# Benchmarks for concept design?

**URL:** <https://forum.essenceofsoftware.com/t/benchmarks-for-concept-design/33>\
**Category:** Uncategorized\
**Created:** [November 24, 2021, 6:01pm UTC](https://forum.essenceofsoftware.com/t/benchmarks-for-concept-design/33 "2021-11-24T18:01:36Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![spdegabrielle](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.essenceofsoftware.com/spdegabrielle/32/21_2.png) [@spdegabrielle](https://forum.essenceofsoftware.com/u/spdegabrielle)\
**Post date:** [November 24, 2021, 6:01pm UTC](https://forum.essenceofsoftware.com/t/benchmarks-for-concept-design/33/1 "2021-11-24T18:01:36Z")

</div>

Hi,

I recently came across this blog post

> “an _expressivity_ and _design_ benchmark. We call it _B2T2: The Brown Benchmark for Tabular Types_ .”  
> – [A Benchmark for Tabular Types](https://blog.brownplt.org/2021/11/21/b2t2.html)  
> [https://blog.brownplt.org/2021/11/21/b2t2.html](https://blog.brownplt.org/2021/11/21/b2t2.html)

Is concept design benchmarking a thing? or am I off-topic with this link?

---

<div class="post-metadata">

**Author:** ![dnj](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.essenceofsoftware.com/dnj/32/17_2.png) [@dnj](https://forum.essenceofsoftware.com/u/dnj)\
**Post date:** [January 20, 2022, 1:33am UTC](https://forum.essenceofsoftware.com/t/benchmarks-for-concept-design/33/2 "2022-01-20T01:33:43Z")

</div>

As I’d expect for a product of Shriram’s group, it’s a really nice piece of work, taking the idea of benchmark—in short, build a suite of examples and known issues as the basis for evaluating a potential solution—and applying it in language design. Not sure where you were thinking this could be applied in concept design. But how about as a precursor to designing a single concept? You collect different scenarios and states, but unlike use cases (etc) you don’t expect these to be consistent with each other, or all of them to be included in the final design. Then the next step in the design is weighing the relative value of the different scenarios. This is of course something we do all the time in design (in all fields) but the benchmark format may suggest a way to structure it more?