[openbeos] Re: Waltercon 07

  • From: "Jorge G. Mare (a.k.a. Koki)" <koki@xxxxxxxxxxxxxx>
  • To: openbeos@xxxxxxxxxxxxx
  • Date: Thu, 03 May 2007 12:35:50 -0700

Hi Simon,

Simon Taylor wrote:
From: "Jorge G. Mare (a.k.a. Koki)" <koki@xxxxxxxxxxxxxx>
Hi Simon,

The problem I have is I cannot imagine a 2 day Haiku event possibly offering enough perceived value to offset
the cost for the "new Haiku developer" audience that
we're trying to attract.
That is analogous to the reaction to the proposition of starting a marketing push for Haiku last year. Look where Haiku is now.

The fact that people thought it was too early to start marketing (not me, btw) 
does not mean you are on the right track whenever anyone thinks something is 
happening too soon. If someone suggested running TV adverts showing Haiku I'm 
sure we would agree it's too early.

I am not trying to probe whether I am/was right or wrong. I am saying that many people have a distortioned view of marketing, and preclude it out of being clueless, as in having an oversimplified view of marketing that equates to advertising and commercials on TV. Marketing can actually take *many* forms and shapes, and you just have to choose the best tool(s) based on your goals and resources and any given time.

The difference is that we have had stuff to say to people for a while - look at 
this project, look how far we've come, find out more information - that is the 
awareness building that has been done very successfully (that you can take a 
great deal of credit for). What we didn't have then, and still don't have now, 
is stuff to fill a 2 day long Haiku-specific conference.

There is already a significant Haiku knowledge-base, and there is really not reason for those intimate with Haiku development and the code hold on the transfer of that knowledge to the next generation of active developers. As a matter of fact, the sooner you start transferring that knowledge, the higher your chances of finding more people that could potentially help further develop Haiku sooner. Now, note that I am not referring to devs that would be interested in developing for the Haiku platform, but actually develop the Haiku platform. This is what Haiku badly needs at this stage, and WC can definitely be used as a tool to outreach and educate such developers.

So, in spite of the fact that there is obviously something wrong, there is this fixation with the same failed model instead of trying something different that has the potential to be really beneficial for the project.
There is something wrong. I think it is the whole concept of a "conference" rather than 
an "informal gathering" at this stage of the project.
WalterCon is a conference, and it was meant one from the very beginning. That's why it was called WalterCon.

And it has never been successful from the beginning - how many new Haiku 
developers have been introduced by any of the WalterCons? It's my opinion that 
it's not the content that has been at fault, it's the whole concept.

It's not the concept, it's the lack of vision and planning, plus poor execution. If you look at the history of how WC has been organized, you will see that the drive to promote the event has been pretty much non-existent (usually limited usually to a couple of emails on the mailing list). You need to do a LOT MORE than that to create and articulate perceived value, both before, during and after the conference (yes, during and after too; it builds up value for the next event).

What puzzles me is that you are basically proposing keeping the same "let's get together" approach. I don't see how that is supposed to change failure into success.

[...] it has maintained the small group approach until today, when the project has *undoubtedly* much more visibility and general recognition than it has ever had, is more a matter of Haiku's inability to recognize/accept the need to adapt to the changing needs of the project.

I'm proposing a fairly radical change - dropping the whole "conference" thing 
and just have a gathering. The time for conferences will come.

You can look at past problems that became very evident with WC06 with a critical eye and try to change course by trying new approaches, or you can choose to pretend that everything is OK and stay with the same failed model. I would rather take the former approach, even if it may encompass some inherent risk.

My recommendations, often stated, are to scale back the importance of WC, make 
presentation-style content available to all via the web on a regular (not annual) basis, 
and have a gathering for interested users and developers. I take exception to that being 
described as "staying with the same failed model".

I think you are missing the (important) psychological factor. :)

Developers will not start creating presentation material for the web just because you tell them or you change the tone of WC. History so far has proven that they need some motivation or inducement to spend their time in doing something that is not directly related to coding. WC is the motivation/inducement that they need.

If Haiku were to follow your recommendation, the only effect will be a low key WC with little or no participation beyond the regulars. It is naive to think that because of that there will be an increase of presentation material on the web. On the contrary, you will end up missing the WC presentation that you would have had with a beefed up technical program. So not only have you wasted the potential *spread the word*, *knowledge transfer* and *dev community growth* effects that WC could give you, but chances are that you end up with less of the presentation material that you would like to see online. IMO, the approach is doomed from the beginning.

