Subtle changes, bugs and feedback

sno...tter, pls. @bowell.

That's not as lulz.

Yeah, well, neither's your face.

I do agree that the goal of an egalitarian forum is noble one.
But in real doubt about what you define a valid 'motivation' for following someone?

I didn't suggest you were interested in building a globally successful site valued at millions of dollars and have a movie made about it.

Simply that facebook uses gamification and is doing well as communities go
(btw the LFGSS tab is missing a on facebook)
https://www.facebook.com/groups/7975444238/

The Facebook group is not public, and requires a Facebook account to view it. I can't even verify the existence of it.

It exists, but it's a bit shit.

True, my face is srs bznz, unlike yours.

a b owel?

A b a owel?

@Dramatic_Hammer and @bothwell

If a URL is:

http://uk.weather.com/weather/today-London-UKXX0085

And we parsed the URL, and kept only the last bit and the domain (in your example):

uk.weather.com/.../today-London-UKXX0085

That works for that URL.

But if the URL is this:

https://www.google.co.uk/search?client=ubuntu&channel=fs&q=london+weather&ie=utf-8&oe=utf-8&gl=uk&gws_rd=cr&ei=8o8HVNWtLIav7AbbsIEI#channel=fs&gl=uk&q=london+weather&spell=1

Then should the answer be:

http://www.google.co.uk/search

Which is meaningless.

And as the page is identified, and the real meaning of the page is actually all within the query string... should we keep that and lose the path?

http://www.google.co.uk/...?client=ubuntu&channel=fs&q=london+weather&ie=utf-8&oe=utf-8&gl=uk&gws_rd=cr&ei=8o8HVNWtLIav7AbbsIEI#channel=fs&gl=uk&q=london+weather&spell=1

Or lose the last 20% of this?

http://www.google.co.uk/search?client=ubuntu&channel=fs&q=london+weather&ie=utf-8&oe=utf-8&gl=uk&gws_rd=cr&ei=8o8HVNWtLIav7AbbsIEI#channel...

URLs are determined not solely by the path. Whilst a lot of URLs can be identified by the domain plus path, a not insignificant number can only be identified by the query string... and the important parts of that could be anywhere within the querystring.

In the example above, the most important part is the centre of the string (that is the URL).

Most of the specs on this, and most of the way in which web sites are constructed, mean that the URLs follow a hierarchy of resources... that the left-most part of a path is a higher level section of a site hierarchy.

The leftmost part does not strongly identify a single specific resource, but it does strongly identify the area to which the resource belongs. This is an important distinction as a basis for trust is why we're showing anything at all.

An example could be reddit... if we removed the middle part of a URL, an untrusted section /b/ might fool you into visiting a URL because you trust the last part of the path (it meets your expectations). The middle part, which you wish to remove, is a stronger candidate for trust than the last part.

There is no flawless way to determine, or guess with any accuracy, which part of a URL is the best bit for trust. But the way in which the web has evolved means that you trust the domain first, the left most part of the path next... and you move rightwards.

This is why we're going to keep on using the most significant part. There will always be exceptions, but without understanding the URL structure for every destination a good default general rule is to keep as much of the leftmost portion of a URL.

I still don't see the problem with displaying the whole lot.
Someone posts a link, and you don't rewrite it. so what?
So what if someone wrote a script to go to that link and bypass your rewriting, cos you put it in the title?
It's only going to end up at the same place as it was going to end up anyway.
It's posted by person a and clicked on by person b. It's their fault, not yours.
So what if there is malware on someone's machine. How does that affect anyone but them?
If you've got spam detection for the rewriting, then you're able to prevent the title being the full URL in those cases. But I still don't understand why you'd want to even then?
I don't understand what anyone would hope to achieve writing a script that swapped titles for hrefs.
I don't understand why this is something you fear but not someone writing a script to prefetch links and get the full urls that way.
I don't understand what possible harm can come of someone having the proper URL of a link.
Linking to pages is how the internet works. No one else seems to have an issue with it. Have a disclaimer about external links if it would help cover you legally.

This is why we're going to keep on using the most significant part. There will always be exceptions, but without understanding the URL structure for every destination a good default general rule is to keep as much of the leftmost portion of a URL.

The internet disagrees with you, though. Y'know, it's not just me and dram...mer.

http://ux.stackexchange.com/questions/40326/best-way-to-handle-long-urls

Loads of websites do exactly this and it's honestly a pretty standard way of truncating long urls.

URLs are determined not solely by the path. Whilst a lot of URLs can be identified by the domain plus path, a not insignificant number can only be identified by the query string... and the important parts of that could be anywhere within the querystring.

