Forum OpenACS Development: "Good" vs "Bad" modules?

Posted by Hawke R on 10/29/01 05:47 PM

Has anyone done any kind of organized eval of the well written vs
poorly written modules?
I've heard pretty consistently that the CMS and Workflow parts are
well written, but that, for example, some of the eCommerce ones are
very poorly written.
Has anyone create a matrix, or spreadsheet ranking each from 1 to 10
or something similar?
Based on code quality, documentation, scalability, etc.
Thanks,

2: Response to "Good" vs "Bad" modules? (response to 1)

Posted by Don Baccus on 10/29/01 07:58 PM

That's a good question. I'm slowly starting to form my own opinions as time goes on but haven't dug into everything.

There's a related (and more primal) question - which packages are more-or-less finished and which are so unfinished as to be minimally useful or worse? Incomplete packages can be well-written, just not done, and the haste with which aD pushed out ACS 4 Tcl led to unfinished work being released. Then the haste with which the plug was pulled in favor of ACS Java meant that some never got complete.

If you or someone else wants to organize a way to track feedback on package quality and completeness, that would be useful IMO. We could use the survey or poll packages here on openacs.org ...

3: Response to "Good" vs "Bad" modules? (response to 1)

Posted by Jerry Asher on 10/29/01 08:19 PM

I would like to see a section, similar to the aD acs-repository, that lists the various modules.

It would be pertinent to display the number of downloads, as well as the results of a poll (have implementors found this module useful...) Better would be a small chart for each module displaying the poll results over time and therefore indicative of module progress or stagnation.

As everyone agrees, there are many dimensions to poll folks on those above as well as: ease of use by the user, ease of use by the admin.

4: Response to "Good" vs "Bad" modules? (response to 1)

Posted by Don Baccus on 10/29/01 09:00 PM

We'll implement the download package on the new, OpenACS 4-based openacs.org site. That will let us rationally break up the tarball. I have to admit it's not my highest priority, however - fixing bboard queries that take 40 seconds on 70,000 entries and things like that are at the top of my list.