One of the problems exposed by the open access debate is that there is significant misuse and misunderstanding of the terms and definitions involved.  It is common to read terms such as “free software,” “commons,” “open source,” and “public domain” interchangeably.  There must be an understanding that besides the ideological and philosophical connotations of each term, the heart of the movement is the distribution of intellectual works through permissive licenses.

The open licensing model is centered on the licenses; without them, the movement is just a project management technique.  The free software and open source software movements have shown the way to follow regarding licensing agreements.  The starting point for non-software licenses will be to learn from the experiences in non-proprietary software development.73  However, there should be an understanding that these licenses are just a starting point, as OSS licenses tend to be specific to software development and, in many instances, they have been drafted by software engineers with little or no intervention of the legal community.74  Furthermore, some software developers appear to display considerable reluctance about external intervention in the decision process regarding licensing decisions.75  This section takes a closer look at the attempts to translate the OSS ideals to the needs of scientific research.

Open access publishing tends to use “some rights reserved” licenses in order to distribute the academic materials involved.  These may include the use of standard licenses, or in some instances, it may include the use of customised licenses.  At the time of writing, the most prevalent open access distribution scheme is through Creative Commons (“CC”) licenses.76

The Creative Commons project attempts to create so-called “intellectual property conservancies,”77 separating a block of human knowledge offered for the benefit of the public, but still protected by intellectual property.  This is analogous to nature conservation areas that exist for the wider social benefit, but have restrictions on certain uses.  In the Creative Commons, the goal of intellectual property conservancies is achieved through the offering of a wide variety of licenses to protect creative works from misuse.  This is done through the application of open source principles, where the work retains its copyright protection, but it is distributed freely78 as long as the conditions contained in the license are met.  The interesting part of the CC licensing environment is that it empowers users because there is a wide range of licenses to choose from.  Creators and authors need only to go to a website and select from different options offered in a few drop-down menus; the system then chooses the license that fits the parameters entered.  These licenses range from offering the work straight into the public domain, to more complex licenses with restrictions as to the commercial distribution of the work and the use of licenses in such distributions.79

Creative Commons licenses maintain a minimum set of standards that are met by all of their offered legal documents, with the exception of the one that offers the work to the public domain.  This could be called the Creative Commons Definition, but it is generally known as the CC Baseline Rights.80  All CC licenses provide these baseline rights:

Licensors retain their copyright; this explains why the baseline rights do not apply to public domain offerings.   

The licenses announce that fair use rights are not affected by the license.  This is a curious statement, as it should be assumed that any clause that erodes acquired fair use or fair dealing rights should be specified in the license.

Licensees will have to obtain specific permission to perform one of the acts restricted by the license.  For example, if the license does not allow modification or adaptation of a work, this action could only be performed with the permission of the owner.  This seems to be a redundant statement, as this is an action that is usually understood in all licenses.

Copyright notices should not be removed from all copies of the work.

Every copy of the work should maintain a link to the license.

Licensees cannot alter any terms of the license.  This seems to be yet another redundant clause, as it should be understood that this is common licensing practice.

Licensees cannot use technology to restrict access to the work.  This baseline right specifically forbids the use of technical protection measures. 81

Licensees are granted the right to copy, distribute, display, digitally perform and make verbatim copies of the work into another format.

The licenses have worldwide application, last for the entire duration of copyright (unless otherwise specified), and are irrevocable. 82

It is important to note that the baseline definition of CC licenses does not mention anything about modification or adaptation of a work, does not deal with copyleft-like clauses requiring the use of similar licenses to distribute the work, does not mention attribution, and does not deal with the distribution of copies for commercial purposes.  This makes the basic Creative Commons definition more akin the open source ideals than to the free software principles exemplified by the GPL.83  Nevertheless, creators can choose a CC license that maintains all of the restrictions mentioned, from all of the options offered.  Authors then can choose from the following options to generate their license:84

Attribution:  The work is made available to the public with the baseline rights, but only if the author receives proper credit. 85

Non-commercial:  The work can be copied, displayed and distributed by the public, but only if these actions are for non-commercial purposes.

No derivative works:  This license grants baseline rights, but it does not allow derivative works to be created from the original.

