God Ram Illusion Hassle

The following was taken out of GodRamIllusion because of its inappropriateness. May the participants continue here if they like.

The only good about it was that they went on the meta level and rephrased their perception of the other. Which exactly highlights the point:

• Top posts something of relevance (good)
• but phrases it in a "relational" language which causes many misunderstands (problematic).
• Then - caused by previous misunderstanding - someone jumps to the conclusion that Top just wants to present his HobbyHorse (what he doesn't, though it looks like it) and posts back a corresponding remark (bad).
• Escalation ensues.

Please if you see Top contribute something do not fend him off (it will not work anyway, he feels rightly hurt). Instead rephrase in typical WikiZen terms and ask if that is what he means. If he disagrees with that, then you may still continue as before. I did this rephrasing on the points in question in the list at the T[^B]top.

-- GunnarZarncke

Perhaps we should back up and ask what the GodRam would be used for. However, I suspect the answer will be "Why, it would help one build a grand Insert-Your-HobbyHorse here, system". -t

---

The offending contribution:

• MultiParadigmDatabase automatically gives every record/object a unique row/object ID.

What does this have to do with GodRamIllusion? Which property - global, secure, infinite, persistent, robust, concurrent, AtomicConsistentIsolatedDurable, and uniformly high-performance - is it aiming to achieve? and how does it achieve this property? (I'll happily restore it above if you can provide a good answer, but it strikes me as misplaced.)

You have the belief that all of those services or features should be built deep into the root of the system and architecture. I don't necessarily agree with that for reasons debated in other topics. You are essentially saying, "If you want feature A, you must also include feature B, C, D, E, F, G, etc." You have not justified that to be the only rational choice combo. If they are all "free" and easy to switch off or ignore when not needed or wanted, I might agree, but suspect that they have performance or complexity overhead that makes them non-practical. Sometimes its better to start out in the market with a web browser rather than a XanaduProject. You seem to want to build a giant do-everything "god engine" before starting any real projects. This is not realistic. Incremental is how most systems test the market and grow. -t

Huh?

{Indeed. Top, your response is strangely non sequitur. Why don't you answer the question -- what does "MultiParadigmDatabase" (as if there is such a thing) have to do with GodRamIllusion?}

The topic is about having something that is like RAM or that can replace RAM, but has more features, is it not?

{In a sense. What does "MultiParadigmDatabase" (as if there is such a thing) have to do with GodRamIllusion?}

If you agree with that, and MultiParadigmDatabase "is like RAM or can replace RAM, but has more features", then it's on topic.

{You mean in a MultiParadigmDatabase, "all objects are in a global, ObjectCapabilityModel secure, infinite, persistent, robust, concurrent, AtomicConsistentIsolatedDurable, and uniformly high-performance address space at all times after introduction." Wow, I'm impressed. That's some bold capabilities for something that doesn't exist. Of course, I suppose something that doesn't exist can be given any capabilities you like, can't it?}

So this topic is really about YOUR pet additions to RAM, and not mine? Then call it BradsExtensionsToRam? or something if others' favorite features are not allowed. -t

{FacePalm}

Insults are not helping anything here, other than perhaps a personal catharsis on your part. Here's an abstraction of the conversation as I perceive it:

You: Your contributions are off-topic

Me: Isn't this topic about extensions to make RAM or RAM substitutes more powerful and useful?

You: Well, yes.

Me: And didn't I suggest ways to extend RAM or RAM substitutes to make them more powerful and useful?

You: That was a stupid reply (per polar bear poster link)

So, where is my insult-deserving error? -t

[The way others see your contributions:

Us: Here's an ideal called GodRam. It doesn't exist, so we do certain thi-

You: Here's my GoldenHammer! A MultiParadigmDatabase.

Us: What's the connection to GodRam?

You: I don't want GodRam.

Us: Then it's not really on-topic, is it?

You: Yes it is.

