Cloud Computing


(Mark Edworthy) #1

Cloud must be one of the most overused words within computing and society. Today, people expect to read their e-mail using online mail services (ie. Gmail, Hotmail and Yahoo), share their files and documents using Google Drive or Dropbox, chat and share ideas whilst using Twitter, Facebook and Minds, create documents (using Google Docs, Microsoft Office 365 or Zoho Docs) and communicate with others via face-to-face video calls, using services like Skype and Google Hangouts.

Most users consider cloud computing as being a collection of online services that can be accessed using a web browser or other client application. However, cloud computing is a misunderstood and overly misused term.

In essence, cloud computing is a concept that is made up of a variety of layered elements or tiers. Whilst there is a number of tiers, there are three fundamental tiers that are differentiated by the level of abstraction that is presented to the user, developer and administrator. Each of these tiers (often referred using the suffix “as-a-service”) provides separate elements to the overall architecture of cloud based services.

Infrastructure-as-a-Service (IaaS)
IaaS is the lowest tier, which provides both the physical hardware (host computers and storage devices), network infrastructure (wiring, gateway devices and firewalls), environmental controls (temperature for the devices and surrounding area, power capacity and location) and any virtualisation / hypervisor (ie. a virtualised computer environment) that is running within a host computer.

Platform-as-a-Service (PaaS)
The PaaS tier is the middle layer in the three tier system. This tier arguably contains some of the most crucial elements of the cloud and provides the operating system, application framework, database management system and other process management services (such as e-mail retrieval and delivery mechanisms, domain name provisions and web service environment). PaaS allows for deployment of services without the need to manage or deploy the underlying hardware (which is governed by the IaaS tier).

Software-as-a-Service (SaaS)
The SaaS tier can be generally considered as the interface which the end-users is faced with when accessing any cloud based service. This element allows users to interact with an interface via a web browser or other client based application (such as Microsoft Skype’s phone software). It is also worth noting that other client based applications such as IMAP / POP3 mail applications (ie. Microsoft Outlook and Mozilla Thunderbird), instant messaging software (Pidgin, ICQ client, BlackBerry Messenger or Adium) and integral frameworks that can be incorporated into existing operating systems (such as the usage of the Webdav storage framework within various platforms) could all be considered (within this context) as partial SaaS solutions (various mobile “apps” such as own Android / IOS tool or Google’s suite of mobile tools could also be considered as partial SaaS applications).

As demonstrated above, some terms that are used to describe “cloud” technologies could be considered as vague marketing terminology and shorthand descriptions to sale / provide online solutions to consumers. This is partially the reason why I personally don’t like using terms such as “cloud” when describing online solutions but as a general rule-of-thumb, most users should consider the term “cloud” as a description of online browsers based (and thin-client based) solutions (such as online file and document storage, online mail applications and various communication platforms).

The usage of the word “cloud” is a fairly recent term and was first used to represent network computing in 1993, when General Magic (company that started as a project which was backed by Apple Inc.) and AT&T used this term to describe their Telescript and PersonaLink technologies. Back in April 1994, Wired magazine featured an article entitled “Bill and Andy’s Excellent Adventure”, in which Andy Hertzfeld (a member of the original Apple Macintosh design team and a primary software developer for the Macintosh operating system) provided the following comment:

"The beauty of Telescript is that now, instead of just having a device to program, we now have the entire Cloud out there, where a single program can go and travel to many different sources of information and create sort of a virtual service. No one had conceived that before.”

However, I partially disagree with the above statement and consider cloud computing as an evolutionary step in what was already a growing market. Since the early interpretations of multi-user computer workstations, many “online”, distributed services where provided by mainframes and network computing infrastructures. Many users where using these infrastructures to access e-mails and communicate with others, as well as to interact with remote services such as Bulletin Board Systems (BBS) and to access other systems such as Prestel and Viewdata (Videotex) services (precursors to the British ORACLE and Ceefax teletext services that where provided by the BBC and ITV television companies).

Many large organisations and companies where using similar technologies to provide system for human resources departments, accounting departments, time management systems and mail client applications. Many organisations (especially governmental agencies) still provide their staff with similar technologies and frameworks. File storage services such as FTP servers could also be considered as “cloud” services.

Many organisations provided users with thin-client computing devices (commonly known as “dumb terminals”). These devices where (and still are) used as replacements for fully-fledged computer workstation and would have had less hardware then conventional (fat-client) personal computers. Whilst most of these devices may only have limited localised storage capacity and terminal (command line) based interfaces, many devices had graphical interfaces which allowed users to interact with various remote services.

Thin-clients are still used by many organisations, companies and consumers. These devices are often provided by public libraries for access by the general public, banking ATM terminals and consumer products (such as Google Chromebooks, Hewlett Packard Stream and various other netbook devices).

Administrators and technical support departments provide these thin-client devices which allows for minimum hardware maintenance. Many modern thin-client devices may only provide bare minimum hardware (limited storage and memory capacity) and may only provide a limited graphical user interface with a browser installed. All other applications may then be accessed by the user as “cloud” services.

Within the next decade, I suspect that we will see many established software providers migrating to “cloud” based infrastructures rather then providing their users with traditional offline software products. These “cloud” based solutions allow the developers and providers to maintain, upgrade and provide new features to their software packages without their customers needing to purchase new license agreements, installation medium or to be subjected to tedious upgrade routines and patch fixes.

