Reproducibility terminology/definitions

[Ed Warnicke]

Pol,

Formal definitions can be quite helpful.  I suspect your ambitions greatly
exceed my thoughts on the subject, but I thought I might share how I think
about it in the hopes it might prove of some assistance.

A build is reproducible if an equivalent set of inputs into the build
results in an equivalent set of outputs.

Typically 'inputs' is taken to include not only things like source code,
but also things like the build tools themselves, their configurations, etc.

Typically outputs are considered to be equivalent if and only if they are
byte for byte identical (although there are certainly folks exploring
alternate or looser definitions).

As to 'inputs' being considered identical, I've seen two popular choices:
(a) inputs are equivalent if they produce equivalent outputs and (b) inputs
are equivalent if they are byte for byte equivalent (I'm sure there are
others).

Hopefully these musings prove useful :)

Ed

On Wed, Nov 8, 2023 at 9:38 AM Pol Dellaiera <pol.dellaiera at gmail.com>
wrote:

> Dear friends, colleagues,
>
>
> I'm writing to express my enthusiasm for the discussions and initiatives
> that took place during the meeting in Hamburg. Although I was regrettably
> unable to attend, I have been closely following the outcomes and I'm
> particularly interested in the effort to enhance the website with clear
> definitions of terms related to reproducibility, a terminology.
>
> As a person deeply invested in reproducibility for work and personal
> purposes, I am keen on contributing to this initiative. I believe that
> establishing formal definitions is not only critical for our collective
> understanding but also serves as a beacon for the academic community and
> those new to the concept, guiding them towards a standardized comprehension
> of reproducibility.
>
> To that end, I'm currentlt drafting a formal definition of reproducibility
> that I hope to contribute. However, before I proceed further, I would like
> to know whether any of you have already worked on formulating such a
> definition. Collaboration or alignment with existing efforts would be more
> productive than working in isolation.
>
> Additionally, I would highly appreciate the opportunity to have my draft
> reviewed by peers within this group. Fresh perspectives and expert insights
> would be invaluable in ensuring the precision and clarity of the
> definitions.
>
> If you are interested in reviewing my work or if there is already a draft
> in progress that I could assist with, please let me know. I am eager to
> contribute and collaborate with you all on this important aspect of our
> work through a visio at any time.
>
>
> Looking forward for your response(s).
>
> --
> -Pol Dellaiera-
>   https://not-a-number.io/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.reproducible-builds.org/pipermail/rb-general/attachments/20231108/eec48f44/attachment.htm>