Forum .LRN Q&A: continuation of dotLRN Governance

Posted by Ben Adida on
Hi all,

After much thought on how dotLRN governance might work, I concluded that the current dotLRN governance plan is not an open-source governance plan. In Arjun's words, we are faced with a battle of the present vs. the future. This plan sounds workable in the present and even enticing from a financial perspective, but it compromises the future of a truly open-source, generic eLearning platform.

As unpopular as my decision may be in the present, I must make it with serious thought given to the future of dotLRN as an open-source platform: OpenForce will not be participating in dotLRN governance because the dotLRN governance plan does not reflect open-source values.

Below is the email I sent Al, Carl, and the dotLRN TAB on Thursday evening to inform them of this personally.

Al, Carl, Don

I've spent the past few weeks re-reading the dotLRN governance plan, observing the formation of the Tech Board and Executive Board, and generally trying to reconcile the principles of open-source development and the finalized dotLRN governance plan. I find them irreconcilable.

The plan's expanded and top-heavy management structure does away with the the meritocratic approach of open-source. The detached selection process for the Executive Board and its total control over the Tech Board membership send a loud and clear message: contributing code to dotLRN will not give one any ability to similarly contribute to the strategic (and even technical) direction of dotLRN. In fact, the current model provides for very little incentive to be part of dotLRN - unless you're already a vendor of the existing Executive Board members. This is different in every way from successful open-source projects, including the oft-mentioned Apache Group whose Executive Board grew organically out of the developer - not the user - community.

In short, the problems your organizational structure aims to solve are unclear and do not justify the proposed complexity and contradiction of open-source processes.

OpenForce's own goals have always been clearly stated: to further truly open- source web development. In this context, I do not believe the dotLRN plan will succeed. I thank you for your offers, but I am turning down the position of dotLRN Gatekeeper as well as OpenForce participation in the Tech Board.

As soon as the Executive Board designates a new hosting provider for the CVS and technology website (including bug tracking), OpenForce will do its best to facilitate the transition, including providing any and all code and database dumps in their latest form.

-Ben Adida

Posted by Jim Lynch on
Hi Ben,

I'm definitely curious about your thought process in this matter, so I wonder if you could provide a breakdown of the dotlrn governance plan with your annotations indicating bias against open-source values?

If you believe some of the plan's effects is to be at odds with open-source values, I most definitely respect and initially support your decision (but I'm a nobody... you don't necessarily need my support :), but please... "show me the code" :)

Posted by carl garland on
Just a close tag ... ignore
Posted by Carl Robert Blesius on

Jim, make sure and take a look at
The discussion has been open for more than a month, and the first we heard from Ben/OF was on Thursday (he was on vacation). We put everything on hold because we all really wanted him to be gatekeeper and for OF to take part in in the initial governance plan of dotLRN. I wish we could have changed his mind... but his post makes it official. If there is one thing that I have learned from all this, it is that Ben is governed by his will on this.

We have put A LOT of energy into this discussion with Ben, but WE MUST MOVE ON. Experience shows that energy we put into this subtracts from the success of dotLRN and we all strongly believe in the success of dotLRN as an open-source platform (regardless if it is a model that fits into Ben's open-source view or not). OpenForce is not essential to the ultimate success of dotLRN... we still strongly believe in dotLRN's success with or without them (although with them would have most likely have been a WIN-WIN situtation).

Don sent off one private mail to Ben and others that have been taking part in the conversation that I wish he had shared in response to this post before he left to climb mountains and band hawks and eagles for the week (far away from any form of network or telephone access). I am going to go out on a limb here and share quotes from this mail with the community without his permission, because I think these are essential parts of the disagreement that must be shared now and I am willing to suffer the consequences.

In response to Ben's definition of the building blocks of an open-source community (Joe Developers) Don wrote, "Some of us take a broader view of those who form the building blocks of our community. I've been trying to foster an atmosphere in the OpenACS community where Joe Documenter, Joe Tester, Joe User and Joe Funder are recognized as being every bit as important to our success as the developer community."

Don emphasized that "dotLRN would not be in existence if it weren't for Joe Funder, and much of the functionality reflects the needs of Joe User as specified" in a contract.

