Computing desk
< July 8	<< Jun \| July \| Aug >>	July 10 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

July 9

Rendering of mathematical formulae on web pages

At "proofwiki.org", the source of the pages contains stuff like "$\triangle ABC$", "$d^2 = \dfrac {b c} {\left({b + c}\right)^2} \left({\left({b + c}\right)^2 - a^2}\right)$" and so forth, and also stuff like "$\displaystyle \frac {BD} {DC}$", "$\displaystyle \implies$" etc. etc. All of this is rendered into very pretty formulae on the page. How does this rendering get done? Also, since the formulae look miles better than anything on Wikipedia, could Wikipedia use the same technology? 109.153.232.72 (talk) 00:05, 9 July 2015 (UTC)[reply]

They're using MathJax. If you have an account, you can enable MathJax rendering in your user preferences, which means MediaWiki will use MathJax to render formulae for you as long as you're logged in. This is not turned on for Wikipedia and other Wikimedia projects by default for reasons related to performance and browser support. There is continuing work being done on MediaWiki's math support, which should eventually allow state-of-the-art math handling to be the default on all wikis using MediaWiki. --108.38.204.15 (talk) 02:01, 9 July 2015 (UTC)[reply]

I see, thanks for the very clear and helpful reply. 109.153.232.72 (talk) 02:44, 9 July 2015 (UTC)[reply]

remote db

Hi, is it better to replicate data needed from a remote db ot to join remote tables? Thank u 188.29.164.64 (talk) 18:46, 8 July 2015 (UTC) — Preceding unsigned comment added by 188.29.165.190 (talk) [reply]

From the point of view of distributed databases, you separate your query into independent queries that pull only what is needed from the remote database. Consider this: select loc.name, loc.age, rem.cost from local.table loc join remote.table rem on loc.name=rem.name and loc.age<10; I don't replicate remote.table. I first query local.table for all names where age<10. Then, I send that to the remote and ask for "name, cost" for all records where name is in the list I've sent. When I get back the "name, cost" set, I use that for my query to produce the final result. The reasoning is rather simple. Network traffic is very slow, so you send as little over the network as possible. If that is what you mean by "replicate data needed," then you are correct. You only replicate the data needed. If by "replicate data needed" you mean "copy all the tables from the remote server to the local server," then you are incorrect. That moves data that will not be used across the network. 209.149.114.69 (talk) 11:52, 9 July 2015 (UTC)[reply]

thank you a lot, but is it better to load needed fields while reading or to transfer them just after their update/ins/del? — Preceding unsigned comment added by 188.29.165.190 (talk) 06:23, 10 July 2015 (UTC)[reply]

The time of transfer doesn't make much difference. It is about the quantity of data transferred. The big trick in distributed databases is to reduce overall network traffic. It isn't about timing (until you start talking about building an ACID compliant distributed database). 199.15.144.250 (talk) 17:01, 11 July 2015 (UTC)[reply]

Program reinstalls

Hi! My computer had a program called OffersWizard, so i uninstalled it. But after a couple of days, it reinstalled itself back. I uninstalled it again, and it reinstalled itself again. How can i prevent the program to reinstall itself? — Preceding unsigned comment added by 217.209.38.254 (talk) 13:19, 9 July 2015 (UTC)[reply]

A program cannot install itself. A program that isn't there can't do anything at all. Something else is reinstalling it -- you have to figure out what. Very likely it is an extension in your internet browser. You can find complete instructions for removing OffersWizard at http://malwaretips.com/blogs/ads-by-offerswizard-removal/, also a few other places. Looie496 (talk) 13:42, 9 July 2015 (UTC)[reply]

Some Malware uses AddOns of the Browser and a program. You need to remove both of it. See also MSConfig --Hans Haase (有问题吗) 15:13, 9 July 2015 (UTC)[reply]

A program certainly can install it self, I'll give you a few websites to visit and then we can have the argument again. You could argue that it wasn't "THE PROGRAM" that installed it self per se, it was the website that installed the program, or it was YOU who installed the program by visiting the website, but that's a purely semantic argument. Also, when you "uninstall" a program, for malicious software it's fairly common for a piece of the software to hide it self from the uninstall program so that you actually aren't completely uninstalling it, but leaving behind a piece that will then "reinstall it self". In both cases I think it's perfectly acceptable to say "the program installed it self". Vespine (talk) 23:40, 9 July 2015 (UTC)[reply]

Visual Studio 2010 C++ LNK1123 Error?

