Gre7g Luterman <[EMAIL PROTECTED]> writes: >> What we want to avoid is adding code to support every format of >> domain or wildcard out there. It's much more sensical to support >> one format, and then have files in other formats translated before >> TMDA sees them. > > But isn't that the point of open source... so that everyone can > modify the code to better fit their needs?
That's part of it, yes. > Obviously I *CAN* modify the code since it is open source, but > without it being accepted, I'd have to make the mods every time I > upgraded. Is there a fundamental reason not to embrace every > standard that someone wished to program in? Is there a benefit to > making other applications have to change to suit TMDA? As maintainer, I can't blindly accept every addition that someone proposes. I have to weigh the benefits of adding that new feature with its costs. Those costs include more code (meaning slower execution time), more documentation, and more support. All these activities are time consuming, and the more of them there are, the less time us developers have for developing. Software is a chaotic system like weather, where each new feature is analogous to that butterfly flapping its wings in Tokyo which causes a hurricane in Miami. I think any good piece of opensource software must maintain a certain degree of discretion if its quality is to remain high. If you succumb to feature creep as Microsoft products (and recently Linux) tend to, you end with up a bloated and buggy product. A veritable ``zoo without a keeper''. I don't like to say ``no'', but occasionally this is necessary for the good of the whole. So, if I see something proposed that can be solved relatively easily using other tools in conjunction with TMDA, I probably won't reinvent the wheel. This synergy between applications embodies the UNIX philosophy. Each tool does a few things very well, instead of many things only mediocre. If you think of a solution as a sum of its parts, you can understand why this methodology tends to work much better. Another reason for my not choosing to include something would be if I don't see a general need or use for it. Of course, I'm often wrong, and I don't know everything (yet :), so feel free to disagree and continue trying to change my mind if you feel strongly about something. Those who have followed this project for a significant period of time will know that I've often been made to ``see the light'' after a healthy debate. Anyway, I haven't rejected your proposal, we're still discussing it. The `from-program' source I mentioned I think would fit your needs, while still being general enough to work in other situations as well. Hope this helps. _____________________________________________ tmda-users mailing list ([EMAIL PROTECTED]) http://tmda.net/lists/listinfo/tmda-users