Don also wrote that OF would "be on much firmer ground if ... [Ben/OF had] done this project on a voluntary basis"

Don continued in saying that he has seen "several developers contributing freely to dotLRN despite ... [Ben's] claim that they've lost their previously clear reasons to do so. We've had community-provided applets, help on porting to PG, bug fixes, etc. I expect that to continue."

Enough. I had to share this. I wish Don was around to share the whole thing and add his two cents... but maybe it is better that he isn't... we have already lost enough time on this discussion as it stands.

I also think that the community should read parts of my reply to Ben's private E-Mail (above) that I would like to share:

I think I can speak for Al as well on one point (because this is the first thing I wanted to know before I said yes to his EB offer)... we are not interested in control... or deciding who is on the technical or user advisory board (yet it is obvious that it needs to start somewhere and what a better way than to pick leaders from the OpenACS community and have them create the initial boards?).


We (as schools) are interested in pooling resources Ben... the structures that are forming will allow us to do that more easily (we are already discussing how we are going to cooperate on external authentication).

I also find it very difficult to compare dotLRN governance and Apache governance (they are two very different animals).

In my view dotLRN is a product that MIT Sloan is opening up for a good reasons (it makes good business sense). I see dotLRN as a product of a very liberal non-controlling client-contractor relationship... between Sloan and OF (although I have not seen any contracts... or been given any special information) and I get the feeling that a lot of the architectural beauty in dotLRN is a direct result of your (and your teams) genius (and the freedom given to OF by Sloan)... not to mention the old ACES system and the 2 years of usage coming from Sloan. Which makes it even more sad and strange to see you drop dotLRN (before it is really finished... e.g. developer documentation... research module...).

I have to say that I disagree with your view about the success of dotLRN as an opensource learning platform... it will succeed. Obviously OF abandonment of dotLRN would throw a huge wrench into the process, making everything harder for all involved, but the MIT name, our name, the building momentum, and the OpenACS community are going to sell this Ben as well as other institutions that are waiting for dotLRN's official release. A lot of people are really excited about dotLRN and it is pushing a lot of momentum back into OpenACS.


Ben... I would like to see you change your mind... this is not in the best interest of all involved (or am I missing something?).

We have made our decision to go with dotLRN -> Sloans experience, vendor independence, a community driven effort, a community made up of schools and universities (to start with), and being built on a killer community framework are just some of the factors that play a part in this. We are sure of success Ben. The one thing that we all want to come of the "governance plan" is structures that promote community cooperation and community participation in development.

It is early to jump off Ben... I think you would be making a mistake and I would really hate to see you do it. Like I wrote above... I would like to see you change your mind (regardless of where the technology site lands).... just like you I would like to see a meritocratic structure and I think this has more to do with misunderstanding and failed communication than anything else.


In closing I would like emphasize that I am going to do my best not to take part in this discussion for a while... my experience has shown that it is counterproductive for all involved at this point. We need to minimize the damage and continue. Ben has blown a little hole in the dotLRN boat that we need to concentrate on fixing so we can sail on.
Posted by Arjun Sanyal on
"If there is one thing that I have learned from all this, it is that
Ben is governed by his will on this."


"Ben has blown a little hole in the dotLRN boat that we need to
concentrate on fixing so we can sail on."

  - both Carl Robert Blesius (above)

Carl, here are a few more things that you, and others, may want to
take away from this:

1) OpenForce is not just Ben. There are two other developers (Yon and
myself) who worked on dotlrn and its foundations for over a year
now. And there are three other employees at OF who did not write code
but contributed significantly. So, whatever characterizations you
choose to make of Ben also apply to _us_.

2) If you mean that our decision is "willful", I find nothing wrong
sticking to the founding principles of our company i.e. writing and
promoting truly open-source software. We feel that we serve our
clients, the community we are a part of, and ourselves best by doing
so. You disagree with our decision, I respect that, but I assure you
that this decision was not taken lightly or impetuously.

