Hacker News Reader

Comment by djtriptych | Hacker News Reader

I think pragmatic businesses that employ finance pros / quants are wise to include Excel as an essential part of their product development workflow. Excel is a great prototyping tool, and people are going to use it whether you want them to or not. These folks are motivated by money and don't give a shit about your tool if (they think) it costs them money (or time). Besides that, how many enterprise applications really offer opportunity for prototyping and running small experiments? Excel is to modeling as wireframing is to UI design.

Someone should write a book about productionizing excel models. It's not going away (and shouldn't).

View on HN

> Someone should write a book

Nah, this is not gonna work... The only way out is to make something that's somewhat like a spreadsheet but with source control, a nicer language etc.

I was writing it and then the firm I was working out threatened to sue me for bullshit IP infringement...

Sounds like Excel needs a git extension!

A quick google search shows some people are already working on that, including https://www.pathio.com

Schematiq - https://www.schematiq.com also handles source control for spreadsheet and a lot more. The platform is about making sheets enterprise grade applications - regression, testing, CI etc and introduces new features to simplify spreadsheet logic.

I remember I wrote not long ago a VB Macro for Excel, where all the VB code inside was imported/exported, git pull/commit+push, etc, all of a button push.

It became an instant hit in that company and I short of a medal :).

Note that Pathio has been relaunched as https://www.xltrail.com - it supports the standard Git workflow and every Git system (GitHub, Bitbucket, Gitlab etc.).

Or seek greater adoption of standards around spreadsheets, such as FAST [1].. Flexible, Appropriate, Structured, Transparent.

[1] http://www.fast-standard.org/

Ehh I doubt most quants are interacting much with Excel except when they need to work with non-quants. There's not much of a reason to use Excel when you could use something like Pandas instead, plus the Python libraries/environment will make everything non-tabled better to deal with anyway.

it's different now but historically quants were not well versed in software engineering. If anything it would have been R 10-15 years ago.

R was almost irrelevant in finance 10 years ago, Matlab was much more widely used. R may have more momentum now, but I’m not sure it has overtaken Matlab yet. And C++ was, and probably remains, king.

I think a lot of the discrepancy is explained by the fact that quant is a very widely applied title. There are quants who spend most of their day in Excel and quants who spend most of their day in C++.

10-15 years ago it was excel excel excel for quants in most investment banks.

Reply on news.ycombinator.com