Forum OpenACS Q&A: Priorities, Roles, and the future of OpenACS
First off I wanted to start a new thread that deals with many issues spread across a few previous threads and I want to start off by saying that while I have not directly contributed much to OpenACS I plan to in the future and would like to dedicate this thread to Jerry Asher whom I think has been a valuable resource to the community.
First I think its important as the 4.0 release draws near that the key porters not be distracted from the main job of releasing a solid codebase asap. Any help that the community can give to the key porters should be given if possible. While this is true, there are many members of the community that would like to help but are unsure of how, or their capabilities in regard to chipping in.
This brings me to the topic of roles. Currently the main roles are the actually coders/porters. While this is a key focus now as the community grows their will be many roles that will be filled.
- Site Maintainers
- Graphics contributors
- Reference Article / Documentation writers
- New Module Developers
- Compatible Technology Integrators
- General Users / Evangelists
- External Article, Marketing Writers
- Steering committee ???
- Other roles I'm sure
While the porters are key members now its important that we encourage the community to take up roles that they can and help where they may. I think the mention of a steering committee and some sort of delegation and ease of community picking up the tasks that they are interested in are desired by the community as a whole. I even imagine somewhere down the line where community members could have as part of their profile what they envision their current status/tasks are that they have set themselves. Sometimes members may go off and explore something that they can bring back to the community for consumption and possible integration. While OpenACS is still quite young if/when it does take off the community can expect lotsa newbies looking to chip in and if the barrier to becoming a useful member is lower all the better.
I myself have submitted an article that Ben said will be posted somewhere in future somewhere on OpenACS on profiling and have other articles in the works. I have a web site that should be live soon and I am going to make OpenACS ad part of my rotation. I gave a presentation on AOLserver at my local LUG and am already signed up to do one on OpenACS. I have posted some thoughts that I thought might be of benefit to those porting but am not really one of the porters. While I dont see myself porting or making any modules I do plan on being an active member of the community through other roles and even consider myself one now but the community may not realize it.
If their were a tasks section / todo? that included not just coding issues but some of these other areas and allowed people to add their own projects (maybe even just currently evaluating software) I think it would add some organization and insight into who's doing what and give the fringe members a chance to pitch in if they see for instance that there needs to be a PR announce written for LinuxToday.
I think OpenACS will benefit if it addresses these issues and treats its lesser members well because the shoulders of many can carry much more than that of a few.... Just my thoughts ... once again thanks for all your work Jerry including on virtual hosting. This project wouldnt have come about without Don and Ben and I hope they remain the cornerstone of the community, but any cool building is made with more bricks than cornerstones.
Well said Carl. Like you, I am a new member of this community. After months of getting to know OpenACS and all its components I feel that I'm reaching a point where I could start contributing.
However I still have a learning curve ahead of me before I can start contributing packages. Working with the newly ported packages though, I noticed that there are several bugs left that require attention. Without the skills to fix them myself I would like to report them so that the porter and other community members can resolve them.
The SDM seems a logical place to report them but doesn't look like it is used by others much. Most bugs haven't been assigned to anyone and it is therefor hard to see they have been worked on. The bulleting board appears to me as the wrong platform to raise these bugs. What seems to be missing is way to channel feedback on the modules to porters and coders.
The tools are there, either the current SDM or the new ticket-tracker would do fine. It is the process that is missing. A simple outline of a process could be a big improvement. Maybe a 'How to give feedback' FAQ?
picking up the tasks that they are interested in are desired by the community as a whole.</i></blockquote>
How is this anything short of replacing the current leadership of the OpenACS project with new leadership, i.e. this steering committee?
<p>If the community as a whole is looking for this, then presumably the community as a whole wants the current leadership to step down.
<p>I know, I'm raising a strawman again. Forming a steering committee and giving it control over the OpenACS project isn't the same as asking the current leadership to cede control over OpenACS project.
<p>At least, that's what Ben and I are told whenever we say that, sure, we'll step down if that's what folks want.
<p>I honestly do not see how putting control of the OpenACS project into the hands of a steering committee isn't the same as kicking Ben and Don out of the saddle, though. Could you please explain the difference to me?
<p>If you have a definition that is truly different, are you sure that
others who favor this step share it?
<p>Don't get me wrong, if the community as a whole does want a leadership change, I won't stand in the way. I should hope I made that clear in the thread which led you to dedicate this thread to Jerry Asher.
<p>But if that's what folks want, don't beat around the bush. Let's talk about it openly with no attempt to sugar-coat the real meaning of
such a change.
current leadership into something more official? It wouldn't make
any sense to choose different people than the current leadership for
those jobs. The current leadership is already performing that
function and a steering committee would just be a formalization of
No, that's not been my impression at all. For starters, the current leadership isn't asking for formalization of this sort, we are more than happy with the current informal situation. If your statement were true, if nothing else you'd expect the current leadership to know this, wouldn't you?
If you're right, it would be nice if those making the proposal would say so. I would then feel even more strongly that a steering committee's not necessary, though. I want to see things remain lean and mean for now.
If or when the project reaches the point where it makes sense to form a foundation or don some other legal costume, we'll have to adopt a certain level of formal structure, of course.
steering committee or leadership list of some sort.
Right now I am vaguely aware of who is leadership and who is not.
The team page says basically "a bunch of people work on stuff". I
want to see a page that says who heads up what projects and so
forth. If such a page exists I want a link to it from the
It is almost certainly the case that having a more formal organization would involve asking you and Ben to have less control over certain OpenACS outcomes. It's not necessarily the case, but it almost certainly is the case. But no one is asking for either of you to step down from leadership roles in the projects that you are interested in, and no one knows what less control will really mean.
There are questions then: what good would that be to the project? What good would that be for Don and Ben? What form of organization would be formed? What would participation look like?
I can make the arguments as to the benefits to the project and community. I truly believe there are wonderful benefits in it for you and Ben, but that will be for you to decide. I believe it is up to the community to determine the form of the organization and what participation would look like.
We've been around this loop several times these past few weeks. It may have started here http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0001us. And then at Ben's suggestion, I started a thread http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=00025b to discuss that. I think that thread got me labeled as revolutionary terrorist. The meme then spread http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=00025d.
