Scrapy Truck Factor

[Mívian Ferreira‏]

Dear Scrapy contributors,

I am Master's student investigating the importance of the idea of truck factors in Software Engineering.

As you know, truck factor (aka bus factor) is defined as “the number of people on your team who have to be hit by a truck before the project is in serious trouble”. By "serious trouble", we mean that it will very difficult to maintain the project (e.g., bugs will take more time to be fixed, new features will be delayed etc).

I would appreciate your contribution by answering two questions:

1) According to your experience, what number best represents the truck factor of scrapy?

2) Who are the developers responsible for this truck factor (i.e., the name of the developers who are responsible for the TF result answered in the previous question)?

Mívian Ferreira

Master's Student

Applied Software Engineering Group (ASERG)

UFMG, Brazil

http://aserg.labsoft.dcc.ufmg.br/

[Alex‏]

Mivian, given that you've received no feedback yet, you might say the bus factor is `NaN` (-:

I think you'll be better of trying to analyze the activity of the project on Github: https://github.com/scrapy/scrapy/graphs/contributors

It says there are over 200 contributors, but you'll have to determine which of them are actually making a difference.



This mailing list is probably not a good indicator of activity, I asked a question a while ago and have had no feedback either. It could mean that:
- the question is trivial and people are silently letting me know I should RTFM
- the real action is in the IRC channel
- nobody knows
- nobody cares

Scrapy has a pretty good documentation and there are plenty of tutorials around, so I believe that makes the bus factor bigger (i.e. key-knowledge is not concentrated in the heads of a small number of people, but is distributed across multiple minds).


Where can I read more about the results of your analysis?

[Paul Tremberth‏]

Hi Mivian,

First of all, sorry for such a late reply.

Thanks for the question, it made us add maintainers for Scrapy's package on PyPI!
We realized that not all channels have the same admins (GitHub vs. PyPI vs. ReadTheDocs vs. IRC vs. mailing-list)

To answer your questions:

1) According to your experience, what number best represents the truck factor of scrapy?

I would say that it's currently 5 or 6.

2) Who are the developers responsible for this truck factor (i.e., the name of the developers who are responsible for the TF result answered in the previous question)?

First the founding team, they are still around though much less involved in the development nowadays, yet they know the project very well:
- Pablo Hoffman (https://github.com/pablohoffman)
- Shane Evans (https://github.com/shaneaevans)

The core maintainers, they contributed a ton and still contribute a lot, even if simply reviewing sometimes, they have in-depth knowledge of many parts of the framework:
- Daniel Grana (https://github.com/dangra)
- Mikhail Korobov (https://github.com/kmike)

Other notable contributors, recently more active, helping users with documentation, support and bug fixing:

- Elias Dorneles (https://github.com/eliasdorneles)

- Paul Tremberth -- myself (https://github.com/redapple)

Obviously, scrapy receives a lot more contributions than from the people listed above,

and some do commit serious improvements to the framework.

I narrowed the list to the people with knowledge, contributions and admin rights.

Hope this helps with your study.

Best,

Paul.