Mesa Is On Track Again To See More Than 10,000 Commits This Year

[phoronix]

Phoronix: Mesa Is On Track Again To See More Than 10,000 Commits This Year

I was curious this weekend how Mesa's development was trending this summer so yesterday I ran some fresh GitStats on the growing Mesa code-base...

http://www.phoronix.com/scan.php?page=news_item&px=Mesa-Stats-20170806

[bregma]

Well, first off, a high number of changes to a project is not indicative of stability and maturity. Sure, you could argue "more changes is more good stuff for us as consumers" and I figure that's what you're implying, but there's also the possibility of code churn, and every change stands a non-zero chance of introducing new bugs.

Secondly comments, blank lines, and documentation are just as much a part of the literature that makes up programming as are the instructions to the compiler. If all the changes to the codebase were to improve the comments, readbility, and documentation that could only be seen as a good thing, so there is no need to discount them.

It would be more interesting to see a report on which parts of the Swiss army knife of a project like Mesa is getting all the attention. The software renderer? The OpenGL layer (keeping up with the ever-changing spec)? The Vulkan layer (same)? The hardware drivers? The CIA backdoor?

[schmidtbag]

Well, first off, a high number of changes to a project is not indicative of stability and maturity. Sure, you could argue "more changes is more good stuff for us as consumers" and I figure that's what you're implying, but there's also the possibility of code churn, and every change stands a non-zero chance of introducing new bugs.

To my understanding, many of these changes ARE bugfixes and/or improvements. Not all are new features. Generally though, Mesa is pretty stable.

Secondly comments, blank lines, and documentation are just as much a part of the literature that makes up programming as are the instructions to the compiler. If all the changes to the codebase were to improve the comments, readbility, and documentation that could only be seen as a good thing, so there is no need to discount them.

True, but users aren't exposed to the comments. Most users just care about how much substance (that affects them) has changed, which is just the code itself. Whether there's a sentence or an entire paragraph of comments, that doesn't affect the compiled result.

It would be more interesting to see a report on which parts of the Swiss army knife of a project like Mesa is getting all the attention. The software renderer? The OpenGL layer (keeping up with the ever-changing spec)? The Vulkan layer (same)? The hardware drivers? The CIA backdoor?

I definitely agree here - I'm interested in what gets the most attention. Not only would it be interesting, but it might get devs to realize what may be lacking attention.