Share-Alike: This is based on copyleft principles.  Derivative works can be created and distributed based on the original, but only if the same type of license is used, which generates a viral license. 86

It is possible to have licenses that combine several of these options.87  The strongest (and most popular) CC license is the Attribution Non-Commercial Share Alike License,88 which is the license that most resembles the strongest copyleft software ones (such as the GPL).  All CC licenses are presented in three formats:  the first is a short and easy to read “Commons Deed,” which explains the terms and conditions of the license in a simple manner; the second format is the “Legal Code,” which is the full license; and the third is the “Digital Code,” which provides an HTML version of the license89 that can be read by search engines and makes it easier to list the content in the Creative Commons directory.

Creative Commons presents a very positive step towards the wider distribution of non-proprietary technology.  It is innovative, thoroughly planned and smartly implemented.  CC delivers open access licenses in the digital domain with scalability, adaptability, and ease of use for those unfamiliar with the legal issues involved in licensing.  CC also offers jurisdiction-specific versions of their licenses to make them more valid in an international environment and to respond to legal requirements in a given country.  

The other major open content license is the GNU Free Documentation License (“GFDL”),90 which is the FSF's non-software license, and it is generally used to protect manuals and other literature related to the FS movement.  However, the GFDL is also used in other open access projects, such as the free online encyclopedia Wikipedia.  The GFDL could be classified as an open access license because it allows the copying, distribution and adaptation of a work, provided the author complies with the conditions included.  These can be found in section two of the license, which states:

You may copy and distribute the Document in any medium, either commercially or noncommercially, provided that this License, the copyright notices, and the license notice saying this License applies to the Document are reproduced in all copies, and that you add no other conditions whatsoever to those of this License. 91

This is an important point of the license, because it specifies that it allows for the commercial use of the works.  The GFDL allows for the modification and translation of the work, provided some specific sections are maintained or deleted; and all derivative works must be licensed using the GFDL.92  This clearly means that this is a copyleft license, perpetuating itself through this viral clause.  The viral nature of the license exists in section 4, which states that “[y]ou may copy and distribute a Modified Version of the Document . . . provided that you release the Modified Version under precisely this License . . . .”93  This is different from the share-alike element in CC licenses, as these require only that the work is released with a license that contains similar clauses and rights.  

The viral nature of the GFDL can be seen in practice through the wide copying and dissemination of Wikipedia articles, which are being used by many other open content providers, such as the Free Dictionary.94  The articles found in this online resource have to be licensed through the GFDL, allowing yet another third party to copy them and use them in their website, provided that they use the GFDL.

With so many creative works that may be subject to protection by OA licenses, it should come as no surprise that there has been a recent proliferation of licenses that allow commercial and non-commercial content creators to adopt the non-proprietary open access model.  One such project is the Open Content License (“OPL”), a collaborative effort that sets a copyleft license, ensuring that shared works will continue to remain free to subsequent users.95  Similar efforts also include music creation via the Open Audio License (“OAL”),96 the SCRIPT-ed Open License (“SOL”),97 and even Open Cola, the world's first copyleft fizzy drink.98

Looking at the vibrant nature of the OA movement and the publication of scientific materials in journals or through other online means, it is clear that there are enough reasons to believe that some part of the open science movement is doing quite well through the creation of growing body of work that is easily accessible to researchers around the world.  However, what happens with scientific research that is not subject to publication?  Can scientific databases, archives, repositories, and patented research be protected through open licenses?  

B.  Open Science Licenses

All of the aforementioned licenses have one thing in common-they protect only works that are subject to copyright.  This is valuable when one considers that a significant amount of basic scientific works is protected by copyright, particularly academic journal articles and other literary works.  This raises the question of whether there can be open licenses that protect other types of work in commercially viable fields like biotechnology and health research-which are usually available through databases or patents.  This has proven to be difficult because, although there have been many scientists and researchers advocating the implementation of open licensing models to the scientific arena, it is difficult to find a patent or database equivalent to the GFDL or the Creative Commons licenses.

The reason for this is that the open license model works better with copyright than with patents or databases.  There are two main reasons for this.  First, copyright subsists in an original work as soon as it is fixed in tangible form.99  Second, copyright does not require any sort of registration to initiate protection, which means that copyright “flows from the nib of a pen,”100 making it much easier and cheaper to distribute through an open license as soon as it is originated.  On the other hand, works that require registration to be subject to protection-such as patentable scientific research-will be more difficult to distribute through an open license, as several steps are required to distribute it with some sort of permissible licensing model.