I also suspect that many companies will stop providing their software products using one-off payment models (ie. one-off purchases for a perpetual license) and start providing ongoing subscription license purchases, which may force users to pay for a predetermined time (ie. monthly or yearly subscription) for access to the provided software and services.

In the last couple of years, we have seen a number of traditionally offline products such as Microsoft Office and Adobe’s Creative Suite providing many of their services as “cloud” based products, and in-turn providing subscription based payment methods. Whilst in some cases, some products (ie. Microsoft Office) are currently available either as a “boxed” product (which can be installed on a computer as an offline product), with a volume license agreement (allows the product to be installed on a predefined number of computers) or as a “cloud” based application with a subscription license. However, other companies (such as Adobe) have already dropped their non-subscription, offline options and have forced their users to pay for a subscription to access their products. Many companies may provide limited access to their software packages for free whilst expecting their users to pay yearly charges to access advanced functions, tools and new features.

There is an argument that would suggest that, by changing to a subscription model, consumers may actually benefit due to the fact that this payment model will decrease the overall cost of software licensing due to eliminating illegitimate uses of software. By lowering the piracy rate, the cost that needs to be passed on to legitimate companies and users can be lowered.

However, I can foresee that some existing users may not like the idea of these new subscription models and may prefer one-off purchases for software products. This change in payment models could adversely effect existing software developers and providers, which may see these providers loosing a percentage of their core users. These changes in payment methods could adversely effect the software providers income and profits.

Also, there is the issue surrounding users privacy and the acquisition of data by third party organisations. Many of us may have already read contracts and agreements that contain clauses that include the right of the provider to share their client data with other “trusted partners” (usually for marketing purposes). There are also legitimate concerns about the increasing raise of governmental oversight and unfavourable interference within privately owned companies (such as pushing of overly authoritarian but vague rules and laws – ie. various countries and states using vague, poorly worded and badly termed “hate speech” legislation).

Another concern that many users may have is in relation to ownership of files and documents stored within “cloud” based storage solutions. As an example, when reading Google Drive’s terms of service, section 2 (Your use of Google Drive) states: “Google Drive allows you to upload, submit, store, send and receive content. You retain ownership of any intellectual property rights that you hold in that content. In short, what belongs to you stays yours”. However within the next paragraph, the agreement states that “When you upload, submit, store, send or receive content to or through Google Drive, you give Google a worldwide license to use, host, store, reproduce, modify, create derivative works (such as those resulting from translations, adaptations or other changes we make so that your content works better with our services), communicate, publish, publicly perform, publicly display and distribute such content”. The inclusion of both of these paragraphs could be considered as contradictory and inconsistent. These statements rises real concerns about the ownership of files and the rights of the user.

In November 2017, it was confirmed that Google was using algorithms to analyse users documents (stored within Google Drive services) and whilst Google claimed that the issue was a “glitch”, this didn’t stop Google from temporary blocking users from accessing their documents due to claims of “violating Google’s terms of service”. Over the last couple of years, there have been claims that Google, Microsoft and Apple (to name but a few providers) have been analysing their users content, as well as being suggested to have provided various governmental agencies with private user information (without agencies providing correct authorisation and legal documents).

The other major concern for many users is related to security of online services. Throughout 2016 and 2017, there were a number of data breaches of some well known organisation websites and infrastructure (affected companies included: Verizon, Yahoo, Tesco Bank, Uber and Equifax), whilst some of these breaches where insignificant, other breaches where found to have had customers and users data stolen. With the increase in “cloud” services, it is inevitable that there will be many new data breaches and security incidents, which will affect millions of users.

Cloud computing is a poorly defined marketing term which means different things to different people. However with the increase accessibility of online software, storage and services, users must consider the various issues (as stated above) before signing up to and consenting to the terms of service agreement of any online service.

References & Other Resources:

This article has been published under the Attribution-NonCommercal-NonDerivs license agreement (CC BY-NC-ND)

All associated (embedded or linked) content is provided with the implied permission of the content owner or provider

Collabora Online in Nethserver-Nextcloud
(Rob Bosch) #2

Very nice and indept article! Good to see some peripheral context in these forums. However as the always pragmatic person I am: what can we learn from it and make NethServer have advantage of this?

There are some distinguished parts already implemented in NethServer. Think of the WebTop module, NextCloud module (with the Collabora online and the OnlyOffice apps) Maybe not ‘cloud’ by definition but at least giving similar functionality as fullblown (commercial) offerings.

My biggest concern towards those big ‘cloud’ services has always been the privacy aspect. And I think the online services NethServer can provide are the answer to those concerns since you have all those aspects under your own control.
And maybe we should focus on this specific issue when spreading the NethServer word?

(Mark Edworthy) #3

Thanks for the comments (which are appreciated).

Your comment about “Maybe not ‘cloud’ by definition” did make me chuckle a little. As I sort of alluded to within the article, the definition of the “cloud” (as coined by marketeers and sales / advertising executives) is in part a subjective definition.

Also your point about privacy made me ponder and I am now suggesting that there needs to be another subject category within this forum related to privacy and (more specifically) security issues.

Could @alefattorini please consider adding a new category to this forum?

(Rob Bosch) #4

Agree with you that the “definition” of cloud is one of those definitions with a lot of different interpretations. Would that still make it definition?.. LOL…
For me cloud defines with an online service provided by a redundant cluster of servers. I mentioned “not by definition” because NethServer can perfectly provide those services, but lacks the redundancy a provider like AWS OR Azure can provide.