3) Being one of the architects and builders of dotlrn, I would like
you to remember that we worked extremely hard to write it not only for
our client, MIT Sloan, but also for _you_. We didn't know at the time
exactly who you were. And you didn't know at the time that you would
want to use dotlrn either. Now we know and we are glad that you find
our work useful. We fought "the battle of the present versus the future" and the future won.

I wish you and everyone else success with dotlrn.

Disclosure: I work for OpenForce.

Posted by Alfred Essa on
For the record...this was one of the emails sent to Ben trying to persuade him to join the dotLRN effort. I will let the community reach whatever conclusions it will..

Dear Ben,

Thanks. I do understand that you need to what you need to do. I do hope that you will reconsider. If not, I appreciate your working with us on the transition.

There are multiple models of successful open source communities and the one you advocate is only one of them. The model you propose fits well with open source projects that have an entirely technical impetus and motivation (developers wanting to scratch an itch). dotLRN did not originate this way nor we can we expect it to evolve if it is left to developers alone. But let us also realize that you are trading on an ambiguity. You keep positioning yourself as a developer with purist open source motives. Please do not forget that you are also the CEO of a company in competition with other companies and developers in the community.

In your model all power resides in Ben and OpenForce. You would be the sole gatekeeper and make all technical decisions about the platform. Your intentions and allegiance to open source principles is irrelevant. There is an inherent conflict of interest as long your company stands to benefit from dotLRN. I am *not* asserting that you or OpenForce will act contrary to the community’s interest. But given human nature, we need to ensure that whenever there is a potential for conflict of interest we have mechanisms in place to mitigate it. Technical disagreements have arisen and will arise within the developer community about dotLRN. Unless we provide a deliberative model to resolve them, these disputes will turn into personality battles and acrimony. We have already seen this deterioration and in the long run it will only damage dotLRN’s reputation. In your governance model, the Technical Advisory Board would have only nominal powers since you would have ultimate say and final authority over *all* technical decisions. Yes, the gatekeeper would be re-elected a year from now. By then dotLRN could also be dead.

Technical governance has to be linked with marketing and sustainability. In your model there is no feedback loop from users, marketing, or strategic objectives of the platform. What’s the business model for dotLRN in your plan? If one doesn’t exist, how will it be developed? Whose responsibility is it? The dotLRN business plan has to tie together technical, marketing, and strategic objectives. Here again there are two fatal flaws in your governance plan. First, technical direction would be set entirely by the gatekeeper (namely you) without any accountability or connection to the business model or users for dotLRN. Second, you advocate a laissez faire approach for the non-technical dimension of dotLRN. This reminds me a lot of the invisible hand of capitalism where somehow free markets are going to take care of everything. We know it won’t. If there is a free for all on the non-technical dimension, then the only one that will benefit potentially is you and OpenForce. OpenForce will have an inside track in ensuring that the technical evolution of dotLRN fits well with OpenForce’s business objectives and strategy. Everyone else will be screwed. Once again purity of motives and allegiance to the open source principles don’t mean anything to me. Where are the checks and balances?

You yourself continue to concede that because of client deadlines OF is not able to do as much as you would like (e.g. developer documentation, release timelines). No one is faulting OF for that. But the point is that we cannot get by in a state of confusion where there is inadequate documentation, unclarity about release dates, inconsistent marketing messages, etc. etc. Everyday that goes by with this state of affairs means loss of business and potential adopters.

You do rightly point out that there also a potential of too much power in the Executive Board and also MIT. This is a legitimate concern and we are taking steps to ensure that there are sufficient checks and balances there as well. We are trying to build a formal organization called the dotLRN consortium and MIT will be transfer authority over to that as soon as we can. Since Univ.ofHeidelberg is now fully on board and contributing, we will be marketing dotLRN as a joint initiative of MIT and Univ.ofHeidelberg. Carl will serve as a co-chair on the EB. This has the double advantage of leveraging the prestige of both institutions and ensuring that there is a counter-balance to MIT/Sloan on the EB. I will be meeting with Berklee next week. I know they also wish to participate and contribute in a strong way. I think TAB is now on a sound footing. Lars, John, Dan, and Don are recognized leaders and contributors in the OpenACS community. OpenForce’s participation would give it even more strength. Michael is also making headway in getting a solid composition for UAB.

