Forum OpenACS Q&A: OpenForce's commitment to the OpenACS free software community

A few months ago, a few baffling rumors started spreading, questioning OpenForce's dedication to free software and the OpenACS community. At first, I chose (and instructed other OpenForce engineers) to ignore them. We were busy working on dotLRN, and I didn't want to let various bickerings get in the way of exciting new code.

The last few weeks, however, have seen these rumors spread and grow beyond anything I expected. I'd like to take this opportunity to strongly reiterate our beliefs, our commitments, and point out the efforts we've made to always be top-notch contributors to the community.

OpenForce is entirely dedicated to free software. Our name says it. Our work says it. We will never waver from this. Our code, whenever possible, will be released under the GPL. Always.

A few important points to support our words with actions:

  • OpenForce was the first company to commit to OpenACS - not ACS. We committed to OpenACS back when it wasn't such a safe bet, back when everyone said "we work with ACS.... oh and with OpenACS too." OpenForce was one of the first companies to develop and host a commercial OpenACS site (
  • OpenForce actively lobbied ArsDigita to provide OpenACS hooks into ACS 4.x, and to maintain code under the GPL license. At every conference where we gave speeches (ArsDigita meetings, the European Free Software Meeting, the O'Reilly Open-Source Conference), we pitched OpenACS as a platform and *every* current OpenACS company equally.
  • OpenForce has contributed *many* critical packages to OpenACS 4.x, including total rewrites of forums, calendar, notifications. Important rewrites, cleanups of file-storage, faq, news. Important rewrites and optimizations of the core APIs. Entirely new packages from the dotLRN work.
  • OpenForce was the first company to *manage*, not just release, a client development effort in an open-source manner - dotLRN. Until Lars's announcement yesterday, we were the only ones. We applaud Collaboraid for taking this even further with their latest Internationalization effort. Open-source development is not always easy. Open-source development of a client project is quite difficult.
  • Although not specified in our contract with Sloan, OpenForce designed dotLRN to be as modular as possible from the start. There are currently four companies doing parallel yet compatible dotLRN-based development thanks to this architecture. OpenForce chose an architecture that created a new market: we effectively created our own competition because we believe in true open-source development.
  • OpenForce designed an architecture for dotLRN that allowed dotLRN to benefit from OpenACS *and* OpenACS to benefit from dotLRN without forcing an install of dotLRN. This is far from trivial. It's the reason for the 3-package approach to building a dotLRN applet. It's the only such client development that has been completed successfully to date: true symbiosis of the platform and client application.
  • OpenForce lobbied for assigning dotLRN copyright to the Free Software Foundation as a means of ensuring that future, diverse interest in the dotLRN platform would never compromise the GPL status of the code.
  • OpenForce managed, wrote most of, and released the PostgreSQL port of dotLRN under the GPL (with the help of some community members like Neophytos and a number of others). This was not part of the Sloan contract.
  • To this date, OpenForce spends thousands of dollars in legal negotiations with every client to make as much of the resulting code as possible available under the GPL. That is our continued dedication to free software and to the community.
We know and appreciate that many others have contributed quite a lot, too. OpenForce did not single-handedly make OpenACS into what it is today. We are part of a community.

So why this list? Because somewhere along the road, some people got the idea that we were no longer contributing. That we were fighting against them. We *are* contributing. We are *not* fighting against anyone. We're working with the community. We're trying to make OpenACS bigger and better.

I'm asking that you judge us by the results we've accomplished, by the actions we've taken. Look at the code, look at the opportunities created by this code.

If you have questions, I am making myself available by email, IRC, and physical presence. I'll be at tomorrow's social that Talli is organizing in NYC. Feel free to ask me about doubts you have. Feel free to question anything we've done, as long as you place your faith in the code produced and actions taken. If you have specific requests about things you think we're doing incorrectly, please tell us!

OpenForce is committed to free software and to the free software community. We always have been, and we always will be.

Exactly how is Open Force's refusal to take part in dotLRN governance going to help the community regain confidence in your company and your ability to be an effective leader?

I'm feeling very dense today. Could you please help me dot the i's and cross the t's? I can't help but think that either I'm a congenital idiot or you've made an incredibly stupid decision that you should reconsider in light of your company's commitment to the OpenACS free software community.