Not much came out those loops aside from my getting tagged as FAQ admin.
I encourage you to (re)read those threads once more. The basic tenet made by many of us is that a steering committee will let us reach out further and farther than we could with an organization of just two dedicated leaders alone.
Here is something (abridged and enhanced) I wrote to another one of us in a private email:
What a steering committee can do is organize initiatives, and focus efforts. As I have said before, it would have been wonderful if there had been some formal xml-rpc/soap initiative formed so that Hafeez, Aaron, and Dave could have dialogued out the best interface and created enough community momentum to get that interface documented.I truly believe that asking for a steering committee is not the same as asking for the current leadership to step down. It's a statement in the success of the project. I would like to see a steering committee formed to discuss the site and the project and the best way to proceed. Something ala the Apache Software Foundation http://www.apache.org/foundation/FAQ.html. Something like mozilla.org http://www.mozilla.org/about.html. Something like jakarta has: http://jakarta.apache.org/site/roles.html.
I know you don't want anarchy. I know you want some control. I know you want the project to succeed. Yet you don't appear to want a steering committee with the usual acoutrements: bylaws, voting, etc. I assume I would be misreading you to think you are happy with a "benevolent dictator" approach then. That's the approach it appears we have at this time.
Anyway, I'll try two more tacts.
One: there are a lot of individuals and companies investing their time (read money) into the OpenACS. Investors, and stakeholders usually have some rights and often ask for some openness and some amount of control. When they don't get that, they start rethinking their investment decisions. I'm not happy that Ben and Don discussed and then dismissed wikis without consulting the community and especially without consulting the folks in the wiki threads. I'm not disagreeing with their decision. I'm not happy with the process and I am asking the process be changed. I would welcome a steering committee that makes such a process open. That determines voting rights and voting procedures. If I were a company evaluating OpenACS technology, I would be worried that my investment is run by a benevolent dictatorship, where processes are opaque. It's not closed source by any stretch of the imagination. But it's just not a serious or professional organization, and it's just not indicative of a thriving community.
Two: How do we as a community reward others in the community? Can we provide "career growth" to members? Many of the folks here could use a resume spiff or two. Yet http://openacs.org/about/team.adp is empty. It's one thing to write on your resume: ported this module, created that module for the OpenACS. But the OpenACS just doesn't have that much brand name recognition yet. It's something else to be able to say: Member of OpenACS steering committee, Secretary OpenACS Marketing Initiative, Founder OpenACS XML-RPC/SOAP Initiative, Head OpenACS W2K Group, etc. These wouldn't be empty vacuous roles. With a bit more formal organization and a steering comittee, these would be well focused projects that would directly benefit the OpenACS, create more news events and name recognition for the OpenACS, provide centrally located projects recruiting for community help, and reward some of our developers both with title as well as recognized, increased responsibility and authority.
But even if you believe it's really just the same as asking you two to step down, can you reread these threads I've pointed out and not see community and project benefits from formal organization?
Ben (I think?) wrote a document describe the roles "the gatekeepers' play when I was traveling across Canada for 3 1/2 weeks. It would give you a little more information, at least.
Unfortunately I can't find it! It certainly needs to be linked to from an obvious place.
At least I think I saw such a document...since I can't find it I'm starting to question my own memory!
As far as the OpenACS 4.x project goes, the status report is the best place to find out who has stepped forward to take responsibility for various packages.
We do need something that fleshes out the informal structure of this (dare I call it) organization for folks. It would be good place, perhaps, to flesh out areas where we need folks to step forward and tackle stuff that's being left undone. Some of the things on Carl's list are, in a sense, metatasks (not attached to any particular project), while some are project-oriented (4.x toolkit documentation is part of the 4.x project, 4.x testing part of the 4.x project, etc).
First off let me restate that in my orig post above I think the priority should be for the core porters/coders to focus on getting the next release out the door and anything that the community can do to help them should be. I don't wont the coders to really focus on anything besides coding right now including getting bogged down in other issues. Now let me try to restate what I was trying to put across but like usual am having trouble expressing myself
First off, in no way would I want Don or Ben to step down, I think they have done a fantastic job getting us to where we are today and I believe I publicly posted a thankyou to them almost 2 years ago you the help I had already gleaned from them. I posted a dedication for this post to Jerry because a) I hadn't earlier b) I felt you were not treating him fairly and I am glad that you seem to acknowledge your lossing your temper and giving him propsJerry's made significant contributions to AOLserver and I know he wants to see OpenACS be a great success.
While he may not be actively porting modules right now I do think he is trying to help in other ways. I also think its a good that there is control so not everyones ideas or work automatically becomes part of the core but it is effort and the community (I think) should always acknowledge / appreciate work made to better it even if it is not incorporated.
My definition of a steering committee was more like Postgres's version where there are contributors that heading up different portions of tasks to take burden off of coders for example today it might have Roberto in charge of Documentation, Tali Web Site, and Ben and Don Coding, Dan Oracle Specialist.
It already is kinda like that today I just think that as the coding aspect closes on this release there are alot of tasks that dont make sense wasting the coders time on. I think there needs to be a mechanism in place the eases the curve of all the lurkers and newbies being able to help out. Most wont be able to code but probably can fill other needs and these needs do need to be done. I am not too worried that decision are being made without community input. I just want an easier way for the next generation to help out, Im not too concerned about professionalism although others Im sure will disagree
I gotta love comments of this sort. There's something beautifully Stalinesque about the concept.
I've served on non-profit boards, steering committees, advisory committes. I've run a biology project for a non-profit. I've also run a software company with fifty employees, long, long ago. So I'm very familiar with various forms of organization and the benefits and drawbacks associated with them.
<p>As far as OpenACS goes, I'm not personally interested in working within such a framework at this time. I realize you're not suggesting we start a foundation right now. You're simply suggesting we start lugging around the baggage that comes with such formalism.
<p>That would be the end of our lean and mean project dream.
<p>We'll probably get there some day. When faced with the inevitable need to do so, I may even participate with enthusiasm. However, I do not think we're there yet. If folks insist upon a fundamental change in the OpenACS project of this sort today, such change will take place in my absence.
