From: AgreeToDisagree: Re: ''I'm with you on this one, Marc. There needs to be a way to indicate a disdain for continuing a pointless discussion (I'm not mentioning any names, but Top comes immediately to mind) without ceding any ground. TenSeven indicates that you have made your point (NuffSaid) and you aren't willing to invest any more time discussing a matter that the other party is not discussing in good faith. Sad, but this is sometimes the case. -- MartySchrader'' I protest. I deny discussing in "bad faith". I am not consciously bad or evil on this wiki. From my perspective, it '''honestly appears''' to me that you guys reject science and instead rely heavily on ArgumentByElegance because certain concepts titillate you personally, regardless of their practicality or empirical road-testing. That is honestly how the on-going conflict with "the academics" here appears from my perspective. That's my conscience reality as it is in my head. If I am delusional and seeing it all wrong, I can't do shit about it. There's no self-correcting mechanism kicking in right now to reboot my outlook. As far as I can tell, I'm perfectly logical and rational in general. (I suspect you are the delusional/misguided one(s), but that's another story.) I do try hard to explain why I see things the way I do, and am not trying to "trick" people. -- top * '''Top''': to me you are completely rational and valid. I think what you're witnessing is an actual undiagnosed insanity that manifests whenever there is some sort of important concept that they (wrongly) feel needs protecting. I've seen this behavior on the Python mailing list. It's very weird and anti-rational, but hyper-rational from their own flawed premises. In fact that is what it is, you're presenting something that goes against a long-standing, internalized, and so unspoken premise, yet they either have not the ability to re-evaluate it or desire. --AnonymousDonor * The evidence-to-zealot ratio is a really tiny value :-) -t I suggest we find a way to '''nail down and document''' our differences in this on-going academics-vs-practitioner battle. It may not resolve the root conflict, but at least clarify where and why we differ. Reinventing half the debate wheel over and over is better than reinventing it all. I don't like the repetition anymore than you do. I'm not against theory; I just want to see it road-tested before being placed in mainstream tools. Some want to skip that step. If that's truly a heretical position, then maybe I am delusional after all. It's perfectly normal within my WetWare. -- top ''Sorry, topper, ol' buddy, but you ''continually'' argue in bad faith. The whole middle section of SoftwareDevelopmentIsGambling shows evidence of that. This is why most Wikizens have just given up on you and dismiss you out of hand. I still read your tripe because you are a pretty smart guy and even a stubborn, closed-minded, jackassian, stupidly self-blinded pig finds an acorn once in a while. However, I have to admit that it gets harder all the time to muck through your drivel looking for the occasional gem. They are still there - but getting harder to find all the time.'' ''So, no, you are not evil as such. Just a waste of time, which is far worse in the litany of sins to which most professionals subscribe. -- MartySchrader'' I believe that the biggest difference between us is that I hold dear to GoodMetricsProduceNumbers, while you don't. If that belief makes me bad, evil, stupid, ignorant, dumb, foolish, lazy, stubborn, a bad-dresser etc., then so be it. I'd like to hear a good argument that such is an '''irrational position'''. I believe I provide the valuable service on this wiki of forcing people to face the dearth of evidence for their sacred cows. People don't like it when their sacred cows are attacked and criticized; they grow defensive and sling insults in retaliation. It's human nature, and having a PhD doesn't make one immune to such human emotions; It's in our genes. The criticism process starts them thinking, and eventually they may become sufficiently self-critical of their work to either develop better metrics or to abandon their sacred cow. '''Bitter medicine is often the most effective'''. Someday you may thank me. -- top ''We've had this discussion to no end before, so it's not worth our time to hash it out Yet Again. Despite your claims to be reasonable, scientific, pragmatic, data-swayed, and whatever, you prove yourself otherwise almost every time you "contribute" to this board. Therefore you '''''earn''''' a TenSeven.'' ''[By the way, top, somebody edited this page with the UserName of MarcThibault, then signed it as you. If you want to remain anonymous because of your job situation I'd suggest you do something about that.]'' {That's GrammarVandal, spoofing UserName''''''s as usual.} Does that apply to all the other recent edits from MarcThibault as well? {No, not all of them. Some are the real MarcThibault, and at least one (I didn't keep track of how many) was the real MarcThibault followed by a GrammarVandal edit.