Forum OpenACS Q&A: How long can the OpenACS community get away with being an island?

I'm always amazed to find just how efficiently Google is able to answer my questions on just about any topic. This time I searched for "open source collaboration software" and gained a number of new insights into the competitive field of OpenACS. I remember how when I was working for ArsDigita a lot of people would wonder "So the ACS is very cool, but does anybody know what alternatives are out there?". I remember being slightly disturbed by the fact that nobody could give a satisfactory answer to the question. Well, maybe two years ago competition was scarce, but apparently today the world is different.

In the Communications Category at SourceForge there is no shortage of Collaboration software. Here is a small sample (mostly from the two first pages out of 45 with the most active projects) of software with functional overlap with the OpenACS:

• Groupware: PhpCollab, phpCrystal, Moregroupware, PhpGroupware
• Task Management: Internet Task Management System, PHP Helpdesk
• Survey: phpEsp - Survey Application (7 developers)

The list clearly suggests that in the Open Source Groupware realm the PHP and MySQL technologies reign supreme. It seems to me most of those PHP based collaboration software projects tend to be more focused (less comprehensive) than the OpenACS, some are on hold, but others are flourishing. A for OpenACS very encouraging conclusion that I draw though is that a clear leader has yet to emerge in the Open Source groupware field. I find that amazing if you think about how long such software has been around now.

My Google search also brought me to The Open Source Groupware Category at the Open Directory (dmoz.org). Needless to say the OpenACS was missing from this category, and in fact was missing from any category at the Open Directory so I submitted it to a number of suitable categories.

Another intersting finding was collection of article and software that Grant Bowman - the editor of the dmoz category mentionied above - is maintaining. I sent him an email informing him about the OpenACS and dotLRN and I am curious to see what his reaction will be.

It seems to me Philip Greenspun and ArsDigita suffered from a technical hubris that led us to not only neglect marketing but also neglect software developed outside the company. We were mislead by premises along the lines of: 1) We have world-class developers and are technically superior to the rest of the industry, and 2) Our technical supremacy will sooner or later be apparent to the rest of the world, so we don't need to market ourselves, people will come to us. During the IT hype Philip was able to get away with this approach, and yes, Philip was a great evangelist and had a amazing talent for attracting highly skilled developers. Unfortunately technical hubris can come with a high price tag. ArsDigita failed to market itself in the corporate world and also failed at building up an Open Source community.

Eventhough this was a very small investigation, and eventhough I am very tired now that I am writing this, I will venture to draw some tentative conclusions. First of all I firmly belive that OpenACS is a *very* sophisticated and comprehensive platform that holds great promise and still stands up well to the competition. The recent push of dotLRN is very welcome and has helped raise the quality significantly of many of our most important applications. But, there are a lot of similar open source software brewing out there and I do *not* think that the OpenACS can maintain its lead unless we drastically start expanding our community of developers and clients. Such an expansion is not likely to happen as long as the rest of the world doesn't even know that the OpenACS exists. So then the non-trivial question beckons: which channels do we use to expose ourselves to the world, and who will have the time and energy to make this happen? I alone cannot answer this question but I hope that this posting will provoke some ideas and help stimulate the improvements that the OpenACS project needs if it is to thrive well into the future.

I think we need to be careful to differentiate between OpenACS.org and OpenACS (the system/platform).

OpenACS.org needs to market the OpenACS system (and its community) in a more coordinated fashion. Some coordination is overdue for addressing nondeveloper issues that hinder the growth of OpenACS' use. For example, OpenACS.org needs to serve a larger community of nontechies and professionals (marketing) as well as continuing the existing developer services. See https://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0005Eg&topic_id=11&topic=OpenACS regarding a discussion about this. This discussion also was touched on in January https://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0003ZW&topic_id=OpenACS&topic=11 .

I'll be glad to volunteer with moderating the new forum(s) (with anyone else who wants to help).

I have been talking to some *IT* execs lateley here in Germany and it's no real news that they are pretty much focused on buzzwords, nicely and graphically presented functionality as well as "best practices" of former clients. What we need to manage in my opinion is the following:

1. Get a *professionally styled* new openacs.org website out.

Maybe somebody knows a professional graphic designer that could start making the existing logo (I like Philip's dog on it) look a little bit more professional... Further we need some professional "powered by" logos. After that we need some nice layout suggestions for the new openacs.org site.