<p>Let us be absolutely clear on that point, OK? It will take place in my absence.
<P>Does this mean that Ben and Don imagine they can do everything the project needs? Of course not. We've both shown a willingness to delegate chunks of the overall project when folks step forward and say
"we want to help!" I've delegated damn near the entire OpenACS 4.x project to other folks, with the exception of updating the status report. I am about as far from "benevolent dictator" as you can imagine. I do a semi-competent job (by my standards) of organizing the project, I don't rule it. Ben's delegated the openacs.org website design to Musea.
The fact that we don't respond positively to folks
who step forward and say "we want to take over!" shouldn't be read as read as an unwillingness to respond positively to folks who step forward and say "we want to help!". And I'm tired of hearing folks claim "we want to take over because you won't let us help!".
<p>It is nice to have the fact that dilution of the amount of control Ben and I have over the OpenACS project is, indeed, one of the reasons you wish to see the formation of a steering committee, answerable to a voting public, formed. It's on the table, now. This should help focus the discussion.
<p>I also suppose that, if asked, you'd be eager to serve on the steering committee? You're not thinking, by any chance, that diluting Don and Ben's role might be a way for you to take partial control over the project, are you?
<p>This is a time for absolute honesty, Jerry. The community really deserves to know exactly why you're fighting so hard to force a structural change in a project that many seem to think has been progressing nicely, from technical and organizational points of view.
<p>I'll be honest with you, Jerry. I don't believe you're trying to force change and to yank OpenACS away from us for my benefit or Ben's benefit, despite
the quotation with which I headed this response. I think you're motivated by self-interest and a desire for some degree of personal control over the project.
because a) I hadn't earlier b) I felt you were not treating him fairly </i></blockquote>
Well ... look, you may not realize it, but there's a coup attempt afoot though Jerry will, of course, shy away from that label. Yes, I lost my temper with Jerry, but the reason I lost my temper with Jerry is because I don't feel he's been playing straight with the community.
<p>So instead of my getting angry with Jerry, how about I just tell you what he really means, at least until he feels comfortable telling us himself?
<p>I don't think Jerry wants personal control over the project. However, he does want to transform OpenACS into something that he can build his consulting business on, and his vision as to what needs to happen in order for him to be able to do so isn't congruent with my personal vision of how I'd like to see the OpenACS project and community grow. He wants to be one of those who control the project, and I'm quite certain would feel most comfortable if the others controlling the project shared his vision as to what this project should be. Since our visions clash it is hard for me to believe that he'll everfeel comfortable if I'm in a position to make my vision dominant, so I'm certain that he wants my role weakened to the point where it won't be.
<p>I should probably say "our vision" since Ben and I hold such similar views, but as usual I don't feel comfortable speaking for Ben when he's not around to pitch in with comments on his own.
<p>Jerry won't say so directly, but I wish he would. I don't feel there's anything wrong with Jerry looking at the OpenACS effort as being a business opportunity, just as there's nothing wrong with my motivation being somewhat different.
<p>Ironically, I don't think my vision precludes being able to make a living doing OpenACS work. Quite the opposite, actually.
<p>I know that Ben and others found the earlier rounds of discussion which took place when I was on vacation to be time consuming, disturbing, and distracting, too.
<p>So, let's not drag this out. Let's not posture. Let's not play coy. Let's lay 'em on the table and get this over with. If I'm out of a job give me my last cigarette, blindfold me, and pull the friggin' trigger.
<p>Jerry, if you lose in your effort to gain some control over the project, are you willing to work with us? I think folks deserve to know.
I understand Jerry's proposition because I advocated it originally. I no longer feel that way for a few reasons:
* I had some private exchanges with Ben and others that made me convinced me reasonably otherwise.
* I saw some of the ugly fights that were taking place on the Jabber mailing list, ones that made these discussions seem civil.
* I studied some of the other open source projects and realized that what drove these applications probably weren't their beuracracy but the money companies hoped to make off of them.
* Ben, Don, Roberto and Dan admitted their leadership positions.
I am personally comfortable with these four guys as the leaders of the community. They have invested more time and energy in making sure that the software worked, there was a space for people to discuss how to use it and organized a massive effort to port a few hundred thousand lines of code to another DB. Also, they built and rolled out new applications to make OpenACS a much better product than what aD abandoned.
For the foreseeable future, I think these guys will continue to do a great job. Ben has evangelized the OpenACS longer and better than anyone else, Roberto has done yeoman's work building the docs, Dan fixes bugs within an hour of their submission and Don has led a tremendous technical effort, one that probably rivals anything else in the OSS field. I commend their leadership and I accept their positions delightfully.
I admit that my initial advocation for structure was due in part to my desire for seeing myself and Musea be recognized as leaders in the community. But I also just wanted to see *someone* established as the leaders so we would now at least where the decisions were being made. Like I said, I'm comfortable with that stuff.
I volunteered Musea to rebuild openacs.org for the similar reasons. We want to be recognized as another leader in the community and in order to convince clients of how OpenACS works we can't send them to the current embarrassment (I think it's fair to call what we have pretty embarrassing considering the power of our software). I think that if and when we fulfill our promise it will be reasonable for us to ask for more decision making power in the direction of the community (not that we don't have it now) since we're willing to sacrifice significant time, energy and money to not just benefit ourselves but everyone who currently and in the future use OpenACS.
If there is anyone who should feel similarly, I think it's Jerry. I do believe that Jerry deserves some recognition for his work as well. He has done a great deal of work integrating search into OpenACS. I don't think it matters whether he did it with 3 or 4. 3 is still a living application and he must have learned a great deal about htdig that will be applicable to when it's made available for 4.
I personally find OpenFTS pretty intimidating. I don't know what the hell a lexum is and it doesn't help that a site it's only used for Russian sites right now. I have implemented htdig on a project and it works great. So I like to see the work he's doing.
Don also has a point since Jerry hasn't made his work totally official and has kept all the demos of his work on his site. That's probably not such a big deal considering he's made it obvious that his site exists, that his work is not done and eventually it will all be folded back.
