AWS, Microsoft and Google collectively made up 65% of global spend on cloud computing in Q1 2022, and their share is increasing year-on-year. At least two of these organisations have shut down users and companies they decided did not align with their ideologies. If you put your data on the cloud, it sits somewhere. In many cases, it's on the servers of these three companies who may or may not decide to cut you off without warning sometime in the future. It is also important to remember these three companies have servers across the globe and if a country decides to remove itself from the pack, it could take a peek at what you are storing there.
- Should there be a time limit on opinions? Yahoo and MSN Messenger started in 1999, Skype came out in 2003, with Facebook the following year and Twitter in 2006. That means as of today, you could have stuff on these platforms going back 16-23 years. Before that, things were mostly written down, said directly to someone's face, or over the phone. Unless someone was eavesdropping, recording the call or somehow copied your diary, these thoughts and expressions were lost to time or memory. Think back to your teenage years. Your prefrontal cortex is still developing with raging emotions, which according to literature continues on up to 25 on average for men and 20 for women. You're in the middle of working out who you are and what you like and don't. You're being bombarded in university with postmodernist dogma and a swathe of new concepts. During that time, you will have changing opinions on everything and say things one week that have been replaced by something else two weeks later.
- Skip forward to now. Some prospective employers will do a full web search to see what you have posted. Some will engage companies that specialise in this. If you decide to get into any form of politics, the dive for dirt will be deeper by some opponents. You will have seen what happens in Supreme Court nominations in the US. Multiply this by a large number and this is what's occurring today. Let's have a time limit on this. While that cortex is developing, anything said is off limits soon after. Anything else from say 10 years back is also irrelevant unless they have repeated the concept since. Wayback Machine has its uses but dredging up stuff from ancient social media history shouldn't be one of them.
- If you have a favourite Apple app that hasn't been updated over the past three years, they will be removing it from the App Store. Developers will get 90 days to update the code or be dropped.
- If you own a tractor, Thomas Calburn will be disturbing "millions of dollars worth of John Deere agricultural machinery stolen from a dealership in Ukraine by Russian Federation forces that has been traced to the Chechen Republic and bricked, it is reported". It appears some modern equipment can be remotely disabled. Late on your payments and working on the land? Your equipment may just stop working. With the advent of chip-based products also comes the chance that they can be shut down without your control.
- For lovers of artificial intelligence and sound mixing, I watched a video on the new plug-in from Waves, the Clarity Vx. The product will allow the removal of noise from a vocal track. Not just the background stuff that others can do but singing in the middle of a construction site, random kind of stuff, even in real time if necessary. The developer described the process required to train the AI system to do this. He used over a million segments of recorded noise in a wide range of circumstances, listening to a decent sample of this himself. He also described how important it is to stop the AI from going too far and attack the materials you do not want affected. In response to a question from a viewer, he also pointed out that the product would not be able to remove, say, a second voice for a couple of reasons.
- An AI is trained to do a certain thing, like recognising something a car shouldn't run into. Unless it has been trained to do so, it may not avoid something falling out of the sky. If the focus is a vocal only, it may not be able to distinguish between one of many voices. There are sometimes side benefits to AI training. In the example of Clarity, it does a good job of extracting the solo singer from a music track. In summary, an AI is only as good as the training it received and the focus of the training will be the only thing it can do, hopefully well. If you create online content in challenging sound situations, then I recommend you check out this new plug-in. I have no personal association with Waves other than owning some of their plug-ins.
James Hein is an IT professional of over 30 years' standing. You can contact him at jclhein@gmail.com.
