Comment by cj | Hacker News Reader

Comment by cj | Hacker News Reader

Parent Full thread

IANAA, but very important to remember that:

1) R&D classification of a single employee should be a percentage. Accountants don't list employees one by one on a tax return, they're grouped together and a percentage of their wages can be allocated to different functions of a business outside of R&D.

2) A lot of things do not qualify as "R&D". For example...

- Fixing bugs

- Maintaining existing features

- Improving existing features (even enhancing functionality)

- Building a feature because sales promised it to a customer

- Refactoring code

- Anything related to devops/infrastructure, upgrading servers, updating dependencies

- Anything related to customer support, onboarding or retention

- Anything related to helping Sales/Marketing close new business or get more leads

- Anything related to supporting day-to-day business operations

I'm not an expert, but I think it will be relatively easy for accountants to do some magic behind the scenes by classifying only a small percent of wages toward R&D.

I think the average engineer spends less than 20% of their time on "true" R&D, and if that's the case, 80% of wages can be deducted as an expense without amortizing.

Do you have a source for this? The law seems pretty clear to me:

> ANY amount paid or incurred in connection with the development of ANY software shall be treated as a research or experimental expenditure.

https://www.law.cornell.edu/uscode/text/26/174

There's a very important clause immediately before that though: "For purposes of this section".

Also the parent clause says: "In the case of a taxpayer’s specified research or experimental expenditures for any taxable year—"

So just don't specify the expenditures as research or experimental if you don't want this section to apply.

I guess there might be a distinction between "development" of software and "maintenance" of software (e.g., bug fixing, refactoring, etc.)?

Without getting into the finer points, my understanding is this: My company needs to make metal squares. There is a defined process for it. I just need a welder to weld all four corners. No R&D here.

I need a website with some features. Asking a sample of professional software developers how to build it, you get a variety of answers.

If anything, that the basics of software development still requires R&D is an indictment of our fields lack of professionalism.

Wait, what? There's more than 1 way to make metal squares, and which you want to choose is gonna vary based upon your needs and the fabricator, even if you, for whatever reason, pre-determine you want them welded!

> If anything, that the basics of software development still requires R&D is an indictment of our fields lack of professionalism.

This is a bit much. "Research" maybe, but not "development". Many things are developed -- land, buildings, etc. Someone still has to do the work.

> My company needs to make metal squares. There is a defined process for it. I just need a welder to weld all four corners.

What are they welding that has four corners to make squares?

> I think the average engineer spends less than 20% of their time on "true" R&D

Based on all of the items in #2, the average software engineer spends close to 0% of their time on R&D - with rare exceptions for those working on cutting edge tech, in research roles or maybe super senior engineers figuring out how to solve complex technical problems

The rest of us just write code and sit in meetings

It would be interesting if this reintroduced "10% time" with strict wording about not working on R&D in your 90% time. Just to simplify the labor involved in calculating it.

Reply on news.ycombinator.com