« first day (1807 days earlier)   

4:52 AM
@MartinSleziak The OP deleted his question, which lead to removal of the tag (at least for now).
Another new tag: .
0
Q: convergence of a complex sequence of function

Urgent$f_n(z)=\frac{z^n}{n}$, where it is uniformly convergent? well, $\frac{z^n}{n}\to 0\forall |z|\le 1$ am I right?

 
 
4 hours later…
8:39 AM
That is the strangest tag wiki ever: math.stackexchange.com/tags/catastrophic-cancellation/info just links to the only two questions tagged, and one .pdf file on a personal website (on a university server). I'm not even sure the tagged should be kept at its current state.
 
 
1 hour later…
10:02 AM
It seems that the tag was created in this question:
10
Q: How does rewriting $x^2 -y^2$ as $(x+y)(x-y)$ avoid catastrophic cancellation?

anonymousWhy is rewriting $x^2 -y^2$ as $(x+y)(x-y)$ a way to avoid catastrophic cancellation? We are still doing $(x-y)$. Is it because the last operation in the second form is a multiplication?

The other question in this tag is
5
Q: how do you compute $\|c-a\| - \|b-a\|$ without catastrophic cancellation?

Don HatchGiven three points or vectors in the plane: \begin{align} \vec a &= (a_x,a_y) \\ \vec b &= (b_x,b_x) \\ \vec c &= (c_x,c_y) \end{align} How do you compute $\lVert \vec c - \vec a \rVert - \lVert \vec b - \vec a \rVert$, i.e. "how much farther is it from $\vec a$ to $\vec c$ than from ...

A user who created the tag-excerpt and tag-wiki is different from the user who introduced the tag.
Wikipedia article on Loss of significance: "Loss of significance is an undesirable effect in calculations using finite-precision arithmetic. It occurs when an operation on two numbers increases relative error substantially more than it increases absolute error, for example in subtracting two nearly equal numbers (known as catastrophic cancellation). ..."
You can find some occurrences of this phrase in Google Books and Google Scholar, so at least it seems that the topic really exists.
@AsafKaragila I have added some intro to the tag-wiki, but left the links in. It would probably better if somebody with better knowledge of numerical mathematics looked at it.
If the issue is important enough, perhaps it could be raised on meta or we could ping the users involved. (The tag-creator and the tag-info creator.)
@suomynonA Could you have a look at the above comments related to a tag-wiki you have created.
@AsafKaragila Or if it is not that important, you can simply edit the tag-wiki.
If we really want a good tag-wiki, mentioning it on meta could be good - since then there is bigger chance that somebody who knows enough about the topic would look at it. OTOH I am not sure whether a tag with two questions is really worth the effort.
 
10:39 AM
@MartinSleziak Really, Martin? Really? I've been on this site for nearly seven years now, out of which I was active on meta topics for at least four-five years, and you really had the thought that I might have felt that this tag is worth saving?
 
@AsafKaragila Well, you mentioned tag-wiki, not removal of the tag.
I do not know enough about floating point computations and numerical arithmetic to be confident enough to judge how important this topic is and whether it should have its own tag.
However - as a user who has "been on this site for nearly seven years now" - if you think that tag should be removed, apart from mentioning it here in chat, either editing the tag from the two question and making a post in Tag management thread are reasonable steps to do.
To be honest, it is not clear (at least to me) from your last message what you mean by worth saving. Whether you mean the effort invested into editing the tag-info into a better shape or whether you mean keeping the tag on the site.
On a less serious note, I highly doubt that it is the strangest tag-wiki :-)
 
10:59 AM
I wish was the only problematic tag on this site.
 
11:09 AM
As not to leave things hanging, I made a post on meta about the tag-wiki:
0
A: Questions about tag-wikis

Martin SleziakIt was recently mentioned in chat that the tag-info for the tag catastrophic-cancellation consists basically only of a few links. I have tried to improve it at least a bit, but it would be probably better if somebody who know more about numerical mathematics had a look at the tag-info. (And there...

If the tag is removed or if you decide to post about the issue yourself @Asaf, I can simply delete my post.
 

« first day (1807 days earlier)