User talk:Miym

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Welcome[edit]

Welcome!

Hello, Miym, and welcome to Wikipedia! Thank you for your contributions. I hope you like the place and decide to stay. Here are some pages that you might find helpful:

I hope you enjoy editing here and being a Wikipedian! Please sign your messages on discussion pages using four tildes (~~~~); this will automatically insert your username and the date. If you need help, check out Wikipedia:Questions, ask me on my talk page, or ask your question on this page and then place {{helpme}} before the question. Again, welcome! Rosiestep (talk) 22:50, 24 February 2009 (UTC)

Copyright concerns, User:Yewang315[edit]

Hi. Thank you for noting and following up on your copyright concerns. I have also tagged the article Arithmetic circuit complexity and issued a warning to the contributor that persistent copyright violators are by policy blocked. Please let me know at my talk page if you should happen to see more suspected infringement from that specific contributor. --Moonriddengirl (talk) 15:15, 1 March 2009 (UTC)

QPL[edit]

Hi Miym. What do you mean with "notability"? I know QPL has been held for several years, and most of the development in quantum programming languages are first published there, however, I don't know which kind of reference is needed to add it to the computer science list of conferences. Cheers. --JanusDC (talk) 23:00, 17 August 2009 (UTC)

Hi JanusDC! By notability I just refer to the usual notability requirement of Wikipedia, see WP:N. In particular, we need reliable, third-party sources, see WP:V.
If you think that QPL is notable enough to warrant a Wikipedia article, then I think the best way is to go ahead, find the reliable sources, and write an article. After that it is certainly ok to include it in relevant lists such as List of computer science conferences, too. Notability is already established in the article itself and its references.
If you do not want to write an article but you would like to list the conference as a red link in the list, then we need some reliable sources in the list itself. As WP:SAL says, lists are subject to the usual requirement of having verifiable sources. See the footnotes that we have in other redlinks in List of computer science conferences for some examples.
Thanks! — Miym (talk) 07:48, 18 August 2009 (UTC)

CoNext[edit]

In the sake of fairness and consistency with our criteria, I agree with you. CoNext cannot be included until external proof or notability is provided. —Preceding unsigned comment added by Nick.dorvas (talkcontribs) 16:03, 28 August 2009 (UTC)

Adding "Distributed" in all sub-categories of "distributed" related categories[edit]

In order to convey the major intention of the sub-category, being related to distributed computing, the word "distributed" is crucial. Putting other categories as subcategories of "distributed" is not useful, but rather does a disservice. Thus also removing "distributed" categories since "a sub-category already exists" which does not explicitly say "distributed," is a disservice. Thanks. Comps (talk) 16:45, 2 September 2009 (UTC)

Ok, I don't mind if you add add the relevant entries of Category:Concurrency control also to Category:Distributed computing problems and the relevant entries of Category:Concurrency control algorithms also to Category:Distributed algorithms. — Miym (talk) 17:39, 2 September 2009 (UTC)
Miyum in the "List of computer Science Conferences" you removed some conferences because they are duplicates. They are supposed to be duplicates.

Many conferences belong in multiple categories, e.g. Usenix Security is "Security, Operating Systems and Networking". Please revert your changes. —Preceding unsigned comment added by 75.189.248.175 (talk) 16:32, 31 October 2009 (UTC)

why are you trying to delete our page?[edit]

hi we are working on promoting wikipedia in the balkans with the sfk. why do you try and get it deleted? please help us make it better, not destroy it. thanks, mike Mdupont (talk) 22:06, 2 September 2009 (UTC)

Hello Mdupont, I am trying to help you make it better, see my comment on Wikipedia:Articles for deletion/SoftwareFreedomKosovo. — Miym (talk) 22:26, 2 September 2009 (UTC)

About the links you removed[edit]

HI Miym,

I am User Bracchesimo

Sorry for my pedestrian approch, I am new to wikipedia. I added a few links that were removed from you, let me give some explanations.

All links are strictly related to article, for example:

In the Nearest neighbor search page I added a link to my library like others did. You may say mine is not worth to be published but at least you should try it before deleting. I can ensure you it works good.