So I'm kinda neutral in this argument. I think we're currently fine how we are, but that the critical mass for a True Community may come soon. In the meantime, these arguments are distracting and potentially divisive. Also, I don't like new or potential members of the community getting scared off by community heavy weights taking swings at each other. It's healthy behavior, IMO, but only if it's just some steam that has to be blown off.
Also, it's not worth saying the community sucks because there's nothing can be found on the website. Everybody's admitted that and we're going to fix it. So that should be a dead issue.
Talli has hit upon the path I favor, though:
I volunteered Musea to rebuild openacs.org for the similar reasons. We want to be recognized as another leader in the community and in order to
convince clients of how OpenACS works we can't send them to the current embarrassment (I think it's fair to call what we have pretty embarrassing
considering the power of our software). I think that if and when we fulfill our promise it will be reasonable for us to ask for more decision making
power in the direction of the community (not that we don't have it now) since we're willing to sacrifice significant time, energy and money to not just
benefit ourselves but everyone who currently and in the future use OpenACS. </i></blockquote>
In other words, I like the "one earns one's leadership position" model, myself.
<p>Yes, Talli, Jerry's done a lot of OpenACS-related work, though as of yet I can't call it OpenACS work because he's not working with us.
<p> Ironically, if he had worked with us from the beginning and hadn't started down this path of trying to yank partial control of the project into his own hands by circuitous efforts to formalize the organization, I'd undoubtably be trying to delegate so much stuff in his direction that he wouldn't be able to keep up.
<p>Let me emphasize something I said in one of my earlier posts:
<p>I do believe the time will come when we will have to formalize the structure of OpenACS. It's not the time and frankly, one doesn't do it in order to settle issues of leadership but rather one does it to enable the organization to accept money, own copyrights, buy servers, etc. It makes organizational life harder, not easier - but at the same time it makes certain things possible that are legally impossible if you don't take this step.
<p>I hope we get there and, when we do, I'll be the among the first to suggest it. Actually, I have once or twice in the past, privately, in the context of suggesting that a certain software company near Central Square ought to partially fund our project.
<p>So I'm not opposed to formalism per se. I'm just opposed to doing it in order to dilute my role in the organization. If that motive were taken off the table, I'd still argue that this is not the time, but that's a different sort of discussion.
Not everyone who has spoken in favor of more formalism has as a motive the dilutio of my influence in the organization, and some of these folks haven't understood that this is one of Jerry's motivations for pushing the issue.
<p>So, once again ... I'm glad everything's out in the open and on the table.
I'm going to attempt to ignore these discussion forums this weekend. I will re-read the many postings, read the personal emails I've received, and think about how to resolve this issue that doesn't seem to have a solution for now.
For the record, yes, I agree with Don that these discussions have wasted tremendous amounts of time and produced absolutely zero results to date. Maybe we've somehow invested tons of time into something that will suddenly pay off big. But somehow I doubt it. Sigh. Signing out for the night.
<blockquote><i>My definition of a steering committee was more like Postgres's version where there are contributors that heading up different portions of tasks </i></blockquote>
The Postgres Steering Committee grew in exactly the same fashion as the current "group of four gatekeepers" have grown in this community.
They're the people who started the (modern incarnation of the) project, who provided resources as well as wrote code. The reason you find a sysadmin on the steering committe is due to this history, i.e. he was the guy who got them the physical machine resources for their website, CVS tree, etc. This happened several years ago when you couldn't start up such a project on a $500 PC and a $30/month DSL line, but rather needed real money or resources.
<p>So it may surprise you that *my* model for the evolution of the OpenACS organization is similar to that model.
<p>We could fit that model by changing the name "gatekeepers" to "steering committee".
<p>The Postgres model is a very lightweight model. They're not weighed down with anything near the kind of organizational and formalistic baggage that Jerry's proposing.
<p>And, the steering committee isn't elected, it chooses itself. If we adopted the Postgres model, it wouldn't dilute my influence, it would strengthen it and make moot any decisions regarding it.
<p>That's certainly not Jerry's goal.
<p>Also, let me quote Jerry directly:
<blockquote><i>One: there are a lot of individuals and companies investing their time (read money) into the OpenACS. Investors, and stakeholders usually
have some rights and often ask for some openness and some amount of control. When they don't get that, they start rethinking their
If you like the Postgres model, you shouldn't like the implications of this statement.
<p>For the record, both GreatBridge and PostgreSQL, Inc have *explicitly* vowed to not try to influence the direction of Postgres in ways that run counter to the community's desires simply to maximize their chances for financial success. Jerry suggests something different, i.e. that companies that choose to use OpenACS should expect some level of control of the project.
that's an awful lot of energy and effort that could have gone into
porting, testing or documenting instead...
I've tried to stay out of all of these various discussions for a
number of reasons. Mainly I feel this is the wrong time to be
talking about all of these potential changes. Let the porting folks
finish their work in peace, and *then* it will be time to discuss
what worked and didn't work about the process this time and
what could be different next time.
However, since the topic seemingly just won't die, I'll throw in my
two cents for the record.
Most everyone knows that Don works for furfly, and so I can be
expected to side with him on most things. But I'll freely admit that
he's not always right (shocking, eh? :) and that he does
occasionally have a bit of trouble holding back his temper.
Some of his posts on these recent threads have been a bit over
the top, for sure.
Now, hopefully having established my impartiality... I would *far*
rather have the benevolent dictatorship we have now than the
bureacracy Jerry is proposing. I don't always agree with every
decision being made, and I do think that it could be made easier
for new folks to learn how to contribute (though I reiterate that at
this stage of the project I believe that making major changes is
likely to be more harmful than helpful). But I know that decisions
will be made, that they will be more or less consistent, and that a
quality product will be produced. Far more so than if more
people and more bureacracy was added to the process.
Furthermore, I see even less benefit to adding lots of "named"
jobs just so folks can spiff up their resumes...
It is perhaps true that, as Jerry says, "Investors, and
stakeholders usually have some rights and often ask for some
openness and some amount of control". But I think that is a
false analogy. Investors hand over money, and lots of it, before
they can consider themselves owed any degree of control, and
once they have done so their money is effectively imprisioned
until such time as it can be repaid with the desired amount of
growth. So they have a right to expect to participate in the
running of the company, since that will (theoretically) allow them
to get their money back sooner. OpenACS volunteers give their
time, and are free to put in as much or as little as they like, or to
stop volunteering at any point. Presumably they choose a task
which they want to do, and will do to both their and the "gang of
four"'s satisfaction. There is no further obligation, and therefore
IMHO no reasonable expectation of control beyond the task they
are responsible for.