Surely this actually supports the opposite point of view - if a querystring can have important info anywhere along it, then displaying the first x characters is as arbitrary as displaying the last x characters, no?
And I would that, of the subset of people who actually check tooltips, if they see a querystring, they would then assess the domain almost entirely - i.e. if I'm worried about NSFW content, it doesn't matter what a google querystring is, as I'll have safe search turned on.

I'm with A B_owell here tbh

On my android the keyboard doesn't allow me to ((((( go ))))) when I've finished typing and actually want to post.

I have to close it, then scroll down find the button and hope my fat fingers don't miss.

1 Attachment

Thank fuck for that. It might stop you posting so much shit.

just sent a pm to multiple people and whilst entering usernames for people who I wanted to send the message to, I'd start typing, get the list of names, and the people I interact with most would be at the top, I'd select their name, hit enter and someone else's name would appear in the send to section, as if their name had been selected instead.

Happened three or four times with the same name, and then with other names. Not sure whether the autofill is a bit slow in populating and so when I hit enter it's still figuring out who should be top, but it is very disconcerting. I've highlighted a name, hit enter and someone else's name appears...

Events: clicking the "Yes, I am attending" button does not subscribe you to its thread, so that you do not see any comments on it in the "Following" list. You have also to click on the "Follow" button to get those, which has surprised me and caused me to miss updates to plans (which tend to go in the follow-up posts and not the original post). I can see a reason for both buttons being there, since people may well want to follow an event even if they haven't made up their mind about attending. But I can't see why anybody would want to go on an event but not see updates. Would it not make more sense to automatically subscribe people who join an event? They can always unfollow if that is really what they want.

I don't get it, Bruce, can you be more expansive?

BTW, I read your post about bars being open in airports, this morning; I swear it came out in my head in your voice. +1, would hear again.

Sounds sensible... I'll do this tomorrow.

As a follow up, is there a setting for getting an email notification about an event, say 24 hours in advance, if you've ticked 'attending'? I've not actually been in a situation to see if this is the case but it would be useful for those of us with memories like a leaky-thing

There's actually a notification setting for that... but we don't know what the message looks like yet.

Please examine your use of the word "comment". A "conversation" with 3 "comments" doesn't have 3 comments. It has one post and 2 comments. Why don't you just call everything a "post". Whilst a comment may well be apropos of nothing that is not what is happening here.

In "messages" I find the use of the term "comment" most confusing.

The way we're going with terminology, and this work is incomplete at present.

• Sites = unique domain name, top level container
  
• Sections = what are called forums or microcosms today, they are spaces within the site into which posts can be made such as conversations, events, and later on polls, classifieds, reviews, etc. "Section" does not imply hierarchy, or categorisation (sections do not need to conform to a convention), it's more inline with magazine, newspaper and web publishing concepts of an area of the site.
  
• Posts = The collection of all types of things that can exist within a section. Can be verbed easily, "Post a conversation" and makes sense for things like events where the type of event would usually change the language involved... "Post an event" is neutral and highly communicative. This is consistent with Wordpress, Tumblr, review sites, some discussion software, other publishing sites, etc.
  
• Conversation = A collection of comments... the conversation is merely the topic that is being talked about, the first comment is just that. This is equivalent to vBulletin "thread" which is a technical term that has roots in the representation of a conversation as a tree structure, and as parts of the tree represents comments that refer to other comments this is known as a "threaded tree", which was reduced to "thread" by the product wwwthreads in the late 1990s. We're calling it what it is... a conversation.
  
• Event = A block of metadata describing something that happens/happened somewhere, and that may have attendees/participants. The first comment is just the first comment, the event is the metadata that preceded it. An event may not necessarily have comments that form a conversation, it's optional.
  
• Message = A private message, directed at a specific individual or individuals... the message is the collection of comments. There is no direct equivalence for what we've built elsewhere, it most closely resembles a spies' dropbox or a Rugby team coming together in a huddle to discuss something... it's a temporal store of a collection of comments shared with specific individuals
  
  

At present... some of our use of these terms are inconsistent. We currently say things like "Post reply", but we should reserve the word "post" so that it is only ever used in relation to putting something in a section.

We currently say things like "Post reply"

"Append a comment"? We've been metaphorically and literally posting stuff since the invention of writing, when you had to actually find a physical post on which to write your comment, or later after paper took over to which you could nail your missive. And of course, it's still the method in http,

<form action="/comments/create/" method="POST">

Totally true.

But on the web, people seem to suddenly freeze when two entirely different concepts are referred to by the same verb.

It becomes a comedy sketch of unnecessary confusion, "Post a comment? A comment is a post? But I posted an event, an event is not a comment? WTF is going on, I don't even!... (╯°□°)╯︵ ┻━┻ ".

Rage abounds... we'll avoid using obviously reasonable English when it clearly leads to some people having apoplectic fits, which only harms engagement.