User talk:Glrx

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


Welcome!

Hello, Glrx, 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! RayTalk 19:29, 2 August 2009 (UTC)

FPGA[edit]

g'day, how are you today? I noticed that you deleted my edits saying "power and speed comparisons not complicated by LUT" and thought I reach out. please, find below some links that help you understand better:

https://www.youtube.com/watch?v=_oHNvJTWX3Q

http://www.eecg.toronto.edu/~jayar/pubs/kuon/kuontcad06.pdf

http://books.google.com/books?id=0S1Y8modSrMC&pg=PA666&lpg=PA666&dq=different+LUTs+for+power&source=bl&ots=w__AjMePtC&sig=CrMAPWu9jXNv41erBg3XKbe9bIg&hl=en&sa=X&ei=it2hU5elKMqcqAaP6IAo&ved=0CFIQ6AEwBQ#v=onepage&q=different%20LUTs%20for%20power&f=false

so, I hope that helped and it seems best if you make your deletion of my edits undone yourself. best beaki — Preceding unsigned comment added by Beaki (talkcontribs) 04:10, 19 June 2014 (UTC)

@Beaki:
This discussion is about the insertion of this material. The material describes 3 manufacturers' current approaches to FPGA architecture; it includes a URL link in the body to chippath.com, and it adds chippath as an EL. I removed the material again. The material is unsourced; it does not even cite to the sources listed above.
We don't know what type of comparisons are being made in the paragraph. The previous (sourced) paragraph was about speed, power, and functionality (area). Such comparisons are easy to do in a specific context. Which FPGA is suitable? How fast does it go? How much power does it draw? Those are not difficult questions. For a particular problem (and particular design tools), different FPGA logic blocks may be better, but custom designs are not constrained by fixed architectures..
The primary links inserted into the article were for a chippath.com comparison tool. WP is not a buyers guide. WP need not advertise chippath's tools.
WP does not consider youtube a reasonable source.
The first paper is a primary source. It points to significant (x35 down to x18 predict x5) area penalties for FPGA vs ASIC, x3 to x4 time delays, and x14 with power. The material I removed does not make these claims and does not cite to this source. The paper also goes for a hybrid approach: hard blocks can be useful for gaining performance -- except when the task doesn't need the hard block and then it just wastes area.
The book appears to be a primary source because it reports the author's findings rather than surveying the field.
The sources above do not seem to be saying the comparison is hard. They say things can be better.
The main statement of the material is not a clear one. It's not clear which LUT architecture is better. But WP does not need to say that one company uses this LUT architecture, another a different one, and it's not clear which is better.
If you want this material in the article, then you should follow WP:BRD: bring it up on the talk page and get a consensus for inclusion.
Glrx (talk) 03:48, 30 June 2014 (UTC)

Thanks for help with Lead zirconate titanate![edit]

I appreciate your help with Template:cite on this page! Kragen Javier Sitaker (talk) 00:56, 21 June 2014 (UTC)

Kelvin bridge[edit]

Why?Freshman404Talk 20:22, 23 June 2014 (UTC)

There was no citation to that book in the body of the article. The reference section is not intended to be a bibliography of every book that mentions the topic. Feel free to add some text to the article and use the book as source (and provide page numbers). Glrx (talk) 01:35, 25 June 2014 (UTC)
I add no more text because same explanation is written in that book-In my book it is in chapter 5, page 113. Please let it remain.(And use PING to notify me.)
Thanks--Freshman404Talk 06:31, 25 June 2014 (UTC)

July 2014[edit]

Hello, I'm BracketBot. I have automatically detected that your edit to Numerical Electromagnetics Code may have broken the syntax by modifying 1 "[]"s. If you have, don't worry: just edit the page again to fix it. If I misunderstood what happened, or if you have any questions, you can leave a message on my operator's talk page.

List of unpaired brackets remaining on the page:
  • diameters better than NEC2 and probably NEC4;<ref>http://www.cebik.com/content/a10/model/fd.html]</ref> this includes different diameter parallel wires, different diameter wires joined at an angle,

It's OK to remove this message. Also, to stop receiving these messages, follow these opt-out instructions. Thanks, BracketBot (talk) 18:39, 2 July 2014 (UTC)

Constant folding and string literal concatenation[edit]

Hi Glrx, in this edit, I’ve taken another shot at discussion string literal concatenation in the context of constant folding. I’ve kept the main text brief, focusing on the similarities (it’s solving the same problem), and relegating the C-specific details to footnotes, so it’s less distracting. How does it look?