Although the specific difficulties of providing a patent solution will be discussed in more detail later, there has been some success in applying limited open science solutions for databases and patented works.  These efforts are listed next.

1.  Database Licenses

The importance of scientific databases for research is an issue that has been well explored in the existing literature.101  There can be little doubt that in the information age, access to the vast amount of scientific data stored in databases is of utmost importance for researchers around the world.  Access to a large number of databases is offered for a fee by a vast array of service providers and institutions.102  The growth in the number and the economic importance of scientific databases has been accompanied by increased concerns about the reuse of the data to provide further works of commercial value.  A report by the National Research Council in the United States points out that:

Currently many for-profit and not-for-profit database producers are concerned about the possibility that significant portions of their databases will be copied or used in substantial part by others to create “new” derivative databases.  If an identical or substantially similar database is then either redisseminated broadly or sold and used in direct competition with the original rights holder's database, the rights holder's revenues will be undermined, or in extreme cases, the rights holder will be put out of business.  Besides being unfair to the rights holder, this actual or potential loss of revenue may create a disincentive to produce and then maintain databases, thus reducing the number of databases available to others. 103

This is of particular worry for those who are releasing genetic data into the public domain, as described earlier.  With publicly available databases, commercial providers would find large sections of readily available information that can be repackaged and resold as part of a commercial database.  This possibility is precisely what has prompted the calls to protect databases through open source licenses.104

Despite the suggested application of the open source licenses and ideals to databases, the actual application has not been met with the enthusiasm that it deserves, which can be explained by two main reasons.  Firstly, most non-commercial information-particularly in the field of biotechnology-is released into the public domain.105  This type of release is extremely useful for future researchers, but it does little to curb the further commercialisation of the data.

Secondly, the legal protection of databases is a subject that is not fully harmonised at the international level, where different jurisdictions apply a wide range of legal figures and levels of protection to this type of intellectual work.  For example, the United States106 has been struggling with the application of copyright law to the subject of databases by extending the definitions of originality of a work.  Earlier cases declared that the mere rearrangement of information was not enough to prove originality.107  Despite the originality requirement, there are circumstances where the courts will award copyright protection to databases due to the fact that there is enough originality in what is done to the data.108  Europe has followed a different path by embracing a “sweat of the brow” approach, where the work and investment that goes into the gathering and arranging of the data is rewarded, even if the data itself is not original,109 which is particularly evident with the European Directive on the legal protection of databases.110  The Directive awards a sui generis right to databases in which there has been a quantitative and qualitative investment in obtaining or verifying the contained data.111  However, this picture is made more complex by the recent rulings from the European Court of Justice,112 which have eroded the database right considerably.  To further erode the European database right, a report by the European Commission about the application of the right has produced negative results, and even the call to potentially eliminate it.113

It is precisely this complicated legal landscape that makes the possible application of open source software models so difficult for databases.  It would seem possible that providers of scientific data contained in a database compiled in a country that provides for the copyright protection of databases (such as the United States), may be able to issue their work through an open access license, perhaps even a Creative Commons license.  This would be possible because in most jurisdictions databases are protected as literary works.114  However, those providers would first have to be able to be awarded protection in the first place, which is not always the case, as evidenced by the originality standards prevalent in the United States.  In countries with a sui generis right, the licensing would have to meet with the very high requirements of the recent cases and the directive, which is not an easy task.  According to Waelde and McGinley:

Suffice it to say many questions remain over the extent to which scientific databases might qualify for the sui generis right.  Whereas at first blush it might have appeared that many might fall out with the necessary criteria, . . . it is far too early to argue that the contents of scientific databases fall into the public domain as a result of the ruling, however much that might benefit scientists and the progress of science. 115

Taking into consideration all of these difficulties, it should come as no surprise that a large amount of online scientific data is still released into the public domain.  Nevertheless, there are indications that the solution may not be found in database protection, but in contractual law.  This is best evidenced by the International HapMap Project Public Access License (“HPPAL”),116 which is part of the HapMap genetic database project.  Unlike all other open licenses, the HPPAL does not assign any intellectual property rights; it is an end-user agreement.  The data can only be accessed through following a registration process, where the user is required to agree to terms and conditions before gaining access to the certain parts of the HapMap genetic database.

