SoylentNews

It's probably a little basic for many of the readers here, but today I put out a new video aimed at coffee roasters explaining how the measurement systems they rely on work under the hood. General knowledge stuff (what's a bit, how does analog to digital conversion work) but most coffee roasters have never looked at their systems from this perspective and a little bit of background knowledge makes it easy to troubleshoot data acquisition problems that can often be solved with a simple change to a hardware or software setting.

YouTube is about to get a lot spammier as changes to the YouTube Partner Program are set to reward channels that can produce a regular stream of click bait trash at the expense of niche channels producing videos of lasting value.

Channels currently in the program will need to meet new thresholds of 1000 subscribers and 4000 hours watched over the past year as of February 20 or get kicked out. This affects my channel as well. I started that in 2010 and have mainly posted coffee related videos such as How to Develop a Roast Profile. In the years since then I've heard from people who have used what they've learned in my videos to advance their careers in coffee, to improve their businesses, and to make their coffee better. I can't go to trade events, even ones on the other side of the globe, without being recognized from my channel.

Still, that doesn't always translate to view time. Last year I put together a weekly series of 1 minute videos. I got a lot of positive email response to the series, but it tanked my average view duration with the result that total view time for the months last year correlated negatively with new videos uploaded. The result is that I'm suddenly expected to drum up double the view time for a good month without much warning to avoid getting kicked out of YPP.

While I'm certainly concerned about my own channel, this is hitting a lot of niche educational channels that either have the view time but not the subscriber count or the subscribers without the view time. It's hitting animators who put in a huge amount of work for short pieces. It is leaving untouched the real "bad actors" at the root of YouTube's PR problems. And it is showing that YouTube does not value the huge number of small channels that make YouTube a site worth visiting on the Internet.

A while back I posted my experience attempting to use Amazon Video Direct. Since then I was able to get a little farther along so I thought I'd provide an update on how that's going.

The first thing that I tried to upload was a video covering my product development process. It's the most popular thing on my YouTube channel with close to 100K views and I've been told there are companies using that as a training video. It's a bit dated now, but it seems like the sort of thing that might be appropriate for this service. My first attempt to upload video was using Firefox on Linux. With this, the progress bar would indicate some progress for a few minutes but then it would rewind the progress bar back to 0 and repeat that a few times before producing an error message. Attempts at the same upload with both Firefox and Edge on Windows 10 resulted in instant failure, same error message. Safari on Mac OS X, however, did allow me to complete the upload despite some unrelated technical problems.

Whatever processing and content review is done after pressing the Publish button but prior to the video being available took nearly 24 hours to complete. I had chosen worldwide availability (where the world consists of the US, Japan, UK, and Germany, apparently). My video became available first in the US and Japan but within an hour of noticing that it was available there it also became available in the UK and Germany, continuing a long tradition of media getting to Europe late for no good reason.

Beyond the slow publishing process and the lack of notifications (yeah, you only get an email if a video was rejected for some specific content review reason according to the documentation. Nothing if there was a technical failure or publishing success unless you're constantly refreshing the web page.) it was at this point that I discovered that the documentation on how to get a link to your video so you can promote that (the ridiculous process documented is to search for your video and then copy a ten digit code and insert that at a particular place under the four regional domains) is wrong. You get a 404 error on 3 of the 4 domains if you follow the instructions. Not hard to figure out but it would be nice if documentation matched up with reality.

One nice aspect of the service is that if all things were equal, they seem to be offering to pay content creators significantly more than I'm getting from YouTube, but things aren't equal. It takes more effort for someone to start watching a video on Amazon's service and the apparent audience, at least for coffee roasting videos, is at least a couple orders of magnitude smaller than it is on YouTube. At the present rate that people are streaming, I should get a $100 payout in a little over 18 years assuming things don't pick up and the service lasts that long. That could change but I'm keeping my expectations low so I can be pleasantly surprised if it does.

There are three other videos on my YouTube channel that I intend to put on Amazon as a series and there are a couple more currently in production that I might put there just so I can get a better idea of how that's working out. I'll try out some browsers on Windows 7 and see if anything works there.

If this is intended as a killer of any competing video service there are a lot of things that need improvement.

Oh, and apparently Amazon thinks I was in the James Bond movie Spectre. I don't remember that at all.

[Update: Attempts at uploading video from browsers on Windows 7 also failed with yet another failure mode. Not the instant failure of browsers on Windows 10 but the progress bar made no progress for a while before the usual error message was presented. Looks like this is for Mac users only for now.]

A new version of Typica, the premier open source software for coffee roasters, is now available. Several new features are aimed at improving quality assurance processes. This is also the first release to be available in multiple languages (English and German). Mac and Windows builds are available on the project site. Source code is available on GitHub. The software leverages other open source software such as Qt and PostgreSQL. More details were posted to the release mailing list and there's also a release video.

Unfortunately, the machine that I use to build the Mac version died shortly after uploading the .dmg. This doesn't affect ongoing work on version 2.0 but it might make additional 1.x releases challenging if I can't fix that.

I recently acquired a camera that can record at 120fps and to test that out I made a video showing bad espresso extraction on bottomless portafilter. Playing that back at 12fps makes it easy to see some of the things going wrong. Video.

I've recently written a little article about coffee blending: the methodology I use and a recently designed example.

Today I came across this article that some people might find interesting to read (but I'm too lazy to do a proper write up/story submission right now). It's about a recently developed mushroom based powder that supposedly can be blended with bitter foods that blocks the perception of bitterness. The hook mentions coffee (which gastronomically is a sweet food so this would probably only benefit the low quality stuff) but the more commercially significant use for now is probably in combination with stevia.

Personally, I think it would be better to foster an appreciation for the diverse tastes available in food rather than try to make everything "universally palatable". The notion that everything must be sweet seems terribly boring to me.

Recently the topic of water for coffee brewing came up in Is Day-old Water Safe to Drink? and I mentioned that there was science behind the advice against using distilled water for coffee brewing. There was a talk at the 2015 SCAA Symposium (an event that preceded the Specialty Coffee Association of America's annual expo in April) by Maxwell Collona-Dashwood about the role water plays in brewing coffee. There isn't a transcript that I know of, but there is a video of the talk that people might find interesting.

In some ways the coffee industry is very quick to jump on fads, but in others it can seem overly conservative. Many are unaware or openly hostile to the benefits modern technology affords, so I decided to write an article on how my record keeping at the coffee roaster has changed over the years from the sort of paper logging still depressingly common in the industry to a modern system and some of the benefits specific changes have brought. It's aimed at professional roasters, but some people here might also find it interesting.

Today I finished putting together a new coffee roasting video (mostly done in QML) in what seems like it's going to be a regular series for me. This time I roasted three batches of a coffee from Costa Rica and tasted samples pulled at different points of each of these roasts to explore the concept of balance. This is mainly intended for people who already have some coffee roasting experience as I don't want to rehash things that I've already covered in other videos.