From owner-freebsd-gnome@FreeBSD.ORG Tue Jan 3 06:14:48 2006 Return-Path: X-Original-To: gnome@freebsd.org Delivered-To: freebsd-gnome@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6E67D16A41F for ; Tue, 3 Jan 2006 06:14:48 +0000 (GMT) (envelope-from mezz7@cox.net) Received: from centrmmtao03.cox.net (centrmmtao03.cox.net [70.168.83.81]) by mx1.FreeBSD.org (Postfix) with ESMTP id 020C843D45 for ; Tue, 3 Jan 2006 06:14:46 +0000 (GMT) (envelope-from mezz7@cox.net) Received: from mezz.mezzweb.com ([68.103.32.140]) by centrmmtao03.cox.net (InterMail vM.6.01.05.02 201-2131-123-102-20050715) with ESMTP id <20060103061211.FXIP613.centrmmtao03.cox.net@mezz.mezzweb.com>; Tue, 3 Jan 2006 01:12:11 -0500 Date: Tue, 03 Jan 2006 00:15:40 -0600 To: "Donald Byrd" References: <20060103045048.9B3CB5C1C@precious.home.tig> From: "Jeremy Messenger" Content-Type: text/plain; format=flowed; delsp=yes; charset=us-ascii MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-ID: In-Reply-To: <20060103045048.9B3CB5C1C@precious.home.tig> User-Agent: Opera M2/8.51 (Linux, build 1462) Cc: gnome@freebsd.org Subject: Re: Libsoup broke (part deux) X-BeenThere: freebsd-gnome@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GNOME for FreeBSD -- porting and maintaining List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 03 Jan 2006 06:14:48 -0000 On Mon, 02 Jan 2006 22:50:44 -0600, Donald Byrd wrote: > Jeremy, > > You forgot to copy gnome@freebsd.org ;) > >> Don't want to upgrade system and trying to use new stuff from ports are >> the most common that cause apps can't be build in our list. So, yes, we >> are expecting for user to upgrade apps first. > > *ALL* apps? Not going to happen. A small machine like a P2 or P3 500mhz > takes HOURS or even DAYS to build GNOME2 and dependencies. You also have > to > consider all the 'config' prompts that come up waiting for a reply. That > adds a lot of time to the building process if you are not on top of > things. > > Doing a 'portupgrade -a' would upgrade everything installed in the > system. > Hundreds of ports could be upgraded. This might actually expose you to > new > bugs in unrelated ports. Not only is it unreasonable to expect a user to > do > this, it is downright *dangerous* and *irresponsible* to ask a normal > user > to do that. After all, if the ports were so perfect and safe, the user > wouldn't be reporting a port problem in the first place. > > I understand the need to know if the user is installing the latest and > greatest, so if you are going to go that route I suggest doing > 'portupgrade > -rR ' instead. Sounds like you should be use package, not ports tree, for those slow machines. >> Hours have never happened to me before, they usually come up fast unless >> the error is very unique. > > Ever notice that when you are super familiar with a subject, searching > is so > much easier? Many normal users wouldn't know what to search on, and > wouldn't > know what to do with the answer when they got it. I don't have to be "super" familiar with any subject to search. Example: http://lists.freebsd.org/pipermail/freebsd-gnome/2006-January/013282.html Let's copy one line of error like this, "nsHostResolver.h:95: error: ISO C++ forbids declaration of `PRAddrInfo' with no type" and paste in google. The third link in the result come up perfect. Also search for "nsHostResolver.h:95" or "ISO C++ forbids declaration of `PRAddrInfo' with no type" or different error lines in mailing list archive, the second link come up perfect with a solution. >> That line I think it should be remove. > > We agree on something :) The question is kind of an advanced or > power-user type thing. Indeed, it is not easy to make that kind of decision. I can ask my team if I can remove that line or tweak it if one of them want to keep it. >> Actually, your attach has never make it to the list as see a proof in >> below link. Next time, let us know that you are sending attach in the >> email and we will asking for re-send attach. > > Ah, and there lies the rub. In order to comply with the process I have to > send an e-mail that doesn't comply with the process to notify someone > that I > intend to be sending another e-mail that does comply with the process. > Oh, I > suppose I could just paste the build log into the text of the message > but I > thought that compressing it would be more responsible. This is getting > complicated. > > FYI: I received this reply to the original message. > >>> Your mail to 'freebsd-gnome' with the subject >>> Libsoup broke? >>> Is being held until the list moderator can review it for approval. >>> The reason it is being held: >>> SpamAssassin identified this message as possible spam > > I don't know who the list moderator is, or if you can see this type of > stuff > or not. I don't think all of people in my team have the access to that. Maybe marcus can access it or not. >> Sorry if you think that I was rude. I just gave a straight answer. We >> get > >> a lot of emails with no useful info, so our answer is getting plain and >> plainer until they resend something useful. > > I hate to say it but that is a closed source attitude. With closed > source, > they have your money and aren't going to make any more by fixing your > problem so they make it hard on purpose (ON PURPOSE). But with open > source, > the valuable commodity is the knowledge that something may be broken. The > end user has it and the developer's don't. So the error reporting process > should be easy, polite, helpful, as automated as possible, and designed > to My first replied was very polite with provided an URL and the URL was very easy to follow in the "2. What to report?", but you blew up with all of those