We know and appreciate that many others have contributed quite a lot, too. OpenForce did not single-handedly make OpenACS into what it is today.
Gee, thanks, Ben. I for one know for certain that OpenForce did not single-handedly make OpenACS into what it is today. It's even possible that I, along with several others, have contributed "quite a lot" more than "quite a lot" into making it what it is today. In addition there are a couple of dozen others who have contributed "quite a lot".

An attitude shift in the direction of humbleness on the part of Open Force, and a willingness to apologize for certain missteps, would do a lot more to improve your relationship with the community and, I dare say, Sloan/MIT than this post of yours.

I want to make it clear to everyone that I greatly appreciate all the hard work OpenForce has put into the project.  I don't think anyone seriusly involved in the community has any misunderstanding of the importance of the work OpenForce has done.
I tend to avoid threads that don't involve code, beer or lingerie but I'm confused here. I haven't heard of any of the rumours that Ben mentioned. I also haven't heard that OF had refused to be on the dotLRN governance board. So I suspect that a lot of this is based on private emails.

But I don't understand the vehemence here. What exactly has OF done that is so bad? They created dotLRN and suggested a governance plan that was shot down by the community. If it's true that OF is refusing to be a part of the new dotLRN governance, is that such a bad thing? Should they be forced to be a part of a governance system that they disagree with? I would love to see them take part, but I don't know the details of what's been going on behind the scenes. I certainly see no evidence that they've abandoned free software nor done anything to lose my confidence in them.

I guess I'm posting because this thread seems so incongruous with my own perspective on dotLRN and OpenACS - most likely because I don't know what the *real* issues being argued are.


I am at the same state as Vinod.  Don't know why Ben posted this.  And why DonB replied that way.  Can someone fill in the details and points?

So no comment on the above posts except... what the heck is going on?

Vinod and Jun:

Your confusion is is my confusion of a few months ago. And your surprise at
Don's attitude is also my surprise. That's the kind of attitude I've been getting
from him for a couple of months now, with strong implications that others feel
that way.

I chose to go public because it got to the point that it interfered with normal
day-to-day development: some, including Don, chose to ignore my technical
recommendations claiming that I had ulterior motives. This is not a good thing
for the open-source development of OpenACS.

Don, once again, you have taken a complex, *private* email of mine and
taken it wholly out of context (when, only 4 weeks ago, you made it clear that
such action is inappropriate when Neophytos requests posting *your* private
email). Not to mention that, as Vinod points out, OF's participation in dotLRN
governance is OF's business and no one else's.

You're asking for OpenForce to "be more humble" in order to "improve" our
relationship with the community, making this assumption that there's
something big to fix. I'm okay with that if you can tell me: what did we do to
hurt our relationship with the community in the first place? What is your beef
exactly? What is this big thing we have to fix? Please point to actions we've
done that have hurt the community and I will be more than happy to address


I can't really comment much.  But for whatever happens I hope things are possitively resolved.  I don't know who else are involved but even if its just OF and DonB there is already a substiantial part of OpenACS contributors involved.  At any rate I hope things turn out for the better, a negative turn out may severly have a negative impact on the project.

Anyway I am sure community members like me who have little knowledge about this and is concerned about the matter awaits further postings by any other community member involved in this issue.  Thanks.


Here is where I think some misunderstanding occured.

Originally the new-portal package was supposed to be part of OpenACS. THe docs (very mininal) specifically said that portlets for new-portal could not rely on dotlrn functionality because they would go back to OpenACS.

The latest code does not have this recommendation (last time I checked, who knows why it is gone.)

You yourself decided that new-portal could not be part of OpenACS, then changed your mind. Again, that is all I know but that much information was posted in the bboard.

Many tcl procs were deprectaed and replaced with a new API with no notice to OpenACS developers. Some of them would be interested in major refactoring of core OpenACS code.

Also, there is a extreme lack of developer documentation on how to implement dotLRN. Right now, OpenForce, if anyone, are the only developers who know how the parts work together.

