Please visit Home (Summary) for the FINAL version of Policy Dialogue Summary
Policy Dialogue: Comments Archive
| Common Understanding: Open Dialogue on Open Standards |
|---|
| Regional Conference on Open Standards: the Key to an Open ICT Ecosystem |
|---|
| Sofitel Central Plaza Hotel, Vibhavadee Ballroom A, Bangkok, Thailand |
|---|
| Chairs: | Shahid Akhtar, Programme Coordinator, United Nations Development Programme's Asia-Pacific Development Information Programme (UNDP-APDIP) |
|---|
| Thaweesak Koanantakool, Director, National Electronics and Computer Technology Development Agency (NECTEC) of Thailand |
| Co-chairs: | Nah Soo Hoe, Advisor, UNDP-APDIP |
|---|
| Sunil Abraham, Manager, International Open Source Network (IOSN/APDIP) |
The purpose of this dialogue session was to explore policy issues affecting open standards and openly discuss and share practices, to come to some common understanding and possible future collaboration.
There were 54 participants representing 13 countries from the region - from government, the private sector, academia, and civil society.
The following summaries are clustered by the topics covered during the dialogue, highlighting the salient points and suggestions for moving forward.
Table of Contents
1 Open Standards attributes
1.1 A good deal of time was devoted to discussing the attributes of open standards. It was proposed and agreed that the attributes of open standards include:
- Openly developed
- Openly maintained
- Openly modified
- Openly accessible1
- Openly implemented
| Comment | Username | Time |
|---|
Perhaps we should note here that we are refering to technical IT standards here and especially software-related standards rather than the entire spectrum of standards in general.
-- Nah Soo Hoe
| DialogueMember | 04 May 2006 - 11:31 |
| Not to draw out discussion, but were these elements actually agreed to? We had much discussion of and around them. For clarity, I think it is necessary to specifically ask if they are accepted as stated or not. -- Jeff Kaplan | DialogueMember | 04 May 2006 - 14:47 |
| Does a standard need to include all the mentioned attributes or either of them to be open standards? Openly maintained, modified and accessible sounds like a must for the open standards, but not in terms of development and implementation. It can be opened after the development by a closed group of developers. Open implementation will be too strong for business viewpoint. It is similar to the case of GPL which later issued LGPL to softhen the agreement. -- Virach Sornlertlamvanich | DialogueMember | 06 May 2006 - 23:07 |
| I believe the above 5 attributes of open standards represent the ideal situation. Depending on who you are and the type or work you do, you might find that only some of the attributes apply to you. For example, if I am developing a learning management system based on some openly available standards, I want to be able to get hold of the standards documentation. I expect the standards to be maintained by some organisation. For software development, I agree with Dr. Virach. There is usually a core group of developers who will work on the project in a closed environment first before they decide to open it up. In any case, a software project can be openly developed but if a person is not up to scratch, he will be left out totally! [Lim Kin Chew] | DialogueMember | 06 May 2006 - 23:55 |
as i suggested these, let me clarify to address VS's comment - yes, all five are needed to be a truly open standard for software interoperability (to clarify NSH's comment) or as LKC suggested for the ideal. also, if we are going to include these five terms some clarification as to what is meant should be added. i've suggested definitions below and these, as everything can be discussed further...
Development: A purely open standard is created in a manner that allows anyone who wishes to participate to do so without discrimination. All proceedings and records of proceedings are public and decisions are made by consensus. No one can prevent anything that actually happened during the development of a standard from being recorded in the minutes. There are no secret agreements. No one person or organization can veto a decision.
Maintenance: Once created, a purely open standard is maintained by a community under terms similar to that which developed the standard. In addition, the community has a responsibility to maintain backward compatibility if it decides to do so. That is, openness gives the maintaining community the right to make significant modifications, but since membership is not restricted, the community may get enlarged at times by a sizable group of people who may either want to force the changes or maintain the status quo. .
Accession: Anyone who wishes to have a copy of a purely open standard may have one at no cost. When possible, the standard is available online and the document format used to represent the specification is an open document standard.
Implementation: A purely open standard can be implemented by anyone in any way desired with no payment of royalties.