2. We need a bunch of "best practices" that explains how the system is being used with *screenshots* etc. We have a lot of prominent examples like:

Education ~ Sloan, Berklee

Knowledge Management ~ Siemens ShareNet

Non-Profit ~ Greenpeace, Urban Parks

And maybe some statements/testimonies of prominent people from these companies!

3. A description of what the core can do in terms of scalability, security, modularity.

4. A short description of all existing modules and at least one *screenshot* to every module.

5. A Powerpoint presentation that basically is a summary of 2-4. Further a pdf version of that presentation... Rafael Calvo already came up with a nice pdf document using a lot of screenshots!

6. We need to place openacs.org within press releases for newly launched sites like the upcoming dotLRN.

Do we need marketing for the Product OpenACS: YES, but why ?

• We need to raise the awareness for OpenACS. If people talk about Groupware, Knowledge Management, Collaboration or E-Commerce they should talk about OpenACS (at least in the OpenSource realm).
• If more people are aware of OpenACS they will come to the site to check it out, searching for:
  • What is OpenACS
  • Where can I get it (Am I the only one thinking about putting RPMs up for RedHat and SuSE to include in their distributions ?).
  • What do I need to do to install it (already there, though it is still to time consuming)
  • How can I put it into use (Use cases, might be fictious, might be case studies)
  • Who is using it (case studies)
  • If I use it, will I be left (numbers of companies, downloads, installations, see seperate thread) dead in the water soon.
• Awareness raises the number of installations. Together with good training material it will raise the number of developers, which in turn will raise the number of available developers for the companies, which allows them to take on new clients.
• Giving speeches at conferences, publishing articles in magazines, getting the OpenACS book out will raise awareness also in decision making IT folks. And it is easier to sell, if your potential client already has heard of OpenACS
• Teaching material would make it also easier to adopt a curriculum for OpenACS in universities. After all that still is the cheapest way to get trained new developers.

• If having marketing material and case studies at openacs.org prooves to be too much of a headache for the community, use openacs.com for this (if Ben is willing to provide the domain for that use to other companies as well)
• Write up case studies with your clients. It is the best marketing you can hope for.
• Write use cases for the toolkit (we are on it)
• Get the new openacs site up and running. If the community (museatech in particular, thanks a lot for working on the new site) lacks graphic design ressources, I can ask our designer to come up with some ideas. But we should first agree on a structure for the site.
• Think about fairs where it would be useful to have an OpenACS booth. Let companies interested jointly use that booth. (I propose Leantec 2003 as one option)
• Go to your local university and ask them to teach a course about OpenACS for free (and / or sell a dotLRN installation to them 😊).
• Write up at least one article each how to use OpenACS for the hot topics out there in the market (for KM I have one version which needs to be revamped to focus more on OACS at http://www.sussdorff-roy.com/resources/km-article. Furthermore I have a german version for groupware (published in a Linux magazine))
• Issue press releases about great news (e.g. the release of dotLRN).

In more detail I am thinking of a little revamp of the samoyed part of our existing logo and secondly a more modern font and color (organizational identity 😊 for "openACS" or "openacs" don't know...

Basically I am also willing to coordinate the Marketing effort!

I'm actually working on this now, but these things take time. However in marketing terms its the holy grail. Get a big name behind it (i.e. Oracle/Vodafone/IBM size) and we're sorted 😉

C'mon folks, someone out there plays tennis with the head of products at Oracle surely ??

AIESEC is also a minor brand which carries a lot of recognition in business circles. And what about working with consulting companies [I admit, I did loose on or two teeth at the doors slamming into my face so far, but hey, we can try it again].

As for playing golf. If you want to get in bed with Oracle, you might want to start sailing 😊.

Agree with those points, obviously and will begin on some of them.

I think however we have a few more fundemental items to address before plunging into marketing activities.

