To the keepers of the holy grail of Open Source

Ben Tilly ben_tilly at hotmail.com
Fri Jan 19 18:26:34 UTC 2001


Jorg Janke <jjanke at accorto.com> wrote:
>
>I would like to raise three issues:
>a) License issues
>b) Compiere license
>b) Open Source Trademark
>
>a) General License issues
>---------------------------------------
>- I am a bit frustrated about the process; I had to submit our suggestion
>three times before receiving the first feedback.

It would help if you sent mails as regular ASCII rather
than formatted HTML.

This is a question of knowing your audience.  HTML can
be made to look nice, which means that it will go over
well with suits.  However it suffers from bit-rot,
displays differently on different platforms, is more
complex for standard text tools to process, etc.
Therefore technical types tend to see HTML email in
the same category as Word and Powerpoint attachments -
a sign that the sender is not technical and does not
have a clue about how the technical community works.

So say it in ASCII.  It may not look as pretty to you,
but it will go over a lot better with us.

>- If you only want to deal with the Fortune 1000 - please say so.

Actually the Open Source community has a lot of people
who are very distrustful of the standard Fortune 1000.

>- The current "ready to use" licenses available are inconsistent and there
>is no guideline when to use what

They are not consistent with each other because they are
trying to address incompatible goals.  What guidelines
would you like?  Any guidelines are opinions based on
current goals, ideology, and position.

>- If you want to take the Open Source License seriously, but don't want to
>deal with the "little guys", I suggest you come up with some templates - or
>do something like eTrust
>http://www.etrust.com/webpublishers/pub_resourceguide.html

Actually the continuing proliferation of licenses is
generally agreed to be undesirable.  People seem to
want to create licenses that protect themselves in
some way.  But if you take code licensed under two
different licenses designed that way, you usually are
unable to borrow.  That reduces the ability to use
opened code, or to easily add interesting functionality
to it.

My strong advice would be to carefully consider what it
is you wish to accomplish and use an existing license.

>- I don't think that people would mind the alternative: use one of the
>following templates or pay a fee for us looking at the license - assuming
>that there are templates available and some guideline for non-lawyers when
>to use which.

Templates for stamping out licenses by the dozen would be
a very bad development.  But it would be good if the
"common knowledge" about what licenses are good for what
were gathered in one place rather than having to be
discovered through experience.

>b) Compiere License (www.compiere.org/license.txt or
>www.compiere.org/license.html)
>----------------------------------
>- The feedback/critique I received was valid, but my reply unanswered.
>- Could you please tell me if I should just forget about it, come back
>later, .... ?

Make your license straight ASCII.

Also I would suggest reading

  http://www.gnu.org/philosophy/bsd.html

and then thinking about your license carefully.  You
are basically trying to trade software for branding.
Unfortunately this is a model that quickly becomes
prohibitive when you share code.  As a result even if
you get the official OSI label, you will find
relatively few people are willing to use your code,
and you will be on your own for deciding what code you
can use.

If you use an existing license then at the least people
will be able to give you advice on what code-bases you
are able to freely borrow from without having to study
your license closely.

>c) Open Source Trademark
>--------------------------------------------
>- As you know, the 'characteristics' of Open Source projects are very
>different.  Recently, there are quite a few companies using the Open Source
>as marketing tool (in addition to the failed commercial projects)

The failed commercial projects are not as failed as you
think.  Relative to the stock bubble silliness they have
come down.  Relative to the estimates that an intelligent
person could have come up with based on the literature at
the time?  They are doing quite well for companies with
their real size and history.

Many other companies moved to an open source model when
it probably was not the greatest fit because of the
bubble.  Many of them are in serious trouble.  OTOH the
same is true through the entire .com sector.

>- I think, you guys need to come up with some guidelines on 'ethical' Open
>Source projects. I realize that there is a fine line ... and Tim O'Reilly
>would not support an Open Book Source project similar to the original
>Napster.

I believe that OSD was intended as such a guideline.  And
Tim O'Reilly is open to experiments, and indeed has tried
some open source books.  Whether or not to do that is up
to the author.  More than one example exists where a
project did the O'Reilly book and also turned the
technical work into online documentation.

>- I suggest that you come up with some guidelines (and even approval
>process) to separate the "good" from the "others".

I thought that that is what the OSI was supposed to do.
They have been fairly low-key though.

>- Some of the major points, I see:
>	- Does it need to compile (out of the box in defined environment) ?
>	- Does it need to be installable (out of the box in defined
>environment) ?
>	- Can it rely on or requite products only commercially available ?
>	- Can you charge a download fee ?
>	- Would you be able to use the product without paid support ?

The OSD covers several of these.

>- To test your criteria, you should have a look at www.opensales.org - try
>to install it, try to do something with it (you can't get support from
>www.opensales.com if you have not paid for the estimated $100,000
>implementation project)

They recently changed their name to Zelerate, and the
latest round of funding came with 40% cuts and the
decision to move to a business model which I think
will be a disaster for them.

You are allowed to try to sell GPLed software for
anything you want.  It generally doesn't work to well.
The fact that there already is a fork from an earlier
version on SourceForge does not bode well for them.

>- Another test criteria would be Compiere. Our intension is to make 
>Compiere
>THE ERP system for the little guy with 50% market share - by giving it away
>and offering support for the not so technically/functionally inclined.  
>But,
>our credo is, that it has to work out-of-the-box without limitations or
>outside help - Our targeted guarantee: "up and running 2 hours after
>download or shame on us". We see the main acceptance in third world
>countries (who would not be able to pay anyway) resulting in an active user
>and developer community.

I know nothing about your target market, except for the
fact that ERP systems do tend to require very substantial
support.

Cheers,
Ben

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com




More information about the License-discuss mailing list