I was curious about the rate of bug reporting in Fedora, and did this quick experiment. I thought it might be interesting to folks here who either work on the infrastructure or are curious about long-term collaboration trends in Fedora.
I checked the date of reporting of every 10,000th bug (bugzilla #1, #10000, etc, all the way to the recent 1150000---see attached data). Some bugs were private so I didn't have access to their info, but I got enough data to calculate bug velocity (increase in the bug number divided by the time interval) over time. The data is a little noisy, but you can clearly see the ever-increasing trend.
See attached plot; it turns out that Bugzilla is receiving presently ~400 bugs per day and rising (a bug every 3 minutes or so).
On Thu, Oct 09, 2014 at 05:39:34PM -0400, Przemek Klosowski wrote:
I was curious about the rate of bug reporting in Fedora, and did this quick experiment. I thought it might be interesting to folks here who either work on the infrastructure or are curious about long-term collaboration trends in Fedora.
I checked the date of reporting of every 10,000th bug (bugzilla #1, #10000, etc, all the way to the recent 1150000---see attached data). Some bugs were private so I didn't have access to their info, but I got enough data to calculate bug velocity (increase in the bug number divided by the time interval) over time. The data is a little noisy, but you can clearly see the ever-increasing trend.
See attached plot; it turns out that Bugzilla is receiving presently ~400 bugs per day and rising (a bug every 3 minutes or so).
This is pretty fun, but doesn't it speak about bugzilla usage in general (ie: Fedora, EPEL, RHEL, Atomic...)?
Or did you make sure you're only getting Fedora's bugs?
Pierre
On Thu, Oct 09, 2014 at 05:39:34PM -0400, Przemek Klosowski wrote:
I was curious about the rate of bug reporting in Fedora, and did this quick experiment. I thought it might be interesting to folks here who either work on the infrastructure or are curious about long-term collaboration trends in Fedora.
I checked the date of reporting of every 10,000th bug (bugzilla #1, #10000, etc, all the way to the recent 1150000---see attached data). Some bugs were private so I didn't have access to their info, but I got enough data to calculate bug velocity (increase in the bug number divided by the time interval) over time. The data is a little noisy, but you can clearly see the ever-increasing trend.
I'm afraid there are a couple of problems with this analysis:
- Automated bugs (eg from abrt) may or may not be considered to be real bug reports.
- Bugzilla has both imported large bug sets from other databases at various times (when Red Hat acquired other companies), and also has had periods when it didn't allocate bug numbers sequentially. At one point IIRC each new bug report incremented the bug ID by 10.
Rich.
On 10/10/2014 06:12 AM, Richard W.M. Jones wrote:
On Thu, Oct 09, 2014 at 05:39:34PM -0400, Przemek Klosowski wrote:
I was curious about the rate of bug reporting in Fedora, and did this quick experiment. I thought it might be interesting to folks here who either work on the infrastructure or are curious about long-term collaboration trends in Fedora.
I checked the date of reporting of every 10,000th bug (bugzilla #1, #10000, etc, all the way to the recent 1150000---see attached data). Some bugs were private so I didn't have access to their info, but I got enough data to calculate bug velocity (increase in the bug number divided by the time interval) over time. The data is a little noisy, but you can clearly see the ever-increasing trend.
I'm afraid there are a couple of problems with this analysis:
- Automated bugs (eg from abrt) may or may not be considered to be real bug reports.
Just a COUPLE? Thank you, kind Sir---I would have said it's rife with problems. For one thing it totally ignores the distinction of Fedora vs. EPEL and RedHat bugs. Still, it gives an order of magnitude estimate and the trend. No matter what the origin of the bugs, they still have to be dealt with somehow.
- Bugzilla has both imported large bug sets from other databases at various times (when Red Hat acquired other companies), and also has had periods when it didn't allocate bug numbers sequentially. At one point IIRC each new bug report incremented the bug ID by 10.
Ah so that's where the spikes come from. I was wondering.
On Fri, 2014-10-10 at 11:12 +0100, Richard W.M. Jones wrote:
On Thu, Oct 09, 2014 at 05:39:34PM -0400, Przemek Klosowski wrote:
I was curious about the rate of bug reporting in Fedora, and did this quick experiment. I thought it might be interesting to folks here who either work on the infrastructure or are curious about long-term collaboration trends in Fedora.
I checked the date of reporting of every 10,000th bug (bugzilla #1, #10000, etc, all the way to the recent 1150000---see attached data). Some bugs were private so I didn't have access to their info, but I got enough data to calculate bug velocity (increase in the bug number divided by the time interval) over time. The data is a little noisy, but you can clearly see the ever-increasing trend.
I'm afraid there are a couple of problems with this analysis:
- Automated bugs (eg from abrt) may or may not be considered to be real bug reports.
~35% bugs are from ABRT : https://jfilak.fedorapeople.org/media/abrt_bz_stats.txt
Regards, Jakub
bug velocity (increase in the bug number divided by the time interval) over time
That quotient is a scalar. Please use "bug rate" instead of "bug velocity".
Easy mnemonic: Velocity is a Vector, Speed is a Scalar. "Velocity" has multiple syllables and multiple coordinates, "speed" has one syllable and one coordinate.
[Irony: the original poster has an address @nist.gov (US National Institute of Standards and Technology), where they should know and appreciate the difference.]
On 10/11/2014 09:44 PM, John Reiser wrote:
bug velocity (increase in the bug number divided by the time interval) over time
That quotient is a scalar. Please use "bug rate" instead of "bug velocity".Easy mnemonic: Velocity is a Vector, Speed is a Scalar. "Velocity" has multiple syllables and multiple coordinates, "speed" has one syllable and one coordinate.
As a physicist, I see your point but I believe you are mistaken in principle. Velocity is indeed a vector quantity, but it is a vector in the parameter space you are considering. In our 3D world, velocity vector is a time derivative of the 3D displacement. In the bugzilla case, the system variable is scalar, so velocity is a scalar too. Normally, speed is the absolute value of velocity---one has to work hard at teasing it out of simple-to-get velocity; in this case they just are the same.
'bug velocity' just rolled off the tongue, simply because 'bug' is monosylabic and it somehow created a nice 'foot'. 'bug speed' sound awkward to me, although I kind-of like 'bug rate'.
[Irony: the original poster has an address @nist.gov (US National Institute of Standards and Technology), where they should know and appreciate the difference.]
Please don't mix my work in it. I don't speak for them, and they don't for me, especially on Fedora issues.
-
Jakub Filak -
John Reiser -
Pierre-Yves Chibon -
Przemek Klosowski -
Richard W.M. Jones