Avogadro started in 2007 … and it took 11 years to hit a million downloads
As part of the 1.98 release, I looked up our current statistics on SourceForge
We’re now around 2.1 million downloads - we would have passed 2 million sometime in April 2023, so less than 5 years after our first million. Somehow, we’re seeing 25-30,000 downloads every October!
Since this doesn’t include most Linux packages, or the >100k downloads on GitHub, it’s pretty amazing for chemistry software.
So thank you. Thanks to an amazing community, from all the contributors, to the users, testers, bug reporters, … it’s amazing to see what’s grown up from a small start.
Don’t worry. We’ve got lots more planned.
Though correlation is not causation, the start of the academic year often is around this time, too. Speculation: there are introductions into quantum chemistry in lines of «Avogadro is an assistant to prepare (and eventually visualize the results of) computations with MOPAC, Gaussian, …».
Little bemol: the statistics by sourceforge is about Avogadro, the download button links to the old branch (e.g.
avogadro-1.2.0.tar.gz for a visit with Linux as operation system). While there are situations where this one can provide functionality which isn’t yet back in the branch leading to Avogadro2, I suggest a more prominent link to avogadro.cc or two.avogadro.cc than a mere «Follow Avogadro». Perhaps «A new version of the program is in preparation. See two.avogadro.cc for current development releases and documentation.», or similar.
Yes, the ~25-30,000 downloads definitely come from the academic year.
There’s a saying that you don’t get a second chance to make a first impression.
I’ve had a number of discussions at conferences, considered all the feedback here, and we’ve done several community surveys.
The main concerns about Avogadro 2 have been that it’s missing certain features and there isn’t documentation as to the changes, benefits, etc.
When we declared a “1.0” release, we were very careful to make sure it was really a good tool, because we knew crashes, bugs, etc. would reflect poorly.
In short, I’d rather keep pushing towards a 2.0 release … at which time, everything will be ready:
- updated user documentation
- plotting spectra
- “auto optimize” tool with constraints