EccoProWiki: OtherOpenSource

Wonder if perhaps NetManage? (and/or others) might want to consider Source Forge where a respected and vibrant open source development community does just what we want, update and improve a great program that has tatters around the edges due to operating system creep..

Richard

Instead of open sourcing ECCO through SourceForge?, another consideration might be that of donating the code to an existing open source community in dire need of a PiM?.

OpenOffice?.org is the largest open source community yet assembled. They have developers and engineers working everything from the desktop productivity environment to devices to server side collaborative computing initiatives. And if ever an open source community had users, it's OOo.

OpenOffice? and Mozilla are also at the center of most cross platform alternative desktop efforts. Dropping ECCO into their lap would guarantee as wide a deployment and participation as possible. The Mozilla community would be another good choice. And given the problems OSAF has had getting Chandler out the door, Mitch Kapor might also be interested. It's also true that when Mitch first announced the Chandler project, he was run over by a hungry herd of ECCO lovers.

The problem with the SourceForge? route is that simply opening up the code is no guarantee that the engineering resources will show up. However, if Netmanage drops ECCO into OpenOffice?.org, the excitement will deafening. The world will hear this roar.

OpenOffice?.org is at the core of quite a few commercial desktop and systems distributions. Novell Office, Sun's StarOffice? and Java Desktop, NeoOffice?, Mepis, Xandros, Linspire, Progeny, the list goes on. IBM recently released their Lotus integrated stack, and at the Workplace desktop core is OpenOffice?.org and Mozilla.

Recently the European Union issued a requirement that all documents and information based transactions involving government resources or government business be in the OASIS/ISO Open Document XML file format.

 This is the XML file format OpenOffice?.org donated to OASIS to become

the foundation of a universal file format. Interestingly, Microsoft came to agreement with the EU based on Sun's offer to write the XML filters (with Microsoft's help and permission) so that Microsoft's proprietary XML file formats can automatically be converted to the OASIS/ISO universal standard.

Through the efforts of the OpenOffice?.org bibliography project, the file format is well on it's way to becoming the standard for the Library of Congress, the Australian Historical Archives, the entire legal industry, and governments the world over. It's the first universal file format standard to be able to accommodate the multitudes of content, data, graphics and streaming media types that make up modern compound documents. Including the kind of embedded scripting demanded by business processes, intelligent forms, and intelligent document workflow routing.

The file format is compliant with the entire movement of open XML technologies, including XForms, SVG, SMiL?, XML events, XHTML, XSLT, Relax-NG and on and on.

The situation at OpenOffice?.org is that there is a tremendous cry for an an integrated PiM? - Project Manager that could become the foundation of the many collaborative computing efforts surrounding the OOo and Mozilla core. Many thought Chandler would be the answer, and maybe someday it will. It certainly is an exciting project. But we need a solution today. And we need a solution that can be engineered into the OpenOffice?.org – Mozilla environment. Not one that is self contained and distant from the rest of the productivity components.

Perhaps most importantly, OpenOffice?.org is dedicated to cross platform implementation and compatibility. There are many Windows based distributions where one can install an entire OSS productivity environment based on the OpenOffice? - Mozilla core. The OpenCD? Project, the MozOO? project, Linspire's OOFf, Ubantu's LiveCD?, the GNUWin CD Project, and The WinCD? Project all offer different Windows productivity environments that have one thing in common. They all feature OpenOffice?.org and Mozilla at the core.

Oh. And one other thing – they all need ECCO Pro.

I also think that the OpenOffice?.org community would make quick work of getting ECCO to run on WiNE?, making it cross platform from the gitgo. Integrating Thunderbird eMail, IM, and document management into ECCO might take some time, but i hope you'll consider the OpenOffice?.org and Mozilla communities as a good home for ECCO.

~ge~

As Gary Edwards pointed out, Openoffice.org is a natural place to look for inspiration for and possibly close collaboration with an OSS Ecco project. In particular, take a look at http://www.openoffice.org/contributing.html for many of the ways that both techies and non-techies can contribute to a serious OSS product.

