This is a blog post in two parts…
- A brief review of the new Screen Record feature for Chrome OS
- A mathematical exercise to justify to myself that I was fired because of bad maths.
Chrome OS Screen Record
In early 2021 Google released a Screen Recording tool for Chrome OS (Chromebooks). It is available in the stable channel (you don’t have to switch to Beta or Developer mode) but it is hidden… a bit. To enable it you need to open up chrome and type…
…to open up the “Experiments” page.
Search through the experimental features for “Screen Capture Test” and enable it. You will be asked to restart Chrome and once you restart you will have a new control in the system tray.
When you click the “Screen capture” control you get a new toolbar across the bottom of your screen.
From left to right the first icon is for feedback to Google (same as Alt-Shift-i), the camera icon is for screenshots (very handy when you can’t remember the hotkeys), the video icon is for screen recordings, the dotted square is for full screen, the dotted square with the + is for capturing a region of your screen, the window icon is for capturing the contents of a window and the cross is to close the tool.
As I mentioned above, for screenshots this is handy because you don’t need to remember the hotkey combination but it will also be handy for taking screenshots when your device is in tablet mode.
I found the controls very intuitive as there were visual cues to assist you such as…
- tooltips when you hover over the different options
- an overlaid message to touch anywhere to record
- a countdown from 3 to start the recording
- a circle on the taskbar with a square inside, the universal signal for “STOP!”
For the following part of the review you can refer to the video at the top of the post. As you will see for this simple test all I recorded was the camera app and Google Keep. The Chromebook I am using is the top tier 2017 Pixelbook with the i7, 16GB of RAM and 512GB of storage. I am also using the Pixelbook Pen.
I have done a little bit of screen recording on Linux, Windows and Mac and one of the tricks to include vision of yourself within the screen recording is to open the camera app in the respective OS. You don’t actually use the camera app to record but you record the GUI of the camera app with the live image of yourself on screen. In this respect I think Chrome OS struggled a little. As you can see in the recording there is quite a bit of lag between the audio and the video. My Linux/Windows machine is quite powerful as is my MacBook Pro and both seem to achieve this recording trick but perhaps the Pixelbook I was not quite up to it.
I found the pen input to be reasonably responsive during the recording but there was occasionally a bit of lag causing me to miss a letter or accidentally write something twice. You may also notice the occasional notification that my “stylus battery is low” so that may also have had an impact on the performance.
During the recording I occasionally had to stop and restart recording because of interruptions (clients coming to the door, phone calls, etc.) and it seems that some of the recordings may have been recording at a higher resolution than others. I have not taken the time to confirm this yet but during the video editing I was experiencing poor performance playing back some videos. The playback would stutter and the audio had a different depth to it. I think some of the recordings I performed in full tablet mode and others in desktop or tent mode so this may have had some effect. I will need to investigate further to confirm this.
Overall I found the video and audio quality to be good but the file was saved as a *.webm which I am not very familiar with. Apparently it is a new open codec designed for the web but for me I had to use ffmpeg to convert it to *.mp4 so that my video editing software could handle it.
That’s all I have to say about screen recording for now so now on to…
The Maths of Being Fired
I took some work in a factory to supplement my income. It was unlike any work I have ever done and very physically demanding. I enjoyed the work but was having trouble achieving the speed they expected from me.
As an example of how demanding the job was, I was eating a lot more food than I usually did and I was still losing lots of weight. My weight was swinging down as far as 10kg as I lost weight and back up as I gained muscle and then back down as I lost more. After two weeks I was able to buckle a belt around my waist that I had not worn in 20 years. After another 2 weeks I could wear it on the third hole.
Another gripe I had was that there was two sides to the job (two person job), one side was very physically demanding and the other side was very light duty. I was told on the first day that the job rotates on a two week cycle however I was kept on the physical side for four weeks straight to try and improve my speed. In my perspective pushing someone harder and exhausting all their energy is more likely to slow them down than speed them up. It has now been over a week since I left the job and my hands are still recovering.
Despite the fact that I pushed myself as hard as I could and was exhausted, by week three I was finding some speed. Wednesday and Thursday I got the work completed in time and on Friday I was on target until the last job when I metaphorically hit a wall. The last job involved one of the biggest models they make. The completed unit weighed approx 35kg and I had to build about 40 of them. The little amount of strength I had left was quickly used up and I didn’t get the job finished. However in light of this shortfall they kept me on another week.
Week four was when it all went wrong. The full details are in the video but in a nutshell, there was a change made to the job process due to a lack of accessory boxes. The supervisor figured that he could take the person usually packing the accessory boxes and put them on the other side of the job. He figured that by having two people building he could almost double the output but leave the boxes open. Then when the accessory boxes arrive both workers could pack accessories until all units are packed off in full.
In theory this could make sense but while we were attempting this new schedule I started processing the maths in my head. The following facts jumped out at me:
- It still takes two people the same amount of time to test and pack the units even if the boxes aren’t being closed (and more space is needed since they can only be stacked one high).
- Both people still need to do their initial picking work, person one picking parts for the build, person two picking some accessories and assembling the packaging.
- Building is the one area where you can “double-up” the work but in a normal cycle person one performs 3/4 of the building and person two performs 1/4 of the building. Putting person two on building for the whole build cycle only gives you 150% of the normal total building time.
If you can’t already see how this new job process is doomed to fail, watch the 20 minutes of maths on the video explaining it 😉