I wanted to leave a comment in this Bill Dollin’s post, but after that it came out to my mind to leave a trackback to it, and going with a my own post on this subject as far as I have several things to discuss about.
In the OSS jungle, it looks there are in the last times more and more solutions and projects based on commercial closed-code software. For example zigGIS, the Open Source ArcGIS connector for PostGIS, in which I am involved, is by itself an OS project tied to proprietary frameworks (Microsoft and Esri). This is many times not so good, as I cannot dig, for example, in Esri ArcObjects core code for understanding why a particular issue is coming out.
At my office we are in the process of taking an important strategic decision.
We are going to replace our actual commercial closed-source CMS (Content Managment System) - that we were using for some portals we manage for several years - for an Open Source solution.
There are many scenarios we have to consider, that could be mostly summarized with these points:
1) Our IT staff is mostly Microsoft-oriented (.NET and SQL Server)
2) We can afford hiring one or two consultants if we would go for a technology different from Microsoft (for example Java or Python), but it would be quiete time consuming to redirect our acutal IT staff in a different technology direction
3) We need an OS CMS supporting enterprise features like multisite
4) The OS CMS we are going to choose should have already implemented basic features we actually need (editors/validators scenarios, forum, blog, IM, versioning…), we have not time at the moment to implement by ourselves
After some digging in the CMS jungle - there are really hundreds CMS based on many many technologies (.NET, J2EE, php, Python, Ruby…) - we made a choice for a .NET solution, and another for a full OS Solution, the latter meaning something that can run totally with OS sofware, included the platform (for example Linux).
For .NET we selected Umbraco, a very interesting CMS environment, totally based on Asp .NET, SQL Server and AJAX. I like very much the way Umbraco is structured, but at another side it lacks - at this time - of several basic features we would need, so we would have to write several extensions by ourselves.
But what is my major concern, it is the Common Public License under whic Umbraco is released, that would mean, in the next future, to have plenty of usefull additional modules, but for a license fee for each module and without getting their source code.
As a full OSS solution we went for Plone, that looks by far the most powerfull OSS CMS out there. Plone, and its Python-based framework - Zope, is totally OS, and it has every CMS feature, even enterprise like multisite, we would actually need.
Even if i really believe that Plone is amazing, the main concern is that in our organization there are not Phyton/Zope developers, and it would be very difficult to turn our actual MS-oriented stuff into this amazing framework.
What to do then?
We have to take soon this decision, and we are in big concern about that…
Hi Paolo. I understand your concern regarding umbraco going to CPL but don’t be affraid. It’s correct that the reason is to open for commercial modules. An oppotunity for ISVs to build great, well-documented and stable add-ons as a supplement (not a replacement) for the range of often un-documented, half-done open add-ons that’s floading (there’s great and done add-on’s as well, but it’s quite often that open source add-ons really needs a lot of polishing).
This is something that should make you happy. It doesn’t make umbraco less open source (CPL is OSI-approved) and there’ll still be loads of GPL and free add-ons distributed I’m sure. But with commercial add-ons you can start to justify making demands.
The change to CPL will give umbraco an even better eco-system. It has been discussed in our forum, but all agree that there’ll always be open add-ons too. And that a commercial add-on only stand a chance if it’s well-documented, hard to make and very polished. For those simple add-ons there’ll always be an open alternative. And even making a commercial add-on, doesn’t have to exclude the option of getting the source code anyway.
Cheers,
Niels (founder, umbraco)
Hi Niels
first of all congratulations for the great piece of software you created.
Having said that, we would be more confident to go for a full OS CMS solution for .NET, but there is not choice in the .NET world.
Umbraco is great, but our major concern about it is that at the moment it has - for obvious reasons - only a little bit of CMS features we would need (versioning, groupware, workflow managment, events….).
I will let you know what will be our choice,
cheers, Paolo
Ruby on Rails
what do you think about dotnetnuke in .net world?
Have you tried http://www.mojoportal.com/ or http://www.rainbowportal.net/ ?
[...] few months back, Paolo posted about mixing commercial and open-source tools. I think this will eventually be the prevalent means [...]