The Lisp Questionnaire
http://www.xs4all.nl/~alemmens/alu/questionnaire.txt is ineffective. If
you look at the database of people who answered the Questionnaire
http://www.xs4all.nl/~alemmens/alu/database/ you will see 235 replies.
The first submission was made in June 2004. 226 of the submissions
occurred in 2004. Only 8 in 2005. Only 1 in 2006, in January.
And this is worldwide. That's PATHETIC.
Since this resource is all but useless, and all but ignored, we should
ask why. If we think surveys are important, we should do a proper one.
And figure out what makes for a "proper" one, so that we don't end up
with something equally pathetic.
Some obvious things about what's wrong:
- I'm not about to dig through a worldwide list of 235 people, one by
one, to find out anything about them. It's an invitation to a big
chore, with no promise that it's going to get me anything of value. Why
would I look at such a list, compared to posting on comp.lang.lisp or
some other forum where people will actually respond to my question?
Even if crickets chirp, they chirp more quickly than digging through a
list of random stuff.
- If people were grouped by geographic location, I'd have a much
stronger incentive to look at their stats.
- If people were grouped by interests....
- If people were grouped by Lisp implementations....
- If groupings in general were searchable, like a proper database...
- If the Questionnaire and Database were displayed on a website that had
a certain level of professionalism, so that it appeared to be well
maintained, well cared for, actively supported....
- If the database were simply much larger, so that it appeared to have
value...
- I'll wager no active effort is being made to get people to answer the
survey. Either that or sporadic efforts were made, and the survey taker
was too swamped to do anything with the results he got. So there is an
infrastructure and automation issue here.
Now, there may be a meta problem about whether anyone cares about these
sorts of things. The last time I think I ever looked at a database of
people's stats was probably on Gamasutra. That would have been 3 years
ago, when I was sticking myself in there as a contractor and was curious
what others were doing. I've never cared since. So my question is:
- is there a site out there, for any language or technology, that
demonstrates "best practices" for taking surveys and using the results
to further some promotional goal?
'Cuz we should study that up.
Cheers,
Brandon Van Every
A minimalist draft of a charter for LispSea is available on the wiki:
http://wiki.alu.org/lispsea
Please continue to send comments via this mailing list.
Let's keep the wiki page clean and simple.
Thank you,
-Daniel
Just in case this is new to anyone:
There is a larger group called the Lisp Gardeners, which
is not tied to a single geography.
http://www.lispniks.com/cl-gardeners/
Their mailing list started only a few months ago, so it's
fairly easy to get caught-up.
Consider LispSea to be /somewhat/ of a Seattle chapter
yet not necessarily tied to Common Lisp.
This implicit relationship also frees us from doing everything
ourselves and allows constructive overlap and a loose local
group (so we need not become burdened with policy or dogma).
There can be much traffic on the Gardeners list. It's
bursty. For a while, I simply read it periodically via the
Mailman interface linked from the above URL.
-Daniel
PS - Please respond to the Lisp Questionnaire:
http://www.xs4all.nl/~alemmens/alu/questionnaire.txt
Brandon wrote:
> If the goal is to get business people interested in Lisp, wouldn't an
> Eastside venue such as Bellevue Community College or Digipen be more
> appropriate?
The business aspect is more of an outreach project. We'll most likely work with the WSA, ACM or perhaps UW to issue surveys. Subsequent interaction might take the form of a 1 or 2 day conference, and Meydenbauer Convention Center in Bellevue might be appropriate.
I'm not expecting business executives to attend presentations by Lisp programmers.
Such a conference might involve the Assoc. of Lisp Users (ALU) for legitimacy and would most likely be a year away.
Promoting Lisp is a longer-term effort. First, we need to resolve important logistic details within the Lisp community itself.
For example, with Python, you have "one stop shopping" for nearly all your needs. With Lisp, first it's which flavor? Then, which implementation? And of course, there is the matter of de facto libraries. (With Common Lisp, ASDF and ASDF-install should be standard but aren't because they arrived after the ANSI spec. With Scheme, you have the Revised^nth game...)
Within a business perspective, these details add complexity, which in turn adds risk.
