discussion about a possible enterprise edition of Lua with support services

["Ray Garcia" <rgarcia@...>]

> -----Original Message-----

> From: Luiz Henrique de Figueiredo [mailto:lhf@tecgraf.puc-rio.br]
> Sent: Tuesday, November 12, 2002 5:17 PM
>
>
> We have indeed considered a model that we would still distribute
> Lua freely
> in source form as usual but also provide an "enterprise edition",
> which would
> be very easily installed and also include more libraries. We have also
> considered making ourselves available for custom support and consulting,
> but we don't know what this would entail exactly.
>
> I liked Rebol too, except that it's not free. Perhaps we should look into
> making a Rebol alternative based on Lua. Perhaps it's pretty easy already,
> being mostly a question of putting together existing libraries.
>
> Thanks again.
> --lhf
>

[response from Ray]

I think Lua has a lot of promise and a potential market it can pursue.

- I suspect that current Lua users are not necessarily representative of a
larger market.  You should certainly gather opinions from this group but
keep in mind that they are very likely to be much more technically competent
than the typical technology consumer.  I my opinion while you will still
need to support people who are proficient in 'C' and want to use Lua as an
embedded scripting language, the larger market are people who want to create
a complete application by scripting the solution.  This market potentially
includes lots of people even ones who are not normally programmers.

- The enterprise edition would need to be packaged first before you would be
in a position of charging for support.  I would think that you could create
a package with Lua 5.0 and lots of the addon utilities that you have on the
site already.  This would include a full library of optional modules (i.e.
OpenGL, Sockets, GTK, Database Access, whatever), plus it needs to include
some kind of Lua IDE.  Of course all of this would need to be brought up to
the same level of quality for consistency.  You would need to solicit
permissions from all of the people behind the addon products as well.

- Once you have an enterprise edition you can model yourselves after the
JBoss.org effort.  What they do is give away the software then charge for
support.  They have a range of programs the highest being an annual support
contract for Corporations.  They will do custom programming and training as
well.  You could add to this an embedding service for anyone who wants to
use Lua in this manner and requires that you support custom libraries.

- To start the support effort you can take an open community approach.
Anyone who has shown the skills to create a Lua add-on can sign up.  They
agree to Lua community governance and to make themselve available for paid
projects.  Each person takes a part of the planet to support.  If you have
more than one person in a location then you get them to agree to split the
work whenever it comes up.  This is to get started.  If this is successfull
then you will need to certify people before they can use the Lua brand
designation.  Basically what JBoss does is establish a trademark and
branding that only approved certified consultants can use.  You would do the
same with the Lua brand name.  Anyone can still publically support the open
source free version of Lua they just would not be able to claim they are
doing this on behalf of the Lua community.

- Now this is key.  You need to distinguish between giving support for
non-commercial purposes and support to Corporations.  Keep Lua free even for
commercial purposes.  The difference is that if someone is profiting from
Lua they should not get free support thus profiting from the work of others.
I'm not talking about a small startup programming shops, these need free
support and are probably struggling coders.  The business that should pay
for support are large Corporations that are not in the software business.
If any very large technology companies use Lua they should pay for support
as well.  Everyone is welcomed to use the public forums and mailing list for
support and the Lua team will continue to participate as well.  The paid
support is done much faster.  Free support comes as a best effort when
people have time to get to it (just like it is today).

- I beleive the Lua is very close to being a Rebol alternative if you
leverage the work that was done with the Yindo project.  They have nearly
all of the support that Rebol has except XML, database, Gui widgets, SOAP
Client/Server.  Yindo is just as fast as Rebol yet much more open and easy
to code.  You already have a CGI version of Lua.  I just found out that the
Yindo project is coming to an end this year and will not be supported.