Re: MetaCard (was: Re: Logo)

[Scott Raney <raney@metacard.com>]

On Mon, 16 Aug 1999, Ian Bicking wrote:

> Scott Raney <raney@metacard.com> wrote:
> > > Or, is it to say that you could create the application, but that users 
> > > could only use it under the terms of the Starter Pack, i.e., with a 
> > > limit on the length of scripts, etc., before they have to pay for a full 
> > > version?
> > 
> > It would be limited by the MetaCard Starter Kit limits, yes.  As I
> > said before, the package I was envisioning would basically be
> > HyperStudio with no HyperLogo (or more precisely, only very limited
> > scripting).  But scripting certainly wouldn't be required to do
> > anything most people do with HyperStudio because navigation in that
> > package is all based on data structures, not custom scripting.
> 
> How are the Starter Kit limits defined?  Is that a hard limit put in 
> the software, or a limit the application has to put in (to be in 
> accordance with the license)?  It seems like a reflective language 
> makes the issue a little funny (assuming MetaTalk is reflective -- 
> but I suppose it must be if it can write its own editor).

The limit is 10 commands in the script of any object.  That is, a
person with the Starter Kit can only edit scripts that have less than
10 commands in them (things like the "on .. end" handler definitions,
and control-flow operators like  "repeat" and "if" and "switch" don't
count toward the limit).  Starter Kit users can view scripts longer
than that (assuming the original developer didn't password protect
them), but they can't change them.

> Would computer-generated code be included in the limit as well?  I 
> doubt much would be necessary, but I can imagine that the 
> dynamic behavior of a HyperStudio-like stack might imply at least 
> a smidgen of code.  Well, maybe not... anyway, if it did, would that 
> count?

The limit also applies to script-generated scripts (it pretty much has
to because this is how the scripts are actually set in the development
environment, which is of course build in MetaCard).  So you can do
some custom scripting, but not much.  It shouldn't be necessary to do
any, really, for the proposed project because you can trap all of the
messages triggered by user interaction in the development environment
and then base behavior on data structures (properties set on the
objects the user creates), for which there are no limits.  Even if you
did need some scripting, it would most likely just be one or two calls
to prewritten handlers (either built into the package, or maybe
available as add-ons from other developers), something the Starter Kit
limits would allow.

> > > I'm not entirely sure what you are thinking about here.
> > 
> > The only unexplored issue here is getting the original development
> > team going.  They would get started by just getting MetaCard K12
> > licenses which would require that they be currently affiliated with a
> > school, or by getting a standard license via a grant from MetaCard
> > Corporation (which is usually pretty liberal about this kind of
> > thing).  
> 
> So you are saying that someone could (possibly) get a MetaCard 
> license for free to work on this?

Yes.  We do grants this all the time for projects that we feel are
important or to attract talented developers who otherwise wouldn't be
able to afford a license.

>  (How much does the K12 license cost?)

It's currently $250 for a 10-seat license.  But we're currently
considering changing it so that the K12 licensing would be done only
on a per-school basis.  The price would probably around the same
(except maybe for high schools which might be charged a little more),
but it would eliminate the pesky license-counting required now (which
most schools probably don't do anyway ;-)  If anyone has any ideas
positive or negative on this, or any other suggestions on pricing or
licensing policy, we'd be interested in hearing them.  We have no
full-time marketing staff on this project (even the summer intern we
had on the project is gone now), and so it doesn't get a lot of
attention around here.

> This would probably be important if it was done by volunteers, 
> which is what makes up most of the Linux community.  Even if, in 
> the big picture, a few licenses aren't that expensive, there's no 
> infrastructure to share payment of such things and it would end up 
> being a signicant block to the project.

Agreed.

> > Once the first release is made, some aspects of the program
> > could be changed by people using the free Starter Kit version of
> > MetaCard (notably doing language translations of the prompts or doing
> > minor customization of the UI like change the size, color, or
> > positioning of the UI elements).  Anyone with a K12 or regular
> > MetaCard license would be able to change any aspect of the program,
> > assuming the original developers choose to allow this (which I would
> > assume they would, at least if were an open source project).  
> 
> This is where there's a real challenge to fit it into the open source 
> development model.  Freely available development tools underly the 
> entire system.  I think something made in MetaCard could certainly 
> be free (as in beer), but it's hard for it to be free (as in speech) in a 
> meaningful way when a license is required to modify it.  In all 
> practicality, it is unlikely anyone outside of the main/initial 
> developers will make contributions.

I kind of doubt that.  The MetaCard community is still pretty small,
but there is a lot of sharing that goes on already.  And there are a
handful of K12 schools using it already, most of which have more than
one individual with a serious interest and experience with MetaCard
and an interest in K12 education.

> That doesn't mean it couldn't be successful anyway.  In reality, few 
> of the users of a HyperStudio-like application would even consider 
> changing the program, not to mention being skilled enough to.  It 
> isn't high on the priority of most teachers -- though there does 
> seem to be some significant (in terms of passion, not numbers) 
> teachers who do like to toy with the programs they use.  The need 
> to get a license -- even if a K12 license is cheap -- would probably 
> exclude them from this potential.

Possibly, although if they could afford HyperStudio, they can easily
afford a MetaCard K12 license.  And of course if they're really
serious, they could join the development team (assuming you all
wouldn't require dues or some sort of ritual right of passage like
having to set up a Linux server to prove that they're worthy ;-)

> It seems like it's mostly a strategic decision.  Even though one 
> would have to forgo most of the benefits of an open development 
> model, it would still probably be the easiest and most successful 
> way to make a HyperStudio clone.
> 
> This all seems very complicated, but I suppose the situation must 
> be quite simple compared with other similar dealings in the 
> proprietary world.  It's all foreign in an environment where the 
> biggest problem in aquiring and changing a program is slow 
> connection times and the occasional compiling problem.

It is different, I'll grant you that.  I don't know of any comparable
proposals in the software world, though probably what comes closest is
the grants many CAD/CAM software vendors make to university
engineering departments.  And their goals are similar to our primary
goal in this project, I think: community building.  The more people
coming into the market with experience with their particular product,
the greater their sales will be to companies that employ those people.
  Regards,
    Scott

> --
> Ian Bicking <bickiia@earlham.edu>
> 

********************************************************
Scott Raney  raney@metacard.com  http://www.metacard.com
MetaCard: You know, there's an easier way to do that...