President Reichardt,
Your letter raises many issues that deserve wider discussion. While you conclude with an upbeat assessment of where things stand and the rosy future facing the OGC, I do not share your view.
From my perspective, the OGC is sick.
By many measures, the OGC is doing wonderfully. The organization is held in wide regard, the standards are widely respected, the discussion space useful for a number of groups. From what I can tell, the finances of the group are doing fine. The TC is, by all appearances, working as effectively has it ever has.
However, from down in the trenches, the picture is not as rosy. The organization has failed to adapt to the wave of openness that has swept the informatic world. The organization will have a very hard time maintaining the reputation it currently enjoys. The organization has failed to improve its production process to ensure that new standard documents are well written. The organization's leadership needs to come up with some way to lead us forwards past the current hurdles. The organization will have to work to restore cohesion following the divisiveness of the current vote. None are likely to happen.
The standards bodies in the informatic world often have an approach which is much more open than the OGC. Many other groups have public mailing lists, published draft documents, and an ongoing conversation with the public. Here we have closed meetings, private wikis, and have to vote just to talk to the rest of the OGC. It is time to open up.
I have written you and Carl several times regarding the utility of the current approach at the OGC. In response to my last mail, you even promised action on the subject but have been silent ever since. I have yet to hear that this current design helps anyone, that any company is participating in SWG work and in OGC discussions but would not if the other members were allowed to talk openly about their work. Absent evidence to the contrary, this seems an artifact of the 1990's and not relevant to the 2010's. At the very least, it is time to justify staying closed but, really, it is time to open up.
It would be easy to do: allow SWGs to choose how they want to work. Let us work with public mailing lists and on a public wiki. Let us share drafts among ourselves without having to go through ridiculous internal votes. Let us join that public forum, the internet. It is past time to open up.
But you seem not to see the issue and have made no progress in these past five years even in defining the need for the current structure. Apparently, the urgency and utility of openness are lost on you. Since you don't get it, there will be no pressure and sustained work to move the OGC forwards on this.
The OGC has gained a reputation as the place for open geospatial standards.
Certainly, there was an easy argument to make early on that the benefits of open standards could be substantial to many organizations and that the standards of the OGC could play a starring role. There was the theoretical SDI picture of CSW as the directory service, and all the other services hanging off of that. WMS could be a visualization service, WFS and WCS data access services, S*S the backend data gathering service, and so on. That is the vision, and it rocks.
However, while the vision was an easy sell, that was the past. Now, reality will set in and the OGC will need to maintain its reputation in the cold, hard glare of that reality. The limitations of the services are becoming apparent, the lack of harmonization between the services becoming visible, the mistakes, the painful process to fix them, and the lack of progress becoming obvious to all. That is the reality and it is painful.
So maintaining the reputation of the OGC, which might be very high right now with governments, international institutions, and external companies, will require a fair amount of work. Some way forwards will have to be sought and sustained energy maintained to make that happen. For instance, it may become time for the OGC to figure out how to push more energy towards standards maintenance than it has done so far.
In your letter, you posit a 'big tent' vision for the OGC, as an organization that accepts any and all standards, and presumably then lets others decide which are good and which are bad. For better or worse, however, the OGC's reputation has been built on having a core set of well built standards that could be implemented directly. Under your 'big tent' vision, governments and other groups are going to be awfully surprised when they discover that they now have to do a whole bunch of due diligence before deciding on using OGC standard because the 'big tent' has welcomed many and various standards but all are doing things their own way. Warning everyone of this situation is work you are going to have to do as OGC President. The cost of evaluating which mix of OGC standards to use will become apparent to everyone, from lawmakers to business folk to the general public, either because you publicize it effectively and they do it up front or after spectacular failures. The latter, you might suspect, could also impact the OGC reputation.
So, we will see what happens. How probable is it that the OGC pull off such a massive educational campaign to make these external groups aware that this extra work is needed and to help them make these assessments? Well, since there is no structure at the OGC for such work, it seems most probable it will not happen.
OGC standards are, by and large, poorly written; all new documents should undergo extensive editorial review before being released.
Outside the OGC, the standards are widely criticized by implementers as being obscure and part of that criticism is well deserved (though much of it is also the hot air of folk who don't know of what they speak). The recent effort towards 'modularizing standards' partly addressed the issue by requiring that new injunctive language be readily identified. That was a huge step forwards. Equally important was the idea that all injunctive language be testable. Wonderful, nay revolutionary! Nonetheless, the OGC still happily votes to release utter crap.
In the last two efforts I joined, I have acted not as a contributor but rather as an editor and only at the last minute. WMTS was approved as a new OGC standard (or almost approved, I forget) when I first read it; it was a total disaster. It took several months of hard editorial work to get the SWG members who had written it to express exactly what it was they were trying to say and face the decisions they had been hiding from. Similarly, the TC recently voted to release a new Best Practice document which, upon reading, I found painfully badly written. Now, thanks to a huge, renewed effort from some core contributors, the document is gaining better form, a more accurate language, and even a systematic structure it had lacked previously. Something is wrong at the OGC if we are voting to release such documents as our final result but everyone sees the utility of a major re-write prior to release.