In the Poker probability (Texas hold 'em) page I added a link to my Open source calculator. What's better tha a software probability calculator in this page? there is a link to some poker tables that in my opinion is much less than this software.

In the Convex hull page I added a link to my Convex Hull Algorithm (Open source), I was also planning to write some text about it, since I judge that section not so good. The link points to a video demostration (very instructive), isn't that good?

I did tha same for a Point in polygon algorithm, Dilation (morphology) and Convolution... They are all striclty related o the article. More, they are much better than some link already present.

I am not a spammer, I would have placed them automatically instead of getting an account, show my face an patiently add them one by one. I am sure if some expert test those codes the link will result appropriate. I thing you were to impulsive removing them, I have a website, I understand your position, I receive huge amount of spam and sometimes I delete it without even looking at it. I know what spam is and this is not.

I am not expert of wikipedia discussions and don't have enough time to learn. If the link will be removed without even see where they points to I wont place them anymore. I am not advertsing, I don't want to insist. I hope you will understand.

My Best regards

Luigi

Don't gratuitously remove red links[edit]

You summary in this edit makes it look as if your only reason for deleting the link is that the article doesn't exist.

If so, that violates this Wikipedia policy.

Red links, i.e. links to non-existent articles, should be left intact if an article with the title that is linked to ought to be created. Michael Hardy (talk) 03:55, 9 October 2009 (UTC)

I'm sorry, my edit summary was a bit too short. Of course red links are fine, but then in a list like this we must have references that show the notability of the red-linked person.
(I'd be particularly careful with the sourcing of red links in this particular list. First, it may have WP:BLP issues. Second, a "list of prominent people" easily attracts vanity entries.)
But thanks for catching this; I'll try to be more careful with my edits. I think I'd have tried to find sources for this entry myself, but the latest addition of this link was an anon user undoing User:Gandalf61's changes without any explanation, so I clicked "undo" a bit too hastily simply based on the edit history. — Miym (talk) 09:37, 9 October 2009 (UTC)
I wandered here “randomly”, but just wanted to note that Wikipedia:Red link is in fact not policy, but a guideline. While guidelines should not be simply ignored, they have less support and weaker force than policies. —SlamDiego←T 12:34, 22 November 2009 (UTC)

Defaultsort[edit]

Hi, I noticed you added the Defaultsort parameter to a lot of articles on complexity classes. I think I understand what default sort does, but I'm not sure I understand why you added lowercase versions of the title to the article as its sort parameter. --Robin (talk) 14:24, 9 October 2009 (UTC)

The problem is that case matters when ordering category entries. See WP:SORTKEY: "However the "alphabet" used here is based on the Unicode character listing, and may give unexpected results. For example, all capital letters come before all lower case letters; modified letters come after all unmodified letters; and spaces come before anything else."
I just tried to fix the sorting in Category:Complexity classes for the articles whose name begins with an F or an N. Some of those articles already had DEFAULTSORT parameters, most didn't, and the end result was a bit strange; I think it is more or less "correct" now. (Did I break something?)
You can see similar problems within the articles beginning with "P" in Category:Complexity classes. For example, Parity P is between PSPACE and PSPACE-hard! This is because "PS" < "Pa" < "Ps"; PSPACE-hard seems to have a "DEFAULTSORT:Pspace-Hard" parameter but PSPACE doesn't have. Adding consistent DEFAULTSORT parameters to all articles should fix it. — Miym (talk) 15:43, 9 October 2009 (UTC)
I understand. Makes sense. --Robin (talk) 17:27, 9 October 2009 (UTC)

Dear miym[edit]

HI Miym, sorry for my poor skills in writing something in "Wikipedian Format". Some weeks ago I added a few links to my website, this has been tagged as an attempt to promote my website. Yes it actually was, but all the link were strictly attinent to the article content. I have readen the Wikipedia terms and I thing thing some of those links are really worth to be published.

Now I am asking you to test them, how can I create a discussion made by people who works in the field?

I saw some links which is greatly poorer than the one I suggested but they weren't removed.