The wording of the HPPAL makes it appear to be an intellectual property assignment of rights (it is after all, called a license), but it is not entirely clear what rights are held over the data that is being offered.117  The HPPAL is very careful not to assign intellectual property rights, so it must be assumed that this is just a user agreement where the author enters into the obligation to comply with the terms and conditions set out in the document.  Specifically, paragraph three of the license states that:  “[y]ou may not access, copy, modify, sublicense, distribute or otherwise use the Genotype Database or the data contained in it except as expressly provided under this License.”118

The most relevant part of the HapMap license is with regards to future patent applications.  Paragraph 2(b) of the license does not allow the patenting of genetic information from the database, with the exception of particular uses of sequences, provided that the patent allows further use of the information obtained from the database.  The paragraph reads:

[Y]ou shall not file any patent applications that contain claims to particular uses of any SNP, genotype or haplotype data obtained from the Genotype Database or any SNP, haplotype or haplotype block based on data obtained from, the Genotype Database, unless such claims do not restrict, or are licensed on such terms that that they do not restrict, the ability of others to use at no cost the Genotype Database or the data that it contains for other purposes; . . . . 119

This is an attempt to provide a viral or share-alike element to the agreement, as if the freedoms protected by this license are protected in the future licensing of patented material.  The HapMap license offers an ingenious way of getting around the problems of database protection enumerated above, as it relies on contractual obligations rather than on intellectual property protection, and may prove to be the way to go as far as database licenses are concerned.

2.  Patent Licenses

If the application of open licensing to scientific databases has been minimal, the porting of OSS licenses to patented research has been almost nonexistent and doubly problematic for reasons that will be explored in more depth in the next section.  Nevertheless, there have been a handful of attempts to provide a workable license for patented material, including a recent draft license.

One of the most promising efforts to provide a license has been put forward by the Creative Commons project.  Because Creative Commons licenses are geared specifically towards creative works subject to copyright protection, a new concept has been designed to accommodate scientific research.  This concept is the Science Commons project,120  which has been created to generate licenses that will deal with intellectual property works that are not covered by existing CC licenses.  The project is ongoing at the time of writing, and it has yet to produce a license draft available to the public.

Another proposal is offered by Hubbard and Love, who explore some alternative models of pharmaceutical research and development to produce new medicines.121  Their proposal uses the existence of free software as an illustration that alternative business models are viable, but unfortunately it fails to make the point of how to translate OSS licensing ideals into the pharmaceutical industry.  Although Hubbard and Love's argument may not connect directly with OA models, their suggestion is important because they propose workable ways to fund the basic research and to generate incentives to companies to distribute their intellectual property to the public, which would be released through open licenses.  But the question remains, which open licenses?

The answer to the question of licenses may lie in the aforementioned CAMBIA project.  One of the most important parts of the objectives of CAMBIA is the use of open source ideals to generate a protected commons for researchers in the life sciences.  This is done through a couple of licenses-the BiOS Plant Enabling Technology License and the BiOS Genetic Resource Technology License.  These schemes can be used for specific types of patented material.  According to the project, the licenses work like this:

Instead of royalties, BiOS licensees must agree to legally binding conditions in order to obtain a license and access to the protected commons.  These conditions are that improvements are shared, and that licensees cannot appropriate the fundamental “kernel” of the technology and improvements exclusively for themselves.  Licensees obtain access to improvements and other information, such as regulatory and biosafety data, shared by other licensees.  To maintain legal access to the technology, licensees must agree not to prevent other licensees from using the technology in the development of different products. 122

The core concept of the BiOS Licenses is that they will be able to provide patented research with a permissive license that operates within OSS and open access principles.  However, this is easier said than done, and the existing drafts demonstrate just how difficult it is to word open source patent licenses.  Earlier drafts consisted of a single document that covered different types of patented technologies, particularly “Crop Molecular Enabling Technologies and associated patents, patent applications, know-how, data, materials, and business, technical, economical and manufacturing information.”123  However, the complicated nature of the subjects prompted a forking of the license to cover two different technologies, such as plant-enabling technologies and genetic resources technologies.  While the subject and the specific application of both licenses are different, the legal principles and structure of both are the same, so they will be covered in the same way.