Principally I'm thinking of:

a) Who are our targets/market? How do we profile potential users/clients. Is there indeed a profile and if not do we need to decide on one. I don't think its wise to pursue the 'its everyhting to everyone' approach.

b) Identify *exactly* the OpenACS is. To be blunt an 'open source toolkit for developing collaborative systems' is about as airy and unquantifiable as you can get. Effective marketing comes from an easily communicable, simple product description. I've had direct experience of this whereby customers can't get a 'handle' on exactly what OpenACS is or isn't.

c) Open Source. Ok, this is going to be contentious, but... we have to think very carefully about how open source is presented from a marketing perspective. Unfortunately I beleive AD made a mistake in that it got a little ' evangelical' about open source. If you push it too much then the customers can't see past it. We need to think about what in open source, the customer is realy interested in. I' aware of all the academic strengths of open source, but realistically its not a fully proven approach in business, and therefore we have a virtual 'education' exercise for potential users. Beleive me we don't need that to contend with. So how do we describe it. Lower captial investment... less supplier dependence... transparent delivery.. tricky.

d) We also need to think international. Many of the users here are European based. In particular (and no offense in intended) AD's previous marketing 'tone' was whey too gung-ho, we're great, this is fantastic, we know better etc... probably all true, but not necessarily a good style for European business.

e) Do we indeed highlight the past relationship with AD? There are benefits, but/... not all their customers were quite as happy as perhaps was made out, not all their deliveries were really all that good, they did have a bad rep for poor quality and of course they went bump... none of these things are good PR.

f) Who are our 'sponsors' i.e. which individuals within each company are we targetting? Is it product managers, developers, sales, marketing.. its vital to understand this. I'm guessing we're primarily appealing to the development sector at present... but is it they who have the purchasing power etc..

and so on....

Let me have a think about some of these issues and perhaps we could start with some kind of discussion doc... something to get us going..

I'll put together some thoughts, lets knock it around and hopefully we'll have a good start point.

Oh... and just a personal experience/opinion... conferences and stands are usually a pretty big waste of time.. lets face it most people only attend them to get out of the office and it usually turns into competitors nattering to each other..

Cheers
Simon

Therefore people with most insight into 4.5 should develop a real roadmap for a "marketable version". We should also give the baby a name... maybe 5.0?

I will most probably start developing an intranet for a German broker in August and would really like to develop with a clear roadmap in mind, so that I can cover some of its ToDos with my work...

Being too tech focussed is not a good idea... If we leave it to the technical people to get the product to a state where their entirely happy then we'll be waiting forever! Developers are largely perfectionists by nature.

The commercial env, does not demand perfection, it demands good marketing, good support, good potential and good ROI... I *know* the OpenACS is already able to meet those demands, therefore it is definately time to begin banging the drum a bit more..

Also, if we're not shouting loudly, that in itself becomes an indictment of the software. i.e. if they're not selling it up it can't be that good..

Have some faith folks, this really is an excellent toolkit (warts and all) and we shouldn;t be frightened of saying so, albeit in a balanced manner.

If get a good vision going forward, I can sell it 😉

However, worth mentioning that if we begin down this road, it will also involve a few cultural changes at openacs.org... everyone needs to understand that marketing is a part of every activity. Every site yo write, ever call you make and every conversation you have is a chance to market the platform.

I and others will be happy to provide the strategy and material, but it needs everyone to spread the word..

• Make an RPM that would get you to the point where you add your personal details for the site. Make another RPM for dotLRN.
• Write up a document how to change the look of your site easily (master template aso.).
• Write a step by step guide for installing packages. Add install options like "collaboration site" or "e-commerce site" that would install a predefined set of packages to get you rolling (like RH or SuSE are doing for their distributions). Write use cases for these options (so people get to know what this is capable of).
• Have training material for future developers ready. This training materials should take them slowly through the process of learning OpenACS while giving people the freedom to try them out on their own problem. Please answer this at the new thread

AFAICT Malte and Simon are now officially on the hook. I'll have to read this thread with more care before I can say I perfectly understand what they're on the hook for ... but they're nailed.

By this I mean those who step forward and volunteer get to lead. I think it's wonderful. My middle name is delegation. Or perhaps it is better to consider the (for lack of a better word) marketing aspect a separate project with its own leadership. I don't care how we look at it as a community - it needs doing, it needs leadership, it needs energy. Let's go!