Modification by Others: A purely open standard can be mined for its good ideas and used in part or in whole in other standards. In particular, a profile can be created that states which parts of multiple standards should be used together to achieve a certain end. [roslyn docktor] | DialogueMember | 08 May 2006 - 21:53 |
The above Roslyn's remark is very clear. I agreed. But if so, I am afraid that only newly developed standards are compiled to open standards. Any proved to be viable standard will not be able to announce as an open standard if it was developed by only a group of people. So it is not possible to adopt an existing (closed) standard to be an open standard though it is open up for public use. This will badly prevent any development of software that has to design a non-existing 'open standard' for a specific purpose in software procurement that follows the concept of open standard. It is not acceptable though it is going to be opened after the development. Comparing to the case of OSS, it is the case of Linux, OpenOffice?.org (StarOffice?) and the standards used in them which are licensed as an OSS after a period of closed development.
In case of 'Implementation', RD's interpretation is one of the meaning of 'open'. I thought that the implementation must be open. Any implementation that adopts an open standard has to be opened i.e. the software, specification that interfaces to the standard. I think it is quite unnatural to relate the term "open" to the concept of "no payment of royalty".
I might be too nervous about the interpretation. Please help me to understand the above cases. -- Virach Sornlertlamvanich | DialogueMember | 09 May 2006 - 17:17 |
| In the second case of the above comment, I mean if we are going to interpret the term "Openly implemented" in the definition. -- Virach Sornlertlamvanich | DialogueMember | 09 May 2006 - 17:22 |
1.2 It was pointed out that the level of openness needs to be defined, especially for individual governments that include the terms ‘open standards’ in policies. Since there is much ambiguity associated with the term ‘open’, consumers (including government) must be very specific in stating their expectations.
| Comment | Username | Time |
|---|
| In contradiction to "software", the "standards" looks better with the term of "free". "Free standards" is free of charge and provides freedom to study, use, modify and re-distribute. I do not mean to void of our discussion but the term "open" is quite ambiguous. And it is good at all to let each body define what it needs for the term "open". -- Virach Sornlertlamvanich | DialogueMember | 06 May 2006 - 23:23 |
1.3 It was recognized that there exists a spectrum for the definition of openness:
| A. Open + Royalty Free | B. Open + RAND2 + Royalty Free | C. Open + RAND + Royalty |
1.4 At one end, open standards are defined as being patent and royalty free open standards, and at the other end open standards are defined with patent and royalty implications, and in the middle are those standards where the associated patents are licensed royalty free.
| Comment | Username | Time |
|---|
| This is where there is a need to mount educational / awareness programmes for the public. I believe many people are not sure of terms like "open" and "patent". Perhaps many people are aware of the term "royalty". Perhaps if there is a not-for-profit consortium on Open Standards being set up, then we can mount such educational / awareness programmes. On this matter, we can learn something from the efforts of the Creative Commons movements. Their website (www.creativecommons.org) contains many interesting educational / awareness programmes. [Lim Kin Chew] | DialogueMember | 08 May 2006 - 22:44 |
1.5 All government representatives, some private sector representatives, and members from education institutions agreed that open standards must be royalty free (A and B). Some private sector representatives were of the opinion that Reasonable and Non-Discriminatory (RAND) with royalty implications must also be included in the definition of open standards (A and B and C).
| Comment | Username | Time |
|---|
It was pointed out by a private sector representative that the definition of open standards which includes licensing of essential patent rights to implement open standards on RAND terms (either with or without payment of a reasonable royalty) is consistent with the definitions of open standards adopted by ANSI, ITU-T, Global Standards Collaboration 10.
-Peter Chong, Microsoft | DialogueMember | 04 May 2006 - 19:21 |
1.6 One private sector representative suggested that RAND usually includes defensive suspension
3 , which protects the rights of intellectual property holders.
| Comment | Username | Time |
|---|
It was pointed out by a private sector representative that open source software products can effectively implement RAND-based technical standards. Examples cited were Motorola Smartphones running Linux, Netscape/Mozilla Firefox browsers, and Apache HTTP Server. Open source software can peacefully co-exist with RAND-based standards.
- Peter Chong | DialogueMember | 04 May 2006 - 19:27 |
1.7 It was suggested that the spectrum could follow a ‘creative commons’ model for labelling the ‘gradient’ of open standards; however, it was also stated that labels may not provide the clarity that governments require for their policies and frameworks. Furthermore, national governments must arrive at a definition of open standards based on their needs and local context.
1.8 It was agreed that two models of development, i.e. market based consensus and centralized development co-exist and there was no clear preference for either. It was noted that no matter what the model or accreditation is, the timing of standardization is critical – we must examine the maturity of the standard.