In an effort to resolve all this, the early sessions I described yesterday are a first step. After a few months, maybe we'll come out with our own InstallShield-built packages or perhaps LispBox will be even more mature.
Or maybe we merely each contribute a blurb to the new Lisp FAQ project:
http://www.lispniks.com/faq/faq.html
Either way, much of the activism is to get Seattle programmers more involved and contribute that much more to the Lisp community at large.
By working together, the amount of work by any one individual remains quite small, but for any of us to do it alone would be more than a full-time job.
I'm a practical idealist. :)
-Daniel
This is for those interested in Lisp programming in Seattle:
As a group, we need to start somewhere, and I'm putting a stake in the
sand. My approach is to keep things simple but not simplistic.
For the time being, the home page is:
http://wiki.alu.org/lispsea
The purpose of this group is to
"promote, nurture & expand Lisp in Seattle".
But what does that mean?
By promote, I mean selling Lisp to businesses-- not a specific tool
chain but the idea that Lisp is perhaps more appropriate than other
mainstream options available.
In order to promote Lisp, this involves contacting various employers in
the area to see if they're currently using Lisp and if not, find out why
and address those issues.
Think of it as illustrating the need so those of you who wish to do
consulting have an easier path.
To nurture: semi-formal sessions will be offered to present and discuss
different aspects of Lisp. This includes specific tools.
I'll be the focal point for the time being and will lead the early sessions.
While I'll be the default presenter initially, my personal tools are
common lisp. Where other Lisps are concerned, either someone else can
present to the group, or if too shy, show me, and I'll present the
topic. I'm always open to learning...
For expanding Lisp, this applies to more than just the Seattle metro
area. LispNYC recently announced that they're providing mentors for a
few projects within Google's Summer of Code. While it's too early to
really worry about the next one, let's entertain the possibility.
Additionally, we expand Lisp by talking about it to others and having
facts readily available. This means as a community, doing our part such
as contributing to the new FAQ effort-- which needs help!
Finally, we can expand use of Lisp by contributing to ancillary efforts
like the "Lisp movies" that demonstrate some unique quality of a
particular library or updating documentation where it might be lacking.
So this element is about more than just Seattle.
Obvious questions begin with how, when, who and so on. First: how.
I'm looking into reserving a classroom at Seattle Central Community
College, even though this group has no direct affiliation with any school.
Regarding the classroom-- think of it as complimentary to SeaFunc
(functional programming; see the above URL), which meets in a tavern.
Rather than make you choose one format over the other, you have the
option of either or preferably both.
I've toyed with the idea of formally creating a free course through the
UW Experimental College or something similar, but let's try a looser
model first.
The objective for the classroom type of venue is to convey information
so we all can expand our knowledge. (clean sound, video projector, etc)
I've found taverns to be difficult when the group grows beyond single
digit counts. ...and yes, I'm planning that by autumn, we'll have those
numbers.
As with other group meetings, the format would probably begin with
social time then move into a specific topic for 20-60 minutes and ending
with more conversation time. (We'll look into sponsorship for the
obligatory stack of pizzas to precede presentation.)
When: Once a location is arranged, we'll set a date. There is a
possibility that the first meeting could be the last week of June but
probably July. Summer always introduces conflicts on people's time, but
this is an opportunity to get the bugs out. (As with the Vancouver
meetings, I'll record my sessions, so people won't miss anything.)
Who: I plan to present at the first few sessions. This isn't so much to
set the pace or tone but to ensure that there is content and to build
momentum.
Topics: While there is LispBox, which does a good job of getting you
started, my first presentation will be to move beyond the learning mode.
Likewise, since installing a free common lisp system involves more
than extracting a tar file or even knowing tweaky 'configure' options,
the first session will probably be this: MacOSX + sbcl + Emacs + Slime
with VirtualPC running FreeBSD (aka, fixing the reddit.com model).
A subsequent session might be Apache with mod_lisp (specifically
different than the web project in Peter Seibel's book, just to give you
options).
By the way, I'm a Unix systems programmer by trade, and by that, I tend
to do software systems integration involving a multi-server
architecture. For more on my Lisp background, follow the links at the
above URL. This gives you somewhat of an idea of where my presentations
will originate, but of course we're never just the sum of our jobs...
Thank you for participating!
-Daniel