For example: on the Convex hull algorithms I added this link:

http://www.advancedmcode.org/fast-convex-hull-algorithm.html

pointing to a video demostration of an efficient convex hull computation. Are you expert of convex hull algorihtm? If not please consider proposing the problem to somebody that is.

Waiting for your reply

My Best regards

Luigi

Latex: Minus vs Negative Signs[edit]

Hello Miym,

In the article on graph coloring, you undid a change in which I changed a minus sign to a negative sign. Your comment says that "Minus signs are taken care of by LaTeX." I disagree. A minus sign and a negative sign are different and AFAIK, LaTeX does not do anything to automatically pick between them. LaTeX always chooses to use a long dash for the '-' symbol. From my experience, the way to tell LaTeX to use a short dash, which is the correct symbol for a negative sign, you have to put the dash in a \text{} command. How do you tell LaTeX to use a negative sign instead of a minus sign?...because you refered the document to a state in which a minus sign is used but a negative sign is desired. Bender2k14 (talk) 00:42, 17 October 2009 (UTC)

Dear Bender2k14, I did not undo your change. Have a look at the history; User:RobinK undid your change.
Anyway, I do have a (strong) opinion about this issue:
  • Real LaTeX (and TeX) systems handle minus and negative signs correctly. You simply type, e.g., $-2$, $k-2$, $n^{-2}$, or $n^{k-2}$ in the source code and the output looks good. These are rendered roughly like −2, k − 2, n−2, and nk−2. All of these use the same minus sign, "−", but the difference is in the spacing: the binary operator is (in some contexts) surrounded by spaces, e.g. "k − 2", while the unary operator is never, e.g. "−2" and "n−2".
  • The support for Latex-like markup in Wikipedia is utterly broken: here <math>n^{-2}</math> produces n^{-2}. First of all, the font size is incorrect. But moreover there are spaces around the "−" sign, and that's why it looks so horribly wrong. I agree with you that this should be fixed.
  • A simple solution is to not use Latex-math in Wikipedia at all. Just write ''n''<sup>−2</sup> and the result looks as good as one can reasonably expect in HTML: "n−2"
  • Using a hyphen (like you did in \text{-}) as a minus sign is always wrong. See, e.g., Minus sign; both the unary operator and the binary operator are long dashes.
Hope this helps. — Miym (talk) 10:37, 17 October 2009 (UTC)
You gave a more comprehensive answer than I would have. Good thing User:Bender2k14 mistakenly posted this on your page instead of mine! --Robin (talk) 06:48, 20 October 2009 (UTC)

Jan Węglarz[edit]

Please see my comments at Wikipedia:Articles for deletion/Jan Węglarz. I think the information that I added shows that Jan Węglarz is notable. -- Eastmain (talk) 23:38, 7 November 2009 (UTC)

Consensus on theorem[edit]

I have put a lot of time and effort into improving the article Theorem. The formulation I came up with addresses all the fundamentals. The total revert is not very cooperative at all. I would like to address your concerns, however wholescale deletion and a note that says "I didn't prefer it" do not help a lot. WHY didn't you prefer it? I am able to respond to that. However I suspect it is a belief that the article is mathematical, and that fundamentals do not matter at all. That's POV. If scholars have discerned the fundamentals, than that's called "subject matter." Theorem is an interdisciplinary topic, and it deserves complete treatment. Please consider, in what way we can preserve the subject matter. Please work your perspective into the formulation I worked so hard on. We aren't taking about a GA article here yet. So I am wondering how we make it one. I think I provided an excellent start. Thank you. Be well, Pontiff Greg Bard (talk) 22:28, 9 November 2009 (UTC)

Hello Gregbard, I think the old lead paragraph ("In mathematics, a theorem is a statement proved on the basis of previously accepted or established statements such as axioms. ...") is much more useful to a layman than your proposal ("A theorem is an idea, concept or abstraction token instances of which are formed using a string of symbols according to both the syntactic rules of a language (also called its grammar) and the transformation rules of a formal system. ...") If a person has no idea what is a theorem, the first version might give at least some idea, while the second version looks unapproachable. Besides, the lead section in your version is very long. — Miym (talk) 22:36, 9 November 2009 (UTC)
Gregbard, did you actually expect your bloviating abomination to replace the standard definition of a theorem. Incredible ... —Preceding unsigned comment added by John.legal (talkcontribs) 05:03, 26 January 2010 (UTC)

Reordering refs in reflist[edit]

Known WP:AWB behaviour request in to have it modified, I will check the status. Rich Farmbrough, 15:10, 15 November 2009 (UTC).