Finally, you keep asserting that Joe Developer has no motivation now to contribute. Instead of saying this a priori, can you please give an example of an OpenACS developer who feels this way? More precisely, what about the governance plan is inhibiting contributions? Our dilemma at the moment is quite the opposite. Because we lack strong coordination we are not able to take full advantage of all work that is being done around dotLRN by Sloan, Berklee ,Vienna, etc. On the contrary I see great momentum around dotLRN and interest because people are beginning to feel that there will be strong backing and encouragement from MIT and Heidelberg. I do want to say also that it’s an important element of the governance plan and the business model of dotLRN that we cannot count on or depend upon voluntary contributions. Voluntary contributions are nice and we will take it always, but dotLRN will not evolve unless we continue to have strong funding.

Once again I hope that you will serve as dotLRN kernel gatekeeper and that someone from OF will serve on TAB. If not, we will begin working on the transition next week.

Many thanks,


Posted by Ben Adida on
Jim: the best explanation of my issues with the current governance plan are in
the main governance thread that Carl refers to. I'll spare you the repetition :)
The most concise summary is that this process in no way resembles existing
successful open-source governance. I don't consider that any one of us is
qualified to invent a brand new open-source governance plan that no one has
ever tested before. That is the big red flag, IMO.

Carl: your message is puzzling. In late July, I presented my governance plan
to you in person in Boston. According to the governance bboard thread, my
first public contribution to the discussion happened on August 7th when I put
forth this plan in writing. I posted 8 times in two days explaining my opinion. I
never wavered. I have trouble seeing how you didn't know what I thought until
last Thursday.

If you assumed I would compromise on my principles, then I can understand
your surprise.

What may surprise you even more is that I never expected my turning down
this new governance plan position to become such a huge issue. But now I
truly fear for dotLRN: should the community  expect this kind of treatment if we
disagree with the rules set forth by the executive board? Participate according
to your rules (however they may change) or be accused of blowing holes in
the dotLRN boat? Comply or become alienated? Do what you say or expect
multi-pronged attacks in public?

A brand new governance plan was just enacted. Prior to this, I tried for weeks
to convince you that it was a poor choice. I failed to do so. My choice was
binary: participate in a plan I consider  flawed to the core, or opt not to
participate. I chose not to participate, and I have trouble seeing the flaw in

You're right, I am governed by my own will, and proud of it. In the open-source
world, this is usually considered a good thing.

Posted by defunct defunct on
Hi All (and Ben in particular), :o)

I've been reading this exchange over my coffee for the last few mornings with some interest. And although I'm reticent about contributing, I do find myself fascinated by the substance (a little morbid maybe)...

Disclaimer: I don't know the details of the governance intimately, I don't use or (at the time of writing) have any plans to use dotLRN, and I *really, really, really* am very much on the fence on this one and have no position to back/assert

But what I do have is a number of observations and questions that I think may be relevant, and that I would really like to find answers to..

I, like Ben, speak my mind, do as I my will dictates and so have (possibly) a good deal of sympathy with his postion.

So here's my somewhat random, and throw-away observations...

  • Open Source is wonderful thing because it takes many forms, but retains many solid principles. i.e. peer review, knowledge sharing, communal approaches and so on....
  • Open Source is not a 'completed' or 'stable' business OR development model. I agree there is much to suggest it is, and much to indicate that it may even become the dominant form of delivery, but at present (as with much in computing) this is anecdotal and 'gut' based.
  • Open Source is about freedom, and its freedom therefore that creates good software, breaks boundaries and acheives new heights. It also therefore stands to reason that its a discipline that attracts visionaries, revolutionaries and ideologists...
  • ... But its proving itself as a viable business method. And when commerce becomes involved the necessities of trade apply. Things like moderation, compromise, management and direction/structure. This (as of the time of writing) is still the case for business.
  • Combining the two previous cases is tricky... in fact its very tricky... and seldom do other industries succeed.
  • Academics are good thinkers but poor do-ers.... Business people are good executors, but poor visionaries. Ok I accept thats a blanket generalisation... but its true in the general sense.
  • Software management and delivery in the non-private sector (education, locel government etc) is notorious for things like delays, wasted funds, late deliveries, systems that don't work and projects run by over-regulated committees (perhaps a side effect of public money/lack of commercial pressure.
  • The same in the commercial breeds Microsoft... I hardly need say anything more.
  • Software developers are a strange professional bunch, by comparision with many industries. They are part construction worker, part artist, part scientist, and quite often part philosopher... Almost a precident!

