I posted this question earlier on
but the Safe-CMS project mysteriously died out, so I would like to ask again.
Is there a reason we can’t have PROTOCOL://DOMAIN.SUBDOMAIN/PATH on the SAFE Network? This ordering would just feel more intuitive to me. E.g.
safe://sascha/index.html
safe://sascha.hunting/deer
safe://sascha.fishing/pike
safe://sascha.fishing/salmon
or
safe://google.company
safe://google.company/contact
safe://google.search/scholar
safe://google.search/images
Just use:
safe://sascha/index.html
safe://sascha/hunting/deer
safe://sascha/fishing/pike
safe://sascha/fishing/salmon
or
safe://google/company
safe://google/company/contact
safe://google/search/scholar
safe://google/search/images
Clear net DNS is for resolving names to servers. As we don’t have servers on SAFENetwork, we don’t need to resolve to them - we just resolve names to data.
I haven’t played around with website creation on SAFE for some time. I created some Web ID:s earlier using the Web Hosting Manager, but as I recall there was an incompatibility issue that made me unable to edit old Web ID:s or something. The Web Hosting Manager hasn’t been updated in quite some time, and I’m not sure what to do if I want to start testing from scratch and create a new, empty page with the address address safe://sascha/index.html. There used to be a requirement for creating a subdomain, but I’m not sure about now. Is there an up-to-date guide?
My question above really just has to do with the order in which domain and subdomain (if one is still needed) are entered/displayed in the address bar.
That is a good question. I think the reason of the chosen order of subdomain(s) before domain (or public name) is only because that is the same order in DNS, but maybe I’m also missing/not remembering other reasons.
And that DNS order is probably because it started with the hostname of the server and later the (top level) domain was appended and subdomains prefixed.
But if you think about it: starting with the public name instead of the sub name(s) seems more logical. Like the path-part.
Another question is if sub names are still necessary.
That is in the Web Host Manager. You can use ‘www’ as subdomain, so that it doesn’t show in the Safe Browser.
But in the RFC that doesn’t seem a requirement anymore (chapter 2.1).
A simple 1:1 conversion from http URLs incl. sub domains to safe:// URLs would help adoption.
safe://com/google/mail/userxyz/inbox for instance.
Yes, but I’m questioning the need to stick with old conventions that don’t necessarily make sense on SAFE. Clean slates are nice.
Intuitive to you perhaps, but it breaks every existing url parser and every existing technical understanding of Resource Identifiers.
I’m not going to agree or disagree with what’s more intuitive to me or to you, but why break an existing well-established standard?
I was WRONG: (see @neo’s post below correcting me) → In the regular web clearnet subdomains have nothing to do with DNS, they are dealt with locally by a web server.
I’m not sure how it will all work on Safe. For instance will there be a tld? e.g. .com, .org, etc.
if there are tlds … would that come first in your idea @Sascha? e.g. safe://com.google/sheets/ouertp235-9a8/
or
safe://com/google/sheets/ouertp235-9a8/
Is the ‘.’ more authoritative! 
I suppose if you were offering some sort of web service though then maybe it’s helpful to delineate between things.
If we had something like that - separate browser plugins each act as a DNS system for individual (or groups of) TLD’s … then I guess you could have a plugin that does subdomains the way you like them … you could even copy an existing plugin for an existing tld group and swap around how it processes subdomains so you get it they way you want it personally.
@dirvine what are the current plans regarding safe dns? Will there be an RFC on this or is it already worked out? I did a search on this forum but didn’t find anything very new (less than two years old).
I thought TLD:s weren’t needed. Anyway, I’m not really proposing an idea here. I’m mainly trying to understand how things work. But if @mav says ordering thing differently would break stuff and create unnecessary work, I’m inclined to believe him.
well … if it’s built into Safe, then yeah, but if managed by plugin, then the world’s yer oyster! 
I should add it’s great to hear innovation and new ideas and improvements to design.
But I’m not sure that the original URI was designed without intuition applied. As in, I think the existing design is intuitive to many people from the start.
So if you can demonstrate the existing URI system is broadly unintuitive then that would be really cool and worth putting more design effort into the safe URI system.
Yes, we have to remove the idea of domains. A domain is an area you “own”. Be it a network of 1000’s of computers, a server, or any other privileged position on the internet. It comes from real world domains such as countries (.au, .uk, etc), or regions, Lordship (of land), master of your own domain (home) and so on.
SAFE is a flat region where you have names for your IDs. Your ID can be made into a web site using safe://name/things/whatevers
SAFE is domainless you do not own a portion of SAFE, it owns you 
and you can have an ID on safe and associate some files to that ID and browsers can display those files making up a web site.
