-
Nothing will fully replace analog. Paper, 2D images, you can’t beat it.
- Again, I reiterate, because this is important! I searched for this and found it, that means it’s important!
The risks of digital data storage, format obsolescence, and conversely the non-issue of optical storage. Specifically, microfilm archival. Simplicity and ease of use. Likewise, “well-established” computer compatibility patterns, or lack thereof.
Computers have been around for quite some time for now, and one unfortunate trend that has become quite proven out in general is the difficulty of data migrations. This was more a problem early on in computers than it is today. Historically, most digital data was stored in proprietary file formats, and any effort to migrate data digitally from an old system to a new system almost always entailed a complex process of reverse engineering the proprietary file format, writing a computer program to do a file format conversion, running the converter, and copying the converted data to the new computer system, which might involve the use of special adapters and communications protocols, of which interfacing both sides between the old and the new systems is difficult. Yes, computer systems have a long history of creating obsolescence.
How does the general trend work? First, a “new standard” is developed. It gains traction for a short period of time, but almost always, eventually there is a “compelling new” technology that is completely different from the old technology. It is so much better, and so many people race to it so quickly, that the fact that this new technology is incompatible with the old one is basically forgotten. Come on, “almost no one” used the old technology, but now “everyone” uses the new technology, what’s the point of writing file format converters? Often times, the way things work out is that the original designers of the new technology weren’t trying to make it compatible with the old technology because they didn’t think it was meant to capture the same market. And they were right. It’s just the specifics they were wrong on: they thought they’d capture a much smaller market, but in the end they captured a much larger market.
So, the compatibility problem? Given the specifics mentioned above, this is mostly looked upon as a sad fact, an element of misfortune. But sure, it is also looked upon as predictable: surely customers of a historically small market knew they would have to do things much differently from the mass market, right? They’d be working in esoteric custom-rolled file formats, so they’d have to write esoteric tools to convert their way out.
Nevertheless, even today, we see the same thing going on, even in markets of the same size. A standard is developed, say an operating system. The operating system has support for a range of hardware built in, via device drivers. But eventually, new hardware with an incompatible interface is introduced, and the old hardware with the old interface falls out of favor/popularity. Eventually new operating systems drop support for the old hardware device drivers and interface. Likewise, the old operating systems no longer have support for the new hardware that has risen in popularity to the point of replacing the old hardware. Suffice it to say, even in modern times when computers number in the billions (via smartphones), the same pattern of incompatibility still holds with virtually every symbolic interface, from hardware to software. One cannot guarantee nor place their faith in the assumption that they will always have a pure-digital lossless migration path for their data.
But wait! In spite of the inevitable fact of breaking change in the world of computers and inevitable requirement to be prepared for format migrations, there are some emerging patterns of where the more reliable and less reliable standards are to be found. Let’s take a look at what these are.
First of all, let’s review our stack of symbolic interfaces used in interactive computer software from Tour de Force.
- Physics (analog data sources)
- Hardware interface
- Network communications interfaces
- Graphics hardware
- Storage hardware/interfaces
- RAM interfaces, CPU sockets, PCI bus sockets, other internal connectors
- Driver interface
- Operating system interface
- OS native library interface
- Network communications protocols
- OS-native file formats
- Platform abstraction library interface
- Platform independent file formats
- HTML 5 Web API interface
- User interface
Now, as we’ve just said, all of these interfaces are subject to change. However, out of this stack where most things are subject to change, there are a few things that are most resistant to incompatibility and breaking changes:
- Network communications interfaces
- Network communications protocols
- Platform independent file formats
- Especially platform-independent file formats that are very frequently used in electronic communication, especially when they are used by a very large number of people
Conversely, there are a few things that are least resistant to breaking changes and incompatibility:
- Network communications interfaces
- Graphics hardware
- RAM interfaces, CPU sockets, PCI bus sockets, other internal connectors
- User interface
So, that should give you a pretty good idea of what choices you need to make if you want to preserve information, and what things you should avoid if you don’t want to be caught by surprise by incompatibility and obsolescence.
What is the most intuitive and convenient way for humans to work with very small objects? Computer-like use. It’s well-established as a practical means of working with very small objects. After all, that’s what computer hard disk drives and graphics displays do.
- But, the corresponding issue when working like that for long periods of time. You need to take breaks to get exercise, unlike working in the macro scale which does not demand such. Taking breaks from over-exercise is what would be done in the macro scale instead.
** Why this is interesting. Shrink it down when you don’t care about it, expand it up when you do. In some sense it is a paradox. You don’t want it anymore, so you get rid of it? Nope, you just crush it down into a tiny amount of storage space when you don’t want to see it, then you expand it back out to thousands of times its physical size when you do want to see it.
-
The fact that America has engineering capacity but not manufacturing capacity… it is very much in the same position that developing nations are in related to engineering capacity. They have the computers, the money to purchase them, but not the engineering capacity to get there. The technology simply comes from a “foreign nation.” Literally, in the meaning of the word foreign. So yes, for American consumer culture too. Americans have the engineering capacity to design the doodads required for consumer culture, but America does not have the manufacturing capacity to feed consumer culture’s consumptive electronics habits at the current economic rates it is expecting. Only a literally “foreign power” can produce that which is demanded by consumer culture, at its desired price.
-
Mistakes in software? The state of software engineering today is very similar to the state of mechanical engineering in the 1800s. Locomotives were unpredictable, fussy, and sometimes dangerously crashy, explosive, and deraily. Nowadays those machines are very predictable and reliable, but we currently have unpredictable and unreliable software that is prone to glitches, crashes, bugs, and errors. So, yes… what I am saying is that the newest technologies come with predictable problems, but older technologies are very well behaved and reliable.