FOSS Adoption in Egypt

Purpose of this Document

The intent behind the composition of this paper is to define a guiding framework for cooperative activity between the EOSBC and the Egyptian Linux Users' Group. Broadly speaking, four concepts will be tackled.

These are:

  1. Examine the current status of FOSS in Egypt.
  2. Describe and rationalize the desired state of FOSS in Egypt.
  3. Identify obstacles to making this transition.
  4. Propose action to realize this transition.

We will be focusing on Free and Open Source Software (FOSS) and not simply open source software for the simple reason that the `free as in beer' aspect is the most immediately beneficial to a developing nation such as Egypt due to its effect of lowering capital investment requirements.


Segmenting by Requirements

Different segments of the market have different expectations of software and may require separate handling.

We will segment where relevant as follows:

  1. The end user
  2. The small to medium sized business
  3. The enterprise
  4. Government

FOSS in Egypt Now

FOSS in Egypt Now

There is no clear picture of the state of FOSS in Egypt at the moment. This lack of information is expended upon in section sub:Information. There are, however, subjective and highly unscientific yet nevertheless indicative observations upon which conclusions can be made.


Government

There is no visible government policy either regarding or involving open source software. Not even a negative one; FOSS is simply not on the map. Contrast with emerging South American government policies stipulating open source software in government contracts.

There is no visible government sponsorship of FOSS development.

There is FOSS being employed in government, but it is spare and very invisible. An example is the meteorological service's work developing weather system models on Linux clusters.


Business

It is probably more difficult to gauge FOSS status within the local enterprise and the business environment than in government. Let us try all the same.

It is a rarity to see job adverts in the Friday edition of Al Ahram1 requiring Linux experience.


The End User

The LUG, the most prominent forum for Linux end users, claims under 1,500 members.Let us say that only 25% of the people who would register with this community are aware of its existence and did so. This makes for 6,000. It is unlikely that all 6,000 are even know what Linux is2. If 80% of people who registered knew what FOSS and Linux is, then this leaves 4,800 persons. Of these 4,800 persons, if 75% (a terribly optimistic figure) have actually tried Linux, this gives us 3,600 persons who know what FOSS is and have tried Linux. If 50% of people who try Linux stick with it, then we have gone down to 1,800 aware Linux users in Egypt. It makes no sense to assume this figure down any further; it is already low enough to reflect the state of adoption of Linux in Egypt.

In any case, it is indicative that the most visible Linux Users' Group in the country can only claim 20 to 40 active members at best.

Obstacles

Obstacles


Lack of Awareness

This is probably the most significant factor hindering FOSS adoption in Egypt.

FOSS is quite simply not visible. End users do not know that there is such a thing as Mozilla/Firegoat. It gets worse; the concept of Internet Explorer as an independent entity which has alternatives is rare. This kind of misconception is not endemic to Egypt, but it is exacerbated by the almost absolute platform monoculture.

Similarly, there is the inability to separate the operating system from the physical hardware conceptually. Windows is the computer. This leads to charming notions such as Linux being something you install over windows.

Part of what has been generating awareness of FOSS as a viable alternative in Europe and North America has been the support of large companies, most notably IBM with their interesting Linux marketing ploys. This is not conducted here in Egypt, partially because the awareness problem extends into these companies. It is perhaps also felt that it is not worth pushing Linux in Egypt because the market is not there. Chicken and egg.

People, additionally, are deluded into thinking that what they have works. If the computer crashes, this is because computers crash and not because the OS is buggy. If Outlook is busy redistributing viruses, this is the virus author's fault and not the e-mail client architecture. If Exchange is a beast to set up, this is because e-mail servers are beasts.

People are not aware they have a choice because the alternative is not visible.

There is also a severe lack of awareness regarding the importance of the code being open.


Misconceptions

Five primary misconceptions can be identified. Several of them are not specific to Egypt, but are mentioned nonetheless since they are factors.

They are, in no particular order:

  1. Free Means Worthless: This misconception exists at pretty much all levels from end user to government and enterprise.It is closely related to `what's the catch' syndrome.
  2. Linux is Difficult
  3. The Liability Myth: This might not be too much of an issue outside of true enterprise-class entities due to piracy. One does not expect liability coverage with pirated software.
  4. Linux Arabic Support is Poor: This is the most optimism-inspiring misconception, since it implies that the person/entity has either tried linux or knows someone who has. In either case, the misconception implies that the awareness deficiency here is not that severe.
  5. The Alternative is Free: Piracy is rampant in Egypt and not only at the end user level. Effectively, this levels the playing field price-wise with FOSS. The price tag of FOSS is a key selling point; Joe Average is easier to motivate financially than ideologically.


