This is interesting, and it's good to see something that's actively developed, but it seems to have some of the same issues as ffmpeg-python:
- It doesn't appear to have any way of specifying filters that do not have inputs, such as "color"
- There is no way to provide flags to Popen, e.g. to specify subprocess.CREATE_NO_WINDOW to avoid CMD windows popping up in a GUI app on Windows. This isn't a big deal for running ffmpeg itself, because you can just ffmpeg.compile() then run it manually, but that can't be done for ffprobe with ffmpeg.probe().
Edit: OK, figured out the source filter thing, ffmpeg.sources.color. Is there a way to use arbitrary source filters, like vfilter/afilter can do for regular ones?
> 2) there's typically no privacy-preserving way to pay
This is exactly my problem. I log into as few services as possible and clear my cookies on browser exit. Of course this is supplemented by ad-blocking and blocking most 3rd-party scripts and cookies to avoid cross-site tracking. I have no way to pay for a service while maintaining that level of privacy.
On Debian (and probably most of its derivatives), the package manager will not remove any running PostgreSQL versions and will let you have multiple versions side by side. It also has its own tool pg_upgradecluster, which can perform the upgrade using pg_dump+pg_restore or pg_upgrade (optionally with link or clone options).
Yes, I definitely miss this in Ruby since it also has optional parens for method calls but lacks sigils. Even the most basic-looking expressions can have arbitrarily complicated meanings:
foo.bar = 1
Is that just an attribute set on an object, is it calling a method foo that returns an object containing a bar attribute with a normal getter/setter, or is bar= a method that has arbitrary functionality?
I also like the more explicit operators in Perl that help show context:
$foo + $bar
This is specifically numeric addition and those operands are probably at least somewhat number-like. Yes you can potentially overload it, but by convention, it will still be something like addition, so you don't have to wonder if you are adding, concatenating, appending a list, or something else, because they have different operators.
dist-upgrade is more likely to break things because it can uninstall packages. So it's sometimes useful to run a plain upgrade first just so there will be fewer packages in the dist-upgrade. This lets you review the dist-upgrade output closer and perhaps decide to cancel it if you need to make some changes before it's safe to run. If you are going to ignore the output or just run with -y, there's no reason to split them.
You can post to YouTube and also another platform, then on Patreon/Twitter/Facebook/etc always link the other platform instead of YouTube. That way you would still be discoverable on YouTube, but hopefully get some amount of traffic funneled over to the other platform in case YouTube kicks you off.
They're not really subdomains. If you try to do a whois against e.g. smith.name you'll see it's not a registerable domain. Only the second level can be registered, xxxx.smith.name. Similarly if you had xxxx.co.uk, you own a domain because co.uk isn't registerable. The only quirk about .name is that common names can't be registered at the top level, but uncommon ones can.