Also, speaking as President/CEO of a company which does
nearly all it's business with the ACS, and plans to adopt
OpenACS 4 for all new business as soon as it is ready, I can
honestly say that I feel no need for the additional control Jerry is
asking for (and no, I have never asked Don to do anything for
furfly in his capacity as project manager). All the OpenACS
project "needs" to provide me with is a solid, well-tested core
and as many modules as there were volunteers to port. We can
take it from there. There are many other things which are
desirable, of course, such as ways to report bugs, fix bugs,
contribute new modules, etc, and I believe we will get many of
those things. But I'm not going to waste much energy
complaining if I can't convince anyone to set up a place to post
marketing materials, for example. I can set it up myself,
publicize it here, and if it becomes popular then I bet I'll get a
better reception next time I ask about having it included at
openacs.org (this is just an example, not a real-life decision).
I will join most of the people who have posted to these threads in
recognizing that Jerry has contributed a great deal to the
community. I use his virtual hosting patches myself. But IMHO
he has also contributed a lot of diversions and divisive
discussions. He seems to like to stir things up, in ways which
often appear to be more about causing trouble and gaining
followers than about making positive and practical changes.
Jerry, you have brought up basically the same control issues at
arsdigita.com, on the AOLserver list, and now at openacs.org. It
seems that no matter where you go, your ideas are either
ignored or actively shunned by those in power. Perhaps it is
time to stop feeling that our fearless leaders are persecuting
you, and to start looking at how you might present your ideas
more effectively? Tiltiing at windmills wastes a lot of time and
energy and doesn't really get you anywhere...
where you go, your ideas are either ignored or actively shunned by those in power. Perhaps it is time to stop feeling that our fearless leaders are
persecuting you, and to start looking at how you might present your ideas more effectively? Tiltiing at windmills wastes a lot of time and energy and
doesn't really get you anywhere... </i></blockquote>
Personally I think that it's time for Jerry to start his own, independent project somewhere, someplace, dealing with some issue or some software package or another.
Then he'd have the control he wants. I can't imagine myself *ever* walking into a project of his and demanding partial control in the way that he's walked into OpenACS and demands partial control.
<p>What kind of person would do that? Not me.
Any large scale development process will encounter conflicts which must be resolved. Often resolution is an arbitrary decision in order to further progress the project. In commercial teams, the corporate hierarchy + performance review structure solves this problem -- How do OSS teams resolve them?That's what the infamous Halloween Memo has to say on the subject: http://www.scripting.com/misc/halloweenMemo.html.
In the case of Linux, Linus Torvalds is the undisputed `leader' of the project. He's delegated large components (e.g. networking, device drivers, etc.) to several of his trusted "lieutenants' who further de-facto delegate to a handful of "area" owners (e.g. LAN drivers).
Other organizations are described by Eric Raymond: http://earthspace.net/~esr/writings/homesteading/homesteading-15.html:Some very large projects discard the `benevolent dictator' model entirely. One way to do this is turn the co-developers into a voting committee (as with Apache). Another is rotating dictatorship, in which control is occasionally passed from one member to another within a circle of senior co-developers (the Perl developers organize themselves this way).
Currently, we've followed the "those that do and have done so the longest share the responsibilities".
We don't seem to be out of the realm of normalcy for open source projects.
Any of the examples mentioned above would exclude Jerry entirely. I suspect that very few open source projects would put up with this crap nearly as long as we have...
Petru Paler for President![ Subject to finishing the AolServer/FastCGI interface of course ]
The ongoing talk about a new and better organization sounds really familiar to me and I feel very comfortable that these kind of fights are similarly fought out all over the world and across generations
The funny thing is that while being a member of the student-organization I was fighting for MORE TRANSPARENCY all the time. Then some friends and I wanted to be VERY SUCCESSFUL with an internet-start-up, which didn't really work out. But one of the reasons it didn't, as I heard from neutral sources, was ironically my inability to be transparent!!!
I still like the idea of transparency and maybe we should spend some hours to get a real project plan up an running, which is listing all the people that are currently working on different topics that help the community to procede.
President Petru would be listed with his fast-cgi project
Talli would be listed with his openacs.org face-lift project
Jerry (if he will still be with us after this pretty tough face-punches) with his search-project
Additionally every member that is listed with a project would have the opportunity to edit his project information by himself and post additional reading material concerning his research and progress.
and so on and so on....
Maybe this could be a valuable module for openacs 4.x ??!!
In the end:
"When you point a finger at someone else, remember that three fingers point back at you."
pretty tough face-punches) with his search-project </i></blockquote>
I think you're missing one very important point. Jerry's search project isn't part of OpenACS, it is an independent project that just happens to run with OpenACS.
Jerry won't make his search project part of OpenACS unless he succeeds in grabbing partial control over the organization.
<p>Thus far Petru, Talli and others have been more than willing to work within the existing OpenACS framework. They're not trying to take over openacs.org as a precondition for contributing to the project.
<blockquote><i>I still like the idea of transparency and maybe we should spend some hours to get a real project plan up an running, which is listing all the people that are
currently working on different topics that help the community to procede. </i></blockquote>
I'd love to have some time to work on this and other project issues. Currently, though, the organization is paralyzed with this fight over who gets to run the project, those of us who started it or a new set of people.
Well I really think that the community should go back to work.
Maybe we implement a feature for bboard posts that cuts down on the amount of words to be written on controversial topics There is people that like to hear themselves talk and there is obviously people that like to see themselves write. I mean that a lot of the posts could be a.) a lot shorter (less time consuming) b.) not repetitive - we are not in a novel contest here!
I'm already emotionally drained by this. It's going to probably take me a week or so to recharge my batteries and get back to work on OpenACS 4.x issues in any reasonable way, assuming this ever comes to an end.
I imagine the folks at aD are deeply enjoying this display of community dysfunction, given all the criticism that I and others have hurled their way over the years.