—Nils von Barth (nbarth) (talk) 17:38, 6 July 2014 (UTC)
Nils, the addition is off topic; it is not about constant folding; it is about the syntax of constants. Even you made the comment "implicitly" concatenated. (The article you link also has Python considering deprecating the operation.)
The idea of constant folding is the language allows expressions that an operation (e.g., a + b, 2 * c, 3 + 4, 2 * 7) be done at runtime, the compiler recognizes the operation can be done at compile time, and the compiler does it. In C, there is no explicit string concatenation operator for runtime. Some languages have an explicit string concatenation operator that expects to be done at runtime. The user might write "Hello "+strWorld or "Hello " || strWorld, and the compiler will emit code to allocate the string and do the concatenation. If the compiler knows that strWorld is a constant, then the compiler can do the operation at compile time and just emit the constant result.
A C/C++ example is not doing that because no runtime concatenation operator involved. Neighboring quoted strings are just the syntax of specifying long strings (or building up a string).
Glrx (talk) 20:04, 6 July 2014 (UTC)
I understand that there are differences between constant folding and string literal concatenation, as you note – they are distinct concepts and accordingly have distinct articles.
However, the similarities are also striking – in both cases, during compilation, string literal tokens (generated separately and evaluated during tokenization/evaluation) are then combined at a later step of compilation, and these two distinct mechanisms satisfy the exact same purpose: allowing a complex value (here a long string) to be built up in a complex way (perhaps using functions or, in the case of C/C++, preprocessor macros) but still evaluated at compile time.
With CF, you might write:
foo(x) { return "foo:" + x }
foo("bar")
…while in C you'd write:
#define FOO(x) "foo:" x
FOO("bar")
…but these both fulfill the same goal in very similar ways.
Thus it would be useful to give readers some way to learn about SLC when reading about CF. It would be possible to just put a link to SLC in the “See also” section, if that’s necessary to avoid distraction, but it’s generally preferred to try to incorporate these into the text, hence my proposed edits.
How would you suggest that readers learning about CF should learn about SLC?
—Nils von Barth (nbarth) (talk) 04:08, 7 July 2014 (UTC)
They are separate ideas, and there's a lot of ugliness in the examples above. If an operation is never intended to be executed at runtime, then it has no business in an article about compiler optimizations. If it could never run, then there is nothing to optimize.
The C preprocessor is an expedient hack that is far away from any compiler optimizations. In the old days, it was a separate pass, so the C parser would never see the defines or their invocations -- just the results. As for instances of SLC, I'd expect them to be handled during parser reductions. A remark about C in the CF article seems headed for confusion.
I have seen programs where programmers use string concatenation operations on literal strings as a convenient method of building up a longer string. The use was esoteric. One would like to see the compiler optimize the string, but there's no great penalty if it doesn't.
I want things to be clear.
In a somewhat related matter, code optimization of string concatenation will not be a big benefit in most situations. As I understand it, the big benefits of code optimization were with array index calculations: lots of mundane index arithmetic could be saved. The same opportunites for optimizing string concatenation probably are not there.
Glrx (talk) 17:10, 7 July 2014 (UTC)
Understood – the C preprocessor is a part of compilation (broadly speaking) that’s not relevant these days outside of C/C++, and is generally confusing and baroque, though these later are still in widespread use (avoiding the preprocessor as much as possible). (Historically it’s part of the transition between macro assemblers and proper compilers.) To avoid disrupting the flow of the article, shall I put a link to SLC in the “See also” section, perhaps with an (HTML) comment to the effect of “please don’t conflate this with CF”?
—Nils von Barth (nbarth) (talk) 02:08, 8 July 2014 (UTC)
BTW, for reference: discussions of removing SLC from Python and D specifically mention replacing it with constant folding, as in:
—Nils von Barth (nbarth) (talk) 04:45, 10 July 2014 (UTC)
In this proposed edit (diff), which I’ve reverted (it’s a proposal), I’ve given a very brief mention of SLC:
“A superficially similar feature is string literal concatenation, which concatenates adjacent string literals during lexical analysis, for example replacing "abc" "def" with "abcdef".”
This shows the similarity, but clearly flags it as distinct and how, and avoids any technical discussion of C.
How does it look?
—Nils von Barth (nbarth) (talk) 04:45, 10 July 2014 (UTC)
I'm opposed to the edit. Your sources do not say the operations are superficially similar. I don't consider them reliable sources about CF. What they say is that SLC may make bugs difficult to detect. I'm not sure I buy into that premise: spotting - instead of + is also difficult. One source also did not want to make CF a required optimization.
The sources don't say what you want. They say SLC is dangergous and could be replaced with explicit concatenation. SLC does not belong in the CF article because it is language syntax rather than a compiler optimization.
I think you like the parallel, but don't talk about it in CF article.
I'm copying this discussion to CF; I should have moved it there earlier.
Glrx (talk) 15:29, 10 July 2014 (UTC)

Copied to: Talk:Constant folding#Constant folding and string literal concatenation
—Nils von Barth (nbarth) (talk) 04:32, 11 July 2014 (UTC)

