Date: Sat, 1 Jan 2000 13:35:51 -0600 From: Karl Denninger <karl@Denninger.Net> To: Will Andrews <andrews@TECHNOLOGIST.COM> Cc: freebsd-ports@FreeBSD.ORG, Steve Price <sprice@hiwaay.net> Subject: Re: ports/15822: Update port misc/HomeDaemon to V0.99 Message-ID: <20000101133551.A7428@Denninger.Net> In-Reply-To: <XFMail.000101142305.andrews@TECHNOLOGIST.COM>; from Will Andrews on Sat, Jan 01, 2000 at 02:23:05PM -0500 References: <20000101130202.A50949@Denninger.Net> <XFMail.000101142305.andrews@TECHNOLOGIST.COM>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Jan 01, 2000 at 02:23:05PM -0500, Will Andrews wrote: > On 01-Jan-00 Karl Denninger wrote: > > I'm perfectly fine with you Steve - while you may think I "went after" you, > > I in fact didn't (and didn't mean to). Will, on the other hand, I did - he > > categorically rejected design decisions I made for *very valid* reasons, > > without taking the time or effort to understand them (and the reasons ARE > > there, in the package's README file), and when I tried to explain again > > I got no reply back. > > I'm sorry, I didn't know about these design decisions you made. You could have > simply mentioned these in your original reply to my thoughts based on your > PLIST, and avoided this entire episode. I _AM_ human.. I did. That was the followup you didn't respond to. > The fact is, however, I did reply to your original reply. Maybe it didn't go > through? Stuff happens. The entire litany of how it works and why is in the README file (along with more detail than most people want on this stuff) > > This port may in fact need NO_PACKAGE set in the Makefile, at least for > > now. I suspect that building a package is probably a bad choice due to > > the fact that as things stand (as of 0.99) there is a compile-time switch > > that has to be set depending on the version of code you have from Dan > > Lancini (the guy who wrote the interface piece that this package needs > > to talk to the CM11A). If you're considering committing the port you > > might just want to go ahead and make that addition - I suspect its > > the "right choice", at least for the time being. > > If you have a compile-time switch, couldn't you allow an override through a > runtime switch (like through getopt())? Just another thought. Getting it set wrong fails in a fairly subtle way and there is no way to test it on start-up, so there's no way to KNOW you have it wrong other than some really screwy things happening (in particular, with the RAMP command). Once I got the report of trouble I was able to identify why it was happening relatively quickly, but the guy who reported it had no idea - and, I might add, the problem was flooding his powerline with repeated commands that were doing nothing. Not good at all. Putting a command-line switch in there is asking for it. Yes, I could, but I'm not convinced that its a good idea. Dan Lancini's driver still has to be compiled by the user, so forcing them to compile and THIS (as opposed to loading it as a package) is not really onerous - you can't USE this package without Dan's (at least not for anything X10 related) so there is little incentive to do that. If you're not "up enough on things" to compile the package you're not going to be using it anyway. That, by the way, is why the PLIST stuff is nonsense in this particular case - packaging the port is pretty worthless as it stands, since you have a dependancy that I can't rationally include AND that dependancy is NOT AVAILABLE in a binary format. Again, all this is clear if you actually examine the package, what it does, and what it needs - all of which is documented in the README file. > > BTW, the "submit followup" button is worthless to me on the web interface > > to GNATS, as I don't do Microsoft-style mail things, and the "mailto" link > > tries to reference something (mailto:) that I don't have set up. > > You can, if you are just going to reply in conversational form, just use your > reply function in your MUA. There's really no reason to use that "submit > followup" form Steve suggested, but it's there. Assuming the conversation takes place in somewhat-close-to-real-time, that's reasonable. If it doesn't the old message will likely be gone from my mail spool. -- -- Karl Denninger (karl@denninger.net) Web: http://childrens-justice.org Isn't it time we started putting KIDS first? See the above URL for a plan to do exactly that! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000101133551.A7428>