Here are my thoughts for what they are worth:
- The current leadership group of Don, Ben, Dan and Roberto started this project and have brought it a long way. They earned the right to lead it as they see fit. Others will rise into leadership roles by taking on the the tasks that need to be done and earning that right. While everyone can contribute their ideas, someone has to make the final decision and at this point in time, these are the guys. Stop whining. When the organization needs to change, I trust they will make that decision. You can choose to contribute directly to the core project under their guidance or pursue parallel or complementary projects as fits your needs. Thats the beauty of open source. You are always free to pursue your own vision. (Xemacs as a branch from emacs comes to mind). We dont all have to agree on everything as we all have different areas of interest and concern.
- OpenACS is not a commercial venture and it should not be one. The demands of a commercial venture are much different that a community software effort even though you can build a sustainable business around it. I have no wish to see investors control OpenACS in any way.
- To help all those trying to build a business around OpenACS, I think it might be worth considering a separate organization or association of developer companies which can address needs of the commercial community. Things like education (maybe update some boot camp problem sets for OpenACS) and even certification or central job registry etc.. The objective being to promote what we find to be good practices and code, provide some means of confidence to clients what we know what we are doing etc and to promote a larger presence that a single small company cant provide by itself (ie if you go out of business the client can feel confident that there are other companies in the association or whatever who can pick it up). This should be separate from the OpenACS effort which is focused on develooping the core toolkit. This would be focused on supporting consulting practices centered around the OpenACS toolkiit.
- Right now we have enough work cleaning up OpenACS 3.X and getting 4.X to work the way it needs to. Working on the intranet in 3.2.5, I am still finding bugs and overhauling the current structure which I hope to fold back into the release when it gets a bit farther.
Ok..said my peace and hope we can move on to more productive stuff
Maybe we should just call the organization here a meritocracy to quell any more attempts at reorganization. The fact that this group has even attempted to port ACS4 to PostgreSQL says a lot about the orgaizers and the community. For the record, I have ported zero code, and therefore, I am entitled to zero say in how things are done around here.
It only meant: I didn't know that his package wasn't part of the "official" openacs
I didn't know that this Stalin wants to take over the rumbling organization (I cannot judge on this!!!)
See, practice what you preach and keep it even shorter (should have left out the "I didn't know that")...
My major concern at the moment is stop talking start working, because I am overall very satisfied with the progress of the openacs toolkit.
Sometimes people feel offended when others try to to improve the few things that are left to be improved and leave out the "GREAT JOB" for what has already been done. So we should start taking such suggestions as SUGGESTIONS and not as major criticism on the project.
Great Job Ben, Don, Dan, Roberto and all the rest that helped on every end - maybe this is exactly what some people want. Having their name mentioned, when others talk of efforts taken for openacs.
Once again a overall project plan might help where you can see who did what from the humble beginnings up to today.
But again let's stop talking, start walking and maybe Jerry can code a project module like the one I have been talking about for openacs 4.x, which might shed some light on the hallways of power at openacs
I hope that I didn't offend anyone with this!!!
Jamie's notion of a separate organization or association of developer companies is an interesting one. I'd hope that we could make this organization function in ways that don't make this necessary, because I happen to think that some of the issues - the need to market, in particular - aren't unique to those who hope to build a business off of the OpenACS effort. Of course, my definition of marketing, or at least our marketing needs for today, might differ from some of the ideas that have been floating around...
Anyway, an association of developer companies within the OpenACS framework would be more appealing to me than the rise of a separate group.
Most frustrating to me in all this is that I've been thinking quite a bit about the underlying problem we face, i.e. how to deal with the explosive growth in interest in OpenACS that has resulted from aD's ditching of the AOLserver/Tcl/ACS product.
I've been thinking about it all along, actually, but have only been acting in the very limited scope of the actual OpenACS 4.x project - first things first. We've been evolving a structure for that project that works reasonably well, especially when one considers that most of the folks working on the project are newcomers to porting and in many cases Postgres and/or Oracle.
I've somewhat naively thought that the community was so eager to see OpenACS 4.x delivered that we could focus on making that happen and worry about managing growth in other areas afterwards, i.e. this fall.
I thought that about the web site, for instance, though by May it became obvious to me (and Ben and others) that we couldn't. A solution to that particular problem, i.e. Musea's stepping up to the plate and taking it over, didn't materialize until recently, but one did.
Perhaps we can't ignore other growth-related issues for the next two months, either. I do wish we could. I also wish the discussion could be limited to folks who genuinely want to improve the organization, rather than include people who have the desire to grab a certain amount of control for themselves regardless of how that might affect the organization.
<LI> I need to learn to use html tags for quotes.... ok not relevant to this discussion
<LI>There are an increasing amount of developers who are trying to establish consulting services using the OpenACS toolkit. It would be good to have common principles and practices and a presence so that we dont have to take VC money to grow big enough to be taken seriously. The "franchise" concept comes to mind as a VERY successful model to scaling business through common procedures and tools and umbrella marketing etc.
<LI>We have increasing number of people wishing to participate who come from different backgrounds and with different skillsets. We need a way to welcome these people into the project and provide them ways to make useful contributions.
<LI>We have larger numbers of developers who are updating or adding code features. We need to be able to review and absorb these code changes more efficiently. The original OpenACS project was hosted at SourceForge and allowed greater developer access than the current model. Should we revisit this? How to we test and QA and coordinate patches and updates as they come in?
<LI>More diverse topics not directly related to OpenACS development are coming up as we picke up more people, yet the site does not allow new topics to be added.. maybe we should revisit this.
Personally I think we should find some consensus of what the vision for OpenACS project is and where it is going. The think the founders have a major say in this with input from the community at large. Our original vision was to be able to run ACS without spending a fortune on Oracle. I think it is time to revisit that.. like revisiting a business plan.
welcome these people into the project and provide them ways to make useful contributions</i></blockquote>
An idea that's crossed my mind is that perhaps we need a volunteer "volunteer coordinator". I'm used to this working model from my background as a board member of a largish (7,500 member) non-profit.
