You are here: /announcements/selecting-offline-computing
Cloud computing was an interesting experiment in human environments. It was initially marketed as a way for people with small IT departments or no IT resources at all to offload their tech requirements to a data center with a dedicated staff of IT professionals waiting to respond to a tech support request at a moments notice.
Statistically, this didn't work out well. Most of theme turned off their phones, and initially resorted to a tech support forum which has a very low level of privacy. More responsible companies turned to email and chat support which works remarkably well. This allowed some organizations to resume phone calls, which were now minimal and for urgent requests or requests that take to long to communicate while typing.
Most people ended up preferring email with 3-5 day response, chat for quick questions, or settings changes (est. 45 minutes), and phone call for urgent requests. Computing users feel the phone calls should be brief and would normally be a paid service or minimal amount of time to get started or ask 1-2 urgent questions.
One of the issues with cloud computing was nosy data center employees or workers. They take in interest in a specific customer and start browsing through their data, possibly even watching them work as they are using the cloud computing systems. This is viewed as a major invasion of privacy and breach of privacy in cognitive processing. Supervisors aren't even allowed to hover over the shoulders at work, why would a data center employee watch someone as they are using the system?
Most computer monitoring software at work watches for policy violations and alerts a supervisor based on rules. A lot of supervisors have found these overly intrusive, accusatory and too suspicious. They have instead turned to eDiscovery with probable cause and management dashboards that allow them to view the alerts all at once to assess the software instead of the employees.
A request has been to apply these same policies to data center workers in cloud computing systems who should mostly be focused on hardware failure, software component failure require installation, and data migration which often does not require browsing through data. Customer support is most often a separate function requiring an open work ticket and they are not to talk about their tickets with their co-workers. There is strict privacy requirements and they must use generalization in open meetings regarding customer improvement services.
Offline computing offers better performance and the concern is often licensing. Traditionally, software expires when the computer hardware gives out. Its normally tied to the motherboard and these can last for 20 to 35 years. The software is not traditionally expensive costing a few hundred dollars, and realistically computers are not meant to be upgraded every few years.
While Moore's law states that computing power will double every 6 months, computer manufacturers have scaled down from the 4Ghz to 5Ghz processors of 2011 to 1.2 Ghz to 1.8 Ghz processors from 2018-2022.
There might be a slight change in making processors 1.5 Ghz to 2.2 Ghz as a second stage computing device for consumers and office workers. Anything beyond that is viewed as an industrial computer that may not be appropriate in a home environment. A lot of computing devices are additionally going to low power, more encryption, and less replacement requirements.
This means that people will need to have hard drives, flash drives, removable storage, and a safe that may not necessarily be bolted to a wall, floor or ceiling. Most places are low crime areas, people on probation are regularly searched for stolen goods, and the computers have developed an interesting behavior when discovering data that has been misplaced or removed without permission or agreement.
Computer interfaces now emit electrical signals and interface with the human neurology system to cause nausea, dizziness, lightheadedness, delirium, auditory hallucinations, and occasionally visual hallucinations when handling data without permission or agreement. This has caused an update in computing policies to preserve human health in electrical environments and to clearly indicate when data is consensually shared and for which purposes it was transferred. The agreement must be typed and agreed to by all parties or scanned into a computer to prevent these electrical interactions through computing interfaces when handling other peoples data.
The way computers work nowadays, makes it feasible to have an offline computing setup that is safe, and free from tampering. Many people have had 0 computer breaches for over 5 years without antivirus or encryption. The courts have additionally proposed administrative processes for examining a computer, stolen data, and chipsets indicating which computers a storage device has been plugged into or was originally installed on.
Cloud computing will most likely be used to lease time on expensive data systems. These systems have been designed to be highly anonimized allowing people to process their own medical records without fear of privacy breach. This allows a business model where the systems provider is leasing you the system and has no interest in your data. This is additionally compliant with economic laws and data privacy laws that globally, are what the people are voting for.
Alternative business models allow trading data usage for systems access and this is falling out of favor but still allowed in some areas. The problem is normally that a person can rarely verify that they are fully authorized to transfer and make agreements with the data they are offering. It can work for some people.
Cloud computing is mostly used by social media in determining which networks or groups to introduce people to. Currently there is no data restrictions, unlimited data, but they have already begun to update policies.
The suggestion is to keep no more than 2-3 years of data in the cloud, to get used to deleting and reuploading data, and to get familiar with data export. The data should be useful to the person without the cloud system, and because it is their data they should be able to use it for other purposes as well. The thing to watch out for is when the system co-creates the data or when there is an agreement restricting use elsewhere. Examine the following scenario:
Two people on a social media platform made an agreement that they will use the data on the platform only. They both exported their data and documented its use. Hypothetically, years later they both deleted the data and have copies of both their data and the shared product. What are some of the things to be aware of?
Most people will be exporting and deleting data that is older than 2-3 years.
What about timestamps and timeline documentation?
See the above section on offline computing. If the data is created on an offline device before it is uploaded to the cloud, that will have earlier timestamps than the cloud version, while someone can infect the computer and potentially delete the timestamps, that will be a different penalty than removing data without permission.
Overall people should be managing their data offline, deciding what will be for long term storage and what are temporary data files, then uploading data to the cloud. The cloud data is for sharing, processing, or timestamp acknowledgment. Once the upload time has been verified, it does not need to stay on the cloud forever. It can be deleted and reuploaded later or used offline, like at a presentation that charges for access or at an event with a ticketing systems.
In other places cloud computing is big and the devices can be easily swapped. Here in the U.S. people have elected to go to offline computing, to use cloud computing as a secondary processing service, and to switch back to perpetual software licensing. The processes for monitoring or examining storage device access will continue to evolve over the next 35 years.