Now, this is not an attack at all, but understand that these things all happened. I am not accusing OpenForce of any intentional slight or ignoring of the OpenACS community. I just think that many OpenACS developers thought there would be a more transparent development of the parts of dotLRN that affected OpenACS.

Also, I appreciate everything OpenForce has done. I have personally benefitted from dotLRN. I also have had to dig into the code to understand how to build a package that runs with dotLRN.

So Ben, if we can involve the rest of the community in important decisions about OpenACS, not necessarily a vote on the OpenACS bboard, but some notification of developers who are working alongside OpenForce to improve the toolkit would go a long way.

I must say that I am truly baffled by this posting. Perhaps Ben can enlighten us on who is spreading these "rumors questioning OpenForce's dedication to free software and the OpenACS community."

As far as I am concerned anyone who is spreading this rumor should be taken to the woodshed and given a sound beating. I will be the first to get in line with the paddle.

Before Ben left for vacation, I asked him to serve as dotLRN kernel gatekeeper. Don Baccus, as chair of the dotLRN Technical Board, asked if someone from OpenForce could also serve on the Technical Board. Make no mistake about it, this would have given Ben and OpenForce *THE* dominant voice in setting technical direction for dotLRN. This was in recognition of Ben and OpenForce's contribution in developing dotLRN. Ben said that he would give his answer after returning from vacation. Note: *THE* dominant voice does not mean the *ONLY* voice.

As many of you know, dotLRN is generating tremendous interest and momentum because of the work of everyone concerned. However, every day that goes without achieving clarity and transparency we lose potential adopters and potential funding. We need to move on with the governance for dotLRN. We need to know *how* OpenForce wishes to particpate in dotLRN. The issue is not whether OpenForce contributed to dotLRN or is committed to open source. Who the hell questions that? However, we need to be clear as soon as possible on how OpenForce wishes to participate in dotLRN. No one expects OpenForce to do all the work. But we need to know who is going to write the developer documentation. When will the code be released officially? How will bugs be tracked? On and On and On.

Do I expect OpenForce to do it all? No. It's precisely because OpenForce can't do it all that we need to know what OpenForce is willing to do and can realistically be expected to do. Whatever it can contribute, the community will be grateful as in the past.

When Ben returned from vacation, we received an email from him declining to serve as gatekeeper or have OpenForce representation. That's a business decision that Ben needs to make, one which he has every right to make, and for which he will be respected. Ben does not believe that the governance plan for dotLRN will work. Carl and I asked him to reconsider. I will post my email message below. I won't post Ben's original message rejecting participation in dotLRN governance unless he wishes to do so himself. We did not receive a reply from Ben. Instead, I saw this posting.

Aside from posting my email to Ben, this will be the one and only posting from me on this topic. I need to spend my energies on dotLRN and all the work that remains to be done, not chasing rumors and innuendoes.

I'm okay with that if you can tell me: what did we do to hurt our relationship with the community in the first place?
If you can't figure it out yourself, Ben, the odds that I can figure it out for you are very low, indeed. Do keep in mind that I defended you against all comers for months, accepting your rationalizations and explanations for OpenForce behavior at face value and asking that others do the same.

Perhaps the fact that so many rationalizations were required in the first place hurt your relationship with the community?

I'm probably not going to post any more on this thread, among other things I'm leaving town tomorrow and won't have 'net access for a week.

Your post is by its nature divisive. I keep silent, division deepens. I speak up, division deepens. Keeping my mouth shut is the easier way to deepen the division, I might as well do that.

This was my email to Ben asking him to reconsider:


I am extremely disappointed in your decision and hope that you will re-consider. dotLRN owes its success to your leadership and that of OpenForce. I know others would also agree with me that OpenForce participation is critical to dotLRN's future. By your serving as kernel Gatekeeper and also OpenForce having representation on TAB, you and OpenForce will have the most influence and power to shape dotLRN's future.

What problems are we trying to solve with the organizational structure? If dotLRN were entirely a technical platform, then dotLRN governance should be of developers, for developers, and by developers. But dotLRN is a vertical application and for that reason it's success will not simply depend upon developers. dotLRN did not get built nor will it evolve because of developer "itchiness". Developers didn't come along and say...."gee wouldn't it be nice if we had an e-learning platform." dotLRN got developed because people put money on the table.