Ok I'm not going to go on endlessley, as I don't intend to justify/argue these points.... so why bring 'em up?

As you can see software development is *still* an industry very much in its infancy!... We have a lot to learn, lots of new things to discover.. and perhaps most importantly a profession to define.

Its perhaps one of the few areas where bright people can make a significant impression on the world, by themselves, and with little to restrain them... Draw the parallels if you will with the industrial revolution in the later part of the 1800's... You really can be a Stephenson or a Brunel... its the birth of a communication and knowledge industry and therefore a time of revolution, explosion and invention....

I admire the attempts to create the governance, and as with other elements of our industry, its early days. You really do have to break eggs to make omlettes! And from that perspective I really everyone to view new initiatives as exciting challenges and experiences. We can't predict the outcome, so we must pursue it...

But..... and I really do mean a big but......

If there's one thing critical to the success of software, the one thing surely guarantees failure in its absence, and thats people, good quality people, visionary people and people with the desire to move it forward.

No matter what happens going forward, no-one should under-estimate the effect of loss of 'intellectual capital'.

Many people contribute to this community in one way or another. But I don't think I have to spell out the fact that there are several key individuals here, who, if lost from the general thrust of the community, would spell its decline and ultimately its demise.

I'm no millionaire, and I'm in no place to claim to be a successful businessman... but every time I get close to success, every delivery thats a 'good one', and every project thats succeeded has always been down to a few key players. A few special, motivated people who've moved it from mediocre to maginificent.

And greatness stands on the shoulders of giants. I, my company and my professional salvation (how poncy is that?) I can largely attribute to Mr. Greenspun... I'm not saying I agree with the guy, I'm not saying he didn't make the most collosal commercial f**k up, and I'm not even saying that the software he initiated was originally all tha good.... but when he was gone, so was aD....

My previous company suffered the same fate. It's success swung around one or two individuals.. without them its an empty shell.

So do I have some sort of conclusion? Yes.

dotLRN needs governance. However it needs governance that includes the right people and does not disenfrachise. It doesn't matter, it *really* doesn't matter whether either form proposed is right or wrong, the point is it is not now inclusive.

And I don't see it moving forward without Ben. Not because I agree with his approach, not because the existing approach is wrong, but ultimately because its the people, not the system, that count. And nothing that governance can provide (I avoid the word 'create') will compensate for that.

My personal interpretation of whats been written (and I can only base it on that) is that there is far too much resistance, far too much intertia and far too much stubborness, both sides :(

For example, Bens somewhat development-centric view of this is far too narrow, and seemingly inflexible.
The Executive Boards approach is too much, to quickly and in one step. Much is made of the fact that they 'financed' the product. But even I can see that by agreeing up-front to the open source nature of the development, that argument holds no weight. You knew in advance you were making this kindof 'public' property, but seem to now fear what the 'public' might do with it? This risks stifling the creativety and freedom that creates that kind of distinctive open source software.

Its this simple.... no-ones got it right yet. If the scheme can't appeal to the most involved parties, then its not workable.. its got to be back to the drawing board.

Ultimately of course it doesn't matter to me, due to my limited interest in dotLRN. But is suspect it matters *a lot* to several others.. A solution should be found.

The issue or Ben or Don or MIT/Sloan.. might irritate you, but the solution is not to say 'well we offered our terms and he wouldn't accept'. Even if Ben is mis-guided (not suggesting he is) you do need a solution that includes him. If you haven't got one, then you need to keep trying till you have. Yes dotLRN has a window of opportunity, and it won't last forever. But it certainly won't succeed without the software! No amount of academic committees and rules will acheive that. Its better I feel, to keep the project moving and, until a satisfactory system can be created, leave the governance where it still is... on the drawing board. I am one of the strongest supporters of governance, and coordinated effort, but not at the expense of the only real resource... people.