In that organization, the volunteer coordinator was the interface point for new faces that wanted to do volunteer work us. It was their responsibility to be on top of what volunteer jobs were available (like many volunteer organizations, we had formal job descriptions for volunteers as well as staff), what skills particular volunteers had, and to do a lot of the work associated with matching volunteers with jobs, including informal job interviews with potential volunteers.
<p>That's *much* more formalism than we need. Someone mentioned the idea of gathering profile information as part of the registration process here, and I find that an interesting idea. I think it would be great to have an inventory of skills possessed by community members, along with a list of tasks that need doing. I think it would be great if someone would step forward and offer to take on this task.
<p>For instance, if we were to draw up bylaws and all that, do we have any lawyers in the house, or would we have to hire one? Right now, we don't know. Wouldn't it be nice if we did?
<blockquote><i>More diverse topics not directly related to OpenACS development are coming up as we picke up more people, yet the site does not allow new topics
to be added.. maybe we should revisit this. </i></blockquote>
Specific recommendations for improvements to this site should be forwarded directly to Talli Somekh at Musea, with a courtesy cc: to Ben.
We have larger numbers of developers who are updating or adding code features. We need to be able to review and absorb these code changes
more efficiently. The original OpenACS project was hosted at SourceForge and allowed greater developer access than the current model. Should
we revisit this? How to we test and QA and coordinate patches and updates as they come in? </i></blockquote>
Well ... commit access to the OpenACS 4.x project is easy to get, there are probably a couple of dozen folks who can commit to various bits of the tree today. Only a few can commit to every part of the tree. The reason for that is simply that we've got a lot of newcomers who are total strangers to the project. I accidently had the tree entirely open at one point in time, and one newcomer started "fixing" code for folks in packages they were working on without bothering to tell them, ask if they minded, etc (and they did, I got a couple of "hey, why's this guy mucking in my code behind my back?" letters).
<p>Today, the crew consists mostly of folks who've been involved now for about three months, and I'd feel comfortable letting them commit freely. Then again no one's asking because folks appear to be comfortable just trying to get their own assigned packages ported.
<p>As far as new projects go, when folks have asked, we've given them a cvs tree to play in. And, of course, if someone working on an OpenACS-based project prefers sourceforge they're certainly welcome to go there instead of here.
<blockquote><i>Our original vision was to be able to run ACS without spending a fortune on Oracle. I think it is time to
revisit that.. like revisiting a business plan</i></blockquote>
This is an excellent point. I personally haven't quite figured out how to reconcile that original vision with our decision to support Oracle as well as Postgres. We were interested in multi-db support all along. The major reason that decision was made was that aD was abandoning their ACS Tcl user base and we didn't want that user base to be left without a community. That's great short-term justification but might not be enough to support a long-term vision.
Geez. I go away for a week and all Heck breaks loose.
Obviously, emotions are running very high right now. It's
particularly unfortunate that we've reached this crisis right now,
when we are perhaps 6 weeks away from a beta. Without a solid
4.x port, all this argument is a complete waste of time.
I've heard a lot of argument, but so far I haven't heard anyone
offer compelling criticisms of the progress being made on the
port itself (although I just got back a couple of hours ago so
maybe I've missed them). It seems to me that, no matter how
grave individual concerns may be about the long-term
governance of the project, anything that slows down the current
port--especially including threats to quit--are hugely
If the community does not finish the 4.x port and fix the most
serious problems with aD's implementation in the next couple of
months, this project will almost certainly remain little more than
a toy for hobbyists. Already, within the ACES world, there are
rumblings that the platform may not be moving fast enough for
some of its most ardent supporters. While we bicker, the
competition moves ahead.
It seems to me that all involved parties have to think hard about
what it is they want to get out of the community. And if the answer
is a competitive software platform that will open up new
opportunities for your business, then we each have to look
extremely carefully at the degree to which our individual
participation in the current crisis increases the risk that we will
fail to build that platform.
You are hereby forbidden to leave, ever, even for a day :)
<p>I don't think the OpenACS 4.x port is suffering at this point, though I don't know for certain. The folks doing the real work, i.e. the people porting packages, working on the OpenFTS integration, in other words everyone involved but me, haven't been participating in this discussion for the most part.
<p>So hopefully they've not found it as distracting as I have.
<p>However, if this continues much longer I would expect it to have a negative impact on that project. How can it not?
As far as leadership, I think we're damn lucky to have somebody of Don's caliber leading the project. I've been involved in many projects over the course of my engineering career, and I think Don rates among the best in this arena. Nobody, other than Ben, has done as much, or worked as hard to make the openacs project a success. I would really be interested to know who else in the community considers themselves more worthy to lead the project than Don and Ben? Who else has done as much for the project? Who has even come close?
These discussions haven't really hurt the progress of the port, because most of the people doing the talking aren't actually contributing anything to the port. It's kind of funny isn't it, that in general, the people that are demanding the greatest change are doing the least to make it happen, while the people that are contributing the most, are for the most part, happy with how the project is being managed.
That's not to say that some things shouldn't change, and I think the one good thing that has come from these discussions in the realization that the way to make a difference is to contribute. Musea caught on to that concept rather quickly, so why is it that so many others are having such a hard time with the concept?
Usually, I avoid these types of discussions, because nothing much ever comes of them. Alot of talk, some wasted energy, a few bruised egos, and then in the end the discussion sputters out with anything changing. If only a handful of the people here come away from these discussions with the idea that contributing to the project is the correct path for changing the project and gaining influence over the course of the project, then I'll consider these discussions a huge success.
Personally I think that it's time for Jerry to start his own, independent project somewhere, someplace, dealing with some issue or some software package or another.
Wow. So if I don't fit your definition of a model OpenACS user, I should leave too?
I don't have either the time or the inclination to take full and entire responsibility for an OpenACS 4 subproject. Does that mean I shouldn't be able to contribute at all? Well... it's your project, and that means you get to make the rules. But maybe there is a better set of rules that is more accommodating. Purely from a utilitarian standpoint, there are probably vastly more people who can make occasional contributions than those who can be package champions, and it would be a shame to lose all that potential help. I think that's at least part of what Jerry wants -- not "grabbing control" over what you're doing, but estabilishing some sort of procedure for the things you DON'T want to do, e.g. cowboy handling.
