Am 01.01.2004 um 04:24 schrieb Ben Hines:
[resend]
On Dec 31, 2003, at 6:11 PM, Max Horn wrote:
Err, sorry, but I don't understand why you react so extremely aggressive. I just re-read what I wrote, and I still don't get why what I wrote makes me a dick (again, apparently). I was merely asking, and IMO not even impolite (although maybe due the fact that I am not a native speaker I am missing something abusive I said, although I have not the slightest clue what that would be). I was (and am) seriously surprised by this approach (tried to show that by my use of three question marks, BTW). I didn't mean to say that you or anybody else involved in this matter is stupid, or whatever else you thought I was implying. I simply didn't understand, and hence did what is usually a good idea in such a situation: ask.
To me sometimes Max, you come across as condescending, abrupt and like you know everything better then everyone else, and that none of our proposals could possibly be valid.
Harsh words, but I have to accept them of course. I certainly don't want to come across like that, but I can't debate your feelings. Let met just emphasize that I never thought "none of your proposals could possibly be valid". I don't get along very well with you, obviously, and maybe never will - but nevertheless, I hold your abilities in high regards, believe it or not. If I didn't, I wouldn't even bother to discuss them with you :-)
Sometimes it does not seem like a technical discussion at all, rather like a lecture from max on why we are so wrong.
I usually tend to play Advocatus Diaboli on "proposals" on any project I participate in. Maybe a bad habit, but so far it served me (and the projects) mostly well... Many many people tend to be rather quick in adopting a solutions, and then only later discover its problems. There are some things in Fink nowadays which IMO came to existence like that and which I am still not entirely happy about. It's usually much harder to undo a design change than to apply it. Thus getting things right the first time is a nice thing, although clearly it's almost never possible... so at least try to get things as right as you can... Hence I like to immediately point out all problems I perceive, or if I see none, I search hard for them. As a result, I guess it's not a big surprise that you got the impression you cite above. Note that I am happy about it :-/. Once again, I can't debate your impression, of course.
Still, IMO, pointing out perceived problems doesn't imply that I am opposed. Yes, it may turn out that the problems are trivial, or not even real problems, or that a nice solution exists, or that the problem can simply be ignored, etc.. Or it may turn out the problem is serious, or even kills the proposal. But pointing it out doesn't entail any judgment (well except that I didn't see any trivial solution :-).
Combined with your comments on a tracker item today (to a simple user who didn't follow the rules - yeah, users do that, no need to yell at them) you seem to be acting more and more like chrisp:
https://sourceforge.net/tracker/? func=detail&atid=117203&aid=860056&group_id=17203
basically sounds just like how chrisp acted before he quit:
http://sourceforge.net/tracker/? func=detail&atid=371315&aid=479267&group_id=17203
What I find interesting about this is that we are apparently so close in some regards, Ben... I am too lazy to search through the trackers, but I think I could easily find similar examples quoting *you*... :-). Actually it's a quite common behavior with hackers. I wanted to quote http://www.catb.org/~esr/faqs/smart-questions.html here, but essentially the whole Introduction section of it describes what I feel, and IMO reflects quite well my, chrips and also your behavior in many cases.
I am extremely annoyed I see for the hundredth time a bug report from somebody who clearly didn't even read the short text they are presented with when they file a bug report. So, they don't even spend 1 minute to read that text? Well OK, why should I spend more than 1 minute to reject their bug report? If they were my customers, I'd be polite anyway. But they aren't - they are wasting their time, and worse, mine.
Clearly, on the long run it would be superior if my reply would be friendly and polite. I actually try to do that usually, but sometimes I just can't take it anymore, and react instinctively (like, IMO, you... that's my impression, at least, I don't claim to know the truth, just as you don't). Not something I am proud of, for sure. But not something I am ashamed of either. I am trying to change myself, but it's quite hard to swallow the bitter pill each time.
This is BTW one of the reasons why for many trackers on various projects I set up standard replies for such things. I can write them once, trying to make them as polite and friendly as possible, and then instead of firing back a flame, use that standard reply. Same for all the people who demand (in sometimes quite hard words) that I unsubscribe them from some mailing list. Etc.
So, I admit that there are superior ways to handle those situations. However, do I ultimately feel sorry for telling somebody that just now, they behaved stupid? Not really.
Do I start to understand chrisp? Definitely. :-)
In my particular situation, yes perhaps there flaws. Point them out in a manner that is not so condescending, please.
OK. May I please ask the very same from you? :-)
Your 'use of the three question marks' is exactly what pissed me off, as if you are asking:
"how could we be so stupid to suggest this??? Don't you know that your suggestion is stupid??? Have you thought this through you idiot??" Asking "if i have thought something through" is rude. Just point out any problems you may see without the commentary.