Instead, put together a compelling plan for WC, articulate it persuasively to the devs and the public, and then you will have much higher chances of motivating the devs to create good presentation material for the event. This can both lead to a successful conference, and have the side effect of all the presentation material created for the event eventually making it into website for everyone to download and see.

I will go as far to say that the most likely source of presentation material that you can put on the web will always come from either conferences or BoF at trade shows, as has been historically the case with the material that we have online (mainly from WC, Google Tech Talk, and NUMERICA).

Even not being techie at all, there are some general topics that I can think of, along the lines of Haiku programming for Linux devs, introduction to the Haiku API, setting up a Haiku development environment, etc. Obviously, there needs to be thought put into finding what the sweet spots for your audience are. I am sure the other Haiku devs out there can think of many more topics and help make a well designed program.

Finally we get to content, the core problem. I don't share your assurance that a program 
could be developed to suit the "new developer" audience (or any audience for 
that matter) - and that is the key underlying my entire position. Here some potential 
audience groups, what content they might like, what problems there are with that at the 
moment and how we might cater to their needs in the future.

Users: Might like to see introductory how-to-use Haiku presentations and 3rd 
party apps in the pipeline. Problems: Haiku is pre-alpha, so how-tos might not 
represent how things will work in the final release, and users will not be able 
to go and try out the techniques immediately. There are no really exciting and 
cool apps in development at the moment. In the future: would be well suited to 
a MacWorld style expo event.

WC does not target (end) users ATM.

3rd Party Application Developers: Would like to know about the process of 
programming on Haiku, and the API. Problems: Haiku is not stable enough to host 
development so people would be directed to BeOS. Might feel cheated to find a 
programming for Haiku conference is actually a 
programming-for-dead-commercial-OS conference. In the future: Would be perfect 
targets for a conference or development seminar event.

Sorry to repeat myself like a broken record (yeah, I am old generation), but who is talking about 3rd party app devs? You are the one who bring up 3rd party devs. What you are doing is like creating a problem so that you can give a solution.

New Haiku Core Developers: These are the people you seem to be aiming WC at. They are people we hope will devote their free time and effort to coding for Haiku. They may be interested in things such as setting up R5 as a development environment, checking out the tree, the general layout of sources in the tree, and coding guidelines. Any other presentations they are likely to be interested in would be very specific to the area they want to work in (VM, Networking, Media, lots of others). Problems: Guides to setting up the environment and checking out and exploring the tree are best followed step by step on the developer's computer - ie it is better to have them online rather than just have a presentation somewhere about them. Any presentation on a specific part of the code will inevitably not cover all they need to know and so the ability to delve into the code independently will be required if they are to be successful Haiku contributors. In the future: Web content is key to this group. They need content they can access when they need to - at the time when they are sat at their computer with the code in front of them. If they become contributors, they may move to the "regulars" group.

Yes, this is the group that WC should be more aware and accommodating of, as they are the ones with a higher chance of becoming regulars actively contributing to developing Haiku. If Haiku wants to even try to grow their base of active regulars, then catering to the educational/informational needs of these newcomers is a must, and WC is the ideal "fast track" venue for that.

Let me make a few comments about the problems that you raise in addressing this audience.

First, I think you have may have a misconception: no educational presentation will ever cover everything that any given person may need to learn. This is true for any conference, and expecting otherwise is, well, unrealistic to say the least.

What face-to-face presentations can do for this audience is a transfer of properly focused knowledge in a condensed and easy to absorb form, plus the opportunity to ask questions and get answers *in real time* directly from the source. This is *extremely* valuable, and irrefutably better than anything else, way better than having to browse the web for bits and pieces of info that may be spread all over the place and/or outdated, and asking questions on a mailing list or even IRC. To address whatever further needs any given individual may have after a presentation, the event program can always be designed so that whoever attended a presentation can spend some time with the presenter and/or other regulars to look into specific issues that they may have. The conference room could have Internet access if needed. It's a matter of being creative. :)

Regulars: Want to meet the people behind the code, and have discussions about 
internals, future direction, and just general social chat. Problems: None 
really, but there is little content of interest to them - they already have a 
good knowledge of the project and know the people and places to ask if they 
need to find out anything. In the future/now: An informal BeGeistert-style 
gathering for Haiku fans is perfect for these.

If all you want to do it for the regulars to get together, then there is no need to change anything. Just go to BG, and that will do it.