2 Procurement policies
2.1 Most government and some private sector representatives agreed to a ‘pragmatic’ approach with a preference for open standards – where they i) exist, and ii) are most appropriate.
2.2 Some private sector representatives were of the opinion that there should be no preference for open standards, governments should maintain a position of choice, instead of mandating technology of specific standards. However, other participants pointed out that governments make technological choices on a daily basis through purchasing decisions.
2.3 Government representatives from India, Viet Nam, Cambodia, and Thailand stated that the choice of standards should be based on the following criteria:
- Risk management/exit strategies/vendor independence
- Cost
- Interoperability
- Competition
- Stability
- Implementation support (also by local vendors)
- Data Longevity
- Performance
- Platform neutrality
2.4 There was generally consensus among government representatives that open standards meet these criteria.
2.5 Furthermore, it was pointed out that governments play an important role in ensuring free competition in the market place through policy intervention – open standards help maximize competition and directed, strategic industry development. Certain government representatives stated that government policies on open standards can improve negotiating positions with vendors.
2.6 It was suggested that governments mandate interoperability instead of developing policies on open standards – looking at the ends as opposed to the means. Certain private sector representatives stated that interoperability can also be achieved by proprietary standards; however, other private sector representatives pointed out that compatibility should not be confused with true interoperability.
| Comment | Username | Time |
|---|
| A private sector representative cited examples of key proprietary standards/technologies that enhance interoperability including PDF, SNA (Systems Network Architecture), PCL (Printer Command Language), Office Open XML file formats, and Java. Proprietary standards does not imply "closed" or the opposite of interoperability. It was also pointed out there are many ways to achieve interoperability, including (1) development of software that is interoperable by design, (2) licensing/cross-licensing of proprietary technologies and IP, (3) specific collaborations with partners, competitors and governments, and (4) implementation of industry standards (including open standards and broadly accessible proprietary standards). Governments should be flexible and maintain a policy of technology choice by choosing from all of the options or ways to achieve interoperability for a given problem. | DialogueMember | 04 May 2006 - 19:56 |
| Above comment that proprietary standards does not mean "closed," there are many ways to achieve interoperability and technology choice/neutrality added by Peter Chong. | DialogueMember | 04 May 2006 - 20:03 |
| in response to the above, expressed at the event as well, there was a discussion on the need to define interoperability as the ability of two or more diverse systems or components to exchange information and to use the information that has been exchanged. I believe Intel suggested consulting the CED report just released that shares "A key benefit of open standards is that they foster interoperability, allowing disparate devices, applications and networks to communicate." The Harvard report is very clear on this, stating that "Open standards are the mortar holding interoperable ICT ecosystems together. Open standards make interoperability possible, turning a principle into active policies and specifications. Open standards enable interoperability in and among diverse ICT ecosystems." [roslyn docktor] | DialogueMember | 08 May 2006 - 22:18 |
| The Massachusette's decision to mandate ODF was also raised, as I recall. Many stated support for it (as IBM did) and that where truly open standards do exist, calling them out is necessary to achieve true interoperability (as opposed to compatibility), control over the information you create, and long term access to that informatioin. [roslyn docktor] | DialogueMember | 08 May 2006 - 22:21 |
2.7 It was agreed that there is a need to de-link free/open source software policies and open standards policies to ensure clear decisions regarding purchases.
| Comment | Username | Time |
|---|
It should be noted that FOSS and open stds are generally good for each other and help popularise each other in many ways as experience has shown. Therefore if a country is keen on FOSS, IMHO it will be good if its FOSS policy can be developed hand in hand with its open stds policy.
-- Nah Soo Hoe
| DialogueMember | 04 May 2006 - 11:27 |
May I suggest modification of this item to be more explicit. For example:
"It was agreed that there is a need for purchase policies towards more openness, where both open standard policies and free/open source software policies are individually developed hand in hand with each other." | ThaweesakKoanantakool | 04 May 2006 - 12:03 |
3 ICT innovation and development
3.1 It was recognized that innovation comes from all sectors, academia, government and the private sector. Open standards maximize competition, they allow companies to have self-directed innovation and not be locked into opaque technology pathway. This is particularly important for local Small and Medium Enterprises (SMEs).
| Comment | Username | Time |
|---|
| I always feel that it is the presence of difficulties or problems that have led people to innovate. I don't think people innovate just for the sake of innovation. Similarly, I believe many open standards that we have taken so much for granted whether it is TCP/IP or USB or PDF all come about because there are some difficulties somewhere. Perhaps another way of looking at this issue is to think of working out some "open standards" whenever we come across some technical difficulties. I believe that this line of thinking will bring about more open standards. [Lim Kin Chew] | DialogueMember | 09 May 2006 - 19:52 |
3.2 In the area of research and development, some government representatives stated that adopting existing open standards saves resources and allows them to concentrate on innovation and build on what has already been developed.