Category: Complexity classes[edit]

Hey, I noticed you added constant time and similar articles to Category: Complexity classes. I thought the category was only for complexity classes. I mean exponential time is an article about exponential time, while EXPTIME is the complexity class. Also, do you think we're duplicating too much by having separate articles on exponential time and EXPTIME, etc.? --Robin (talk) 23:39, 1 December 2009 (UTC)

Thanks for the comments! I agree that we should merge articles such as Exponential time and EXPTIME. (By the way, what Exponential time describes seems to be more like E (complexity) than EXPTIME; perhaps we could merge all three to clarify the confusion?)
I also agree that I'm slightly abusing Category:Complexity classes. In fact, the reason for doing that was exactly what you said: we have overlapping articles such as Exponential time and EXPTIME. I thought that putting all these articles into one category would help us to identify such candidates for mergers.
Anyway, feel free to re-categorise if you can find a better category. Perhaps we could use Category:Computational resources instead of Category:Complexity classes for articles like Constant time? — Miym (talk) 00:10, 2 December 2009 (UTC)
Yeah, I'm not sure what would be the right thing to do. On the one hand, sub-exponential time is turning out to be a pretty informative article, and I've always heard people calling them sub-exponential time algorithms or quasi-polynomial time algorithms, as opposed to saying "in the class SUBEXP or QP." Moreover, some concepts like polynomial time are reused, not just for defining the class P, but also for defining FP, and to define efficient algorithms from an analysis of algorithms perspective. On the other hand, the article on exponential time is terrible. I agree that "Category:Computational resources" might be more appropriate than Category:Complexity classes. Unfortunately WP:CS is an almost dead project, so we can't expect much input from there. --Robin (talk) 00:23, 2 December 2009 (UTC)
How about a new category for things like "polynomial time" and "constant time"? Something which describes what they are, like maybe Category:Computational resource bounds. --Robin (talk) 04:02, 2 December 2009 (UTC)
A different idea: How about combining all the "x time" articles into one article on time complexity? I'm sure a lot of people type "time complexity" into wikipedia and are directed to the analysis of algorithms or complexity theory article, both of which aren't that helpful. --Robin (talk) 04:52, 3 December 2009 (UTC)
I'm becoming more and more convinced that this might be a good idea. (The total length of all "x time" articles is fairly large, but there is a lot of overlap in them.) — Miym (talk) 22:07, 3 December 2009 (UTC)
WT:COMPSCI has been non-dead recently, maybe you could write a proposal there? Here is a (partial?) list of terms that could be changed into redirects that point to "Time complexity": Constant time, Logarithmic time, Linear time, Linearithmic time, Quadratic time, Cubic time, Polynomial time, Quasi-polynomial time, Sub-exponential time, Exponential time, and Double exponential time. I think the target article could be, in essence, a fleshed-out version of the table in Big O notation#Orders of common functions. — Miym (talk) 17:59, 16 December 2009 (UTC)

Steiner trees[edit]

Hi Miym,

Thank you for adding details on the Steiner trees and the contribution. However, some of your edits are not clear: 1. "Steiner tree in graphs" is not something defined in Approximation algorithms by Vijay V. Vazirani (cited as the source of that paragraph), but rather the "metric Steiner tree problem". I'm not sure what you mean by this term. In any case a complete graph that meets the triangle inequality should be equivalent to the metric Steiner tree problem. Not sure why you decided to edit this out as I believe it clarifies things to the reader.

1a. In particular I don't understand the sentence "we do not require that the edge weights correspond to distances in a metric space". It doesn't seem related to anything written in those pages.

2. I added information on a simple 2 approximation algorithm (also from Vazirani). The fact that there are better algorithms should not mean that this information should be deleted. Some of these algorithms are by far more complex than this very simple algorithm. I see no reason to edit this out. It may also be useful to show this info before going into PTAS which is a more general term.