Quick aside. Ben, I'd be very, very interested in understanding your view of what open source principles really are, and how open source works best. This issue vexes me a lot, and I think its fundemental to the success of my business (I expect its the same for you). I'd be keen to know your thoughts, but mail me privately, don't won't to open up another grand debate ;)

Posted by Jeff Davis on
Maybe we should ban <b> tags. I can't take all the shouting.
Posted by defunct defunct on
I think you'll find that CAPS implies shouting... Bold, I would argue,implies emphasis ;)
Posted by Michael Feldstein on

What is this thread about? The subject line is "continuation of dotLRN Governance." If this is truly about the governance structure then let's talk specifics. Let's talk about the details of what is or isn't "disenfranchising." The community will have to decide whether they are satisfied with the discussion of this topic on the previous thread or want to continue it.

If, on the other hand, this thread is about Open Force announcing their decision about the role they have decided to take (or not take) in the dotLRN governance, that's fine too. OF has every right to take whatever stand that they feel is consistent with their own sense of ethics, values, and self-interest. I had hoped for a different resolution but I respect their right to choose.

However, I think there is a dangerous trend developing here where a number of people on both sides of the governance debate seem to feel that they have to defend their honor and reputation. They have done so by responding to unspecified rumors and by publishing fragments of private email conversations. This can only be bad for the community. The fact is that most participants in OpenACS have shown themselves to be smart enough to judge people more by their actions than by their words. We all have to have a certain amount of faith that people will see through any rumors, falsehoods, and half-truths over time.

The last thread on this topic made it amply clear that many of us have passionate views--and passionate disagreements--on the governance topic. Such disagreements are healthy. As Thomas Jefferson wrote in defense of the participants of the Shay's Rebellion, "The tree of liberty must be refreshed from time to time with the blood of patriots and tyrants. It is our natural manure." Passionate disagreements are what democratic community is all about. But let's keep our passions focused where they can be productive.

On the topic of consensus, while I, like Simon (and like most everybody else, I think), feel that OF has been critical to the success of dotLRN, I recognize that we all have to do what we must. It goes without saying that we all feel grieved at this potential schism.

That's pretty much all I have to say about OF's decision or about the issues of integrity. I have more to say about the specifics of the governance plan but will hold my tongue until it becomes clear that governance is, in fact, what this conversation is about.

Posted by Alfred Essa on
Governance or no Governance, there is work to be done. We plan to focus our energies towards a stable release of dotLRN, coordinating various development efforts, developing documentation, and finding additional funding. Anyone who wishes to participate can. Anyone who wishes to contribute can. We are moving on.
Posted by Malte Sussdorff on
Hi Al,

thanks for this posting. I definitly think we need to focus on the product and how to get it out, fast.

Just accept Ben's decision, however it was motivated, but continue to listen to his advise and if you don't think it is valid tell him accordingly and why. If Ben (but not only him) thinks his advise is not heard, state your opinion on the bboard. This way the community has a way to see how the TAB makes up it's mind and why, if, they reject proposals.

And last but not least, maybe I'm wrong, but is "I will not participate in the governance effort" equal to "I will only contribute to dotLRN if paid for it {by the governance board}".

Looking forward to work with the EB on marketing issues and help fix a couple of bugs once we get our sites over to dotLRN / OACS.

Posted by defunct defunct on
Governance or no governance, we are moving on...'

Al, does this mean the governance issue is being sidelined/developed in parallel, whilst development continues? i.e. are you going therefore to allow more time to get the whole issue sorted to 'everyone's' satisfaction?

The reason I ask is that as you know I feel something along these lines is needed for OACS as well.. however, I suspect the outcomes of the dotLRN effort will have a dramtic impact on what and who would be accepted into such an effort for OACS...

Posted by Carl Robert Blesius on


Thank you for your contributions to this discussion.

