Re: Relative Links Fail on TW 2.2 and 3.0 using Windows 2000, 2003, 2007
To add to what Marc said...
"I feel your expectations are not in line with the context."
The context is that the project/product is advertised as compatible with Oracle and Windows servers, and none of the developers use either of these. Your development team, whether volunteer, martian or monkeys typing, should claim only what has been tested and is known to work. That is not at all out of line with decent expectation lest someone using Windows servers and Oracle be lured in to waste thousands of hours trying to get a product to work...
My perception is that database/OS compatibility was more complete in earlier Tiki releases, but gradually the code has gotten more MySQL-specific due to the interests of the people involved, and the feature lists (especially on other sites) haven't been updated to reflect this.
"You are just as responsible as everyone else to solve this problem."
That might assume I am a programmer able to define and make the change. Are the people using this 'project' only supposed to be programmers? If so, mention that in guidelines for use. I am alerting the community to a problem, and to me that is a rational action and the way in which I can probably best help the community, rather than hacking code that I don't belong in.
Of course the great majority of users aren't programmers, but I think Marc is talking about the process in a community of volunteers who, if they are programmers, mainly are already very busy with their own todo lists. The project benefits by this coincidense of needs and solutions. People with needs, who aren't programmers, then have the option to...
"Hire a consultant."
Who if they are allowed to make changes in code (as everyone is) might have motivation to smear it so that their services could be worth more to more people.
I can't imagine this scenario happening. Anyone who would be that malevolent probably has bigger fish to fry, and anyway code contributions (I assume the fixes would be contributed to the code base) are watched carefully enough that "smears" would be spotted.
"Who says this is a Tiki bug? It could very well be a Microsoft bug. Have you complained to Microsoft because the program is not working like it should?"
Inconsistency and a lack of standards seems to have led to this issue. Well-formed code would not have the same inconsistency.
Though you say you aren't a programmer, if you are aware of problems in the code (or if anyone is), if you report it specifically as a bug, that's a good contribution to solving the problem.
"About misleading claims:
It obviously works for some people. I regularly install Tiki with XAMMP and it works very well."
I don't doubt it works in particular environments. List those that have been tested in the product claims. Not those that have not. This is responsible reporting that will keep people from expending inordinate amounts of time trying to get something to work which has overt claims of being proven, when it has not.
Personally, I think the supported platforms on some of the wiki-comparison feature lists have become inaccurate over time, which is more a matter of negligence in keeping them up to date rather than anyone being willfully misleading. But apart from those lists, I'm not aware of "overt claims of being proven" that Tiki will run flawlessly on IIS and Oracle.
There've been efforts recently (introduced by Marc, mainly) to either bring support for these other databases and OSs up to speed or to accept that there isn't sufficient interest by present and would-be participants to do that, so this is an issue that the dev community is aware of and working on. The PostgreSQL community was approached, for example, and there have been results (code commits) from that. Unfortunately, not much happening about IIS as far as I know, which may be telling.
-- Gary
;){kind=link}
;){kind=link}