Remarks on the IBM Standards Principles

Print Friendly

Podcast on Remarks on the IBM Standards Principles

It’s been an interesting day as I’ve watched the various responses to the publication of the IBM Standards Principles. As I hoped, they raised some issues and started some debates. I suspect that longtime readers of this blog will find themes in the principles that I have discussed here before.

The principles themselves are meant to be taken as a whole and are written, I think, in a positive, constructive way. For example, some of the words and phrases used include:

  • “create economic development platforms”
  • “deliver services to their citizens”
  • “quality and openness”
  • “adopt open and global standards”
  • “participate”
  • “advance”
  • “collaborate”
  • “clear, simple, and consistent”

These principles demonstrate IBM’s continued and expanded support for open standards, as well as the importance they hold for our customers. We are strengthening our commitment to open standards and the value they bring to all.

Let me add a few comments to the principles themselves, in reverse order.


Help drive the creation of clear, simple and consistent intellectual property policies for standards organizations, thereby enabling standards developers and implementers to make informed technical and business decisions.

I discussed this in the seventh prediction I made at LinuxWorld 2008. In that blog entry I said

Over at the Creative Commons, they’ve worked out six licenses that allow for several types of “free” use of creative content. The website allows you to check a few options, answer a few questions, and voila, you have a license.

I think something similar could work for standards organizations. Even though mathematically the different possible options for each element of a license could lead to a combinatorial explosion, in practice only one or two of the options are in common use.

We need to develop this small number of license templates and we need to get standards organizations to commit to using them, without change. No tweaking. No saying “Well, we’re special so we’re going to modify this area.” You’re not special.

If there were fewer licenses, they would be better understood. Approval to participate in standards efforts would become simpler. We would better know how to create products and software that mixed and matched standards from different sources. We would better understand how standards licenses lined up with open source licenses.

Where appropriate, we also need to employ ex ante, the early notification of intellectual property that applies to a standard. Disclosure up front so there are no unexpected surprises later.


Collaborate with standards bodies and developer communities to ensure that open software interoperability standards are freely available and implementable.

This is consistent with the software interoperability pledge that IBM made in June, 2007. More than that, it’s making sure that standards developers and software implementers, be it for open source or proprietary software, share requirements and work together to maximize use of the highest quality standards that will do well in the market. It also means that the right policies need to be in place and well understood by everyone.


Advance governance rules within standards bodies that ensure technology decisions, votes, and dispute resolutions are made fairly by independent participants, protected from undue influence.

Back in the 1990s I was a member of the W3C Document Object Working Group. One source of entertainment at some of the meetings was when two employees of the same company would argue strenuously for positions that were diametrically opposed. At the other extreme, having a large group of standards participants voting as a bloc, seemingly under the control of one party, does not bring balance and equilibrium to the situation.

That’s not to say that different people coming from their own positions can’t agree and therefore vote the same way. The problem comes when decisions are reached in ways that don’t reflect real independence of the decision makers.

I’ve asked before in this blog if we don’t need some sort of full disclosure from standards participants. In the wiki IBM facilitated last summer, there was a good discussion of the notions of open government and how these might apply to standards making. Over time various votes on standards will be won or lost. I think an open, transparent organization should help users and other stakeholders understand who voted how and why. This is especially true for organizations that represent countries. We must have and understand accountability.


Encourage emerging and developed economies to both adopt open global standards and to participate in the creation of those standards.

When we have a minimal number of high quality standards that many people around the world can use, then implementing software and other products can become easier. We get better interoperability and we can exchange information globally. We get more competition, wherever it might come from. This drives more and better choices of software and other products for customers.

Standards should take into account requirements coming from anyone who might use the standards, even if that person or organization is on the other side of the planet from you. Conversely, the more countries or regions that use global standards instead of creating local silos, the better connected we all will be.

A standard that can be used with great success globally does not necessarily have to be an “international” standard. Standards organizations need to respect the needs and opinions of technology creators and users from both emerging and established economies.