Clearly, we had very different interpretations of what "???" signify here <sigh>. That's one of the cases where I'd wish we'd not just have written word, but spoken. Maybe the way I'd have accentuated (? is that the proper word?) the question would have made clear that it wasn't meant as an insult, nor as a rude remark, but rather as genuine surprise.
Again there is obviously a middle ground, not all people are so easily offended and set off as I am, some wouldn't be bothered at all, and the large center ground would simply read it and say 'that guy was a bit harsh eh? He was just making a proposal'
Once again I have grin at how similar we are in that regard, actually. :-)
Hm, in the light of what Ben Hines that (namely that for 'developers', one can 'hold' packages), does that mean that we essential will start to tell developers (like Ben or me and many others) that they now have to start manually fixing their Fink installation whenever changes like that are made? I mean, since for example I would have to hold those -dev packages mentioned above, that would mean in the alternate world (where we implemented the changes as Martin and Ben suggest them), I would have had to manually fix up my freetype2 installation. Do I understand that right so far?
I don't understand your concern here... no you wouldn't have to manually 'hold' or 'fix' anything. Fink would know that you manually had installed those packages (via a 'fink install' or apt-get command for example), and so would attempt to keep them installed at the end of any fink run rather than removing them. You wouldn't have to explicitly hold them at all or manually fix up anything. This is not a developer only feature. It would work best as a default option - most users would NOT want to keep build time only deps installed. Even as a developer, i only want packages i have asked for via fink install, even -dev.
My concern is that (as I understood it) Martin suggested that we could start doing things in Fink which rely on all the buildonly packages being removed. However, if we did that, we'd run risk that those changes may break for people which *do* have buildonly packages installed. So either we do not start making changes that rely on it, or we have to accept that certain people will have to do some extra work in some cases, in order to keep their Fink installation going.
Once again: I don't claim that this is the absolute truth, this is just how I take what Martin wrote, I don't mind learning that I misunderstood something. Nor do I say that this is a show stopper, or insurmountable. I merely bring up an issue. If it's trivial in the end, great! If not, let's at least talk about it.
To clarify again my 'that is planned' statement simply meant I certainly plan to attempt to do this, i didn't mean to say it was a done deal as you have interpreted.
OK. That is indeed miscommunication then. In my mother tongue, I'd roughly differentiate between, "that is planned" meaning "we want to do this, it will be done", while "this is in planning" means "we are working on making a plan to realize this, but haven't yet finalized the details" and finally "this is being researched" means "We are looking into this and consider doing it in the future". So yeah, I was additionally shocked by the formation, because this is really the first time I heard about this "plan". Wouldn't be the first time an idea has been hatched out on IRC and then it was "forgotten" to mention it on fink-devel :-)
I can see how you might have take it that way from others actions on fink in the past, but hopefully you know that Ben Hines generally doesn't ever commit things to fink without bringing them up to everyone in detail. For example when i implemented the mirroring stuff I posted a design for them and solicited comments.
I actually never feared that you (or anybody else, for that matter) would do any rush commits on this (although, indeed, sometimes I have been badly surprised in the past <sigh>, but that's a different matter). And let me state that I do consider you, Ben, as a very capable developer, despite all our differences.
Doesn't mean I can't voice my concerns, does it?
It is the other people who change dep engines without asking. :)
*cough*. Let's not talk about that... <sigh>
Anyway maybe in the end it will turn out to not be feasible, but I certainly don't see blockers here.
Good. The other way around, I never said it is impossible. And thanks to some of your explanations, I see the proposal in a more positive light then initially, too. Like for example by learning that you do consider some of my concerns already.
A pity it didn't start with proper explanations, I might see it all in an even more positive light :-)
If the solution is a full chroot, so be it - that will have many other benefits too. All the problems you mention i believe can be overcome. Packages could still be switched out if you were holding something with a "build depends hold".. fink could simply know to put back the ones you wanted at the end of a run and ask you what to do if there are issues. Anyway I didn't plan to fully implement this right now, i'd like to work on getting fink to mix binaries and source dependencies better first since it is a more common user request.
Fine. Also a very interesting subject, albeit somewhat orthogonal to this one :-). Looking forward to any results on this (I have been looking into this a little bit during the past dependency engines discussions, too, e.g. while considering to actually use the dep engine from apt, and making apt-get aware of how to "install" stuff via fink... but I am digressing :-)
Good morning, I am going to bed now,
Max
------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click _______________________________________________ Fink-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/fink-devel
