|
Size: 3061
Comment: Revert spam
|
Size: 1408
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | Line 1: |
| This is a sketch of UX and protocol considerations for a client which wishes to help users use gateways. | Hello there, |
| Line 3: | Line 3: |
| In every place where the UI would ask for a Jabber ID (most notably contact add, but possibly also invite to chatroom, create chatroom from 1:1 conversation, start voice/video call to non-contact, search view where a Jabber ID can be used, etc) it should scan the user's roster for JIDs which have `disco#info > identity[category=gateway]` and show some way of choosing to enter an identifier for that gateway instead of a Jabber ID. The simplest example might be a drop-down with `disco#info > identity[type]` and optional disambiguating `disco#info > idenity[name]` such as "sms (Cheogram)". | There is a collection of online snapshots you may not know about, and it can contain a copy of the site you thought was gone forever. I'm not exaggerating. The Wayback Machine has been quietly saving versions of the web for decades, your old pages, images, and files are very likely sitting in there right now. |
| Line 5: | Line 5: |
| If a user has unambiguously selected a gateway, use https://xmpp.org/extensions/xep-0100.html#addressing-iqgateway to get a prompt: | The catch has always been getting them back out in a usable form. Scrolling through old snapshots page by page is fascinating but useless when you actually need a working website again. That's the exact problem Restorix solves. It crawls the archive for every surviving capture of a domain and rebuilds the whole thing into a complete, cleaned-up site. |
| Line 7: | Line 7: |
| {{{ <iq type='get' to='gateway.tld' id='gate1'> <query xmlns='jabber:iq:gateway'/> </iq> |
What convinced me was how simple it is: you paste in a domain and it tells you the exact file count and price in seconds, free, with no account required. You only pay when you decide to restore, and if anything fails you get an automatic refund. Most restores finish the same day. |
| Line 12: | Line 9: |
| <iq type='result' from='gateway.tld' id='gate1'> <query xmlns='jabber:iq:gateway'> <desc> Please enter the Whatever of the person you would like to contact. </desc> <prompt>Whatever</prompt> </query> </iq> }}} |
If you've ever lost a site to a bad update, an expired host, or a crashed server, do yourself a favor and run the free check at |
| Line 23: | Line 11: |
| This text may then be used as part of the UX as makes sense to help the user feel comfortable. Getting a success here and not an error also means that `jabber:iq:gateway` can be used in the next step. | https://bynd.li/restorix |
| Line 25: | Line 13: |
| NOTE: it is also acceptable (and in some UX may be preferred or even necessary) to pre-emptively ask all gateways for the prompt, instead of only once one is selected. | Restore a website under a minute from Wayback Archive (Public Archive.org Backups). |
| Line 27: | Line 15: |
| If an identifier is (or might be, depending on UX) one for a particular gateway, use the following procedure to map it to a Jabber ID and then perform the in-context relevant action on that Jabber ID as though it had been entered directly: | |
| Line 29: | Line 16: |
| 1. If a prompt was got for `jabber:iq:gateway` above then use that protocol to finish resolution: {{{ <iq type='set' to='gateway.tld' id='gate2'> <query xmlns='jabber:iq:gateway'> <prompt>(555) 123-4567</prompt> </query> </iq> |
|
| Line 37: | Line 17: |
| <iq type='result' from='gateway.tld' id='gate2'> <query xmlns='jabber:iq:gateway'> <jid>+15551234567@gateway.tld</jid> </query> </iq> }}} 2. If an error is received in response to `jabber:iq:gateway` protocol, print the error message to the user and allow them to try again. This allows the gateway to provide useful feedback about format errors to the user. 3. If an error was received in response to fetching the `jabber:iq:gateway` prompt, then the gateway does not support this protocol. Look for a `disco#info > feature[var="jid\20escaping"]` and if found construct a JID using XEP-0106 escaping on the user input concatenated with `@gateway.tld` 4. If `jabber:iq:gateway` and `jid\20escaping` are both unsupported, replace any `@` in the user input with `%` and concatenate `@gateway.tld` |
|
| Line 47: | Line 18: |
| NOTE: If the result of this procedure is a string that is not a valid Jabber ID, it should be rejected just as if a user manually keyed in an invalid Jabber ID. ---- CategoryHomepage |
When you decide not to receive additional messages from us, simply fill the form at brnd .li/delist webpage with your domain address (URL). 6 Rue St Ferreol, Bellerose, CA, USA, 93879 |
Hello there,
There is a collection of online snapshots you may not know about, and it can contain a copy of the site you thought was gone forever. I'm not exaggerating. The Wayback Machine has been quietly saving versions of the web for decades, your old pages, images, and files are very likely sitting in there right now.
The catch has always been getting them back out in a usable form. Scrolling through old snapshots page by page is fascinating but useless when you actually need a working website again. That's the exact problem Restorix solves. It crawls the archive for every surviving capture of a domain and rebuilds the whole thing into a complete, cleaned-up site.
What convinced me was how simple it is: you paste in a domain and it tells you the exact file count and price in seconds, free, with no account required. You only pay when you decide to restore, and if anything fails you get an automatic refund. Most restores finish the same day.
If you've ever lost a site to a bad update, an expired host, or a crashed server, do yourself a favor and run the free check at
Restore a website under a minute from Wayback Archive (Public Archive.org Backups).
When you decide not to receive additional messages from us, simply fill the form at brnd .li/delist webpage with your domain address (URL). 6 Rue St Ferreol, Bellerose, CA, USA, 93879