Lars's point about the toolkit being more sizzle than steak, more promise than delivery at this point is true. Yes, many of us have built real sites on top of OpenACS 4 but ... it's still weaker in some regards than OpenACS 3. There's a lot of confusion within packages, some of which truly suck, most of which don't really make proper use o f the core tools (due mostly to a lack of guidance and communication between the core team and package implementors when parallel development was happening at a very rapid pace).

So, Simon, this isn't really a matter of the mania of developers for perfection. There are serious issues to be addressed. For instance last week I spent a day contracting with Sloan to help out with scalability issues uncovered by dotLRN. We've got lots o' stuff like this to deal with in the core.

That doesn't mean that it is unusable or that we need to hold off trying to attract attention to ourselves until all such things are fixed. Not at all. A roadmap that identifies brokenness and who might fix individual things and when would be a big help. It's also the kind of thing that will help keep new developer types around. If a new hacker stumbles across something that's strangely whacked out and broken in psychotic ways, but sees that it has been identified and is on a list of things to be fixed with a name attached to it, the hacker's less likely to decide we're stupid and our toolkit worse.

I've been talking for a couple of weeks now with a potential delgatee who is interested in taking over the task of managing the nuts-and-bolts of organizing the building of things like roadmaps, package ownership and status docs, etc etc. More later, perhaps in the next couple of days, by next week latest. This would be a great help to the project and a great help to me (because I don't currently have time to do the basic mechanics needed to communicate these basic items to the OpenACS 4 world, and haven't for months as has been obvious to all).

Building the roadmap needs to be a collaborative event among those who know the core well and presumably have been thinking about it. Lars and myself are two who've participated in this thread who have already discussed some of the issues both in person and via e-mail (and Peter was with us when we talked in person). There are others.

Having a parallel effort on the other end, led by the likes of Malte and Simon, would be fantastic ...

Peter's comments about technical hubris are well-taken. I would hope that we, as a community, don't suffer from that. We aren't the only smart kids on the block. We just have more fun :)

I'll post a bit more when I get back from a meeting...

talli

As far as the logo goes, it WAS designed by a professional graphics designer. The only direction given was that it should include Alex. Designing a logo for an open source project is much different than designing one for a private corporation. It was meant to have a fun feeling to it.

Talli, I agree that your site is MUCH nicer and cleaner than the out of the box OpenACS. Why are we not using it already?! There was a flurry of energy about it, and then it kind of vanished (the flurry, not the site).

The sad truth is this...while OpenACS is great, and has been used for some great projects, it is lacking on some of it's promises. Instead of re-re-rebuilding the marketing/design/graphics wheel I would suggest going with the Musea site and then concentrate on hammering out the bugs of the current system! I know that's what the whole dotLRN and 4.x push is about. But until it's a stable solid system that can handle ecommerce, it's not going to get a huge amount of exposure.

I would also concentrate on some of the key modules such as bboard or ecommerce to the exclusion of the others. Get those FULLY decked out and working PERFECTLY. Make them user-friendly and very customizable and people will flock to them. THEN you can "sell" them on the whole system, which will be being built at the same time.

Don't take this as an attack, I've been a MAJOR evangelist of OpenACS through the years (www.edulix.com, www.royal.com, www.mobio.com, www.sv2s.com are a handful of OpenACS sites I've built) but even I have been looking at PHP solutions lately.

Let's get things GOING!

You're right about this:" Marketing for business folks? I continue to believe that this remains the job of the various OpenACS companies, not of OpenACS.org or the developer community."

..but they need information about OpenACS for the presentations etc. for example: capabilities, limitations, strengths, weaknesses.  The OpenACS.org community knows these answers best, and should have them presented (honestly, accurately) in a kind of marketing-situation-analysis in the beginner docs.

If info on competing systems is unavailable, then OpenACS is in an even better position. Business types don't like risks associated with unknowns.

Maybe if one of these companies was an Oracle or a Sun... but realistically we aren't..

Even clients like Sloan, whilst I'm sure have some weight in your part of the world, count for very little over here. In fact a direct quote from a product manager in one of the largest cable operators over here was...