3.3 It was stated that some of the problems (Avian flu, Tsunami, etc.) facing the world today cannot be solved by individual corporations or nation states, and that publishing data using open formats will create global collaboration that could rapidly address these challenges.
| Comment | Username | Time |
|---|
| I want to add the example of the Sahana Project in Sri Lanka during the Tsunami disaster. The Sahana Project was developed using FOSS (e.g. PHP, MySQL?, Apache) to provide all the logistical support. In times of disasters, using open standards based formats really speed things up. We should anticipate future disasters and build up scenarios in which other open standards can be deployed, e.g. using e-learning standards or FOSS-based learning management systems to ensure that e-learning contents can be delivered to school children during periods when schools have to be shut down because of avian flu. [Lim Kin Chew] | DialogueMember | 06 May 2006 - 23:22 |
3.4 It was agreed that the Internet is the best example of an innovation based on open standards that itself has become a global platform for further innovation based on open standards.
| Comment | Username | Time |
|---|
| There are other open standards which have led to some interesting innovations. For example, the SCORM specification in e-learning has given rise to other innovations like Pocket SCORM, SCORM for Interactive TV. Anyway, it would be wonderful if some organisation can take the trouble to document the innovations based on open standards. When we have more examples, we will be able to convince more people to go for open standards. I am sure there are innovations in other domain areas. [Lim Kin Chew] | DialogueMember | 08 May 2006 - 22:50 |
4 Free/Open Source Software
4.1 It was pointed that FOSS and open standards share a symbiotic relationship and there is an overlap between the definition of FOSS and open standards. Some participants were of the view that the ability for FOSS to implement a standard is a measure of the openness of the standard.
| Comment | Username | Time |
|---|
| Certain participants seemed to suggest that ability to implement a standard by FOSS was a basic requirement (as opposed to just a "measurement") for a standard to be considered open. Or should it just be existence of multiple implementations without reference to free/open source? I thought this issue was more specifically discussed. -- Jeff Kaplan | DialogueMember | 04 May 2006 - 14:54 |
| The way I look at open standards is that firstly we need some documentation on the standard or the specification itself. This can be downloaded from some website without any charge. Secondly, we need some implementations of the standard itself. Thirdly, we need some kind of conformance/compliance tests to ensure that the implementations do indeed meet the expectations of the standard. Fourthly, we need to run the implementations on some other platforms to ensure that interoperability does take place. I would say that there must be systematic procedures and the processes must be transparent for true open standards. If there are no implementations and no tests for interoperability, then the so called "open standards" are still under development. [Lim Kin Chew] | DialogueMember | 09 May 2006 - 17:46 |
4.2 There was general consensus that government policies on FOSS be delinked from policies on open standards. However, some participants pointed out that open standards policies help to level the playing field in the domain of software
4.3 It was stated that in developing countries both FOSS and open standards are a means to providing universal access to ICT.
5 Strategic planning and architecture frameworks based on open standard
5.1 There was consensus that governments need a high-level interoperability framework. Such frameworks should:
- be forward looking and examine scenarios and use cases
- consider the industry-wide perspective and impacts on industry
- guide government in delivering government-to-government and government-to-citizens (e-government) services and functionalities
- include performance and service level agreements
- involve stakeholders and include guidelines for vendors
- include risk management and exit strategies
- consider data longevity/archiving
- vendor and platform neutral
5.2 It was agreed that there is no single interoperability framework that can suit all countries, and that a framework is formulated from the needs and requirements of individual countries.
| Comment | Username | Time |
|---|
On Government Interoperability Frameworks, a private sector representative recommended that "pragmatism be the rule" instead of a unilateral preference for open standards and that "limiting procurement to only officially adopted open standards can limit the ability to exploit new technologies" (Berkman Report, pp. 4, 24). It was pointed out that private industry was united on the policy that there should not be mandates for OSS. Procurement should be based on merits and functionality. OSS and proprietary software are part of the software eco-system. Likewise, open standards and proprietary standards are part of the spectrum of standards and ways to achieve interoperability. The best way for Governments to maximize interoperability is to maintain "choice" on which option is the best way to achieve interoperability in that case, choice on which open standards and/or proprietary standards to use and choice between open source and proprietary software.
