So with the code-context, the dictation in Listener is getting okay-ish. It's still pretty frustrating and error prone, but I can use it maybe 1/4 of the time (mostly for doc-strings). Part of the frustration is just that the language models are not yet well tuned for some commonly needed phrases that the tokeniser didn't ...

There are Python packages that are basically a PITA to get compiled into your virtualenv. The one I constantly hit when doing GStreamer, DBus, IBus, etc is gobject-introspection (a.k.a. gi). So, there's now a trivial package available:

source path-to-venv/bin/activate
pip install venvhpl
venv-hpl gi

So a few years ago, as I was working on a project called Listener that used PocketSphinx to allow dictation into the Eric IDE, I got some feedback suggesting that the only practical solution these days would be to use a neural network based dictation engine. That led me down a rabbit hole from which ...

Figured I'd try to update the old Dell 15" XPS with the broken keyboard as a machine for the kids to use as a desktop (for at-home schooling, etc) and do a bit of testing on the new release while I'm at it. The machine has been on 18.04, was updated and rebooted right before ...

So having finally written down the thoughts on a carbon tax, that kept distracting me from actually working on Open Source, I finally got a bit of work done on Open Source on the last night of the vacation.

What I started work on was getting a sampling profiler format supported, and for that ...

So if we (Canada) were to set flat carbon cost at a reasonable $.6/kg ($600/ton) instead of our current $.02 rising to $.05 in two years with the big polluters excluded, what would happen? Would it be effective at changing people's behaviour?

Fuel

• Crude would go up by $258/barrel (currently $60)
• Oil Sands crude up ...

I've pushed out the PyOpenGL 3.1.5 release. It has some (relatively minor) fixes for wgl and egl operations, and one for what looks like a change in numpy scalar handling in the latest numpy. Available on PyPI now.

So it has been a long time since I forced myself to boot into Windows to get support/development work done for the platform. Other than one hard-freeze where the machine updated its drivers, rebooted, and just stopped recognising the keyboard or trackpad, it pretty much just worked (turning it off and on again did, as ...

So it turns out that there's an extension for getting the DRM name for an EGL queried device that seems to work on Ubuntu 19.10. With that it should be relatively easy to target an off-screen render to a particular device. (The extension allows `eglQueryDeviceStringEXT` to respond to `EGL_DRM_DEVICE_FILE_EXT`). Happy New Year all.

So I've now spent the last day and a half playing with getting EGL offscreen rendering working on Linux. There are two major ways to do off-screen rendering with EGL and OpenGL. In the first, you use a pbuffer surface, that surface is basically a purpose-defined surface-type for off-screen backing of a renderer. When I ...