Nothing wrong with that per se. But in not pursuing more than just a friendly and informal get together, Haiku is wasting a great opportunity to grow its dev base, and in a way it is also denying potential newcomers the opportunity to absorb the Haiku knowledge-base and to engage with the existing community of developers. This is a real waste.

There will come a time when Haiku is a stable platform
when we want to reach out to people and get them using
and developing on Haiku. Then we can advertise a conference
where people can learn about using and programming for
Haiku, the API, UI guidelines, meet other interested developers, learn a little about the internals. That is the point when we have an audience comfortable with going to
conferences (professional developers, perhaps even funded by
their companies) and a solid idea about the things we need
to tell those attendees. That's when a "conference" makes sense.
Let me see if I understand: are you saying that Haiku has no interest in finding more developers that could potentially work on stuff that needs to be finished before R1?

Errm, nope. I'm saying a conference is not the way to find those developers.

What can I say? In my experience, conferences are one of the best "fast track" venues to engage engineers. :)

I don't know what you base your assessment on, but your reluctance to even entertain the possibility that WC could potentially those developers seems to me quite analogous to the reluctance to marketing for Haiku in the past...

And why do you keep coming back to this professional developers thing? I am not talking about professionals or businesses. I am talking about devs that could be enticed to attend if they were given a good reason to.

You're talking about the "New Haiku Core Developers" group I mentioned above. I don't think a 
conference is the right channel to attract them as I have explained. The professional/business people I have 
renamed the "3rd Party Application Developers" in this discussion - they might well be attracted to 
a conference but the time is not right for that "Developing for Haiku" conference just yet.

Again, I don't why you waste your time writing about business people, as they were nowhere in any of my previous messages. You are the one who keep bringing them up. What are you trying to prove?

Absolutely agreed. The "right channel" for this target
audience is very definitely the web. It's true we could
use some better content (I *really* like the online
presentation series idea, by the way).
The web is very important, and I don't question that. But the web is not the only channel to promote your conference:

I'm not talking about the web as the channel for conference promotion - I'm 
talking about the web as the channel for conference material. The advantages 
are huge:

- Free access to presentations as and when visitors want.
- Free choice of specific technical presentations of interest.
- Much lower barrier to entry - no time or money commitments required.
- Can be watched sat at the dev's computer with the code in front of them.
- Presentations can be added on a regular basis which will both build 
understanding and maintain interest of subscribers.
- Contributions from all over the world are possible.
- Questions arising from the talks can be added to the same page so everyone 
benefits from the insight of all who have watched it.
- Save the planet by not flying around the place.

If you, or anybody else for that matter, can successfully drive such an initiative, go for it! It would be great. But this does not have to be done as a replacement for what can be a conference venue to successfully attract new talent.

If it's not a conference, as you say, then the name of the event needs to be changed. Perhaps something like WalterFest may be more attune to what you envision for WC. Otherwise, the name does not reflect the reality and is therefore misleading (one of the BIGGEST no-no in marketing).

Agreed. WalterFest sounds OK. If the content as I describe is added to the 
website then it really just becomes a social event (obviously people are free 
to present stuff live too - this might even be easy if the presentation 
material has already been created for the web audience). I'm OK with just a 
meet-up event.

There is no point in presenting *old* presentation material at a conference. Nobody will care. I think you are turning the world upside down to make your point, but that's not the way the world is. :)

Obviously having all the content that was going to be in the conference 
available for free online reduces the value proposition of the conference as an 
event. That's not a problem if overall the result is positive for Haiku, as I 
firmly believe it would be.

How is a low key event and the resulting less presentation material on the web positive is a real mystery to me...

With regards to the informality aspect, I suppose you are referring to a casual atmosphere (correct me if I am wrong), which I also think is preferred. But one thing is a casual atmosphere, and another informality in the form of lack of organization and/or a well defined program; the latter is very detrimental, and a real show-stopper.

The key thing for the regulars is the social aspect. Interested local Haiku 
watchers may also be persuaded to attend to meet up with some of the community. 
That's all the event should be at the moment.

Again, when the time is right for a "Developing for the Haiku Platform" 
conference we will need to think and plan much more carefully.

It's not about "developing for the platform"; it's about "developing the platform". By limiting educational and knowledge transfer activities to the web, you actually deny the new potential talent from moving faster into a stage where they can actively contribute to the development of the platform. A real waste, if you asked me.

Writing these lines I realize one more important thing that I don't think has been said so far: a successful WC *really* needs an organization team.



Other related posts: