"Design investments" for the Future

It is, perhaps, the most flagrant example. I suspect many people replace/upgrade phone products every couple of years.

But, the same is also true of laptops, computers, monitors, etc.

Here, the cost to fix something easily outweighs the cost of replacement. I.e., anything less than $100 you don't even *consider* "hiring" someone to repair (unless it has some *other* value).

I routinely have folks bringing me 50" LCD/Plasma/LED TVs just because it wasn't worth the hassle -- and uncertainty -- of a repair. Instead, "buy new" (what a great excuse: the old one is broken! :< ) They actually don't *want* their old TV repaired! Rather, just assume I will tinker with it and then find it a new home, somewhere.

The last point is the important one. People would rather buy new (even at a much higher price and even though the "new" doesn't provide any more value than the "used").

[Keep in mind that a machine is only worth that much if you have already partially disassembled it -- so the sheet metal can be sold to the scarp metal dealer, the copper/aluminum heatsinks to another, the circuit boards to a third, etc. I have no idea what a machine "as is" would be worth -- probably a few dollars less!]

There's a difference between interest payments and, for example, selling an unlimited number of apps, music, books, etc. targeted (and often *tied*!) to that particular device.

I.e., like selling toilet paper -- there's a continuing market! :>

Think about it. How do *you* set the price of an item that *you* sell?

Imagine a "yard sale". You know your Beatles album (sorry, I'm thinking about vinyl this morning) is "worth" more than your Monkees album. How *mush* more? 2X? 3X? etc.

Assuming you can come up with that multiplier (and, that your market can be made to agree with your assessment), now where do you set the *absolute* value? X?? $1? $0.10?

People look at prices in a relativistic manner. This is more than that. How *much* more is fuzzy. As is the absolute price.

E.g., when debating a capitol equipment purchase, I think about how much time it is likely to save me over its expected lifetime (or, a single project if I don't foresee any ongoing use). Then, I look at what I can *actually* earn in those hours.

But, when it comes to a new vehicle? How do you decide what that is "worth"? Car dealer sets a price. You figure how much you can move him off that price point. But, how do you decide if this new price is "right"?

E.g., I drive < 1,500 miles per year. If I monitored it more carefully, it's probably closer to 1,000 miles. My insurance is relatively inexpensive ($600/yr). Assume I get 25 MPG so there's 40G of fuel -- at $3-$4/G is ~$150. If I change the oil annually (despite low mileage) and add a few dollars for incidentals (car washes, wiper blades, windshield wash, etc.) I'm in the $800/yr ballpark -- assuming no unexpected repairs, tires last forever, etc.

At 80c/mile, *I* decide it's not worth the hassle -- we can share a vehicle and leave mine in the garage (she won't drive a "big car")

Other folks may claim they *need* a vehicle (12,000 miles/yr is roughly the norm, here -- perhaps higher in this part of the country). Yet, they also seem to "need" the GPS, hifi, heated seats, power mirrors, etc. What are each of those "worth"? Do they even consider it??

But source code takes up very little (relatively) storage space. Imagine imaging your development system after each project (disciplining yourself to ONLY keep those things related to the project on the system). How quickly would things grow?

E.g., if I have to design and layout a board and then write the code to make it all work, I have to save the datasheets for each of the components, the tools to design/layout the board (in case I need to make a change to it next year), the compiler/IDE *and* my sources.

I used to just keep CD's (of music, software, etc.). But, the number of disks just got to be outrageous. Trying to find *a* disk was a chore. Enter big, cheap disks. Copy 2500 CD's (300 DVD's) onto a 1.5TB $99 disk, unplug the disk and let it sit for "whenever" you decide you might need it.

I currently keep two copies of "stuph" on magnetic media along with the original optical media. "Memory (disk) is cheap". A backup of the magnetic media is a few clicks and then wait for the bytes to move onto the new medium. Backing up the original media is far more time consuming.

I had a 400 CD (music) changer. It was the size of a 5 or 6U server! The magnetic disk that holds an equivalent number of CD's takes up far less space *and* is "searchable".

The same is true when updating my NetBSD sources, downloading new "packages", etc. The days of 56KB UUCP packets are long gone! :> How often do you update your Linux system(s)?

I'm currently instrumenting the house. Do away with phone lines, TV antennae, etc. Push everything down the network. Wanna watch a movie? Listen to some music? It's just bits on the wire. SWMBO wants to browse her photo catalog from her studio? Just pull the images down from the server! Phone rings and you want to answer it in the kitchen/garage/back porch/on the roof/etc.? Why have complete devices everywhere you might want one -- just put the user I/O there!

As I said, your requirements change to meet the technology.

If I was limited to 10Mb *shared* fabric, I would never consider these things, realistically. A single video feed would saturate the medium. But, at 100Mb *switched* fabric, the video server is the bottleneck, not the fabric itself!

--don