3. Because it is possible to show a polynomial time transformation between the Euclidean problem to the metric problem and back, that information should be there. I see no reason to "censor" this information as it helps with approximation of this problem.

4. I would appreciate a discussion on these matters rather than editing out text that is based on acceptable resources. —Preceding unsigned comment added by Gshaham (talkcontribs) 13:56, 6 December 2009 (UTC)

Hello Gshaham, thanks for comments. A quick question about "a polynomial time transformation between the Euclidean problem to the metric problem and back". Metric Steiner trees are APX-complete while Euclidean Steiner trees admit a PTAS. Are you really claiming that you can take an APX-complete problem and transform it into an equivalent problem that admits a PTAS? In effect, showing that P = NP? — Miym (talk) 14:02, 6 December 2009 (UTC)
The fact that one is APX-complete and the other is PTAS is not a contradiction. I'm talking about a constant c optimization, not PTAS. Since it's possible to prove that after the polynomial transformation the optimal solution does not exceed the original one then we can show carry over results from a c constant optimization from one problem to the other. — Gshaham (talk) 00:09, 8 December 2009 (UTC)
Ok, let's focus on constant-factor approximation algorithms then if it's easier. There is a polynomial-time 1.01-approximation algorithm for Euclidean Steiner trees. And you claim that there is a polynomial-time approximation-factor preserving transformation between Euclidean and metric Steiner trees. Doesn't this imply that there would be a polynomial-time 1.01-approximation algorithm for metric Steiner trees as well? — Miym (talk) 00:22, 8 December 2009 (UTC)
That's my understanding from Theorem 3.2 in Vazirani. Basically you construct the metric space Steiner tree by computing the shortest path for each edge. This guarantees that the optimal solution in the metric problem would be at most the optimal solution in the Euclidean problem. Thus you preserve the approximation factor. - Gshaham (talk) 12:25, 8 December 2009 (UTC)
I think you have misunderstood the definition of the Euclidean Steiner tree problem; it is not a graph problem but a geometric problem. Theorem 3.2 in Vazirani's book does not provide a transformation between the Euclidean Steiner tree problem (geometric problem) and the metric Steiner tree problem (graph problem). It provides a transformation between two graph problems: (general non-metric) Steiner tree problem and metric Steiner tree problem (both of these are described in Steiner tree problem#Metric Steiner tree). The Euclidean Steiner tree problem is a very different thing; it is strictly easier to approximate than the graph problems (metric or non-metric Steiner trees) and there is no approximation-factor preserving transformation between the geometric problem and the graph problems. — Miym (talk) 13:17, 8 December 2009 (UTC)
I see, I thought we were referring to the same problem in Vazirani as Euclidean Steiner tree problem. Probably the original article prior to your cleanup caused this confusion. Now that we are on the same page I would suggest that perhaps the article first describes the weighted Steiner tree problem (same as http://www.nada.kth.se/~viggo/wwwcompendium/node78.html) which is APX-complete and then describe the metric problem (which is equivalent) and Euclidean problem (which is PTAS). The unweighted version described in the beginning is simply a restriction on the weighted version and I hardly see any references to that problem description (where all the weights are 1). It would be better if Wikipedia defines the problem similar to other books and articles. We could reference the case where all the weights are 1 and 2 leading to a constant approximation factor of 1.28. - Gshaham (talk) 13:43, 8 December 2009 (UTC)
About "Steiner tree in graphs": I did not add this term in the article, it was there already, I just tried to clarify it. Steiner tree in graphs = Problem 3.1 ("Steiner tree") in Vazirani's book. Theorem 3.2 in the book shows that we can focus on metric Steiner trees instead of general (non-metric) Steiner trees. I hope this answers your concerns 1 and 1a. I can try to clarify the wording. — Miym (talk) 14:08, 6 December 2009 (UTC)
I changed the Steiner tree problem#Metric Steiner tree section a bit. Hope it addresses some of your concerns; among others, it now mentions the triangle inequality and avoids the problematic sentence that was difficult to understand. Feel free to edit the text; I agree that the article needs a lot of work! I'm not trying to censor anything, I just wanted to quickly fix some factual errors and add enough content (specifically, PTAS vs. APX) that should avoid the same confusion in future. — Miym (talk) 14:29, 6 December 2009 (UTC)

