Tuesday, June 14, 2022

Bias in AI?

AI itself isn't biased, since it doesn't really have consciousness and judgement. It just uses data. 

It goes without saying that we need to avoid feeding it biased data but, even accurate data can lead to conclusions we don't want. For example, if there aren't a lot of minority or women candidates in a certain tech field, the AI might conclude that it's an undesirable trait.

To counter that, we might actually have to add bias into the system. For example, pre-add weight for underrepresented groups in tech. I guess, sort of how bookmakers set odds so that all bets don't just go to the favored sports team.

© 2022 Praveen Puri

Arming Teachers?

I'm not saying it's the best way but, if a local government insists on arming teachers, I think they shouldn't do it haphazardly:

1. Choose a body, such as the state police, to design and run the program.

2. A small group of volunteers from teachers/staff are vetted and receive special training.

3. The state police are responsible for maintaining the guns, which are secured in certain locations of the school, along with a bright ID vest which identifies them to students and responding police.

4. At any time, some of the volunteers are "on call" and have keys.


© 2022 Praveen Puri

Monday, June 13, 2022

Unnecessary Tech: Digital License Plates

Example of an unnecessary tech product looking for a market:  Michigan has become the latest state to approve Reviver digital car plates.  

Today, you have simple metal plates. You pay a yearly fee and you get a yearly sticker to put on. Simple.

With this, you get a digital display that has toxic substances, uses power, you need to also pay a monthly subscription to the company, they can display messages, and you can be tracked. Not to mention, what if it gets hacked?

So, for an investment of time and money, you get an environmentally unfriendly, more complicated, security risk that duplicates the low tech you currently have. That's not Strategic Simplicity .

© 2022 Praveen Puri

Wednesday, June 8, 2022

Bumble Bees are fish!

I saw an article in today's WSJ that, in CA, farmers sued to block bumble bees from being added to the state's endangered species list because it just covers "bird, mammal, fish, amphibia, or reptile."  

Well, the fish and game commission appealed and a state appeals court ruled that bumble bees qualify as "fish" under the CA legislation's legal definition of fish, which is very vague (basically, no backbone).  

You can't make this stuff up!

© 2022 Praveen Puri

Tuesday, May 24, 2022

Exploding Seatbelts!

Hyundai is recalling 240,000 cars because the seatbelts might explode! 

They have a technology called "pretensioners" to automatically tighten seatbelts in case of an accident. It relies on "small explosive devices".

Get this: "...pretensioners supplied for the seatbelt retractor mechanisms may not perform the way Hyundai's safety engineers intended, and an uncontrolled explosion may release shrapnel into the cabin."  

WTF!

Definitely not technical Strategic Simplicity®!

© 2022 Praveen Puri

Thursday, May 5, 2022

Today's Inflation is Another Reason to Practice Strategic Simplicity®

With today's inflation increasing their supply chain costs, companies are either eating the difference, raising prices, or reducing the amount supplied.

There is another option, rooted in Decision Simplicity, which is part of my Strategic Simplicity® framework.  By reducing the options and choices you offer customers, you can buy larger quantities from fewer suppliers, thus increasing your leverage for negotiating costs.

© 2022 Praveen Puri

Thursday, April 28, 2022

Why Do I Need A Data Lake?

This webinar from my colleague Douglas Squirrel is a great strategic introduction to Data Lakes: https://squirrelsquadron.com/events/2022/02-24-why-do-i-need-a-data-lake.html

They have great strategic value because they allow you to store data that you don't currently use, but may need to analyze in the future, in a raw state (structured enough so that you can retrieve it).

Keeping the data in a data lake can be better than the alternatives: discard it, or store it in a processed state in a data base / data warehouse (where it's silo'ed, slows down performance, and has lost raw info during the formatting/cleanup process).

© 2022 Praveen Puri