We need an editor and you should figure out how to get a good one working for us, probably by pulling them on staff. All standardization documents should undergo an extensive editorial review to be clear, focused, concise, and intelligible. English is a much harder code to wrangle than are the informatic languages most of us technical folk are trained to write. English is openly ambiguous and interpreted by a legion of non-standardized interpretation engines: our readers. Give us the help we need.
But we have never done such a thing before and it sounds hard to find a world class technical writer to help us write our standards and more staff would cost more money, so there's no reason to think we actually will undertake this editorial review or other work focused on writing better standards.
The OGC has no leadership.
There is, to my knowledge, no shared vision for what we are doing together, no sustained aim driving us forwards, no primary goal to measure ourselves against this year, over the next five years, or the next decade. We are not, for instance, intent on building the standards for an open SDI. That would have us focused on identifying the issues we still need to solve, and giving primacy to those issues. Right now, we are still in passive mode, waiting for any three members to start a SWG on whatever issue they feel like writing. Then, in the testbeds we go chase the novel, the sexy, and forget the past. We follow no shared goals, we follow no inspired leader.
The OGC Architecture Board (OAB) has become an oversight board of the TC but its activities and commitments are unclear. The public face of this group, http://www.opengeospatial.org/projects/groups/oab mentions october 2010 as the last election. If memory serves there has been one non-election in the interim where candidates were merely proclaimed members of this board; I don't know, it is doing its own thing on its own time frame. The board did publish the Mod Spec but seem to have gotten bored with that are have not shown much energy in maintaining that document or in leading us towards better standards. Where would the OAB have the OGC focus? What are the key issues which are not being addressed by the current structure?
The Board of Directors at the OGC is ... Well, if you elect people to that board that are openly hostile to new standards and to new versions of existing standards because it does not serve their personal commercial interests, it is hard to imagine productive leadership will come from there. I will continue to do my best to ignore that board. Thankfully, you mostly stay out of TC business.
The OWS Common effort is dead. There has been no sustained leadership to push us forwards on this front.
We have tried to tackle REST, RESTful designs, and some related issues but these efforts have all fizzled. There is lots of interest and lots of energy but no way to make progress. Leadership on the issue would have found a way to get three competing approaches under way, to test out the issues and advantages of each, to explore the solution space. 'Passiveship' is still waiting for someone to do something somewhere sometime.
So, from here, the OGC seems like a meandering morass without much vision or aim. We went through a generational turnover from the initial contributors to the people who are currently active but we did not adapt to the new reality and have suffered for it. For example, can you, as president, tell me what you would encourage me to be thinking of during every SWG meeting? What is a shared goal over the next ten years we can all embrace? Which issues are key for the needs of the users of the OGC Standards? I can't formulate what the president or the Architecture Board thinks is important; it therefore seems clear that that leadership is not impacting me.
Your letter seems to be taking some form of leadership role; however, I find that mostly surprising, as I discuss below.
Leaders inspire, forge ahead, find ways to make progress, bring out the best in us; you who would be leaders are not doing that. Perhaps you need to work differently, perhaps others need an opportunity to make progress. However, the leaders have no reason to take on this new work; they are at the top of the hierarchy so everything looks good from up there and it's sexy to have this leadership role on your resume. I have no illusions that the OGC will gain effective leadership in the near future.
First, let me point out that I do not care what actually happens with the vote. In the structure you have designed for the OGC, I don't even get a vote so at that level this is the problem of others, not mine. As I see it, there are no elegant solutions short of ESRI pulling its document for the good of all and there is no reason whatsoever for that to happen. The harm to us as a community has largely already happened: the SWG members feel put upon, the non-SWG members feel used, the camps are entrenched. Since I expect the standard to pass, after all there is power and money behind it, I have already geared myself up for the consequences. I personally would reject the document on purely linguistic grounds; even if you agree with the intent, the technical design, and the utility of this document, its text needs so much work that the document should not be foisted on any readers in its current form. Poor implementers! However, to me, the vote fundamentally no longer matters.
What interests me then, is a realistic analysis of the situation we are in.
On technical grounds, the candidate suite is a terrible design. If we started from scratch today to build a mass market suite of geospatial services for the future, no one would do things this way. There are many other technical issues which could not be tackled given how the work group was established. So this new standard is, in reality, a standard for a legacy system. On technical grounds then, the OGC has no reason to feel proud of this series of documents.
On language grounds, the documents are a disgrace. I've pointed out many of the linguistic issues in various comments, including injunctions that do not say what exactly is being enjoined but hope the reader can figure that out for us, tests which cannot be implemented, other tests which simply say 'please test everything here', an introductory section which is meandering through a sea of buzzwords with no visible purpose---I certainly learned nothing from reading it. The document is not any worse than many other OGC documents but certainly shows that the OGC is not making progress towards writing clearer standards. So the OGC has no reason to be proud of this text either.