If we try and squeese SAFE back into a hierarchical structure (of ownerships etc) then we bring back the problems associated with that including the need for registrars and many other issues we are trying to eliminate.
In safe the name is “com” and its associated with just one ID. The rest is sub directories of com and google is a sub directory and owned by the person who has the ID “com”. So its not doing what you may consider it doing by thinking of domains.
Subdomains have everything to do with DNS. There is a DNS record for every subdomain. Just not necessarily on your ISPs or google’s DNS servers, but on DNS servers none the less.
True
I have not really thought of this before but we never own any of the SAFE network, we may own our Node but never any of the data/files/whatever nor the actual network. If we take our node and leave then the network just readjusts and forgets us. When we store files it is really us asking the network to keep this file safe forever and guarantee to allow us either exclusive access to it or some/everyone access to it.
Domains was a way we could own a portion of the conceptual space on the internet controlled by a bunch of registrars who followed their own guidelines as to who got what. And also marked out our territory on the internet. And we wonder why companies are doing everything to own us.
It seems you don’t like the word “intuitive” any more than I do. (At least I didn’t say “more logical”.) In any case, it’s hardly objectively quantifiable.
To be honest, I don’t really understand the difference between subdomain and subdirectory, especially on SAFE. Maybe I’m just puzzled by the current design of the Web Hosting Manager. But what @neo says above makes “intuitive” sense to me.
EDIT: Domains only apply to the current internet and not SAFE.
Subdomain is the same as a domain in almost every fashion. Its the application of it that muddies the water on the internet. www. is a subdomain of the domain following it and web servers allow www. to be mapped to the following domain or the following domain to be mapped to www. or no mapping.
Note the following is handled somewhat differently now with deregulation and commercial pressures.
.com - these domains got the title TLDs (top-level-domains) handled by internet authorities (eg .com .org .au .uk .nz .info and so on)
.e - these domains handled by the registrars associated with the TLD they handled
.d - these domains are what is usually referred to as sub-domains (eg www.) and handled by the “owner” of the .e level domains
.c - another domain sub to .d level handle by the .d domain “owners”
.b - and down another level
.a - last level of domains that the person created in the DNS, but it can go further.
The principle that each level is a domain in its own right and a sub-domain of the level above it.
There was a time in the past where users could have their website as a sub-sub-TLD (ie a sub-domain of the website hosting them). Often done by ISP for their customers.
SAFE is simply condensing out the domain structure into one name. The sub-directories are the same as for any web-site.
You saying you haven’t thought about this before even makes me a bit uneasy. Few things about SAFE have been so clear to me from the very beginning. You certainly understand SAFE better than I do, so I just hope I have been correct in my own understanding all along.
Why this is one those thought exercises, it does not change the reality of SAFE. And was not related to the second quote you quoted of me.
The thought exercise was “do we really own anything”? Typically we own something (car, food, house, whatever and then discard it or pass it on to our children when we die. What is “ownership” anyhow. When we have exclusive and guaranteed right to something we typically say we own it, but do we? If I own a plot of land, do I really own it since all it is is a “title” to that plot of land and the laws of the land say that I have rights to live there etc. But in Australia I have little rights below a few meters or above a few meters. And then some authority can remove me from the land for various reasons. What is ownership?
Please do not take a thought exercise into something that distracts from what SAFE is.
In normal thinking you are paying for uploading data to SAFE and provided with guarantees to that data according to the request. Public, private, AD data, etc
To me ownership is a different thing. If you have control over it, then you own it … if you can destroy/erase or otherwise make the data inaccessible, then you own it. If not that, then what can ownership mean? for example, do you own the atoms in your body, or do they belong to the universe?
Also domains on Safe CAN be a thing, so can DNS … as I pointed out if someone makes a plugin for the browser that operates as a DNS system would for the user … then there can be domains, subdomains, subsubdomains … just about anything really.
I get it. And I have some thoughts on that too. But it’s five o’clock in the morning here and I haven’t slept, so I’d better save those for later and for another topic. Good night, everybody.
P.S.
Agreed.
But why do we want elements of the broken internet. Domains is one of the parts that allowed the current internet to own us.
A flat naming system is all we need.
Even most web hosting companies went away from the sub-sub domain system to a simple sub-directory to allow free hosting etc.
Domains is a broken system that derived from the education system of universities in the 80’s etc and was not useful for the internet we know today.
A flat naming system and using sub-directories like all web sites do now because it is better overall