} ---- Since Top counters (eventually) every discussion of programming practices and tools other than his own with demands for his idea of acceptable evidence, I'm left wondering how many of his own preferred programming practices and tools meet these standards -- and if they don't, then on what basis he prefers them and rejects others. -- DanMuller ''I'll let long-time listeners see if they can predict my response to this. -t'' [I'm a long-time listener, and I've no idea what your response to that will be.] I generally don't claim my pet approaches are ''objectively better''. It's up to the objective betterment claimer to provide evidence/proof, not alternative tools and techniques. Some of you seem to think there's an '''equal burden''', but logic and debates don't work that way. The difference of value between two tools is UNKNOWN by default. If I claim you don't have objective evidence that A is better than B, that's NOT the same as me claiming B is objectively better. And, the "mind" in my handle (TopMind) is based on my long-running and controversial suggestion that outside of machine performance and duplication factoring, the value difference between tools/paradigms/languages/techniques (tools for short) are largely tied to WetWare, AKA psychology and physiology, which is often individual-specific. The computer doesn't care how you write the software, only the humans do. Computers don't maintain software, humans do. And, I will do my best to explain my personal WetWare processes where appropriate. * From OnDecomposingSystems, "I'm confident a document processor built with a RDBMS would score equal or higher on most representative numeric maintenance metrics thrown at it." ** ''It is being ''personally'' confident. Yes, it's ArgumentFromAuthority and I admit it. And, note that I don't belittle others personally for rejecting my ArgumentFromAuthority claims. It's low on the EvidenceTotemPole and I take that into consideration. Anyhow, the default is not that any given technology is better than the other. Probably nobody has an OfficialCertifiedDoubleBlindPeerReviewedPublishedStudy for their pet GoldenHammer. -t'' * From AreOoAndRelationalOrthogonalDiscussionThree, "OO and relational tend to have a higher overlap in my observation of their usage in practice. For example, should "employees" could be modeled as objects, or as relational tables. To avoid violating OnceAndOnlyOnce (a MirrorModel), we ideally don't want employee-related attributes to be cataloged and tracked under both relational and OOP." (From context, the "higher overlap" is compared to that between procedural and relational). ** ''The word "score" is perhaps inappropriate. My apologies. -t'' ** So what do you think an appropriate word would be? ** ''I reworked that paragraph. Please see PageAnchor Overlap_57.'' ** Okay. I've changed the above to reflect your changes. ** ''Okay, so then what is your specific complaint after the rework?'' ** It's not a complaint. It's evidence. ** ''Of?'' * From ComplexityOfOutputtingDuplicateTuplesInTutorialDee, "I'll accept the "complexity" of Nulls and bag output because it better fits the messy real world and simplifies human effort needed to create output." ** ''If your abstractions assume a clean world which doesn't exist, then they can get in the way instead of help. A BigIdea can be ''too'' big for a given need. Related: EightyTwentyRule. (Oracle and IBM agreed with me.) Related: BagVersusSetControversyRoadmap. -t'' That being said, I still value the idea of numerical metrics because it can help us make better choices by understanding the impact of our design choices on the economics of software creation and most importantly maintenance. Specific tools may work somewhat better under different design assumptions. But overall, I don't think any are a net GoldenHammer. (I also believe that there is value in mastering a few select tools rather than try to master combining gajillion tools.) The value of such metrics is generally within the variations offered by a given tool, not in choice of tool. And possibly to "prove" to others that tool choice does not matter much from an objective standpoint, which implies that WetWare does. In other words, '''WetWare is the overriding factor when objective differences are minor'''. It's unfortunately that WetWare is not easily measured and/or inconsistent between individuals. But I'm just the messenger. Software engineering is about a model's interaction with the human mind. It's not really about machines. I wish it was about inanimate objects or rules, then we could analyze it without getting into the messy world of psychology and WetWare. But reality is not always clean and nice to us. -t ---- You're right, TopIsNotEvil. Top is simply wildly out-there. I'll keep out-there ideas in the archive. You never know when some rare tool is just the tool for the job. --BottomMind ---- I also believe, that TopIsNotEvil, he has just a different point of view from everyone here. At least in my case, his comments made me stop being a ObjectOrientedWeenie and start thinking that the RelationalModel has some very interesting advantages (even if TopMind himself does not seem to like the pure approach of projects like Rel, I have also been able to find the occasional gem in the things he writes, so I hope he keeps doing it) --LuxSpes ---- TopIsNotEvil, just UnconsciousIncompetence ''I just have a different idea of what constitutes "evidence" than a good many WikiZens, and this is the largest reason for conflicts. I have much fewer conflicts with those who believe in the same principles of evidence as I do. If you demonstrate that I am inconsistent with the kinds of evidence I accept (my "evidence model"), then you may have a case that I am nutzo in the head. Otherwise, it's just a '''different root philosophy''' when it comes to what kinds of evidence to use for contentious subjects.'' ''My evidence model is heavily influenced by economic theory, the scientific process (propose, model, measure), and RaceTheDamnedCar. Some have argued that economics and measurable empiricism is the "wrong way" to settle things; but so far I have not seen a good justification. I mostly see ArgumentFromAuthority and excessive ArgumentByElegance. I also believe in catering to WetWare, which is a squishy gray-area of science. --top'' ---- See: AssumeGoodFaith, or perhaps AssumeStupidityNotMalice SeptemberTen ---- CategoryComparisons