Begin or end participation in standards bodies based on the quality and openness of their processes, membership rules, and intellectual property policies.

IBM belongs to many, many fine standards organizations and we look forward to long and productive relationships with them. These organizations and their work are strategic to IBM’s business and the products and services it offers to its customers. IBM is proud to be a member of these groups.

Today IBM has various processes in place that determine if it joins a particular organization or a working group within it. I think these work well and they include both business and legal analyses. Usually IBM doesn’t join a particular effort if the work it is undertaking doesn’t match IBM’s business model. For example, I don’t think IBM is in any standards group that is creating specifications for, say, umbrellas.

IBM may choose to not participate in an effort if IBM thinks it unnecessarily duplicates previous, presumably better work. This was the case with OOXML in ECMA. If IBM gets involved with a standards effort it is because IBM thinks it is important and is something it will likely implement. This involves a lot of resources and, as with any organization, those resources need to employed efficiently and in the right places.

With this principle, IBM is saying that it will increasingly look more closely at issues like the openness and transparency of a standards organization, as well as the modernness and consistency of the processes and intellectual property rules. IBM did so before, but it will do more in the future. IBM will sharpen and communicate its criteria to those involved in a cooperative manner.

Notwithstanding the various sexy headlines I’ve seen today, leaving a standards group would be a last resort. Though IBM is but one company, it hopes to use its experience to help resolve problems that are found in a constructive and collegial way, before the situation becomes too dire.


To reiterate what I said at the beginning, IBM is more committed to open standards than ever before. IBM believes that open standards are good for customers since they provide more options, better products, and insurance against being locked in by any one vendor or provider.

I hope others will find value in these principles.


  1. Bob,

    Very nice. I work on a couple of standards at UL and ANSI. There is a fair bit of overlap in staff working on the standards at both organizations (the standards are different, but related), and they are a great bunch that I’d go out for dinner with afterwords any time (and have many times). However all it would take is one bad apple to mess things up. Hopefully that never happens, but…

  2. Is IBM going quit standards bodies? Anyway i hope it’s just a strong voice for changes in standards organizations….

  3. @Tad: Only as a last resort would IBM leave any, as I mentioned in the blog entry.

  4. IBM Human Resources were in the Hursley cafeteria today, handing out IBM Reward Gateway vouchers.

    They enable IBMers to buy Marks&Spencers gift tokens at a 6% discount. M&S is a bit like Macy’s.

    Us slightly-autistic developers wonder what place they have in the commercial scheme of things. They surely are not confidential to anybody. But if everybody in the world had them, it would cause the sort of ‘financial and commercial meltdown’ that is taxing the President to the tune of $700B to try to avoid.

    If Coca-Cola were handing them out, they’d be for 5 cents off a bottle of Coke, and the Coke salesman would be trying to get them to all of Pepsi’s best customers.

    Do they have a commercial value, or not ? Can you sell them for cash, or only give them away ? Does anyone have a justifiable commercial complaint, whatever may (or may not) be done with these vouchers ? Is it ‘par for the course’ in the commercial competitive world if it causes Pepsi’s market share to collapse ?

    Is this by any chance what is meant by “For IBM Internal Use Only” ? I thought we’d scrapped that classification a while back.

    On with business. I don’t think IBM will sell any more copies of OS/2 or Lotus SmartSuite. Not with IBM Lotus Symphony on free download to everyone in the world.

  5. Hello Bob

    IBM clearly does believe leaving standards bodies is a last resort. Never mind ISO – the company is still a founder4 and sponsor of the body which has had most criticism over OOXML: ECMA

  6. [quote]If IBM gets involved with a standards effort it is because IBM thinks it is important and is something it will likely implement. This involves a lot of resources and, as with any organization, those resources need to employed efficiently and in the right places.[/quote]

    Hmmm, I thought IBM was mostly spending resources getting involved in a format it did not want to implement and spent a ton of resources on that that htey could have spent on a format they were actually supporting like ODF

  7. hAI – How have you been! Long time, no comment!

