I think there is a future in cryptocurrency (crypto). However, I don't think there is a future in the crypto that is currently available. Although I don't understand all of the mathematics behind it, the economics has major problems. The way new coins are created creates scarcity, which drives up prices to ridiculous levels. It's a speculative investment, not a currency.

A successful currency should be widely available, and have a stable value that gradually loses value, not the other way around. Therefore, I propose the following:

First, this currency relies on statistical principles to function. Therefore, the miners of this coin do not get to select which transactions they process, and which ones they do not. Transactions are selected randomly, to meet statistical requirements.

Next, a trendline will be calculated on the value of past 10,000 entries to the blockchain over time. If the slope of the transaction value trendline is not decreasing at a rate of 3% per year, (I use the US federal reserve target for inflation to obtain this target) a new coin is minted and rewarded to the miner. This "interest" is the miner's reward for doing the work to make transactions possible. It also discourages hoarding of coin as it will lose value at a known rate over time.

I chose 10,000 as that should get enough certainty to measure inflation to within 1%. If there are less than 10,000 entries to the chain, the trendline computation will not be run, and no new coin will be minted. There will need to be some bootstrapping to get the process to the first 10,000 transactions.

I have no idea how to do this, and I especially don't know how to do it securely. I am also concerned about developments in quantum computing blowing up the entire cryptocurrency ecosystem. I just want to get some thoughts written down.

I've been coding in MATLAB for over a decade. In the past couple years, I've transitioned to Python. However, my training is in mechanical engineering. In school, our curriculum was very much focused on making a program work, and not making it robust. No time was spent on programming best practices. Are there resources to learn best practices, without going back to square one?

I have a bunch of old MP3 files that have varying perceived loudness. I used some tools to correct this over a decade ago, but for some reason they didn't get archived.

I tried using the Linux utility Normalize-Audio to correct the loudness issue. This software seems to be full of bugs. While it evaluates a batch of files, it only corrects the last file in the batch. That means each file has to be corrected individually. Upon correcting a file, there is no discernible difference. Running the program again outputs the file still needs correction. Bottom line, it doesn't work at all.

I dug up an old .deb file for mp3gain. Unfortunately, I had similar results to Normalize-Audio. It seems to set a flag in the MP3 tag, but it has no effect on the file's playback.

The best way to normalize an MP3 is Audacity. It actually decodes the MP3, normalizes it, and reencodes it. It can be done very simply with a macro GUI. Don't bother with other options. This is the only way.

Audacity's denoise effect is also amazing. Just highlight a quiet part, where the noise is obvious, run the tool, and click the noise selection tool. Select all and run the tool again and click ok.

Sharing a file or folder on Windows can be done accidentally. In Linux, it's a disaster. Upgrading from Mint 19 to 20 broke my Samba configuration. Getting it running again was a major project, compounded by Systemd issues. It seems that Samba might work great for an enterprise that has a lot of resources to devote to getting it running well with all of its features. However, setting up file-sharing for home use is a disaster.

Edit: The issue was found to be with a change to Samba in Mint 20.

My Wife recently tried to create a PDF on her Windows machine and realized there was no software on her machine to do so. Back around 2008, good free PDF printer software was ubiquitous. However, it seems the market has dried up, and the the open source options have become littered with adware. Have any Slashdot readers come across good options?

Slashdot is full of very smart people. I'm sure most have been reading up on all of the latest research on COVID-19. What interesting information have you found?

For example: Peak contagiousness appears to be ~2 days before the onset of symptoms. Or the average incubation period is 5-6 days, but can be as long as 14.

What else have you found out?

My oldest child is in first grade, and learning to read. I noticed she was coming home from school and making some interesting statements about how she reads. One particular phrase that bothers me is, "I can read pictures." Recently, I heard a radio show on NPR about whole-language reading instruction, and how it's a terrible way to learn. I've since learned, that this is a hotly debated topic.

I learned to read in a phonics only setting. To me, this is the only way to read. I don't look at pictures, or the rest of the sentence unless I am completely clueless about what a word is. This whole-language approach just seems wrong.

Have any Slashdot members been through this experience with their children? Did anyone find good research supporting one way or the other, not just opinion? What is your opinion on whole-language versus phonics only reading instruction?

My neighbors and I were discussing buying electric vehicles. One thing that had us concerned was the cost of upgrading the electrical service to our garage. Our homes have detached garages, and our circuit breakers are old and full. I figure the cost to upgrade would be several thousand dollars using a licensed electrician. Do any Slashdot readers have experience with this?

After reading about Geely's purchase of Lotus I began to wonder about another maker of light weight cars. Team Edison2 won the Automotive X-Prize with their Very Light Car (VLC) concept. A quick Google search shows no results after 2013. It also appears their website has been shut down. What happened to them? It appears the team's CEO returned to real estate development.