We in Heidelberg felt we had to go to Boston earlier this year because there were too many unknowns. dotLRN governance was only one point of many on my list discussed on an afternoon filled with meetings (where we met with Ben as well). What we briefly talked about that day was a start (I wish Al could have been present on that day). What was clear to me is that certain things had been neglected, because of slipping deadlines. Transparency was another very important point on my list (and in my opinion we would not been at this point right now if both OF and Sloan had put a little more effort in this department). There is still a lot of work to be done on BOTH of these points. I think everyone realizes there might be problems in the technicalities of the present governance document... and we take Ben's and the communities fears and comments very seriously. Another thing should also be clear though... we are dealing with a gray area here because dotLRN as it now stands is not the result of a volunteer effort (although it is clear how important volunteer work is) and OF is not a contributing individual but a company that was hired with dotLRN in mind.

Like MIT vendor independence is very important for us and like Ben I think us making a "huge issue" out of OF turning down participation is a red flag that we must carefully evaluate.

There are still a lot of unknowns... but all documents up to now are in the public sphere and almost all of the discussion has been pulled into the open now (for better or worse) and there is a lot to digest for everyone right now. I totally agree with Al... governance or no governance there is work to be done and transparency is something that we are going to push ASAP.

Arjun (and others at OpenForce),

I realize that OpenForce is much more than just Ben and I am sorry that we haven't had a chance to speak more with you all on this. Thank you for making it clear that Ben has acted as spokesperson. It makes sense to me that you put a lot of thought into this decision as a company (I am sure it was difficult). I also realize that you all have helped make what dotLRN is (for _us_ all) beyond what SloanSpace is. It is WELL known that you (OF) are respected for being VERY good architects and I think Al made a good choice in selecting you all to lay out the dotLRN framework (independent of any deadlines).


I respect your strong will and adherence to your principles. My original intention was not to attack or alienate, please accept my public apologies if it came over like that. I have put a lot of time into trying to understand what OF is trying to achieve/communicate. I think you have made it very clear that you are very strongly dedicated to an open-source cause. If you do not feel you can take part in the governance of dotLRN because it doesn't fit into your view of what open-source is, I can respect that. I was not attempting to alienate you (that would be even more detrimental to dotLRN than your choice not to take part in the governance).

I have seen an honest and pure attempt from MIT/Sloan to promote a mix that leaves room for a much wider dotLRN community representation, without compromising the technical direction (another one of our core disagreements that I do not want to reopen). Once again I can respect the fact that you do not agree and I think you have made it clear that your decisions are based on an honest dedication to the OpenForce cause.

I was not expecting you to compromise your principles Ben, but it was a surprise how you have chosen to protest. There are a lot more shades of grey in your very black and white idealistic picture (we have also talked about this at length).

There are a couple of reasons I think dotLRN has solid footing and one of the main reasons is because dotLRN is clearly a subset of OpenACS (a very organic open source project that permeates upward into dotLRN and makes it what it is, adding its own flavor of open-source stability). Another thing that is important are the ideas and energy that have and will continue to flow into dotLRN from in and around MIT (MIT has a history of disruptive ideas with practical results). Due to the dismal E-Learning landscape where a monopolistic situation is slowly taking shape dotLRN is poised to make a big splash and I strongly believe that dotLRN will quickly gain ground in Europe (in addition to the states) and I am looking forward to watching this happen.


P.S. I am going to be dropping out of this discussion for a bit and I also totally understand if Al does not respond much on this for a while either.

Posted by Ben Adida on
Malte: no, for OF and me, this is not an issue of having the EB pay for those
positions. It's conceivable that the EB would want to do that given the huge
workload that the TAB will be given in the current plan, but that is not a factor
that would change OpenForce's approach: we've been offered to compromise
on open-source principles in the past in exchange for money or power, and
we've always turned those offers down.

Carl: I appreciate and accept your apology. It seems we agree that we
disagree on certain crucial points, and that's really all there is to it.

Posted by Alfred Essa on

I agree. Adherence to opensource means that technical decisions need to be based on merit and subject to community criticism and examination. The dotLRN Technical Board must and will take into account developer views. Their decisions have to be made in the "open" and subject to community review and endorsement at all stages. The Technical Board will also need to coordinate everything with OpenACS leaders and community. Again, anyone who wishes to participate or contribute will be able to do so. The primary role of the Executive Board will be to evangelize and secure funding.