International Conference on Logic Programming[edit]

Why did you remove Category:International conferences from this article? Clearly it is an international conference. Vegaswikian (talk) 19:04, 12 December 2009 (UTC)

Sure, but we already have the category "Computer science conferences", which is a subcat of "Academic conferences", which is a subcat of "International conferences", so it is redundant. See also Wikipedia talk:WikiProject Academic Journals#Conference categories for related discussion. — Miym (talk) 19:10, 12 December 2009 (UTC)

Passive and Active Measurement Conference[edit]

It is a quite notable conference. Next time please check appropriate sources, before you arbitrarily remove entries. Being in the ERA list as B does not mean the conference is not good.

http://academic.research.microsoft.com/Conference/1443.aspx?query=passive

I would also replace the libra.msra.cn links with the new academic.research.microsoft.com site. —Preceding unsigned comment added by John.legal (talkcontribs) 03:58, 25 January 2010 (UTC)

Sure, I believe that PAM is notable, but we need reliable sources that show that this is the case. ERA rank B is not a reason to remove, but I don't think it's a reason to add, either. There are almost 1000 conferences in the ERA listing with rank A or B, including dozens of conferences related to networking – we can't list everything, and hence we need to be a bit more selective. As a rule of thumb, the conferences listed in List of computer science conferences should be among top-10 conferences in respective fields; ERA rank A indicates that this might be the case, while ERA rank B doesn't tell us much about it. I removed PAM without prejudice; feel free to put it back when you find a reliable source that indicates that show that PAM is clearly notable. Having multiple independent sources is naturally the best option. (Whatever you do, please do not add incorrect references, like you have done with PAM: it obviously isn't "ERA: rank A"!) — Miym (talk) 08:34, 25 January 2010 (UTC)
My bad to cite era-a, you are right. Like the CORE ranking, the ERA rankings are pretty much BS. They list not so good conferences as A and good confs as B. Nevertheless just the fact that a conf is include in the A or B lists means that it passes a standard of quality. The most reliable way we have to determine whether a conference should rank high is MS academic search. You mention below that PAM does not have many citations. You should not look at the number of citations but the citation ratio (citations per paper published). We list ICC which has almost 1:1 citation ratio and we would not list PAM that has almost 6:1?
Regarding libra.msra.cn links, I couldn't find the word "libra" in the sourcecode of the page. Could you perhaps be a bit more specific? — Miym (talk) 08:42, 25 January 2010 (UTC)
It seems you have already replaced libra with the new MS academic search links, great.
And regarding your link to Microsoft Academic Search: I did check their ranking in http://academic.research.microsoft.com/CSDirectory/conf_category_14.htm but PAM seems to be the 42th conference in the list, and the number of citations is much lower than in most conferences that we have listed in the field. Again, I'm not saying that this makes PAM non-notable; I'm only saying that this is not a particularly convincing positive indicator of notability. Some other sources are needed, I think. — Miym (talk) 08:48, 25 January 2010 (UTC)
the problem with MSAS rankings is that they are based on an eigenvector-based metric. They pretty much compute the pagerank of a conference, but instead of web links they use the citation/authorship/published-in etc relations. This is a very good metric but it is unfair for new venues that do not have many incoming citations. The best way to resolve this problem is to actually look at the citation ratio of the conference. I believe the position PAM is currently listed in is fair and I have substantiated it sufficiently. I am not alone in this belief: http://mybiasedcoin.blogspot.com/2009/10/ranking-networking-conferences.html , where you can find posts by well-known networking profs. To this end, I now cite its citation ratio with the indicator to search for PAM in MSAS.

Categories for discussion nomination of Category:ACM publications[edit]

Info non-talk.png

Category:ACM publications, which you created, has been nominated for deletion, merging, or renaming. If you would like to participate in the discussion, you are invited to add your comments at the category's entry on the Categories for discussion page. Thank you. Mike Selinker (talk) 02:03, 23 January 2011 (UTC)