Tom Riley • about 2 years ago
Top-Down vs. Bottom-Up Design
Top-Down vs. Bottom-Up Design
We need more design ideas in the hack. So which approach is best? A lot of time is wasted on this argument. The correct answer is “All of the above”.
First you work on the overall design concepts, usually because you need to sell the project; then you write some code; then you farther develop the design expanding distinctive elements; then you write some more code. Back and forth; back and forth. Hopefully you will meet yourself in the middle before the deadline.
Because most of the design work has to be done off-the-clock and most on the too short fused benchmarks are only for code written, it looks like software work is all coding and no design at all. This is a misconception.
Coders hate the sales requirement in their job and even worse they hate the bad ideas sent down by management from on high. If you do not sale a good idea up, they will send a bad idea down. What is critical is to understand that the “elevator talk” and other sales “junk” are, in fact, the only top-level design anyone outside your team is going to listen to.
To write winning sales stuff, you first have to work out the actual top-level design clearly both in your own mind and to your co-workers. You cannot sale it if you do not understand it yourself.
In the Born to Storms, AI project this design process has gotten completely out of hand. The design is now book length and the code is not yet started.
Enjoy, Tom Riley, Born to Storms AI
Comments are closed.
5 comments
Tom Riley • about 2 years ago
Design Timing
Unproven design ideas are a dime a dozen. It takes at least two years of development on an idea to get to the point that you can even estimate its value in the marketplace. It takes three or four years before you are actually selling a product and thereby have real data on the value.
Designing for what is hot today is useless. Your product will be obsolete before it reaches market.
Guessing what will be hot in three or four years is very hard, a real throw of the dice. You pays your money and you takes your chance.
My hack project, Born to Storms AI, is designed to address the problems of our climate crisis. Although cold right now, I think this area will be hot by the time a new product could hit the market. Specifically, as soon as this pandemic passes (a secondary climate crisis event), the general population is going to wake up and smell-the-coffee-burning. They will then be right for new products and tools but will want them right that minute.
Enjoy, Tom Riley, Born to Storms AI
Tom Riley • about 2 years ago
Design in the 21st Century
Design of new tools and products in normal times is hard enough. The 21st century is not a normal century as our climate crisis is defining it.
To design for the future you must have a vision of the future. Quite a bit of modelling work has been done to build this picture but that picture is not clear.
The artic, for example, is warming much faster than the models show. This hastens sea-level rise and releases vast amount methane into the atmosphere. Since the beginning of the industrial revolution the sea level has risen about ankle high. The models put it chin high by 2100. We can deal with this but we must deal with it.
Many novels set in the 21st century are dystopias. Sociality has already broken down and people are making do in rubble. This image is useless for design work. It is defeatist.
Assuming business as usual for decades yet does not work either. We need to get ready for a massive effort or we will simply be overwhelmed.
It is time to design the tools our young people will need to face the 21st century. A great many of these tools will include Artificial Intelligences. Please join us.
Enjoy, Tom Riley, Born to Storms AI
Tom Riley • about 2 years ago
What exactly are you trying to do?
The key question for nearly all design work is: “What exactly are you trying to do?” One question but you need a hundred answers.
What the Azure people are trying to do with this hack is to increase the number of experienced programmers with hand-on experience in Azure. I am fine with that and would be happy to help move that idea forward.
What most participants are trying to do is to build experience at the professional level, perhaps win a badge for the old resume, and win a few bucks.
What I am doing exactly is different. I am trying to build recognition for the Born to Storms idea and in the process determine if the AI described in the novel, JanetA, is in fact possible a the currently level of technology.
I wish us all luck,
Tom Riley, Born to Storms AI
Tom Riley • about 2 years ago
Out-of-Box Ideas
Management always says they want to hear your out-of-the-box ideas, but that is rarely true. What they want is ideas that are soundly grounded in the box but have one toe sticking out. True out-of-box ideas are just too hard to explain and too risky. It is also nearly impossible to get the criticality of the idea across in a short slogan.
My work on “Born to Storm” is a clear example of out-of-box design.
I spent a couple years building up a model of the 21st century from the best available estimates.
For example, the sea level is expected to rise about one meter by 2100. The recent warming in the artic now makes this nearly certain. This will drastically challenge everyone who lives near the sea.
The artic warming has also disrupted the jet stream that until recently went round and round at high latitudes. The flow now loops both north and south. Cold fronts drive deep into the US and a Paris hot spell made it all the way to Greenland. We will not be able to count on the weather.
This picture then has your young people facing problems of historic intensity. The design problem then is to provide them with the tools they will need to make this a fair fight.
The best way I could find to develop this design was with a 56K word climate fiction novel, proudly out-of-box.
The question I am trying to answer with this Hack is simply will the Artificial Intelligence tool I proposed in the novel really work?
Enjoy, Tom Riley, Born to Storms AI
Tom Riley • about 2 years ago
Design Exercise:
Design is really the skill of imagining the world, as it is or as it will be, then adding one idea to that image. Does the idea fit in? Is if a wanted product or service?
Imagine Exercise: Imagine that in just a few years there are thousands of young people out in the field taking on the historic problems of rising tides, great storms, and summers of fire. Imagine that each has on his/her shoulder a cell phone with an AI interface. Each has trained for so long that the human and AI have become a symbiotic pair. Imagine that they all fight the good fight!
Let us design that AI, today.
Enjoy, Tom Riley, Big Moon Dig AI