Posted by Alfred Essa on

We are not shelving Governance. We are just being pragmatic. There is a lot of work to be done and we are focusing our sights on that. As they say, the proof is in the pudding.

Posted by Tom Mizukami on
On a purely pragmatic note - is anyone currently working on developer documentation? Is OpenForce going to complete this? What is the short term plan to keep this moving forward? Thanks.
Posted by Andrew Grumet on
I very much hope that OpenForce will provide documentation. With all due respect for this company's contributions to date, I think Arjun was a bit hasty when he declared victory (see "the future won" comment above). As I'm sure we all recognize, good design is a necessary but not sufficient condition for success.

This thread suggests that documentation is in-progress, but the current CVS snapshot does not contain any files ending in htm, html or xml. The best possible outcome would be for Ben and co. to complete this work and deliver to the community something they can be really proud of. If pride of authorship is not sufficient motivation, the consortium members should seriously consider whether a little cash might help the effort along. We all gotta eat.

A lesser but still positive outcome would be to provide copies of their docs to date---even if incomplete---so that the community can finish the job. These could be made available through the public CVS repository or even posted to /new-file-storage. I believe at one time a "dotlrndoc" package existed, but it seems to be gone. As a last resort, I can probably dig up a snapshot of these docs to provide a starting point.

Finally, as to the short-term plan. The push to release will henceforth be driven by Sloan in coordination with the TAB and community. We will be undertaking the technology transfer (CVS, installation docs etc) starting next week and will hopefully have a publishable schedule by late next week. Harass us (well, me anyway) if things don't get more concrete in the next couple of weeks.

Posted by Peter Marklund on
Andrew, I think documentation is an important point, the tradition of documenting packages from the ArsDigita days has been lost, not only in dotLRN but also in OpenACS in general.

A point that I would like to see considered by the Technical Advisory Board and the Executive Board is to merge bug fixes and improvements done in the MIT and Berklee trees (and at other universities - Missisippi, Vienna...) into the official tree. Also, and maybe more importantly, is there any compelling reason why MIT couldn't use official code base? The code base would gain much more credibility that way. Andrew?

Posted by Andrew Grumet on
Also, and maybe more importantly, is there any compelling reason why MIT couldn't use official code base? The code base would gain much more credibility that way. Andrew?
I'm not sure exactly what you mean, but perhaps my comments will shed some light. During the weeks and months leading up to the Sloanspace launch---and I believe this is still true---running dotLRN required us to be at the HEAD of the OpenACS tree. So there was no "official" code base to build on. My preference is to reach a point where we can say something like, "dotLRN 1.0 will run on OACS 4.6 or better". I have yet to clear this with others, but in Cambridge we've been talking about scheduling the release for a couple of weeks after the OACS release, so that we can adequately test dotLRN against 4.6.
Posted by Peter Marklund on
Thanks Andrew! So both the Sloan system and .LRN run off of the latest CVS sources of OpenACS, that's great. What I was referring to was that the Sloan system is, and correct me if I'm wrong, not based on the latest CVS sources of .LRN (

Would it be possible to run the Sloan system off of the official .LRN sources? How big is the diff?

Posted by Andrew Grumet on
Sloanspace is most definitely "based on" the dotLRN CVS, but it is not an identical copy. Same goes for OpenACS. We're all waiting for the day when you can build a site solely through configuration without the need for editing distribution files. Today that approach is impractical for Sloanspace. So for now we are importing OpenACS and dotLRN as CVS vendor branches (this document describes how we do it). It is working quite well. We've merged 7 times now
symbolic names [for readme.txt]:
        dotLRN: 1.1.1
Each Head represents a merge of the latest code. Or next merge will be against the 1.0 release.

How big is the diff? It's manageable, at least, or we wouldn't have been able to merge so many times. But this is a great question for Caroline to answer. What I can say is that it is very much in our interest to take merge the non-Sloanspace-specific improvements and bugfixes (here is a particularly cool bug) back into dotLRN. As I type, we are working with OpenACS team to do this for OpenACS. The smaller the diff, the happier we are.

Posted by Andrew Grumet on
With regard to my "things getting more concrete" comment above, please see this thread, which is the first step in that process.