Poor Availability

The relative ease or difficulty of obtaining FOSS is a factor not only in deployment and adoption, but in visibility also. Let us assume that Linux is the most frequently sought after FOSS product.

Linux can be gotten either in physical media or through an internet download.

Physical media requires distribution points and these exist but are sparse and ill-advertised. There are perhaps two or three shops carrying Linux, but these are exceptions. The LUG maintains a distributors' list; this is a list of volunteers in various locations with contact information whom you can get in touch with to obtain copies of linux. Needless to say, even if this list were operating smoothly (it isn't) it is hardly a solution to the availability issue.

Internet downloads are not a solution in a country where telephone lines are not yet ubiquitous and where DSL is still in its infancy. One does not download the latest three-CD Mandrake distribution over dial-up.

See section sub:Improving-Access for a proposed solution.


Lack of Requisite Infrastructure

Non-end user adoption of FOSS requires several things which don't exist.

The pool of available FOSS-skilled labor is too small3. If an enterprise is aware of, for example, the plentitude of excellent FOSS enterprise and business-class software and wishes to deploy one such item, it will most likely not be able to find someone to deploy and run such a thing. Proposed longer-term remedies are in section sub:Infrastructure-Building.

Additionally, there is no institution-backed liability coverage for FOSS, or it is not visible enough. The business will want a supplier to sue when the database eats all the historical data, and you do not get this when you download and install postgresql, for example. Of course, part of this is related to the business overestimating the extent to which existing software suppliers are willing to assume liability (covered in section sub:The-Liability-Myth).


Inability to Perceive Viable FOSS Business Models

This is primarily an issue at the government level and the development-oriented business. There is a strong impetus to export at the governmental level, and domestic software output is not exempt from this; there is the perception that FOSS means zero price, which is not true. One effect of the anticipated dominance of FOSS is the commodification of software which, when taken at face value, works against government efforts to boost revenue from software exports. In reality, this is far from certain; `exported software' could translate to programming tasks outsourced to Egypt which would make the development task an in-house affair and thus exempt from the GPL requirement for redistribution of code. In any case, this point is likely to be a complicated one to tackle.

The Desired State of FOSS

The Desired State of FOSS

The project must have measurable goals such that progress may be measured. This section is a list of proposed indicators to assess this.

Common Benchmarks


Awareness is Achieved

  1. 85% of the segment can identify one reason why FOSS is superior to proprietary non-free alternatives
  2. 70% of the segment can identify two reasons why FOSS is superior to proprietary non-free alternatives
  3. 50% of the segment can identify three or more reasons why FOSS is superior to proprietary non-free alternatives

Misconceptions are Educated Down to Acceptable Levels

  1. 85% of the segment4 can credibly argue against one of the main misconceptions.
  2. 70% of the segment can credibly argue against two of the main misconceptions.
  3. 50% of the segment can credibly argue against three or more of the main misconceptions.


The End User


The end user segment will be regarded as requiring no further action when: Availability is Ubiquitous

  1. The price to the end user of FOSS does not exceed 125% of the cost of media and duplication5.
  2. At least one of the three bookstores/stationery stores closest to the end user's residence/place of work carries linux and FOSS software.

The Small, Medium, and Large Sized Local Business


FOSS is either deployed to cover 85% of business needs for which a FOSS package exists which is at least equivalent to the most logical proprietary alternative, or plans to deploy to reach 85% are in place.


The Foreign Multinational


FOSS status within this segment of business is beyond the scope of this document.

Tools for Change

Tools for Change


Information


Information is needed in several areas to better appreciate the status quo and to quantify progress made in this initiative to promote FOSS. These areas are:

The End User

  1. How many computers are there in Egyptian homes?
  2. What operating system is installed?
  3. What software is installed?
  4. How many of these operating systems and how much of this software is properly licensed?
  5. How many home computer owners are aware of FOSS alternatives?
  6. How many home computer owners are aware of the ideological aspects of FOSS?

The Small to Medium Sized Business

  1. How many small businesses are there?
  2. What does their IT infrastructure look like?
  3. What are their IT requirements?
  4. How well does their existing infrastructure match their business requirements?
  5. How many small businesses are aware of FOSS altenative solutions?
  6. How many small businesses are aware of the ideological aspects of FOSS?

