI and DataHerald?
评论 #39970978 未加载
Log_out_大约 1 年前
That &quot;natural language&quot; will magic and away complexity mindset has done so much damage.
jhoechtl大约 1 年前
&gt; As of the current version, MQL is designed to work exclusively with PostgreSQL
评论 #39970886 未加载
kelvinjps大约 1 年前
isn&#x27;t SQL already a way to query your DP with natural language?
评论 #39972161 未加载
roydivision大约 1 年前
Or one could, you know, learn SQL.
评论 #39973146 未加载
评论 #39970919 未加载