Your revert in Euclidean algorithm[edit]

Hi I have reverted your revert for the following reason: Presently, MathJax rendering is buggy for formulas containing option ALT (<math alt=sometext> ...<math>. It displays "sometext" before the formula, as it would be a part of the formula. As "sometext" is not really useful, and most readers prefer to read the latex source than confusing descriptions of formulas, I have pushed "sometext" into a comment. This allows the MathJax users to read the formulas, and the rare readers interested in such comments to accede to them. For users of PNG rendering, both versions are equivalent. Thus, I do not see any reason for your revert. D.Lazard (talk) 13:29, 9 July 2014 (UTC)

Your edit is a bad idea. Presumably, MathJax was working last week. Somebody broke it; it no longer follows the spec. Instead of fixing MathJax, your solution is to insert a workaround on every article that uses an alt attribute in a math tag. That will tickle a lot of watch lists. We had a system manager who advocated that approach; he broke something and started to "fix" it by modifying every user's init file; it was the last straw; I fired him.
Furthermore, the math works fine for me because I'm not a MathJax user. My impression is MathJax is the minority. Glrx (talk) 15:21, 10 July 2014 (UTC)

Your revert on Captain Richard Winters[edit]

Binkleyz (talk) 15:10, 10 July 2014 (UTC) Curious why you undid my captialization of "Captain" (Richard Winters). Captain is a title and therefore a proper noun, and should be capitalized.

This is about edits on 28 and 29 May at Richard Winters. Your edit went through and capitalized all the ranks and positions. My revert restored the original lowercase. For the reason, see MOS:JOBTITLES. My immediately following editsmoved the picture you added out of the paragraph text and reinserted a clarification that you made. Glrx (talk) 15:43, 10 July 2014 (UTC)

A year and a half after you opposed my RfA[edit]

I am inviting you to leave me some feedback, 18 months after you opposed my RfA. Do you still believe I am not fit to be an admin? Do you believe I have been able to improve past the concerns you have brought up? Do not be afraid of being too harsh, I am specifically welcoming criticism as I believe it is the best way to improve and I am always looking to learn from my mistakes. I am particularly looking for feedback as to whether you have objections to myself lifting the self-imposed 1RR restriction I had agreed to towards the end of my RfA. If you don't have time to comment, don't fret it either, this is nothing I'll lose sleep over. :) ☺ · Salvidrim! ·  19:48, 20 July 2014 (UTC)

RF connector[edit]

re this revert at RF connector

The edit you undid was not mine, it was User:Rfconnector's. Cluebot's choice was clearly wrong. I don't have an opinion on the content of the edit. jhawkinson (talk) 01:16, 14 August 2014 (UTC)

I'm mystified too: I don't know why or how Cluebot thought it was vandalism. Glrx (talk) 01:45, 14 August 2014 (UTC)

Lisp[edit]

I notice you reverted my edit. And you are not alone with that practice on the Lisp page. Please have a look at this. And I have changed the article to what the source says. -- Zz (talk) 20:15, 21 August 2014 (UTC)

your revert on Proximity Fuzes[edit]

I placed that video reference in the article along with a copy of the cover for the tape box because it seemed to me relevant that, as I described my entry, this was "in popular culture". We've (that is, Wikipedia) certainly got many, make that MANY, articles where such entries are common. Please reconsider your objection. wiki-ny-2007 (talk) 05:09, 1 September 2014 (UTC)

About my revert of wiki-ny-2007's edits to Proximity fuze and, indirectly, File:Fuze-102.png.
Including the image of the video's cover art appears to be a copyright violation. Copyrighted cover art can be used in limited circumstances, but this usage does not appear to be one of them. We don't need to see the cover to identify the work. Neither is the article about the video.
The information about where to purchase video is not relevant to the article. "Copies are available through the PBS clearinghouse" sounds in advertising; such a statement does not belong in an article's text.
You added a popular culture section with the video as the only item. Popular culture sections are not for documentaries about the topic. History is not popular culture. Popular culture sections are used when the subject of the article is used as a story item or plot device in a popular work such as a book, movie, or song. Compare Ark of the Covenant#In popular culture.
Documentaries are used to supply content for articles; in that case they are legitimate references. The video is not used as a reference for any content in the article.
Documentaries can also be included in further reading (watching?) sections if they supply content that is not currently in the article. Does the video offer content that is not available in the existing sources? Does it have interviews with principals?
If you think the video should go into the article, then bring it up on the article's talk page. See WP:BRD.
Glrx (talk) 15:56, 1 September 2014 (UTC)

Statement by Glrx[edit]