Government

  1. What does the current government IT infrastructure look like?
  2. What public plans exist for IT infrastructure modification and enhancement (such as e-government initiatives)?
  3. How is FOSS represented in the infrastructure?
  4. What level of awareness exists regarding open source, software commodification, and exchange/interaction based on open standards?


Supporting References


A number of documents will be needed to be able to quantify some of the above-mentioned points.

  1. A FOSS Equivalency Document which describes as broad a range as possible of proprietary software actively employed and its FOSS substitute, where the substitute is at least equivalent to the proprietary package.

Action

Action


Awareness Generation

Visibility is the key. Misconceptions cannot be handled before there is awareness, so this should be the key area of action.

Stimulate Media Coverage

  1. Lobby the editor responsible for the weekly computer column in Al Ahram to run more Linux coverage. Propose also a weekly review segment on a free open source application.
  2. Facilitate television show interviews/discussions related to FOSS. Shows such as `Sabah ul Khair ya Misr' are assumed to be open to good suggestions so it is a matter of contact. Have proposed interviewees and topics ready; someone from the Arabeyes project to discuss the concepts of free and open? Someone from the EOSBC? Torvalds? It is all a matter of initiative. There are plenty of journalistic talk shows to target.
  3. Supply the highest-circulated arabic computer magazines periodically with fresh copies of Linux, as well as FOSS for the windows platform. This has succeeded before with the magazine named `Online'. This tactic is best accompanied with an article on the ideology of FOSS and possibly some head to head reviews such as Internet Explorer vs. Mozilla.

Get the Government Involved

  1. Obtain permission from the persons responsible for the Suzanne Mubarak Reading for All initiative to use one location to deploy Linux on, train the staff there in using it, and conduct a usability study. Report back to them with the results and findings. Longer term the objective will be to give them the technical ability to conduct further Linux deployments themselves once the pilot shows its success. The report on the pilot will be valuable; it can be used in many ways. It can be circulated amongst MCIT personnel, to businesses, it can be disseminated to the press and the press will run it if it is sold properly to them (i.e. sensationalize the fact that the Reading for All libraries are at the cutting edge of technology, &c.). Favorable coverage of such a report in the foreign press would generate positive administrative karma within the MCIT, giving further momentum to pushing FOSS.
  2. As a longer term initiative, investigate the possibility of lobbying to introduce tax breaks for savings gained by employment of FOSS (and possibly also local proprietary software) to replace the hard currency-drain of proprietary license aquisition. It is not entirely unreasonable to think this may be achieved; tax code amendments tend to be cryptic and might not generate enough attention to alert proprietary software vendors.
  3. As another longer term goal, successful lobbying could perhaps result in pro-open source legislation similar to the Peruvian initiative.

Involve FOSS-friendly Suppliers

  1. Stimulate major Linux suppliers to become active Linux-wise. Start with IBM.
  2. Use the EOSBC initiative as a forum for soliciting opinion and coordinating activity; involve suppliers as well as companies which have implemented FOSS intrastructure. Institutionalize this utilization of the EOSBC to form an industry pro-FOSS lobby. This organ will be useful in stimulating government involvement. It might even be advantageous to involve major FOSS-friendly suppliers who do not have a presence in the Egyptian market; they are likely to be interested in involvement in a group which aims to create FOSS-friendly circumstances in one of the largest markets in the Middle East and Africa.

Leverage Ideological Compatibility of Islam and the GPL

Traditional Islamic views on intellectual property is said to mesh very well with the spirit of the General Public License. In a country which is overwhelmingly Muslim, this angle is begging to be explored.

Since this is a potentially sensitive area, any work here had best be conducted by a respected Islamic legal scholar.

It is a good topic for discussion, certainly; there are several prominent Islamic-themed websites where one can solicit the advice of a sheikh.

Establish Mindshare in the Small to Medium Sized Business

Why small to medium sized businesses? Enterprise-class business in Egypt will tend to be multi-national in which case the decision making will occur predominantly abroad. One of the benefits of reliance on FOSS is lower capital investment requirements, which will be felt more acutely at smaller scales.

One possible approach is to think backwards: use specific FOSS packages to introduce the concepts of FOSS rather than using the concepts of FOSS to stimulate adoption of specific packages. To take an example, promote postfix/exim/qmail as a mail solution and use its adoption to educate about FOSS as opposed to aducating about FOSS such that businesses may use postfix/exim/qmail.

