On Jul 28, 5:13 pm, Rey Bango <[EMAIL PROTECTED]> wrote: ... > plugin authors, a chance to see if anything breaks. And the code is in > SVN so there's no reason for someone to not be able to test it.
That's a key phrase, "to not BE ABLE to test it." Certainly authors are able, but we must recognize that most plugin authors write a plugin to scratch their own itch, and as long as it works for them, they have little incentive to go out of their way to make sure that it works for others. i think we're all guilty of that at some level. > As with any development effort, testing is the part that everyone hates > but we really need you guys in the community to pound on the early > releases to ensure that your code doesn't break. i'm just playing devil's advocate here: as a developer, if "my" code "works for me", where is the incentive to spend my energy testing use cases which i won't personally use? The point is that nobody can *expect* any developer to continue to re- test their code under conditions which they don't personally use/ experience. e.g. i don't own a copy of Windows and cannot/will not test using MSIE. Nor will i install Opera just to make sure that my code works for the 1% of users who choose to use Opera. If an Opera user wants to test it, great, otherwise it won't get tested. > The jQuery team is > small as it is and we like it that way because it allows us to have a > streamlined approach to things. And that team's only "responsibility" (interpreted lightly) is to make sure that their code works "as advertised", and not to ensure that every (or, for that matter, ANY) plugin works with their core code. > we need to continue to rely on the plugin authors and > the community to test things out. With so many plugins in the repo, its > just too big of a task to test all of them out. Amen, borther. But i think people expect too much of plugin authors in general (i.e., take too much for granted). They (the authors) write a plugin to scratch a personal itch and then release it for others with the same itch. We cannot *expect* the authors to then be pro-active about checking against newer jQ releases. Of course, many are quite active/pro-active, but we cannot *expect* it, nonetheless. Am i going to go back and test my plugins against every new jQ release? Only if i happen to be using that release somewhere. And i think most plugin authors take the same approach. Perhaps i'm splitting hairs or arguing over semantics here, though. i admit that i have a hard time viewing the world through a normal user's eyes, as 98% of my time is spent in hacker-land, and coders and users often have much different expectations. :D