'Sloan... Academic types, well thats hardly a good recommendation'

Not that I'm defending such ignorance, but I'm afraid its somehting of a truism. Business folk are often sceptical about anything outside of the pure commercial arena.

Just to push that example a little further, one of the previous problems we had in getting classic ACS into my previous company was the 'Academic Evangelism' that AD pushed so vehemently. It just doesn't sit well with the business community.

So, in order that OpenACS does not disappear down that road (i.e. serving charities, education, amateurs and little else, although I accept they are important) we absolutely must give thought and effort to its commercial marketing.

My ex boss had a saying.. 'If it looks professional it might well be, or it might not.. but if it don't it almost certainly isn't'

***

However I can see the difficulties in trying to merge a commerial and a technical entity into one. And with that in mind I would certainly suggest a two pronged approach. An OpenACS.ORG and an OpenACS.COM as it were..

But, it still requires the leading lights from the former, to actively contribute and assist the latter..

And of course in comparision to development, marketing is cheap! Anyone in this community can help, anyone can make calls, anyone can produce collateral and so forth.. by effectively marketing this community so the technical arm benefits...

Would you not agree with this?

The presentation has a lot of interesting snippets on categorisation, UI, search, keywords etc.

Quite an interesting presentation, and probably belongs in the design forum, but again highlights the fact that making information accessible to users will take more than a couple of new bulletin boards, and logo

We already have the URL. We already have a willing volunteer to put it together (ooops that'd be me) so if we have some space/location for an openacs install... we're away...

Although a site in itself won't help, but it will be a gateway to progress. If there's some agreement (and space) then I'll put together my thoughts/vision for what it might be and pass it around for consultation.

To give you a flavour, imagine if you will a portal for the 'business of web development' a gateway of information for commercial e-development with OpenACS as it basis...

However... one or two is not enough... I'd like to see a few more volunteers on the commercial side.. gotta be some talented marketeers, sales and industry savvy folks out there....

How do we feel about 'becoming the de-facto technology for e-development' ?

Thats gotta be the best solution... after all two sets of users.. what a pain.

"How do we feel about 'becoming the de-facto technology for e-development' ?"

I know that I do not want marketing of this type associated with OpenACS. If an individual company wants to advertise it's services this way, no problem, but I don't feel it is appropriate to assocaite this type of message with the community.

That is, if a half-dozen or so OpenACS consulting companies want to collaborate and create a "vision" for the marketing of OpenACS, I don't want it to conflict with the goals of the community.

So, information about the technical benefites of OpenACS, and how it can make it easier to develop web applications using OpenACS is just fine. I just don't want to see dot-com style "branding" of OpenACS going on. Recall the various stages of the ArsDigita home page.

Some time ago we have also been talking about a new name. Maybe we can have a new name with the old Alex logo! We've had *MANY* suggestions, so we don't really need to start all over... I mean dotLRN isn't called openacs-learning either!!!

I also disagree with Ben that marketing should be to the consultant... It looks way more professional and has way more synergy effects to have an organizational marketing strategy and material. Maybe these materials can be modular so that every consultant can put them together for his specific needs (education, knowledge mangement etc.)

It's funny to see that the European guys are so keen on marketing. I thought it was an american stronghold 😊 ... and I will be on the business marketing / .com team!

I understand you point,but... I don;t see my comments as conflicting with the 'goal of the community'

As far as I was aware the 'goal' was to effectively serve all its members.. that includes commercial organisations.

If you don't want OpenACS to become a leading light in this area, then may I ask exactly what you do envision?

Commercial organisations need to feel confident about the product and the goals... hence the slight separation of tech from commerce.. whether you like statements like that or not, it happens t be part and parcel of what it takes to promote products...

What I meant was what is "e-development"? How does that tell people what OpenACS can do for them. I am all for promoting the strengths of the OpenACS platform.

What area do you see OpenACS as a "leading light" in? I would like to promote OpenACS for many different uses. But as Ben said we need focused messages to these various markets.

So I support a marketing effort, I just want to make sure it is focused and actually explains what OpenACS is.

I think it's a great example of the split between a site focused on toolkit developers/users and one focused on prospective decision makers.