The problem we are trying to solve is simply this. If dotLRN is to succeed we need the widest possible adoption and as quickly as possible. In my estimation, we have a small window for success, twelve to eighteen months at most. In order for dotLRN to succeed, we need wide adoption. In order to have wide adoption, we need:

  • transparency. this point has been made eloquently by Carl and I won't make it again. this comes up over and over again with everyone I talk to. as soon as one starts to talk about "organic" processes, the decision makers head for the door. it's as simple as that.
  • legitimacy. having the MIT and Univ. of Heidelberg brand helps.
  • sustainability/funding. dotLRN will not succeed if we expect developers to code and thereby scratch an "itch"; most developers could care less about an elearning platform; they do care about having bread on the table. for me the primary goal of the EB will be develop a business plan so that dotLRN development can be sustained and that there is adequate funding to do the things we all want to do.
If you can persuade me that your governance ideas can achieve transparency, legitimacy, and sustainability, I am all ears. I have no interest in making technical decisions; neither does Carl; and neither will others on the Executive Board. We all share the same goal. We want dotLRN to succeed. All I ask is that you give us the benefit of the doubt and reconsider your decision.



Let me say, that dotLRN is a superior platform. It is well written and took a temendous effort. I should have sent that last message to Ben as an email. He asked, and I explained what I have seen.
Don and Al beat me to it.. not much more to say.

I am still hoping that Ben will change his mind about being gatekeeper (it is still very early for dotLRN and I feel his experience would also be valuable in building the dotLRN Consortium)... yet time is of essence and we need to move on (we have been waiting for Ben's response for more than two weeks).

I would like to emphasize that one of the major selling points of dotLRN will be vendor independence. dotLRN is going to succeed with or without OF (partially due to their commitment to open source software)... I hope with. If Ben ultimately decides against it, it will set dotLRN back... but dotLRN will survive. If Ben does not want OF to take part in the governance I hope that they take part in other ways.


P.S. To prevent problems related to transparency in the future the i18n agreement between Collaboraid and Heidelberg has specifically made transparency a requirement. I hope that this can serve as a model for future dotLRN/OpenACS development. Please follow and contribute to decisions here:

If you can't figure it out yourself, Ben, the odds that I can figure it out for you are very low, indeed.

You're dodging the question. Smearing someone's reputation and then refusing to justify the smear is an incredibly sleazy thing to do.

Don't explain it for Ben. Explain it for Vinod, Jun, and the rest of the community. Even if you don't think Ben could understand the issue (not that I believe for a second that your claim to this effect is anything but an excuse to avoid answering a question with which you are uncomfortable), the others deserve an explanation.

Full disclosure: I work for Ben. But I disagreed with him on the issue of dotLRN governance, and spent a fair amount of time trying to convince him that Al's plan was the better one. And though I disagreed with him, I strongly believe that Ben's rejection of Al's plan was entirely principled, and that he had no ulterior motive for that rejection or for refusing the dotLRN gatekeeper position.

But I disagreed with him on the issue of dotLRN governance, and spent a fair amount of time trying to convince him that Al's plan was the better one. And though I disagreed with him, I strongly believe that Ben's rejection of Al's plan was entirely principled, and that he had no ulterior motive for that rejection or for refusing the dotLRN gatekeeper position.
I made no comment about Ben's motives. I simply said it is a stupid decision. Since you disagreed with Ben on the issue, you might be willing to go so far as to agree that it's a suboptimal decision, perhaps?

As for the rest, as I said above I'm going to keep my mouth shut. If that makes me "incredibly sleazy" in your eyes, then so be it. The community's more important than my reputation and a public slugfest will do no good whatsover.

Ok thanks for the added postings atleast somehow we get a little background why the sudden posts about this.

As for my comments.  I think no further details, comments, posting, etc. are needed.  Because it seems that this issue is a gray area.  There is no clear right and wrong... in fact there might not be even right or wrong.

I still hope that things will be resolved for the better even though it does not look like it will be.  Be it be resolved in public or private, I would not care.  Each of us have our decisions, principles, etc. and lets live with it.  Further dicussion on this issue on this topic is nothing bug negative time.  So as for me I will stay away from this topic.