  8. IBM does not propose to differentiate between the impact of similarity and compatibility standards. Simply, patents that read on similarity standards (e.g., a patent on a design of a plow) have provided a reasonable balance of private gain and public good. Patents that read on compatibility standards (e.g., a patent on a wireless air interface) seem unbalanced toward private gain. The reasons for this are developed in The Entrepreneur and Standards( Attempts at policy (legislative or standardization) to address this issue have floundered, and are likely to continue to do so, when proponents of patents raise the success of patents that read on similarity standards, and proponents of a lack of patents in standards raise the problems that are caused by patents that read on compatibility standards, and neither side explains the difference.

    Microsoft exerts control of its software APIs in many ways. One of the more significant is that Microsoft can change such APIs at any time with on-line updates. This prevents other companies, unless supported by Microsoft, from maintaining compatibility with the Microsoft APIs. So any standardization process that hopes to offer “openness” must provide a means to control such updates as part of its process. I don’t think this point is addressed in the papers I have scanned.

    Taking a technical approach to solving these problems has a much higher chance of success in the near term. OOXML and ODF is a sad state of affairs in the standardization world, but not for the reasons often noted. Fighting to create a single standard is an outdated standardization approach. When memory is very cheap it becomes practical to support two or more ways to implement compatibility (OOXML and ODF are both standards for document format compatibility). What is necessary is a standardized mechanism to identify, negotiate and select which way compatibility will be achieved. I term standardized mechanisms that support all three functions (identify, negotiate and select) “adaptability standards.” The Fundamental Nature of Standards: Technical Perspective ( offers one lower layer approach to adaptability standards.

    In internet connected systems, once adaptability standards are used it may not even be necessary to have compatibility standards. Consider a 5G cell phone (using a software defined radio) programmed for some new feature by an new entrepreneur. If adaptability standards existed in 5G (a suggestion I have made) the base station on identifying an unsupported capability in the entrepreneur’s cell phone goes to a “known” internet web site and downloads the necessary software to provide compatibility. This is an example of an “open system.”

    Consider also RIM paying over $600 million for rights to patents that read on a compatible interface RIM designed. If RIM had designed an adaptable interface and such patents were identified the interface implementation in each BlackBerry could be changed to avoid patented approaches the next time any BlackBerry connected to the server.

    Once an interface is produced in high volume, patent trolls will search for patents that may read on that interface. The likely hood of such patents being identified rises rapidly with the volume of the implementations and their profitability. Not supporting adaptability standards seems to me to be a very risky approach for IBM.

  9. The problem I have with the standards process is accessibility. I do a lot of work in the open source world on a popular content management system. The slow evolution of web standards, and the incompatibilities and other artifacts of vendor battles that we have to deal with are frustrating and limiting.

    I have been involved with standards bodies in the past, and I’m certainly not suggesting that committees should be flooded with representatives from thousands of small projects. However, it’s equally preposterous that this segment of the community can only attempt to influence standards processes by proxy though a large organization.

    In my experience, Open Source communities — at least those that don’t result from significant single-source sponsorship — are relatively free of bias and thus have a strong inclination to work toward optimal solutions to problems.

    The current method for organizing standards bodies effectively blocks the participation of highly qualified people working in a truly open environment. As it stands, any prospect of participation is tied to relatively large financial contributions that can and will influence their positions.

    I am not sure what will solve this problem, but certainly a top priority should be a major overhaul of the model of having large corporations fund the participation of their staff. If every corporate sponsorship was both prepaid for a long term (say, 8 years) and was sufficient to fund the participation of 1.5 to 2 non-aligned positions in the process, and those positions were filled on the basis of merit, then perhaps we would start to see better and more timely results from these bodies.

    [My apologies for hacking the post ID this comment]

Comments are closed