Date: Sat, 01 Jan 2000 14:14:18 -0500 (EST) From: Will Andrews <andrews@TECHNOLOGIST.COM> To: Karl Denninger <karl@Denninger.Net> Cc: Steve Price <sprice@hiwaay.net>, freebsd-ports@FreeBSD.ORG Subject: Re: ports/15822: Update port misc/HomeDaemon to V0.99 Message-ID: <XFMail.000101141418.andrews@TECHNOLOGIST.COM> In-Reply-To: <20000101121727.A35145@Denninger.Net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 01-Jan-00 Karl Denninger wrote:
> You never retrieved the port, you never looked it, you took a look ONLY at
> the shar file and made your call - without examining the evidence.
The port can be viewed in the shar you attached. Perhaps you meant that I never
retreived the distfile? In that case, you would be correct.
In any case, I still think that where possible, hier should be followed. I have
patched many programs to force them to conform to hier through gcc -D, which is
why I suggested using #defines and such. (Note: It was JUST a suggestion, you
needed only provide a reason why that suggestion couldn't be used and the topic
would then proceed to a better solution, which can of course be your original
port!)
I didn't really mean to put this up as a "roadblock" of a sort, but rather as
something to point out. Like, maybe you forgot to take hier into account?
People are human, after all.. (and I've made many mistakes myself)
>> What's so hard about adding a few #define's as necessary and then using gcc
>> -DXX=\"blah\"? Maybe I just don't know how your program works. :)
>
> You don't. It has external dependancies, among other things, and further
> it has location-specific data that the user will be VERY unhappy to lose
> down the road. THAT was the reason for not putting it in "share"
> originally (to isolate both its DEPENDANTS *AND* the localized stuff).
Dependent software should be installed independently.. the ports collection
takes to doing things this way. If you have software that this program is
dependent on but is not part of this program's distribution, you should make
separate ports for those programs/libraries, and use the appropriate
LIB_DEPENDS, BUILD_DEPENDS, and/or RUN_DEPENDS. This way you can place them in
separate ${PREFIX}/share/* directories and thus isolate them.
> Fine. Your call, of course. The code will and does live without you folks
> - after all, it was never intended as a PORT in the first place.
Geez, if you're going to interpret my messages this hard, you need some valium
or something. My comments ARE JUST TRYING TO BE HELPFUL, and you are blowing
the whole thing WAY OUT OF PROPORTION!
If your program DOES INDEED _require_ being installed into a directory in
${PREFIX} that breaks hier, so be it. You will have to convince a committer
that it's for the best.
BTW, why do you keep saying "your call" when it's clearly not something for me
to decide? I am NOT a committer - get that straight!
> Gee, you're assuming I keep PR numbers around in some kind of database.
> I don't. I made a contribution; if the [Gods of FreeBSD Ports] don't want
> it, then fine - reject it and be done with it. If you do want it then commit
> it and be done with it.
You get replies from gnats-admin whenever you send in PRs. I would hope that
most people keep these until the PR is resolved. Else they can just search the
PR database @ http://www.FreeBSD.org/query-pr-summary.cgi?query. Simple as that.
Plus, the replies people send to you; you can just `reply' to them in your MUA
like you normally do to regular mail. I was just trying to provide an example
of when using your normal MUA might not be appropriate, and using send-pr would
be more appropriate (where MUA's don't preserve the content of your replies
correctly, for example, for sending a patch to your Makefile, which requires
tab delimitation in most cases).
[ all of the below is my reluctant reply to pure flamebait ]
> There are no misunderstandings here Will, and this little snit-like response
> of yours in that last line says EVERYTHING about what pisses me off on
> occasion (and has over the last several years) about the little band to
> which you belong.
>
> If you're not a committer, then don't reply and act like one.
>
> I'm not responsible for keeping an "annointed list" around my machine and
> checking off one or another box when I get a reply to something I send in.
> I refuse to do that - its simply not my place or position to do so.
>
> If you want to speak on behalf of that part of GNATS, then expect to be
> treated as if you have - you did, and were. A complaint rather than a
> "open->closed" reply says all that needs to be said. Holding a PR open
> without cause doesn't solve anything either.
I am not being a snit. I am trying to help you make a port that will benefit
others in the FreeBSD community. You seem to be taking my point of view
harshly, and your reaction is certainly not necessary. All I ever wanted from
you was an explanation why your port breaks hier. I never, EVER, not ONCE
_rejected_ your port! Merely, I was just trying to provide another point of
view, so that you could consider improving your port or defending your version
of it. You are not on trial here, remember.. you're just here to show us (well
at the very least myself) why your port as you've presented it is what should be
committed to the tree, and why it cannot/shouldn't be made to conform to hier.
The fact I'm not a committer does not stop me from helping other people with
their ports. Your blunt statement would offend a large number of people here,
since a very small portion of -ports is actually comprised of committers. Most
of us are just plain ole porters who enjoy porting stuff to FreeBSD.
There is absolutely no reason why a non-committer can't be allowed to comment
on someone else's port (or any other contribution).
> BLUNTLY:
>
> Someone who DOES have commit privs please make a decision on this submission.
> If you're going to reject the port then PLEASE SAY SO and I'll go on my way
> and update both my local repository storage (which I ALSO fixed to comply
> with your numbering systems and naming conventions) as well as the web page
> so people know what's going on! I can also terminate the way I have been
> maintaining the code for the last few days (which I happy to personally
> dislike, but will maintain for compatability reasons if this is going to
> be a port).
>
> If you're going to COMMIT the port then please do so, in that I'd like to
> know how I'm going to be maintaining the software from this point forward,
> and this decision on your part is germane - in fact, critical - to that
> process.
>
> I thought I'd do the community a favor and add something to the base of
> code that is available to run on FreeBSD from the "official" sources.
> I certainly didn't need to do so, and in fact there were some compromises
> that I had to make in order to do this, some of which will probably upset
> the rather sizeable base of EXISTING users of the software (in particular,
> the move of control and executable files will certainly be a "gotcha"
> for them at least once)
>
> If the port is unwanted as it stands then tell me so and that's fine - no
> harm done - the mistake certainly won't be made twice!
>
> Pardon me for "intruding" into your little treehouse on a new year......
>
> Sheesh!
By all means no one here is going to reject your port if you are willing to
consider other people's point of view and provide an explanation for why you
did something this way or that instead of doing it the usual way. Just as I
have taken much pain to point this out to you..
All this controversy makes me wonder if you read the "Making a port yourself"
section of the Handbook at all.
I still can't believe you derided everyone here as a "little group" and by
implying that our "treehouse" is full of elitists. Because in fact, we are all
human just like yourself.
--
Will Andrews <andrews@technologist.com>
GCS/E/S @d- s+:+>+:- a--->+++ C++ UB++++ P+ L- E--- W+++ !N !o ?K w---
?O M+ V-- PS+ PE++ Y+ PGP+>+++ t++ 5 X++ R+ tv+ b++>++++ DI+++ D+
G++>+++ e->++++ h! r-->+++ y?
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?XFMail.000101141418.andrews>