@Jon Ericson
Yes, exactly: Jeff’s Venn diagram is incorrect for it doesn’t represent all the possible intersections between the 4 sets, only some of them.
It couldn’t show apps that lie in the intersection of Wiki and Forum but have nothing to do with Blog or Reddit, nor the ones at the intersection of Blog and Reddit which are not even remotely like Wikis or Forums.
Jeff: if you don’t like the models shown in the Wikipedia example, projecting this image of yours onto a Sphere would render it correct (the 4 circles would over the North Pole and intersect again at the South Pole, showing all the possible connections at the cost of a little redundancy).
hi jeff, didn’t know where to post it, but i would like to point out that one of the most active users, the user aku, has ussr’s hammer and sickle in his icon, which is a little bit offensive for some people, like using nazi’s swastika as icon would be offensive for jews.
The incompleteness of the n=4 Venn diagram (yes, I noticed too, heh) is a non-issue if the emphasis is solely on the region of complete intersection, though. If Jeff wants to make a point of mapping, graphically, the intersection of Digg and Blogs that intersects ineither/i Wiki nor Forum factors, then he’s got a problem, but otherwise we’re just overnerding things.
And if we’re igonna/i overnerd this sucker, I’m wanting to see at least one site/application that fits each possible region of intersection.
(Or a compelling argument for the impossibility of pure-opposites intersections, which would justify Jeff’s diagram as functionally complete after all. In other words, establish why you ican’t/i have a Digg/Blog/No-Wiki/No-Forum site or a Wiki/Forum/No-Digg/No-Blog site.)
By the way Jeff: when posting an answer here using Firefox 3, I’m always getting a message saying in an effort to curb malicious comment posting by abusive users… before being able to post again.
If I go back to the original page, I can see my comment has been posted. If I wait and post again, I get a double post.
Nice to finally see the site after hearing so much about it. I had some interesting answers to my question that had no right answer. You did a nice job on the UI. I missed a couple of features because I wasn’t expecting it to be that user-friendly. This will definitely be a site I check regularly for some food for thought, and a site on which to bounce off my periodic software neurosis.
The only problem I’ve seen is that I think I saw one of the answers to my question disappear. Can users retract their answers?
I vote for OpenID too, its great. (admiteddly I do not use Yahoo’s implementation so I might be using it right)(Hint: go to myopenid, they’ve done a lot right).
What I cannot understand though, is all the people asking for a ‘throwaway’ login id. Isn’t OpenID exactly this? You don’t need a single openid account for all sites you know, you can have as many as you like. So go to myopenid, create your throwaway userid/password (or certificate, or phone callback) there and use it to login to stackoverflow.
I reckon they’re afraid they’ll realise they can reuse the same id on other openid sites and then start liking it. Won’t that be a terrible thing!
Sorry, but openid is a tall cold glass of fail as far as I’m concerned. If that’s the only way to properly get to grips with stack overflow then I won’t be getting to grips with stack overflow.
It’s pretty fun to go through all the unanswered questions, but as a few people here have said, they get answered pretty quickly, usually while you are in the middle of trying to answer them. Since there are so many people eager to earn reputation, questions very quickly lose their ‘unanswered’ status, even if none of the replies really adequately answer the question. I would suggest being able to filter unanswered questions a little more leniently, such that, say, if none of the answers have been upvoted at all, it will still be considered unanswered. I’d suggest filtering by ones that the original poster hasn’t marked as The Answer, but it seems like the majority don’t ever get that marked.
Maybe if the original poster can check an option that negates its ‘answered’ status, even if there have been responses, if none of them are appropriate, so questions that genuinely haven’t been answered will still be easily found. Sort of like bumping their question, but without having it bubble to the top again. Maybe they could be allowed some small number of times they can perform that action per question, to limit it.
So I’m all for programmer art… hell, I have lots of programmer art.
But you’ve got to do something about that layout. As it stands, I can’t stay to look at the site for more than a few seconds because it’s designed so terrible. I see a lot of the comments about that have been shut down immediately with little to no explanation.
Listen, while I hate, hate, hate Expert’s Exchange, this site is not going to be the EE killer unless you do a far better job with design.
Tried it, liked it. Found myself answering questions that over-zealous moderators had removed before I could hit submit. Found comments from moderators closing questions they deemed too subjective or off-topic. Trying to force the group, rather than filtering your own view, is fail. Won’t be going back for some time.
OpenID works well with me in SO but I am seeing a lot of complaints from others about OpenID. Therefore SO should not rely solely on OpenID for authentication. A regular username/password should also be used. OpenID should be an option but not the only option.
Fine layout. Easy to use. Great idea to edit own answers. Nice gimmick of badges. Search function great. Already enough users to build a broad base.
Some problems:
Both highest rated answer and accepted answer should be shown.
Some people ask hasty questions and accept answers which are a bad
solution (unsupported hacks etc.).
philibertperusse has already mentioned it: People are eager to
earn rep and therefore are not inclined to downvote questions.
Idea: Only people who haven’t already answered may up- and downvote
a question. If question is upvoted, every answer earns an extra
point, if downvoted, every answer lose a point. So good/bad
questions are marked out, but the very small gain for everyone
stops people from cheating (socketpuppetry for questions and
answers).
The identity is irrevocably lost if you edit anonymously. And I have
something to say you possibly don’t want to hear:
I.don’t.want.OpenID. OpenID.bugs.me.It.is.a.real.pain.
It is a solution for a non-existing problem. Joel has already wrote
an article about synchronization platforms like Windows LiveMesh
who wish to fulfill one place for all my files.
OpenID wants to fulfill one identity for the whole web. Guess
what, noone needs both of them.
If you always use the same identity, you can already auto-login
with the browser (and you don’t care about security
anyway). If you don’t use the same identity, OpenID is wholly
pointless. Please give an alternative. Please.
I can’t bow down to OpenId yet, but it appears you are after a self-healing/blog/wiki/twitter upgrade/ source of info, which I think is a good concept.
Execution will be rough for a while while those who can/cannot refine the answers gets sorted out, and you separate fluff from that’s nice from solid bottom line info.
