That could indeed be an aspect of a search engine that functions on top of a relative naming system. But I think you’re teetering on the brink of a very good point.
Anecdotally, would you say that you use search engines to find results that only contain the keyword(s) in their domain name?
Yes, this is the point I was making, why advantage the rich, who have all the resources and reserve their seats - it’s the opposite of what Safe’s about!. It’s not even just that - they could easily buy up all the Mr Nissans etc - that’s why I said you could also end up with the rich squatting.
This rewards resources- rather than resourcefulness. [quote=“smacz, post:35, topic:6388”]
This type of system gives too much power to moneyed interests.
[/quote]
Absolutely - spot on 
Edit:
Lol…can’t believe you are arguing this - so are the small photography studio or blogger’s site. Big business should be given preferential treatment to the detriment of small business entrepreneurs and individuals…yeah…right.
LOL…nice one
Yes, of course. It would be part of the network logic, as is Safecoin handling, and in time other features as they are added.
Having thought more it seems feasible to me. I’m not proposing it, I’m saying it is technically feasible, so if we think it’s a useful thing to do, we shouldn’t rule it out.
Ah, lol. I use SMTP to send directly from Linux mail, but still use a hosted server for everything else.
My typical use of search engines involves keywords that are often unrelated to the domain name + a tld to limit searches to particular types of sites. If I know the domain, I go straight there and, if needed search on that site directly. Only occasionally will I use a search engine to search within a site. I expect on safenet that will be even more true.
In some ways I think perhaps the domain name isn’t even very relevant to me - but the tld is very relevant/useful. Hence, having a wide variety of standard tld’s might come in handy. In the same way, when I search for files on my computer, I want to limit the search to a video or a picture or a particular document type to limit the results.
I dont think eliminating DNS is condusive [sic] or practical.
What is it not conducive to? You have a gaping hole in your sentence there.
Condusive to the usability of the network, it would be like reverting back to typing ip’s in. Are you really arguing for a regression of the internet? Any sort of decentralized rating system to delinate which google.com will give you the search engine and which google.com will land you on a malware ridden site will be abused to the upmost and give the network a bad name.
Lol…can’t believe you are arguing this - so are the small photography
studio or blogger’s site. Big business should be given preferential
treatment to the detriment of small business entrepreneurs and
individuals…yeah…right.
Except that Mr. Macdonalds photography studio only pays one salary, Macdonalds fast food Inc pays 420,000 peoples salaries. now who stands to benefit more from the domain?
It does indeed seem that the domain name itself plays a small role in that scenario. What I mean is that the domain name doesn’t really mean much unless there is some information about how it applies to the situation.
What I need is a referral.
It can be a referral from “past me” to “future me” (bookmark/petname), or from an outside source (search result/emailed link). But the fact that navigation is typically done based on referrals means that they can only mean something once the referral is received.
In other words, the domain name is meaningless without the context of the why. Now - with that context - mapping personal context of why onto the domain name becomes important.
But when you visit that site you’re familiar with - by going straight to the domain - you’re forced to utilize the domain name that someone else chose (a referral with a global name), instead of being able to denotate the site as what it would mean to you. You’re forced to remember both the why as well as the what, instead of combining the two into one namespace.
safenetforum.org, while my bookmark for this site - what I type in my browser to get here - is just safe (my local referral). But it’s the only safe that I have for myself; it only points here - and nowhere else. [1]In this example, it’s more of a referral from my past self, and it only applies to me, locally. I could have also denotated it as safe.forums, or safe#decentralizetheinternet, or anything else that would make sense to me - I could have come up with my own scheme for myself (based on the domain name or not).
Likewise, when I receive a referral from an outside source - a search result, or an emailed link - I’m still forced to utilize the domain name that someone else chose, but I don’t have to remember it, it’s just a referral. I don’t have to worry about the domain name in that context.
PythonListOfLists instead of https://stackoverflow.com/questions/11487049/python-list-of-lists. If I don’t, there’s no harm done in seeing the location of that information as that long, unwieldy URL.So in any given namespace system, does the relevance come from the domain name? Or does it come from the why (the “why I want to visit this site”) - with the global name (the what) as a “suggestion”?
[1] For more information on why bookmarks only solve half of the problem though, see the section “Browser Bookmarks” here.
P.S. I want to remain on the subject of “The What and the Why” for at least another back-and-forth, but to address the rating system (which is a tangent of what we’re discussing) you can refer to the first paragraph of the “More Detail, and Interactions” section of the link above.
Basically, while that may be true (paying to push up ratings), if you create search engines that are centered around the why of the search, the domain name itself becomes more and more irrelevant.
Re: Mr. McDonalds - You are still arguing that “might makes right”. This is flawed thinking.
As for the rest, as you can see in the post below yours, typing in URLs is becoming an increasingly rare occurrence, and I believe that they can be safely eliminated altogether.
In fact, it’s bound to reduce the surface area of social engineering attacks such as typo squatting, phishing, and other mimicry attempts. (Goodbye Cryptowall)
Since there is no globally unique human-memorable addresses in the Petname System, URLs/URIs will be deprecated. They will be replaced with a Key input to reference unique data. This will be used sparingly, but it is still relevant in order to be able to navigate to a specific site without any digital Referral.