The first important common element of the licenses is that the patent owner always retains control over the technology, and what is established is a permissive chain of distribution.  However, there is a possibility that that the licensor could be a licensee himself.  The reason for this is because the BiOS Licenses contain a viral clause that allows licensees to sub-license the material, as long as the same rights that are contained in the license are preserved in the vertical agreement.124  Paragraph 2.1 is the license grant, which gives licensees:  “a worldwide, non-exclusive, royalty-free right and license to make and use the IP & Technology for the purpose of developing, making, using, and commercializing BiOS Licensed Products without obligation to CAMBIA, including a right to sublicense . . . .”125

Similarly, improvements to the patented technology are allowed as long as those are communicated to licensor, along with all improvement patent applications.126  When compared to existing open source and open access licenses, the elements and clauses present in the BiOS License make it akin to a CC Attribution Commercial Share Alike license.

This is a worthwhile effort to create a viable “open source” license for patented materials.  However, even in its draft stages it is easy to see that the language seems stretched and unclear in many instances-something that could turn away some potential licensors who could find the complex explanation of the terms and conditions difficult to navigate.  Another question that arises from the draft is that it is not very clear what would be the role of the initial licensor and his place in a lengthy chain of sub-licensees.  Paragraph 2.1.2 for example, states that:  

BiOS LICENSEE shall be responsible to ensure in said sublicenses that any Improvements produced by sublicensees are considered to be Improvements hereunder and that such sublicenses require the performance of all applicable obligations due to CAMBIA and other BiOS Licensees under this Agreement and any associated Agreements.  BiOS LICENSEE shall provide a list of sublicensees to CAMBIA in writing at least once a year, which CAMBIA is entitled to post in CAMBIA's Protected Commons website (BioForge) or make available to BiOS Licensees. 127

Does this really mean that the sub-licensee has some duty of care with regards to the actions of subsequent sub-licensees?  How can such responsibility be exercised?  Does this give the licensee the same rights and obligations as the licensor?  What about the obligation to notify the licensor of any improvements?  It would seem that this would fall into the obligations of the sub-licensee.  Is the sub-licensee in any obligation to notify the original licensor?  

It has been pointed out that the BiOS License is a work in progress and requires more fine tuning to be ready for consumption, so no further analysis of the drafts is wise at the moment when the final texts may change considerably in the near future.  Nevertheless, the draft licenses as they exist are tremendously complex, they represent a considerable legal document, which small and medium research facilities are supposed to sign to with little or no legal assistance.  The patent filed is subject to minute legal technicalities, and adding a complicated license to the equation may increase the legal uncertainty that already exists.  Research centres may be tempted to simply stay away from all open licensing efforts to reduce their liabilities.  But there are other inherent problems with patent licensing that make the porting of OSS extremely unlikely.

3.  Trouble with Patents

The difficulties with the one existing patent license scheme and the lack of other open science licenses indicates that there appears to be an inherent problem in porting a licensing model that has been designed to work with copyright into a system that would have to work with patents.

There can be little doubt that patents offer the strongest short-term protection of technologies emanating from costly scientific research.128  It has several advantages in order to protect certain technologies, particularly because some types of research may produce outputs that would not be suitable for copyright protection.  While this is precisely how the commercial world operates, those interested in making their work available to the public under some sort of open license model will have to do it by protecting their work, as the entire system is based on the threat of infringement suits brought against those who had not shared the work according to the required clauses.

The problem then for some institutions wanting to release their research is that they will have to obtain a patent in order to license it.  This can prove to be an expensive endeavor.  Some studies estimate that an average biotechnology patent application could cost an average of $7,500 in the United States alone.129  Because patents must be applied separately in each jurisdiction where they will be exploited,130 the costs for a small research institution could be prohibitive.  Even when the patent has been obtained, the enforcement of patents is where the costs are steeper.  The cost of defending a patent in the United States where the dispute is less than a million dollars can range from $300,000 to $750,000.131  This means that even if a research institution obtains a patent to protect their research, the right holders would find it extremely expensive to defend their intellectual property against misuse-particularly considering that those likely to use open source licenses may be small research institutions, or even individual researchers.132  The problem would be more pronounced for researchers in developing countries, as they would possibly have to enforce patents abroad.