lousy excuses when you didn't know that your attach has never make it to the list and think that I ignored it. BTW: I asked people in IRC if I was rude to you and they said 'no'. > accept the fact that most of the time the user doesn't know the > difference > between PCI and his keyboard. Bug reports that can't be reproduced or > even > understood should still get filed away so maybe someone very > knowledgeable > will someday notice a pattern. > > The whole reason I was having this libsoup problem is that I was trying > to > update GNOME to fix a crash on startup problem. I didn't bother trying to > report that one because I don't know enough about it to know whose > problem > it was. I made the mistake of clicking on the "inform developers" once, > only > to get a face full of "if you don't know what you are talking about, > don't > submit bugs..." attitude that I (obviously) hate. There was no "if you don't know what you are talking about, don't submit bugs..." attitude in single of my email. I only commented, "Don't bother to send us an email if you are not planning to give us the more informations."; it's a big difference when someone refused to give info of very simple things such as 'uname -a', 'pkg_info' and etc as stated in the URL[1] because of not have time or so. Turned out that your attach never has made it to the list and you didn't tell us that you have sent an attach until third or later email. [1] http://www.freebsd.org/gnome/docs/bugging.html > This problem I thought was different in that it was likely just a port > dependency issue. I've encountered those before and have always had very > positive experiences when reporting them. There is no way for us to able to guess which dependencies without know what error you are getting. In end, give enough info always will get better respone from us with no wild guess. Anyway, we should move on. Cheers, Mezz > Have a great new year, > Don. > -----Original Message----- > From: Jeremy Messenger [mailto:mezz7@cox.net] > Sent: Monday, January 02, 2006 11:04 PM > To: Donald Byrd > Subject: Re: Libsoup broke (part deux) > > On Mon, 02 Jan 2006 21:44:18 -0600, Donald Byrd wrote: > >> *sigh* Typical. >> >> Before I sent the e-mail I was unaware of this web page so don't take an >> attitude with me. >> >> You expect me to do this before I contact you. This is very, very, very >> time >> consuming. What if I *don't* want to upgrade every port on my system? > > Don't want to upgrade system and trying to use new stuff from ports are > the most common that cause apps can't be build in our list. So, yes, we > are expecting for user to upgrade apps first. > >> How many hours do you expect me to spend searching mail list archives? > > Hours have never happened to me before, they usually come up fast unless > the error is very unique. > >> Who am I to decide if a "bug" is OS/User/Platform specific or not? >> Unreasonable >> expectations. > > That line I think it should be remove. > >> http://www.freebsd.org/gnome/docs/bugging.html >> 1. When should I make a bug report? >> . After running any build failure output through gnomelogalyzer.sh. >> . After running cvsup to obtain the most recent ports tree. >> . After running portupgrade -a to ensure that all applications are >> up-to-date. >> . After searching through the FreeBSD GNOME Mailing list archives to >> see if the problem has already been reported. >> . After deciding whether the problem is FreeBSD-specific, or is a bug >> in an application that would affect all users, on all operating systems. >> >> >> As for the information requested in a bug report, of course my e-mail >> was >> not specific enough to constitute a bug report. I wasn't aware that it >> would >> be so considered. I have sent casual messages to port caretakers before >> and >> have been either directed to a specific bug reporting web app, or I was >> just >> asked questions. Typically, the correct course is unknown until you ask. >> That's all the e-mail was doing. I suggest that you put text in your >> (fresh) >> ports description that you only accept e-mails that conform to the spec >> outlined in your debugging web page. >> >> As for my reply, like I have already said, the bug reporting >> expectations >> are an unrealistic burden upon the normal user. >> >>> Of course, they will asking what error are you seeing (that one you >>> have >>> never show us), what is version of Windows, what were last thing that >>> you >>> did to the Windows and how did you reproduce it. >> >> For the record, my original e-mail I sent (not the one you replied to) >> had >> an attachment with the entire output of the build process including the >> error. The second e-mail was just providing additional information. Get >> your >> facts straight before you go throwing stones. > > Actually, your attach has never make it to the list as see a proof in > below link. Next time, let us know that you are sending attach in the > email and we will asking for re-send attach. > > http://lists.freebsd.org/pipermail/freebsd-gnome/2006-January/013286.html > >> Don't worry Jeremy, I won't bother trying to report anything to GNOME >> anymore. You ignore the information I do send, you will ignore the >> problem >> until I trash my system after spending 3 days rebuilding every single >> port >> on the box, and you are rude. > > Sorry if you think that I was rude. I just gave a straight answer. We get > a lot of emails with no useful info, so our answer is getting plain and > plainer until they resend something useful. > > Cheers, > Mezz > >> Have a great new year. >> >> /out. -- mezz7@cox.net - mezz@FreeBSD.org FreeBSD GNOME Team http://www.FreeBSD.org/gnome/ - gnome@FreeBSD.org