– My Petname System RFC (PR)
Also, the rating system is not an inherant feature of a relative naming scheme. That is something that search engines can, but need not include in their search rankings.
The big Corporation. Now I can’t believe you’re continuing the argument! 
Nice try…MacDonalds the Corporation benefits - the question for the “employees” is whether working for MacDonalds benefits them? A growth in MacDonalds does not correspond to growth in employee benefits.
So, an even bigger Corporation, should be even more preferred, based solely on how many people it “employs”.?
So a Sweat shop employing millions at slave wages, raping the planet etc should be preferred even more?
Anyway, just a differece of opinion/viewpoint - depends what your priorities are.
The naming system is not the right vehicle for putting big corporations in their place.
Ranking domains based on popularity is just a more dynamic system of what we have now, but without the centralization. The phonebook approach is ideal. If most people expect to see McDonalds Inc. when they type in McDonalds.com, then the corporation should indeed get McDonalds.com. The farmer will have to resort to McDonaldsFarm.com, or something, just like he would today. As it turns out, that’s pretty intuitive for people anyway, and they’re used to it.
The phonebook approach implies that two people can suggest that they be referred to by the same name, even though they’re two separate entities.
Then you go on to use a globally unique human-meaningful identifier approach - which is not aligned with the phonebook approach.
So…what support exactly was your post intended to convey?
EDIT: See the next two posts for clarification.
Then you go on to use a globally unique
No.
human-meaningful identifier
Well of course.
which is not aligned with the phonebook approach
Au contrare. I was mainly responding to the previous post, but what I am supporting is the phone book approach - that is, multiple addresses per domain are allowed - but dis-incentivized.
You see, unlike the phonebook, the results won’t be in alphabetical order. Instead they’re ordered by a popularity metric, such as traffic. For example, picture yourself typing staples.com in the address bar of Firefox, with the autocomplete search results appearing below. If you hit enter, you’ll automatically be taken to Staples.com, but you can also arrow down a time or two before hitting enter to go to other matching pages, whose order is inspired by your browsing history.
Now imagine that same mechanism used for the phonebook instead. Without a second thought, “just hitting enter” would take you to Staples, the office supply superstore - but now you can also go down to another “result” in the phonebook, with the same domain but not quite the popularity of the corporation - for example, a trending blog about a cat named Staples. You can tell them apart by the brief descriptions shown there.
If you consistently check the cat blog, your browser will promote it to the preferred result. But until then, the default behavior aligns with most people’s expectations.
The incentive for the blogger, of course, is to set up another, less competitive domain like StaplesTheCat where he will surely dominate, albeit with a longer name. He would now have two domains for the same site, to reach lots of visitors.
No uniqueness required. But as you can see, uniqueness is indeed incentivized.
I’ll just say this: The way that they are ranked will be the subject of much debate. I firmly believe, though, that this would be the responsibility of the search engine, not the network itself. Which is good, because it leaves the door open for competing algorithms and approaches.
And you go on to touch on a very good point:
You know what? That kinda just says it all right there. Well done.
I think you’ve missed earlier discussions on this…
Creation of accounts and ergo safenet domains is free.
This is not true - it is the case now, but only during testing. Read up by all means, but I don’t think there’s a need for a new discussion at this point. When the details are being decided maybe.
Hello there.
I want to raise an issue which in my view is of crucial importance for the widespread adoption of SAFE network. That is of domain name reservation.
Facts:
Risk:
It is possible to reserve a virtually unlimited number of domains. Moreover, the process of account creation could be easily automatized (bots), in which case there is no real limit to domain reservation. That will grant to a few individuals/companies - for almost no cost - the right to a huge number of domains, including high-value domains like facebook.safenet, google.safenet, etc etc. This would represent exactly the opposite of one of the founding the principles of SAFE network, which is, decentralisation.
Why this might be a problem? Because if you want important migration from www domains to safenet domains, a high cost of buying domain from opportunistic owners will discourage them to do so. Ok, maybe not big players, but certainly other minor ones.
Notice: this is already happening. Sites like facebook, amazon, apple, google, cnn, ferrari, nestle, realmadrid (!), etc, are not possible to create anymore.
What would be the solution? Under which criteria could domains be reserved? Should creating domains have a cost?
I open the discussion.
How is account creation being limited these days? I coudn’t find the most up to date solution on the forum
It seems to me that any sort of PUT to the network should cost safecoin or some sort of proof of work, otherwise the network is open to spam attacks.
Last testnet it was limited by an “invitation” system.
In the network when (test)SAFEcoin is active then a coin will need to be spent to buy a “PUT” balance and account creation requires at least one PUT from that balance. It would be all rolled into one so that you don’t have a put balance without an account or account without a PUT balance.
That is the plan (see above) and the coin spend will ensure you have a PUT balance to upload some data.
Now the question is how do we allow people who want an account who do not have a coin.