As far as Jerry's "crimes" of writing search code and proposing a search committee, go, I think these are really separate issues and comingling the two only unnecessarily muddies the water.
Vis a vis the search code, it's clear from the bboard record that Jerry has been working on this for some time, and making his progress on it quite public. No, he didn't go through Don before starting the code, but give me a break, do I need permission from Don before trying a proof of concept too? This sounds like such an honest mistake that looking for sinister motives is ludicrous.
As for the infamous steering committee proposal, I'll just say that I don't like formal comittees, and I don't agree we should form one, but let's not run someone out of the community for raising the issue!
I'm not trying to fan the flames. I'd just like to see insults (veiled and otherwise) kept to a minimum. :/
contribute at all? </i></blockquote>
No, not at all. In fact, there have been a couple of folks interested in the progress of the OpenACS port who've told me they don't have time to port who've responded positively when I've said, "well, if you could free the time to try a virgin install on Oracle, that would help. Could you?" And then when I later came back and say, "OK, we've hit this little milestone, would you mind testing an Oracle install?" they did so.
Likewise there are lots of little documentation and testing tasks that need to be done. If you can't take responsibility for a package port or other significant subtask, that's fine.
If you can't help at all, that's fine, too.
<blockquote><i>Personally I think that it's time for Jerry to start his own, independent project somewhere, someplace, dealing with some issue or some software
package or another.
Wow. So if I don't fit your definition of a model OpenACS user, I should leave too? </i></blockquote>
No, that's not what I'm saying at all. Jerry has a strong personal vision of what he'd like to see this project be: a place where he can pad his resume, where the needs of the community at large are subordinate to the financial success of the subcommunity who is trying to make a living off of the project, etc.
That's not my vision. Jerry's odds of transforming the OpenACS project into something that fits his personal vision are very low - it will only happen if I leave the project. On the other hand, if he starts his own project he can build it around whatever vision he wants.
<p>And if he does start his own project I can promise that I won't show up and try to yank "vision control" from his hands and place it into a steering committee elected by his user community.
<p>I'm just not the kind of person that would ever steal control of a project from its founder.
<blockquote><i> Jerry won't make his search project part of OpenACS unless he succeeds in grabbing
partial control over the organization. In fact, it's rather vicious. </i></blockquote>
Well, Jerry's been fairly vicious as well.
<p>After all, the act of trying to remove Ben and my authority to control the project and to place that authority into an elected steering committee shoved down our throats over our protests is a vicious act directed personally at us.
<p>Ben and I started this project and have poured about a year and a half of our lives into it. I came back after a 3 1/2 week vacation to learn that Jerry's trying to yank this project out of our hands.
<p>Have you or any of Jerry's fans thought about just how shitty it feels to have to deal with bullshit attempts to screw up this organization at the most crucial point in its short life thus far?
<p>As Michael implies, if we fail to deliver OpenACS 4.x in a timely fashion there won't be anything here worth fighting over.
<p>So ... *I'm* the bad guy? For losing my temper at someone who's trying to steal control of the project we started and have worked so hard on? For losing my temper at one who has caused so much distraction that it threatens to derail progress on the single most important thing we have to deliver? For trying to snuff his petty little revolution once and for all?
<p>Jerry's just an angelic little Lenin, eh? Bah...
I'm just not the kind of person that would ever steal control of a project from its founder.
Have you or any of Jerry's fans thought about just how shitty it feels to have to deal with bullshit attempts to screw up this organization at the most crucial point in its short life thus far?
I can see where you're coming from because it sounds an awful lot like, "Instead of going through Don et. al. let's go through this committee instead." But that would just be stupid; since when are committees fast moving? I don't think Jerry is stupid.
I guess at this point I'm beating a dead horse, but I read Jerry's proposal as, "here's some things I'd like to see the OpenACS leadership address that aren't being addressed, and here's one way to bring other people into 'official' positions where they can help." Explicit delegation, if you will, rather than the more darwinian "meritocracy" we are operating under.
Me, I prefer the meritocracy approach. But was this a mad power grab? I just don't see it that way.
I'll take the merited-consultative-dictator model any time over a committee.
What's the problem with OpenACS that this proposal is trying to fix? Having been involved with a number of OS communities (esp around Jakarta) this one seems pretty healthy; motivated contributors, experienced leaders and mentors, and most importantly, a strategy focussed on quality releases rather than bleeding-edge.
But people can't expect to be assigned important tasks if we don't have some evidence that they can complete them - this is really where the "meritocracy" part comes into play. We don't mean that people have to claw their way over their fellow community members to a position of responsibility for a task.
Take a look at how the RPMs were created. Someone put them together, said "look, this was easy, why don't you clowns do this this as part of your release process?", we said, "well, if it's so easy then why don't you take responsibility for this task" and, presto! We've got ourselves an RPM czar.
I don't understand where this process is broken ...
Why isn't there general comments on the installation docs, then I could add a comment on where the RPMs are and how I installed ACS in < 2min. with them. All of this has been brought up multiple times in multiple bboard threads.
Its fine to say if you want to see something done just do it. But I do understand when some people feel some of the non-coding processes could be better. In general I think everybody is doing a great job, but we shouldn't be stoning members that are suggesting possible improvements.
Seriously, that discussion should be over with. If you don't like the website, forward your suggestions for improvement to him, with a courtesy cc to Ben. If they're working too slowly for your taste, you can tell him that, too, with a courtesy cc to Ben.
We've delegated this task, OK?
Me and Ben have worked to get comments enabled in the documentation. But there's a problem: the HTML code Jade generates creates tags like these:
<BODY CLASS=blah SOMETHING=foo >
Which is not the way the general comments module expects. It's a simple fix, but I moved to a new apartment and ever since friday I've been dealing with boxes, packing, unpacking, setting up, cleaning, paperwork, etc. I only finished getting settled today, so I'll fix this ASAP.
The community grew significantly recently, and we're still adjusting to a more explicit way of reporting things. Before it was a small team of people who kept in touch regularly, so everybody knew what one was working on.
So please be a little patient. We're working. Now that you know the problem, if you'd like to send me a patch to ad-html.tcl, I'd gladly take it