Hello everyone. I am trying to compile a C++ project in Visual Studio 2010. However, whenever I try to compile it, it errors out giving me an error message saying, "error LNK1123: failure during conversion to COFF: file invalid or corrupt." I have read that this is caused by a version inconsistency in the compiler and that this can be fixed by updating VS with the SP1 update. Does anyone know of any other way to fix this? I am using Visual Studio 2010 Pro running on Windows 7 Ultimate X64. An a point of interest, this error manifest it self for every C++ project I try to compile. Thanks for your help in advance. —SGA314 I am not available on weekends (talk) 14:10, 9 July 2015 (UTC)[reply]

Does anyone know the answer to this? —SGA314 I am not available on weekends (talk) 18:26, 9 July 2015 (UTC)[reply]

See this StackOverflow thread and this MSDN article. Tevildo (talk) 19:28, 9 July 2015 (UTC)[reply]

Wo. No need to be like that. I wasn't claiming anything. I really can't access MSDN. I thank you for your help very much and meant nothing by what I said. I just really can't access MSDN or Microsoft's site. —SGA314 I am not available on weekends (talk) 17:37, 10 July 2015 (UTC)[reply]

OK, if you genuinely can't get to MSDN, we need to fix that problem first. You may be the victim of browser hijacking or have some malware on your system. What browser are you using? What happens if you type http://go.microsoft.com into your browser's address field? DO NOT type it into the search box, or anything similar that's in the vicinity - and be sure to include the "http://". Do you get an error page? Does the URL change, and, if so, what does it change to? It might be an idea to try a different browser, and see if the results are any different. Tevildo (talk) 17:47, 10 July 2015 (UTC)[reply]

I got the impression SGA is working with some sort of highly restricted internet services, restricted by an employer, not malware etc. I could be totally wrong, if so sorry to SGA. (There have been a few other ref desk users over the years that say they have highly restricted internet abilities - I usually take them at their word, but it can be problematic - if they can't see refs we post, then we have to paste content here to make it useful, and that process can often be very time consuming.) SemanticMantis (talk) 17:56, 10 July 2015 (UTC)[reply]

Yes, I can understand not being able to access YouTube or Facebook or the BBC - I myself can't access Wikipedia from my work machine - but I find it hard to believe that any rational boss or ISP would block access to MSDN and allow access to Wikipedia. This leads me to suspect that SGA has a non-human-related problem with his (or her) setup. Tevildo (talk) 19:03, 10 July 2015 (UTC)[reply]

No, SemanticMantis was right. It is my IS. Not any malware. I just wanted to know if anyone has had this problem before and they found a way to fix it(other than updating Visual Studio). —SGA314 I am not available on weekends (talk) 19:41, 10 July 2015 (UTC)[reply]

Well, what you need to do is disable incremental linking - Project Properties > Configuration Properties > Linker (General) > Enable Incremental Linking -> "No (/INCREMENTAL:NO)". But I still think it's worth trying to get you access to MSDN, where the answers to your questions can be found. What _does_ happen when you try and get to microsoft.com? Tevildo (talk) 20:19, 10 July 2015 (UTC)[reply]

Incremental linking is already disabled. There is a web filtering system that's setup on my ISP. Its called Sophos. And there is no way around it. I have read the log file and heres what it says,

Invoking CVTRES.EXE:

   /machine:x86
   /verbose
   /out:"C:\Users\dainen\AppData\Local\Temp\lnkBA6F.tmp"
   /readonly
   "Debug\1.res"

LINK : fatal error LNK1123: failure during conversion to COFF: file invalid or corrupt

This happens after the it loads the referenced libraries. I think I might be able to fix this by changing the targeted .NET framework version. How would I do this? —SGA314 I am not available on weekends (talk) 14:40, 13 July 2015 (UTC)[reply]

Why was 0-1 not used for 'S' in Hollerith code?

'A' = 12-1, 'B' = 12-2, ... 'I' = 12-9, then 'J' = 11-1, 'K' = 11-2, ... 'R' = 11-9, so there's a pattern of when the Numeric punch gets to 9, then you move to the next Zone punch (note for alphabetic encoding, 0 is considered a zone punch) you start at 1 again. But not so when you come to 'S', which suddenly is 0-2, not 0-1. Why? Was it so that 'Z' would neatly land on 0-9? 20.137.7.64 (talk) 17:36, 9 July 2015 (UTC)[reply]

It began with 32 characters, ending with "...LMNOPQR/" (the slash was the last character). Then, the entire alphabet was added, but to be backwards compatible, the / was retained after the R. So, it goes "...PQR/STUV...". That is why there is a jump between R and S. 209.149.114.69 (talk) 18:04, 9 July 2015 (UTC)[reply]

Makes sense, but I'd like a citation so I can read more about this. --174.88.133.209 (talk) 22:51, 9 July 2015 (UTC)[reply]

Have you ever wondered why sites that show the coding, like this, place a / between R and S? That isn't a typo. 199.15.144.250 (talk) 13:34, 11 July 2015 (UTC)[reply]

I'm not sure of the exact reason - but punched cards were read at phenomenal speed - and there were issues where some code combinations would make the card liable to tear in half when accelerated rapidly from the stationary card deck and into the reader. Also some early punches used air, blown through the holes in the card, to operate switches (rather than an optical detector)...in that case, air could leak through one hole and out of another under some perverse conditions inside the reader. So much of the hollerith code is an effort to work-around those kinds of mundane electro-mechanical systems. SteveBaker (talk) 22:27, 12 July 2015 (UTC)[reply]