Forum OpenACS Q&A: Migrate Vignette to OpenACS
I work for a medium sized company that uses Vignette. It was purchased long ago in a different era and now we are looking to migrate to another solution.
We have thousands of TCL templates which run our site and recreating that functionality on another platform is prohibitive.
1) Has anyone attempted a Vignette to ACS migration? Is it worthwhile to consider this route?
2) Is there a wrapper library that recreates or bridges the Vignette API over ACS? If we had the source for the Vignette API, how difficult would it be to write a wrapper lib in the groups opinion?
3) I understand ACS has an advanced data model. We have a homebrew data model which is fairly complex as well. It seems like it shouldn't be too difficult to adapt our datamodel to the ACS way of doing things, however we are a "community" site and make heavy use of our member tables. Is it possible to extend the member table or use a table with a foreign key to hold other fields?
4) A lot of what we do with Vignette is Content Management. I know this is and open ended question, but how feaure full and robust is the OpenACS CMS?
5) Is there an upgrade path from OpenACS to the Java version of ACS? Management here likes Java, so an upgrade path would be a selling point. Is the RedHat CMS/Portal server even being developed anymore?
Thanks for your patience and help,
I can not address all your quesitons, but it sounds like one of your needs overlap with things that are being worked on for .LRN that will be flowing back into OpenACS over the next couple of months/weeks.
So let me take the first stab at your questions...
1) I have not heard of anyone doing it, but OF COURSE it is worthwhile (doubt you will get a different answer asking here)
2) Leaving #2 for someone else, but this would be a great contribution and it sounds like something that would draw collaborators quickly
3) There is work on site registration tools and data collection tools going on right now in OpenACS that will be used in .LRN to collect different kinds of data related to individual users and displayed based on roles within the system (e.g. in .LRN a course admin might want to ask some questions and sort users based on those questions, e.g. at the time of registration on a site most organizations want to collect unique user data and display it in different contexts in the future). The assessment package will be the data storage point for this kind of information. http://cvs.openacs.org/cvs/openacs-4/packages/assessment/www/doc/ is the source of the most recent documentation for that part of it.
4) OpenACS CMS (the package) is not being pursued right now in the community, but pieces key pieces of that package have been extracted/reworked for content management use cases. I am not on top of this part of the toolkit right now, but the last time I heard bcms/bcms-ui-base/bcms-ui-wizard (the combination of these packages) was the path the OpenACS community was going down. Maybe someone will fill us in on OpenACS cms tools in this thread?
5) Is there an upgrade path? I doubt it, but if you have a will I am sure there is a way. FYI: I have heard nothing (good) about the ACS Java version yet (other than management seems to like Java) and the last time I looked at the website for any signs of activity I ended up complaining to RedHat for including case studies that are actually running on OpenACS (they have taken them down in the meantime I assume). Update: You motivated me to take another look at it (it is hard not to tire of the Java line). It seems "byline" is what they call the ACS Java version now (the version Redhat is not making money on writing individual customizations I assume... i.e. maybe an alternative upgrade path if your management has the cash in the future?).
1. Not much on the applications side of things yet: http://wiki.objectweb.org/byline/Wiki.jsp?page=Applications so upgrading any time soon might be hard.
2. They are pretty bug free: they only have two open bugs and a single closed bug (although they do point to redhats bugzilla for "core bugs" rather than this third party project hosting solution that they are using)
3. There is not as much going on in there forums (12 posts since September?)
If you want to chat with me about byline any more you can join their freenode IRC channel (I would appreciate it... I am all by-myself on the channel they list on their site right now)
My synopsis for your management: the people using and creating the software is what makes or breaks it, not the language (and things are going well in our universe).
On behalf of the OpenACS community: welcome (or good luck finding something equivalent)
I would strongly suggest you simply do the minimal work to port your Vignette Tcl to vanilla AOLserver first. Then, determine what value you'd get out of OpenACS, vs. just continuing to run and operate your Vignette apps under AOLserver alone.
I have plenty of Vignette experience (heavy R4.0 through V/5 5.6) as well as a fair amount of AOLserver experience.
If you want to discuss things in more detail privately, feel free to email me -- firstname.lastname@example.org.
However you decide ... good luck!