There's a script that looks for "suspicious" voting patterns between users. The details are kept intentionally vague, and normally you shouldn't ever be aware of it...
But it is possible - and in this case very probable - that 63 up-votes cast on your answers by another user were nullified early this morning.
Sadly, as one of the top users on the site, you'...
This is not a bug, but accurately reflects the reputation gained over a particular period of time. If you look at the Users page you can select the period of time:
So it's entirely possible for a user who gives a lot of bounties to have a negative reputation when calculated over one of these smaller time periods.
They are simply different, so this is not a bug. You have one quota of comment votes, and a different quota of question votes. Also, voting on comments is less "serious" -- it doesn't influence reputation, and one cannot downvote -- than voting on questions/answers.
See e.g. 'Daily vote limit reached' only for comments
AFAIK only users with more than 200rep are taken into account. The rest don't count yet as full site members. If you go to http://stackexchange.com/leagues/29/ you will see in the right side bar that only about 10% of the users have 200rep or more, which explains the factor of 10.
Being a user with good reputation doesn't mean you will never see a CAPTCHA.
It does mean that you won't see it as much as a new/anonymous user.
There are a lot of different factors that go into when it comes up (including posting many posts in a short time period).
You do not have enough reputation to add another bounty. You added a 250 rep bounty on December 5th. For you to add another one, you must double the reputation. You do not have 500 reputation to offer such a bounty.
The Data Explorer infrastructure relies on past data, not real time. In fact, it uses an anonymized data dump of all creative commons questions and answers from the Stack Exchange family of websites.
We can see the last data update in the front page:
Once in a while, the data is updated. There are also backups in the XML format available for download in ...
For the purpose of pretty much everything where it matters, a "new user" is anyone who doesn't have the remove new user restrictions privilege.
If you have less than 10 reputation and post an answer to a question 30 days or more after it was asked, your answer is going to end up in /review. We'll dequeue it if your reputation exceeds 100 before the post is ...
It is status-bydesign, as per-site metas (or child metas) don't actually have an exclusive profile. :( The only exclusion is Meta.SO, since it covers network-wide meta-related content, and have individual user profiles separate from StackOverflow.
As reference, see Hat not showing on site meta.
I believe there's an answer on MSE that addresses the issue. The important bit is:
A colon and %3a are semantically identical in URLs per the RFC.
If you look at an example on Gaming (and read the comments), you'll see that the problem should actually be fixed by the site that serves pages with colons in the URL. This is actually a bug at Gitorious.org....
From the Chat FAQ:
Will these rooms exist forever?
Rooms will exist indefinitely, so long as there is at least one person
actively talking in the room. A room is considered worth retaining if
it has more than 15 messages by at least 2 users.
Rooms not worth retaining which are inactive for 7 days will be
deleted. Rooms worth retaining ...
Flag weights above 500 are by design displayed with four digits after the comma. Up to a weight of 500, a "helpful" flag increases the flag weight by 10, but for higher weights, the increase for a helpful flag is less than 10 -- IIRC, it is exactly 1 for a weight of 625 and 0.1 for the maximum flag weight of 750. Because of the somewhat counter-intuitive ...
This seems status-bydesign. Based on the current time of 2014-09-10 14:31:43Z,
"today" is defined to be the time range 2014-09-09 14:31:44Z to 2014-09-10 14:31:43Z (or the past 24 hours) and is listed in a "just now", "Xs ago", "X mins ago" or "X hours ago" fashion;
"yesterday" is defined to be the time range 2014-09-08 14:31:44Z to 2014-09-09 14:31:43Z (or ...
This happens because Markdown treats such syntax as an ordered list:
Ordered lists use numbers followed by periods:
It’s important to note that the actual numbers you use to mark the list have no effect on the HTML output Markdown produces. The HTML Markdown produces from the above list is:
Markdown tagging seems to be site-specific, even across the main-and-meta components. While the same holds on Meta.SO - where the separation between main-and-meta is more apparent since each has their own reputation system - a generic network-wide setup is probably why it works this way. Another motivation behind this site-specific tag markdown is that it ...
As Oded mentioned, there are a number of factors that may cause a CAPTCHA to appear. One of these is time. Perhaps the time was very short.
I've typed up answers while riding the bus, and switching to a different network before posting made CAPTCHAs appear as well. Not sure whether there might be linkage with the locality (network) when it comes to saving/...
This is no bug, it is a normal behaviour. Quoting the meta.stackoverflow.com post "What is flag weight?":
Why is my flag weight not an integer?
Because flag weight above 500 does not increase in a linear fashion,
decimal places are shown for precision purposes and to show you how
your flag weight is changing.
There was a bounty on this post (visible to 10k users) worth 50 points that is not included in your profile because it is deleted. It was there and it was spent, we do not show these items in user profiles though.
Pluralizations is hard, and fix are tricky; this has been discussed many times before. Intriguingly, there may be some better options for this coming up in the medim-term future, depending on how some internal changes pan out. But at the current time this is by-design.
You should really, really take a deeper look on the equation! Is it perhaps caused by a bad monitor resolution?
The same background is used on the main site TeX.SE. For the background of this background you should read Site Design Ideas (updated with mockup) (and later New Design Launched ).
For the equation see especially this answer: https://tex.meta....
We now double the minimum "spend" for successive bounties on the same question by the same user. So if you started at 50: your next bounty on that question has a minimum of 100, the next 200, the next 400, the next (max).
This has been designated status-bydesign on Meta.SO. And, while the questions linked there have been successfully adjusted to reflect updated user names, some posts still hold an incorrect username in the revision history (like the one linked in the question).
That comment has probably been deleted before you went to read it. Seems they don't get deleted from the inbox; I think I've encountered this behavior before as well. Perhaps a mod can confirm this, as they might be able to see deleted comments?
The current thinking is that badges are supposed to be somewhat exploratory in nature, so being extremely explicit about what to do runs counter to the design.
The descriptions are sufficient to get people started, and those that are extra curious will, I am sure, be able to find other resources without too much difficulty.
This feature is not part of the stackexchange feature set. New features can only be requeseted on the primary meta site https://meta.stackexchange.com/. Feel free to post your request there.
If you want to have certain comments under your post removed, you can flag them for moderator attention. A moderator can then delete them. The neutral way is to flag ...
This was reported by Joseph and it is not considered a bug but by 'by design'.
If it is still an issue I guess it would be best to make a new post about it since the post that Joseph's is marked as a duplicate of has been removed.
This comes down to a question of "how valuable is this info?" vs. "how much noise does it add?".
In this case, it adds more noise than value. We show the latest date of that string of events, e.g. "upvotes on post X". How much does it matter exactly when each vote happened? Not that much 99% of the time, so we went for simple - it's an intentional design ...
Given that people can change their minds, this isn't a problem (in the same way that you can retract a close vote).
There isn't a good reason to disallow close votes - given that edits or additional comments may push the question to something that you'd rather close (for example).