Us: Really? It has the properties we're talking about?

You: No. You should change the title to reflect that you're talking about the properties you want.

Us: FacePalm. ('Cause the title already reflects it.)]

That's a curious interpretation. For one, where did you get "I don't want GodRam"?

[You said [emphasis added], "You have the belief that all of those services or features should be built deep into the root of the system and architecture. I don't necessarily agree with that for reasons debated in other topics."]

But why do you believe that only those features qualify something as "God RAM"? You seemed to agree with this topic being about "ways to extend RAM or RAM substitutes to make it more powerful and useful". Now you seem to be saying this topic is about "RAM with and only with the following list of added features: A, B, C, etc.". Why the apparent recent narrowing? -t

[Because GodRam is defined that way (see the very first sentence on this page). BTW, No one has agreed that it's about "ways to extend RAM or RAM substitutes to make it more powerful and useful". It's always been about making something that looks like RAM with a particular set of attributes. The closest to agreement anyone has come was a rather lukewarm "in a sense". And that only meant that GodRam was a way to make something that appeared to be a more powerful RAM substitution. I.e. This page's topic is a subset of that topic, and as a subset, not everything that is on topic for the latter is on topic for the former.]

PageAnchor: generic_alpha. A topic called "GodX" usually means an "X" that either tries to do everything, or has boatloads of features or attributes. It's too generic and common a linguistic concept on this wiki to claim your own personal definition for any X. Again, if you want to limit this discussion to a particular personal sub-set of extensions, then call it something like BradsExtensionsToRam?. Is this not logical? Otherwise, any WikiZen can rightfully expect to consider their own features and extensions to X on topic GodX. As an alternative, you can give your set a working name, but others' extensions are still "on topic" here.

[Because the circumstances that allow for a title to tell you exactly what a page is about are extremely rare, it isn't logical to assume that a title tells you everything. The logical response is to actually read the introduction. And, unless you want to know who's idea it is, GodRam gives more indicators as to the actual subject, and is shorter and easier to remember. Therefore, it's a better title.]

[Now, if you wanted to talk about something else that could be called GodRam, this page can certainly accommodate that. But the way to go about it isn't to insert it into a list of things that attempt to accomplish the original properties. That is what you were originally called out for, and that is why you were told it was off-topic.]

I don't agree with your implied wiki-etiquette rule, but do not wish to formally challenge it at this time. You were not originally clear on the scope of "off-topic", I would note. -t

• [How was the charge of "off-topic" not clear? There was, (and still is) only one scope on that page. Obviously, the charge applies to that scope.]

Ah. I guess the definition of GodRam wasn't close enough to the top of the page. I'll ask Ward if he can alter the HTML style to push the start of the content closer to the page title.

Again, you personally don't get to define "GodX" terms.

Says who?

I explained it above, page anchor generic_alpha.

[At that page anchor, you unilaterally declared what GodX meant. Funny how no one seems to agree with you. It also fails to take into account how it is common practice to give a term a working definition when one wishes to talk about a specific topic, even if that word already has been defined. (And, before you start stating that you don't agree with the wiki-rule, you should note that it's been in practice for thousands of years).]

I don't believe you took a fair survey of WikiZens; you are guessing out of your caboose. And a working definition doesn't preclude other working definitions. And probably only 2 people on earth give a shit about the capability security model: you and the person you bribed. Don't be so territorial; allow other viewpoints.

[Why would I have to survey WikiZens? The "rule" predates Wiki. And of course there can be more than working definition of a word. That's why I suggested that if you wanted to use a different one, you define it in a separate section rather than inject your comments that depend on your definition in the middle of a discussion that assumes the other working definition. Finally, your last statement is just plain false. The wikipedia article on capability security lists six people who have done research into the capability security model. Last I checked 6 > 2. It also lists a number of OS and hardware projects that incorporated (or at least provided support for) the capability security model.]