At AE. You have lost me over there, the "The PHR report also showed that ninety four per cent of internally displaced persons (IDP's) had been victims of some form of sexual assault". is sourced to Women, Migration, and Conflict: Breaking a Deadly Cycle p50, "94% of displaced households", how are displaced households not displaced persons? The other point you mention is also confusing me, "215,000 to 257,000 victims of sexual abuse", is cited to Women Under Siege, Physicians for Human Rights estimates that during the conflict, between 215,000 and 257,000 of them were subjected to sexualized violence, how is it not accurate to what I have written? Darkness Shines (talk) 07:22, 4 September 2014 (UTC)

RE: Wikipedia:Arbitration/Requests/Enforcement#Arbitration enforcement action appeal by Darkness Shines
We have tried to explain these distinctions to you. At AE, it was pointed out that a household is not a person but rather several persons. A displaced household is several displaced persons. If someone in your family was a victim of sexual assault, that does not mean that every member of your family was assaulted. The sexual assault figure is also absurd on its face. The AE discusion pointed out that the 94 percent figure was about violence. Murders are violence, but they are not sexual violence. The discussion at AE also pointed to specific pages in PHR report that gave vastly different figures. A page had a breakdown of the types of violence. We gave you a pinpoint citation in the PHR Report that had one-tenth of your figure for sexual violence. The figure of 94 percent of internally displaced persons were victims of sexual violence doesn't even sound right even when acknowleding 45% of the IDP population is male.
Nobody is disputing the 215,000 to 257,000 victims of sexual violence. That was also spelled out at AE. The problem was with your citation: it went to a page in the PHR Report that did not support the claim; other pages did. The figure also does not support your 94 percent claim. Do you know how many IDPs there are? For the 94 percent figure to make sense, the IDP would have to be less than 300,000. The PHR Report says the IDP population is 1.0 to 1.3 million; 257,000/1,000,000 is not 94 percent. (PHR Report page 4 fn5.)
I get the sense that you quickly jump to the conclusion that you are right and others must be wrong. Do you believe that Callanecc, Sandstein, and I are all wrong? You wanted me to apologize for my comments. You labeled Sandstein's comments a personal attack (even if Sandstein were wrong, a factual error is not a personal attack). If someone had challenged your edits at the Rape in Sierra Leone Civil War article, would you have reverted to keep your interpretation in or labeled those edits personal attacks?
Glrx (talk) 16:44, 4 September 2014 (UTC)
No, I would not have reverted, I would have asked for an explanation, which I have now gotten. Thanks. Darkness Shines (talk) 06:58, 5 September 2014 (UTC)

Your revert on Van Gogh[edit]

I made a post on the discussion page regarding your revert of my edit. Uchiha Itachi 25 (talk) 14:09, 6 September 2014 (UTC) https://en.wikipedia.org/wiki/Talk:Vincent_van_Gogh#Starry_Night_is_his_most_well-known_work

I had replied at Vincent van Gogh#Starry Night is his most well-known work.
Uchiha Itachi deleted this talk page post saying he had fixed his error.[1]
However, it just changed the source of the best-known claim. I agree it is one of his best known paintings, but I don't think it should be labeled his best known. Glrx (talk) 15:08, 23 September 2014 (UTC)

Planimeter[edit]

What's the problem with the reference to the Lego planimeter. I found it really interesting. Nijdam (talk) 22:00, 9 September 2014 (UTC)

re Planimeter and my revert of external link to self-published planimeter made from Lego
@Nijdam: It's a cute hack, but the source is a blog. What content does the EL have that is not already in the article? Advice about using friction gears rather than crummy Lego gears with lots of backlash isn't germane to the article. The Pritz Lego planimeter is also cute, but accuracy claims for both (10 cm2 and 2 cm2) are not great. It's cute, but I don't think it belongs in the article. As an extreme argument, should every WP article about an object have a link to a Lego implementation of that object?
Why do you think it should be in the article? Glrx (talk) 15:57, 11 September 2014 (UTC)
Well, it amused me, and I see no harm in showing. The argument about "every article" is not very adequate. A colleague of mine argued: it should not be allowed to take the train to London, then suppose everyone would do so. Nijdam (talk) 20:42, 11 September 2014 (UTC)

September 2014[edit]

Hello, I'm BracketBot. I have automatically detected that your edit to Grill (cryptology) may have broken the syntax by modifying 1 "<>"s. If you have, don't worry: just edit the page again to fix it. If I misunderstood what happened, or if you have any questions, you can leave a message on my operator's talk page.

It's OK to remove this message. Also, to stop receiving these messages, follow these opt-out instructions. Thanks, BracketBot (talk) 02:21, 22 September 2014 (UTC)

Reference Errors on 1 October[edit]

Hello, I'm ReferenceBot. I have automatically detected that an edit performed by you may have introduced errors in referencing. It is as follows:

Please check this page and fix the errors highlighted. If you think this is a false positive, you can report it to my operator. Thanks, ReferenceBot (talk) 00:22, 2 October 2014 (UTC)