Thanks for giving some backgrounder of the issue, atleast we are not left scratching with our heads.

Important points:
  • my original posting has nothing to do with my opinion of the dotLRN governance issue, which I was waiting to resolve with Al before publicy discussing.

  • Don, if you have no issues to point out, then I have no answer to provide. If you have some "rationalizations" in mind, then I'm sure you know what issues these were supposedly refering to (I sure don't). The only reason I brought this up publicly is because you've been bringing it up privately, causing divisiveness privately, and refusing to hear my technical opinion privately.

Now, I'd like to address Dave's points which are clear and which are very helpful to my understanding of the situation.

  • new-portal. OpenForce has always developed this to be OpenACS- dependent and not dotLRN-dependent. I never advocated that new- portal should be dotLRN-dependent. One of my private emails was grossly misquoted on the issue, which understandably led to this misunderstanding. What I advocated was the idea that new-portal shouldn't be distributed with the base OpenACS tarball and stuffed down every developer's throat, because I believe that is serious bloat. I suggested distributing it separately, as an independent package. I went back on this decision after I realized that, given current OpenACS architecture, this would make people's life more complicated, not less. That said, I'm still very worried about OpenACS bloat.

  • new Tcl procs: we cleaned up a lot of the Tcl API but we never deprecated old procs, we made sure they were always functional. We tried to give as much help as possible to point this out, but we probably didn't do enough. We'll work more towards that, this makes a lot of sense.

  • documentation: yes, we have not documented as much as we should have. It's very difficult under client deadlines to do so. That said, we've provided as much person-to-person help as possible. We've offered our help to MuseaTech, Furfly, and Collaboraid in case they hit any snags. We'll work harder on this, this point also makes sense.

  • generally increased communication: we'll work harder on that, too. I think this is a general problem that we all have, apart from the few courageous documentation gurus like Roberto and Vinod, without whom we would have code and no docs.
Dave: your comments are not at all interpreted as an attack. On the contrary, I find them helpful. It's helped me identify one massive miscommunication (new-portal), and a few points where we should do a better job (generally communication and documentation).

Thanks alot! I agree that we want to have a leaner core OpenACS. I do understand and I also have had any questions I had answered while working with dotLRN.

Let's see ... I can win this argument, but doing so will probably split the community into warring factions with no chance of reconciliation in the future.

Or I can sit back and let Ben cheerfully paint himself as being, as always, the innocent victim of a misunderstanding that he himself bears no responsibility for.  Making me look like an ass in the process.

Hmmm...I'm in an enviable position, don't you agree?  Fucked if I do, fucked if I don't.  Wonderful.

OK, Ben, I guess you win.


This is not an argument. I'm not looking for a winner and a loser. I am not
trying to prove you or anyone else wrong. On the contrary, I am trying to
resolve an issue that has been festering and which could become highly
detrimental to the OpenACS community.

I've identified a serious problem: over the past few months, my technical
recommendations have been ignored (not disagreed with - plainly ignored)
within the OpenACS community because of rumors concerning OpenForce's
dedication to free software.

I've spent weeks and month trying to solve this in private with you and others.
I've heard no specific complaint, only general rumors repeated and spread. I
went public when I was left with no other choice. It would have been
irresponsible of me to let this issue fester any longer as I think it affects the
level of technical discourse that has always been the pride of this community.

So once again, if there's an issue you have in mind, please bring it up. I can't
say that I understand why your bringing up any issue would split the
community into warring factions, but, if you think that's the case, bring up the
issue with me in private. And not just you, but anyone else who wants to bring
anything up! You can even log in anonymously to this bboard and post that

I'm trying to hear people out, just like I heard Dave out, and take these
criticisms under consideration so I can fix the problems. I didn't make anything
up: your postings clearly show that you think there *is* a problem. Let's get it
out on the table and resolve it.

(Note: in order to give Don a fair chance to respond to my post after he returns
from his trip, I will not post again to this thread until next week).

Catching up on bboard posts...

I wasn't aware of these OpenForce rumours, and I really don't know what are the issues and events involved in Don's disagreement with Ben.

I will now go ask the Jedi council for guidance from the force, because it seems the only way to find out what's going on.