We would recommend firefox, since we really like the Mozilla foundation, but the performance of it is so bad we have to steer people away from it.
It is a sad state of affairs.
I would have thought it you had more data than that then the points will all be rammed together and hard to distinguish, so it would be better to switch to density plots.
I once checked out chromium in order to test out whether file writing would be easy to do. The size of the project alone meant it took me forever to find out where to patch it.
(Edit: This was something like 5 years ago before I started to grow a distaste for Chrome)
I've sometimes wondered whether the barrier to entry is the biggest thing holding back FOSS.
I'm an experienced developer, quite capable of diving into the kind of code these projects run. Now and then I have enough spare time that I might be able to make a useful contribution. I would be happy to support some of these projects in the spirit of giving something back.
Every single time I look at a major FOSS project where I'd like to help, the very first thing I run into in the developer documentation is an installation/build process a mile long with absurd overheads. Large FOSS projects often require a whole set of custom tools. Many don't comply with normal conventions on the host OS; I'm on Windows, and assuming every developer in the world runs Linux is a particularly common problem. Quite a few require installation of a specific version of a certain compiler toolchain.
I understand that portability is difficult. I've worked on projects that had to be compiled on many different platforms. I've worked on projects that had to be cross-compiled from one platform to run on another. The issues raised by these environments aren't trivial.
Even so, I have never worked on a successful professional project that managed to make things as complicated as large FOSS projects almost always do. And while I'm happy to help out if I can, I simply don't have time to ensure that I've got this week's project's specific preferred version of a VCS, Bash, GCC, and so on installed so they clutter my standard file system, environment settings, registry, editor/IDE preferences, and all the other things that any developer surely wants as clean and tidy as possible while they work.
I was about to say that for the two projects I've tried this with, Filezilla and Firefox, it was actually quite reasonable. Especially Firefox, given its size, was surprisingly simply even if compiling took longer than I thought.
But then I read you are on Windows. Yeah, that's hell. I tried a few hours on Filezilla before deciding that it's just too messed up.
Do these projects expect that everyone will just set up a dedicated VM with exactly the right development environment for their code or something? I think that's probably the most practical solution in that kind of environment, but presumably it's a level of of commitment that only dedicated, long-term contributors are likely to make. Certainly it's a disincentive for anyone to contribute one-off fixes or improvements in specific areas, and a significant barrier to entry even for those who might become long-term contributors if they get that far, which surely can't be helping these projects.
Actually, I have. I know it's not small, but Firefox' code was quite structured.
There are platforms where developers need to stop developing. iOS was the main one; it has extreme lock in, an unfriendly development environment, an extremely unfriendly publishing environment and is actively doing evil things. But no, all aboard the ios apps train!
Windows I could say similar things about but given the competition I really can't. Android... Definitely not. Some of the most free mobile OSes run atop of android.
But when it comes to developing on platforms in general, developers absolutely need to use themselves as a voice to tell other platforms what is good, what is needed, etc. There are excellent general desktop features that people have found on mac os, windows and linux alike. Without developers to make use of them, regardless of the "open" or "closed" state of the platform, users are worse off.
Extensions are more powerful on Firefox but I think they are coplicated as well. There are many popular extensions that just don't have updates on Firefox. One example is Evernote. I used it a lot on Chrome but their Firefox extension is almost 1 year old version and Chrome/Opera extensions have much better UI now. I hope this improves soon.
Creating an add-on for a product with a smaller market share is not that bad. But for a product with 60%+ that also belongs to a company that "owns" Search, Maps and Email and is known to play dirty it's a different story.