The sheer costs involved could be enough to dissuade small and medium research facilities to stay clear of the potential liabilities involved with the patent system, and continue releasing information through more traditional means.  Nevertheless, there may be a viable solution for the problem of the enforcement of patents held by individual organisations.  The problem of enforceability of OSS licenses is similar to what has been described in the previous paragraphs.  In software, many small software developers do not have the resources to enforce their copyrights.  For that purpose, the FSF recommends that all those programmers using their licenses should assign copyright of their works to the FSF because in that way they can enforce the license better in case of infringement.133  This scheme could be replicated in open science licenses, and has been accepted by CAMBIA as one of its potential roles by stating that BiOS will serve as “a collective defense of the commons and non-assertion of IP rights against other members of the commons.”134  Therefore, collective organisations could be in charge of the enforcement of research held by individuals.

Another possible problem for the use of open licenses of patented technology is that it could be argued that open licenses are incompatible with prevalent patent policy goals.  An often stated goal of a patent system is to encourage the distribution of inventions through the utilitarian justification that allows for the economic reward.135  An open license model might clash with this objective because it would stop inventors from being able to economically recuperate investments in future research related to the patented one, particularly if a non-commercial viral license is used.  For example, imagine a patented gene sequence that has been licensed through a copyleft license containing non-commercial viral restrictions.  Researchers who would want to patent improvements based on the licensed sequence and then exploit them commercially may find that they would be in breach of their licenses, as they would not be able to sub-license their work commercially because of the viral restrictions.  But what if those researchers incurred considerable expenses to produce the improvements?  They would have to license their research using a non-commercial clause as well, which would defeat the utilitarian justification for patents.  The BiOS License does not appear to have this problem because it allows commercial use of the patented technology, so there may be a case that there cannot be a non-commercial open license for patents.

The apparent incompatibility of the patent system with open source is well known in the groups that advocate for its implementation.  The statement of purpose of the Science Commons project enumerates some of the problems faced by the project in trying to translate the CC ideals into working licenses for works that rely on patent protection.136  The Science Commons proposal goes as far as to point out that “[m]any of the things that we have learned in forming the Creative Commons do not translate completely to the world of science policy.  We dealt primarily with copyright-here the issues would also involve patent and trade secret.”137

The potential incompatibility between patents and open source licenses is difficult to resolve.  Even with the early drafts of the BiOS Licenses, the nature of the patent system seems to offer insurmountable obstacles to the possible adoption of a viable open science license dealing with patented technology.  Those industries that are willing to incur the cost of expensive research will want to see their efforts rewarded, and a large patent portfolio will give research-heavy institutions an excellent bargaining position when dealing with other competitors within the industry.  The race to decode the human genome has served to demonstrate that there are significant economic interests at play, and this is a fact that will not go away overnight.  This is why the best option for smaller research facilities and public interest oriented institutions is to disseminate works through releases into the public domain.  This type of dissemination has the effect of widening the accessibility of the research by other small industries.  Another effect of the release of materials into the public domain would be to pre-empt future patent applications, because the research has already been made public.  Eisenberg explains this tactic:   

In addition to making it difficult for publicly-funded investigators and their institutions to file timely applications for patents, the Bermuda rules also lead to the prompt creation of “prior art” that could potentially defeat patent claims based on similar DNA-sequencing efforts in the private sector.  No one can get a patent on something that was already publicly disclosed before the patent claimant discovered it. 138

Despite this seemingly watertight solution, there is still a real potential that the information that has been made available for free could be copied and then used to make patent applications about that same material.  The chaotic state of patent applications in areas such as software and biotechnology139 provides a warning that patent offices cannot be trusted in identifying whether a patent application is innovative, or if it is based on significant prior art.

It would be fair to assume that there are too many problems, and it could be suggested that perhaps open science should be scrapped-at least in open science for patents.  The lack of licenses at the moment makes the possible implementation difficult, while the few efforts that have been proposed still seem to fall short.  Even the strongest proponents of open science and open biotechnology recognise that the movement cannot go forward without viable licenses.  Hope comments that:  

Key issues for advancing the open source biotechnology analysis will be developing open source patent licenses and other licenses appropriate for biotechnological subject matter, assessing the importance of higher capital costs in biotechnology development and establishing whether or not there exist secondary markets for biotechnology services or other commercial offerings that might support business models along the lines that have proved successful in the software context. 140

It is perhaps time to look at different options.  Those who believe in ensuring wider access to scientific research and technology should not be daunted by the difficulties encountered, as there may be other solutions that can provide a viable manner to harness the creative and developmental strengths of the open source model with other ways of dissemination.