- Peter Chong | DialogueMember | 04 May 2006 - 20:22 |
6 Proposals and ideas for the way forward
6.1 Form a ‘group’ modelled after the Berkman
Roadmap for Open ICT Ecosystems exercise to explore specific areas of interest in open standards in the Asia-Pacific context.
6.2 Foster regional cooperation towards developing government interoperability frameworks – at the national and regional levels. This would include the development of local definitions of open standards and guidelines for policy.
| Comment | Username | Time |
|---|
| Perhaps a better way would be to do a quick comparative study of some of the Government Interoperability Frameworks that have been made available publicly. I did a quick check on Google and found that we can get GIFs from the following countries: UK, New Zealand, Australia, Denmark, Hong Kong, Malaysia, Australia. In the case of Canada, the Government has their interesting Government of Canada Metadata Framework. As I understood it, this Framework has metadata at the centre of their strategy. [Lim Kin Chew] | DialogueMember | 09 May 2006 - 08:31 |
6.3 Undertake capacity building activities on open standards targeting CIOs (government), developers, and the general public.
| Comment | Username | Time |
|---|
| Again I must stress the need to define the scope of open standards. We can have open standards for the Internet, for operating system, for documents, for e-learning, for information exchange, for multimedia, etc. I feel that before we talk about capacity building we need to specify the scope of what we want to do in open standards. Obviously, we cannot do everything. We need to be selective. [Lim Kin Chew] | DialogueMember | 08 May 2006 - 22:36 |
6.4 Hold a region-wide competition to promote the development of open standards and to recognize the interoperability of government projects based on open standards.
| Comment | Username | Time |
|---|
| I would think that it is better to organise domain specific plugfest events rather than holding competitions. In fact, there are many of such industry oriented plugfests, e.g. USB Plugfests, WiMAX? Certification and Plugfest, ADSL Plugfest. When we talk of interoperability, industry members are keen to know how well their products and systems match against those industry standards. Having been involved in two learning standards plugfests, I would definitely recommend that we hold plugfests rather than competitions. [Lim Kin Chew] | DialogueMember | 09 May 2006 - 10:14 |
6.5 Undertake a study to analyze existing government interoperability frameworks, and understand how they impact competition in the market place.
| Comment | Username | Time |
|---|
| Perhaps this can be a one-off exercise which can be outsourced to some academic institutions or research bodies to do. [Lim Kin Chew] | DialogueMember | 06 May 2006 - 22:44 |
6.6 Establish centres of excellence, involving all stakeholders, to develop proof of concepts and case studies. For example, these centres may monitor the use language-based open standards.
| Comment | Username | Time |
|---|
| This is a good idea but we need to select the type of open standards we want to champion. Which open standards do we want to champion? For some, I think there is really no need to champion them as the standards have been very well accepted. I presume we will only want to deal only with the technical aspects of the open standards. I have set up the E-learning Competency Centre in Singapore to champion e-learning standards in Singapore. You need the following: funding, competent and passionate people and support from the government. Standards work is very dry and generally not exciting. You need some imagination to make the open standards work more interesting. [Lim Kin Chew] | DialogueMember | 06 May 2006 - 22:40 |
| Web Language Engineering project is to monitor the deployment of standards in creating web documents. De facto standards and the limitation of authoring tools mainly affect the development process. To understand the situation and informing of the standards are neccessary to keep thing in its place. Fundamental modules i.e. collaborative crawler, language identification and archiving with search ability are being studied and implemented with 3 partners. The minimum servers and trained experts are good to start the collaboration. We are scheduling to train the fundamental modules and to release them as OSS for future improvement and extention of the language coverage. -- Virach Sornlertlamvanich | DialogueMember | 06 May 2006 - 23:48 |
| | UNDP Asia-Pacific Development Information Programme | Bangkok, Thailand, 3 May 2006 |
Footnotes
1. It was suggested that access to specifications and documentation be free of charge.
2. Reasonable and Non-Discriminatory licensing – used when patents are included in an open standard.
3. Defensive suspension - the license can be suspended if any user of the standard sues the licensor for patent infringement.
go to TableOfContents
PolicyDialogue Web > Summary
Log In
PolicyDialogue Web
Create New Topic
Index
Search
Changes
Notifications
Statistics
Preferences
Rss, Atom
Security