On maintainability grounds, a new community has been formed. Since the participation of the OGC membership at large has been so effectively excluded due to the requirement for 'backwards compatibility' there is little likelihood that the next version will be warmly embraced by the larger OGC membership. So maintenance or improvement will fall on the original SWG community. Presumably that will happen at the speed and desire of the main implementation and we will be in the situation that we have one vendor with its very own OGC standard to lead forwards as it sees fit. Not a point of pride either.
At the level of the OGC community, this is a debacle. The KML fight was pretty bad though I was mostly unaware of that discussion and cared even less. This seems measurably worse. We've gone from 'lets build together the standards for tomorrow's spatial data infrastructure' to 'Let me pass the standards I can'. This harm, of course, has already been done; the only thing left to see is who gets to feel relieved and who gets to feel bitter by the result of the vote. This is not our proudest moment.
So the reality, as I see it, is that this situation teaches us that there is a structural weakness with the OGC, one that you should be bending over backwards to find a way beyond. But it is unlikely that the OGC undertake the work required to get itself back on track to the former vision of itself as the designer of services for an interoperable SDI; we have been drifting for too long.
Finally, I find your public letter, at this late date, a little puzzling and the use of TC-Announce confusing.
Your letter starts with three paragraphs of self-congratulatory sap. Then you thank the SWG members for their work. Then you thank the voters for being engaged and have the gall to call the discussion 'constructive'. Eventually, you take a stand, or maybe take a stand. Your position seems to be that
It is far better for this kind of dialog to occur within the OGC
and its expertise base than with any other process.
which is so clearly wishful thinking that it feels like you are in a different organization than I am. Can you name two technical points of 'dialog' in this voting process? All I see is points of 'dissension'. Then you go on to suggest that the future will be hunky dory since we can all work to make things better, failing to acknowledge any of the obvious power imbalances in the situation, or the disgust of many in the wider membership at the lack of participation, involvement, or interest from the authors in the expertise of the group. (And of course we have yet to face two groups that decide to take an existing standard forwards in different directions. First filer wins it seems. If I were nefarious I would get organized to form the successor SWG the day after passage of this document just to lead the 'standard' in a completely counter-productive direction. The system is gameable in lots of directions.)
So it seems like your letter has the obliquely stated aim to support the standard. If so, it would have been more polite to state this front and center rather than hint at your support deep in a late paragraph. You seem to be taking this stand either personally, or in your capacity as president, or (partly?) on behalf of the planning committee, the OAB, and the board of directors; that is unclear and the extent of dissension within those groups not stated. So what is your letter doing? Is this a Please vote for the standard from me or a Please vote for the standard from the OGC President for the benefit of the OGC or perhaps a Please vote for the standard on behalf of the OGC boards or maybe even the more neutral Please recognize that there are also benefits to the standard it would be nice to know. And why are you using TC announce, is this an official letter from the board of directors or some such? Will you also use TC Announce to share the letter from dissenters, like the letter from OSGeo to the TC Membership?
A final point of confusion on my part is that you seem to consider it to be your role to be to weigh in on the votes of the TC. I marvel at the length Carl goes to to maintain his neutrality in all things TC. From that I had thought that the policy of OGC Staff was to stay 'hands off' in technical matters. So I am a little surprised to see what appears to be a stance. Of course, being active and partisan on TC Votes would be fine, if clearly and directly stated. The position should be clear and those taking the position obvious. In such a case, however, it might be useful to describe why the position you argue for would benefit the OGC membership, or perhaps just the OGC itself.
What is not fine is pretending that there is no profoundly destructive issue!
This process has been divisive to the community as is obvious to anyone. The process has gotten people riled up across the globe. The vote will not resolve those matters; whichever side looses will be deeply affected by it. Recognition of this fact is the first step towards dealing with it. Y should be planning for the aftermath now, figuring out how you will respond to the position of the loosing side, figure out the consequences for the long term for them, for the OGC, and for the public at large. You will have to act to help get us working together again, no matter what the outcome of the vote. The first step is for you to articulate to yourself what is really going on.
No, Mr President, all is not hunky-dory at the OGC. Your organization needs you to step up to plate and open up the OGC, needs you to act on the public platform the positive reputation has given you and educate the public on the consequences of our votes for them, needs you to raise the quality of the standards we produce, needs you to lead us forward past our current roadblocks, and needs you to help pick up the pieces from the GeoServices travesty. It is not likely we will make progress on any of these is it, let alone on all?
This letter is not meant as an indictment of you or anyone else. It is a merely, an opinionated, forceful statement that the OGC needs to face its weaknesses if it is to stay effective, productive, and true to its vision and, thereby, to help the world build interoperable, open ways to harness its geospatial self.
sincerely, Adrian Custer, Individual Member