My experience, coming from the business world, was as follows: I read something interesting in an article. I go to their site. I look at some screenshots. I read the about article, maybe go on to the functionality overview. Look around for some hints how this might be USED in my organization. Leave.

If I find an answer how I can use the tool for my organization I'd have someone check it out (or I'd do it myself), and when I really thing thats the way to go, hire some of the active members or companies of that community.

Up so far, no information in that direction can be found on OpenACS.org. Is this commercial marketing? Not necessarily.

From what I gathered so far there is a tension between the need for using buzzword driven marketing for executives and other bypassers while containing enough fun and freedom of expression for the rest of the community. For now I'd just leave the buzzwords out and start with the direly needed marketing towards a greater recognition of the toolkit (like I described above).

Well, last but not least, taking in account how many people want to actively participate in marketing shows that we also have a potential for this within the community, people who want to give something though their are not the most experienced developers. And from what I have gathered so far, developers tend to be very lazy when it comes down to user documentation or materials praising their work.

It's not going to be like in a busines situation where the marketing/sales guys grossly overstate what the product is capable of doing and the developers just have to eat it.

So instead of killing the flower right in the beginning, let us start and if the material we produce is too much out of the way for openacs.org to host, we still can come up with a subsite for openacs.com. This is taking for granted that Ben would be willing to lend the domain for that purpose.

I think we also need to give thought to why most of us are here in
the first place. I think one of the best points Philip Greenspun
ever made was that magnet content is what drives the success
of most websites.

CONTENT is what drove most of us to to ACS. Philip's book, the
ASJ articles, the great discussions on web/db.

I think what's happened is that we've reached a size that we're
pretty comfortable with. We have "enough" developers, we have
"enough" business. We're doing fine.

But we're not effectively reaching out towards

1) bringing new developers into the "fold". Malte and many of the
previous comments address this very well.

2) making new developers successfully grow into expert
developers. I think besides working on documentation, we also
need to make it as easy as possible for people to plug into
openacs as developers. We need documents that really explain
how to become a good developer, how to set up CVS and use it,
how to contribute to an open source project, how to make
patches, how to make bug reports, how to how to how to.

For us to concretely make a difference on these two issues, it's
going to take some work and time. I'm very encouraged to see
people stepping forward to help out on this.

I'll have a lot more time to devote to this once we eventually move
our ACS 3.4 based system to OpenACS (I'm thinking about
doing that at the beginning of next year).

Realistically, unless we change how we're doing things,
openacs will never be anything other than a fringe product.

Some questions new people are going to bring when they come
to the openacs.org website is:

- what does this toolkit do?

- why is it better than competing products?

- why is it worth my time to learn such a diverse body of new
things: Aolserver, Tcl, Postgres or Oracle? What does this buy
me?

- does this product have a future? What does it look like?

Stephen, in the best of worlds, or no, rather in a quite dull world, consensus would be possible, but in this life - we will never have one. We still need to decine on a roadmap though with action items and deadlines. However, I think a lot of us have our own agenda and we cannot really afford to wait for the overall roadmap to emerge before we contribute.

I don't want folks to think I'm going to suddenly slap up a website with nothing on it and start spouting Mar-Speak..

I'm merely saying that the balance at the moment is wrong.. i.e. all developer, zero commercial..

Ok, I read this to mean there is general agreement on having a more commercial focussed effort and therefore I'm going to proceed with that remit.

I'll start by getting a plan/roadmap/discussion thinumy out there, and we'll move on from there.

Cheers
Simon

Further expanding on Ben Adida's comments :-

I would be willing to QA and improve the quality of existing packages but what is sorely lacking here is quality CORE documentation as touched on by Don Baccus.

A Basic breakdown of this is how messaging should be handled session persistance and other bits and bobs, other wise you end up with the ACS plague of 10 peoples version of a core routine rather than using core functionality.

As well if the core functionality is 95% there rather than re-invention, some corespondence with the core maintainers could possibly evolve the function so that it can do what you require without become a branched version.

If memory serves ACS Templating was a classic example of this.

So if a bit of time was spent saying these are the core api's and this sort of issue should use this api etc etc, some of you package problems would dissapear

I apologise if this doc exists and I have not found it, I am looking to convert our site to openACS soon so hopefully will become a more informed and active member soon.

Regards