One way that folks on this list could begin contributing right away, is to collect and catalog a "wish list" of Ecco improvements (probably most of these have been expressed on the list many times, in the vein of "if only Ecco could ..."). Maybe eccorocks or some similar site could host a Wiki or other collaborative software to hold this information and support this effort.

Also, as a techie and Ecco user since the early '90s, I'm definitely interested.

Don Dwiggins

While I agree a vibrant engineering environment is desirable for Ecco Pro, I'd just caution that if thrown into a place where there already are so many cross-currents, it might be changed in a release in a manner so to make it unrecognizeable. I would very much oppose such a change to the interface or adding on really unnecessary interfaces and links.

Obviously, some would be welcome, but "updating Ecco Pro" is bound to be a delicate undertaking.

 -jtg

I agree. Making it an independent project under the SD license would allow OpenOffice? and others to make use of the code if they wished but retain Ecco's independent identity.

Soulhuntre

What license are you referring to by SD? Did you just mean one of the Sourceforge approved licenses?

To my thinking, there are a couple of different factors that are important here. One is how and by whom the main project is administered. The other is the license.

With regards to administration, this has a great impact on the timing of release cycles, how code contributions are integrated, the order and priority of development, etc. This hosting location is where the community develops and how the evolution of the product is (loosely) controlled. My guess is that there would likely be at least two different camps regarding what direction the code should evolve, and I would predict a rather early fork of the project. On one hand would be a "conservative" camp that wishes to simply fix a few of the limitations that have been found in Ecco over the years, and will place a high premium on maintaining the rock solid reliability. There would also likely be one or more "liberal" camps that would want to integrate the capabilities with Open Office, Chandler, or whatever. These would likely be more radical changes, longer release cycles and would take time to return to the same level of quality. (By the way: for those not familiar with open source licenses, any provision in the license that would prevent a fork in the project would mean that it is not truly "open source".)

The second issue is the license. The Open Source Initiative (OSI) lists a large number of licenses that meet their criteria for being "open source". Sourceforge lists projects that use 54 different OSI approved "standard" licenses, 990 projects that use "other/proprietary" open source licenses, and 1,561 that are "public domain". Of the OSI approved licenses, I simplify my view into two groups. The first I would call "truly free" such as Apache, BSD, and MIT. The other I would call "Puritanical" such as GPL, LGPL, MFL (Mozilla), etc. Of these alternatives, I see the Apache example as having a lot of value, especially if the "Ecco" name were to be protected in the same way that the "Apache" name has been protected. This would allow "Ecco" to remain Ecco and follow the conservative "fix the resource limitation, but don't break anything else" camp, and would allow OOo to use the base code and features to evolve an "Outlook killer" under the Open Office name. [When I say "Puritanical" I am admitting that this is somewhat of a religious issue, and I think that the example of the Puritans is appropriate. Being from Rhode Island, I follow Roger Williams' lead and side with true freedom and find the "Puritans" to be somewhat hypocritical.]

An interesting thing I stumbled over at some point is that Ecco appears to already contain some MIT licensed intellectual property. Certainly interesting as an anecdote, but also illustrative of some of the problems that may come up as NetManage? looks at moving the code to a different license. All of the existing licenses would need to be honored and may prohibit the code being re-licensed without modification. A good historical example of this is Mozilla. When Netscape opened up their source code, they needed to do significant amounts of development to replace code that they had licensed from others when the owners of those licenses were unwilling to allow their code to be open sourced. In order to open the Ecco code, all of the licenses for code used by Ecco would need to be inspected and possibly re-negotiated. This could be very expensive for NetManage? and might be impossible to accomplish without leaving holes in the code. It might turn out that the first task would be to fill in the holes just to get back to where we already are... [BTW: This whole morass is a good example of why I think that the highly restrictive GPL is not a good idea... I rarely even look at GPL code because in the true spirit of the license, anything I learn (even at a subconsious level) from looking at that code could then only be applied to code that uses the GPL license. That would be an unacceptable restriction for someone who makes a living from software development... So where's the "freedom" in that?]

So my inclination would be to recommend a Sourceforge type hosting and an Apache type license that keeps the Ecco name with the original project. And good-luck to NetManage?, this might be more difficult than many observers imagine.

My $.02 Dave G.