The advantages are:

  1. The business realizes the benefits of FOSS and gains first-hand knowledge of the ideological advantages6.
  2. It is likely to require far fewer resources to promote a specific software solution than to promote a paradigm-shifting ideology to traditionally conservative business.

There is one readily apparent way of implementing this.

Run training programs covering specific packages serving specific needs, such as an `Installing and Administering Linux with Postfix' course. Market this program to businesses. Develop some form of rudimentary certification to cater to people's needs for scraps of paper describing what they can do. Conduct such a program under the aegis of a respected entity such as the EOSBC to establish legitimacy. The resources needed for this are minimal; some organization, an appropriate venue, a few capable instructors, and some effort. The FOSS message will be integrated in the curriculum. There will be a strong enabling focus to such a program; the instructee will be taught how the distributed support model works and will be tested on ability to tap this resource; he/she will additionally be taught that this specific package the program covers is but one of thousands7.


Misconception Correction


Linux Arabic Support is Splendid

The best method of tackling this misconception is probably letting the person try for themselves. This will happen if access to linux is easy and inexpensive (refer to section sub:Improving-Access); curiosity will kill this misconception.

Windows is not Free

The BSA branch in Egypt is not moving quickly enough to stamp out the idea that windows is free. Since piracy is commonly equated with theft, perhaps a religious edict8 would help.

Addressing Businesses

There exist several very professional and freely redistributable propaganda documents on the internet9; these could, at low cost, be reproduced in hard copy and distributed to businesses. This would optimally be carried out by a reputable entity (the EOSBC?) to establish legitimacy.

Concerns regarding liability are likely to be restricted to enterprises (as argued in section sub:The-Liability-Myth), though in the longer term reduced levels of piracy should spark such concerns at the sub-enterprise level.


Improving Availability


Improved availability is perhaps the sine qua non in this document; note three of the five listed misconceptions in section sub:Misconceptions can be educated down with improved access to FOSS. Availability primarily addresses the end-user, and much adoption of Linux in Egypt at the business and government levels depends on end-user adoption.


Infrastructure Building


Education

Just as access to Linux and other FOSS is a key factor, so is access to relevant documentation. There are several high quality and freely redistributable Linux books available. These need to be translated into arabic and disseminated. The translation will require sponsorship; it is not a trivial task.

To take a cue from Microsoft, the time to capture mindshare with the generations of system administrators, IT decision makers, and programmers is when they are learning the tools of their trade: in university.

This prompts a series of proposed actions:

  1. Sponsor periodical Linux CD give-aways10
  2. Sponsor periodical Linux book give-aways, hence the need for a translated version
  3. Lobbying for Linux administration and usage courses. This can happen in several guises, though two are more likely than the rest. The first is having the classes integrated into the official curriculum. This is the difficult approach, though it would be the most beneficial; it forces the IT-specific academic community to acknowledge the prime example of FOSS and institutionalizes its presence. The second option is to have extra-curricular classes (which might very well need to be sponsored) on the campus.
  4. Periodical on-campus LUG-driven events, such as InstallFests.
  5. Supporting the formation of university-specific LUGs such as the Cairo University LUG11. The more the merrier. People are more active and visible when they organize, and this strengthens the distributed support foundations which FOSS relies on. Such support could take the form of providing meeting venues, sponsoring the occasional InstallFest, &c.
  6. To promote the employment of FOSS as academic development platforms, sponsor a prize for the best graduation projects released under GPL and developed on Linux. This idea can be expanded upon almost infitely; best contribution to an existing FOSS project (acclimatizing future programmers to involvement in distributed projects and imprinting the culture of open development in them by participation), &c.

The above-mentioned point regarding printed documentation such as books is also more generally applicable and can be worked in very nicely with the availability suggestion in section ite:Kickstart-an-independent.

Enterprise and Business Support

Where support is concerned, it is probably advisable to let nature take its course. One of the (admittedly quirky) key selling points of FOSS is the distributed support model; one has to already be fairly deep in FOSS culture to experience this, unfortunately.

Eventually, if FOSS takes hold, commercial support businesses should arise in response to demand12.

If the grassroots are properly nurtured and the awareness, access, and infrastructure building issues are handled well, this issue should resolve itself; companies will find themselves internally capable of supporting linux without any specific effort13.

Admins and moderators teams

Administators team:

Moderators team:

Update March 6th: DesertPanther quitted the moderators team today.