Computing desk
< August 12	<< Jul \| August \| Sep >>	August 14 >

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.

August 13

WhatsApp

If I saw that one of my WhatsApp contacts has disappeared, the person has generally deleted my phone number from their contacts, right?--95.116.217.118 (talk) 00:30, 13 August 2013 (UTC)[reply]

Yeah, either that or they've uninstalled WhatsApp altogether. Note: it doesn't mean they've blocked you though, as noted here. --.Yellow1996.^(ЬMИED¡) 01:24, 13 August 2013 (UTC)[reply]

Simultaneous copies to USB thumbdrive very slow

It's been my experience that when doing two or more simultaneous copy file operations to the same USB thumbdrive, the data transfer becomes very very slow. Why is this? Simultaneous copies to an SSD don't see any noticeable slow-down. (at least in my experience) --157.254.178.141 (talk) 01:06, 13 August 2013 (UTC)[reply]

USB connection is the bottleneck, it isn't as fast as SATA connection 140.0.229.26 (talk) 01:31, 13 August 2013 (UTC)[reply]

USB3 is almost as fast as the fastest SATA. But anytime I've had more than one input/output operation going on between drives, it is much slower than doing one and then the other. Bubba73 ^{You talkin' to me?} 01:43, 13 August 2013 (UTC)[reply]

Flash media likes to write in large blocks, because a block needs to be cleared and then rewritten when a change is made. This means it is very fast for large sequential writes, but much slower for several small writes. If you're copying to two locations on the drive at once, you're asking the drive to jump back and forth between two write locations. USB drives are usually marked as removable media, which means Windows tries to get data out to the drive as fast as possible just in case you decide to pull it out. With non-removable media such as an SSD, the OS knows it can take its time getting the data out, and send write requests as complete blocks. Jessica Ryan (talk) 12:14, 13 August 2013 (UTC)[reply]

Our various articles on buffers and caches don't really cover it very well. Just to clarify, in both cases Windows Explorer is sending out interleaved writes to both files. On a removable drive, the interleaved writes get pushed out to the drive as quickly as possible, causing the drive to issue repeated writes on the same few blocks, which is a very slow operation. On a non-removable drive, the writes are still requested by Explorer in the same order, but the disk driver knows there is no hurry. It uses system memory to buffer the writes until it can send it out as a few large whole-block writes rather than several partial writes. USB hard disks (with platters, non-SSD) are likely to use a drive with an onboard buffer, so even if Windows is pushing out writes in an inefficient order, the drive uses its onboard memory to get the same speed benefits that disk drivers give non-removable drives. Jessica Ryan (talk) 13:24, 13 August 2013 (UTC)[reply]

I noticed this too, sometimes i'd be copying one file that would take 1 minute, start copying another file of the same size to the same place and all of a sudden it's going to take 5 minutes. And it would, it wasn't just windows famous time estimate. So I started using teracopy. Vespine (talk) 00:38, 15 August 2013 (UTC)[reply]

Firefox and Software versioning

Do the authors of Firefox know about Software versioning? Every month or two there is a release with a new major version number but with only very minor changes. Bubba73 ^{You talkin' to me?} 02:18, 13 August 2013 (UTC)[reply]

Yes, haven't you been paying attention to software versioning? For over two years, Mozilla's been on a "rapid release schedule, and they manage their release versioning thusly. Because many parts of Firefox are open source or free software, and the final web browser is an agglomeration of numerous projects with independent schedules, Mozilla (the organization who brands and develops the final Firefox product) have found this versioning system to work more effectively. Don't get hung up on the magnitude of the version-numbers; there are no units for software-version; it is a nondimensional quantity. Nimur (talk) 04:27, 13 August 2013 (UTC)[reply]

With a "major" release every month or two, soon they will be at version 914.0, and from my end, there is essentially no difference from one version to the next. Bubba73 ^{You talkin' to me?} 04:30, 13 August 2013 (UTC)[reply]

Did you notice anything different from FF22 to FF23? All I've got so far is they changed the logo! --.Yellow1996.^(ЬMИED¡) 16:18, 13 August 2013 (UTC)[reply]

Why are you concerned? If you can't tell the difference between versions, then you're losing nothing by upgrading, and you won't notice or care if you don't upgrade. If you do not wish to upgrade, nobody requires you to do so. Mozilla does not bill you for software updates. Plenty of other people are taking advantage of new browser features, bug fixes, and the opportunity to select various levels of product stability. Those people benefit from a quick turnaround.

Some people upgrade their kernel every day; others are still living on Windows 98 or Linux 2.3. Part of the benefit of free software is that you may use it, modify it, and upgrade it (or not upgrade it) in any way you like. Nimur (talk) 04:49, 13 August 2013 (UTC)[reply]

Periodically, they break everyone's addons. It's hard to predict whether a given update will do that. Then of course they get swamped with addon updates, and it takes ages until the fixed versions appear in the Mozilla addon collection. So you have to go hunt down the homepages of your addons and update them manually. Whee! Bobmath (talk) 05:28, 13 August 2013 (UTC)[reply]

Every release can have security fixes and some bug fixes, but otherwise I don't see much that helps the average user. Bubba73 ^{You talkin' to me?} 13:38, 13 August 2013 (UTC)[reply]

I'm not as upset about FF as about Windows. Their versions do not come in monthly intervals, but still... XP was a bug fix to 2000, 7 was a fix to Vista, and neither was free of charge. And that doesn't even include the infamous pieces of software which refuse to install unless you have the Service Pack the programmers are in love with. Heck, (most) users pay for that kind of software!

Just my $0.02. - ¡Ouch! (^{hurt me} / _{more pain}) 06:45, 14 August 2013 (UTC)[reply]

Resolved

references to fiction versus non-fiction in search results

How can I quickly determine if the search result I am looking at refers to a fictional or non-fictional fact. Can an AI search process determine this for me? Is there a way that I can exclude all references to fiction, games, and other non-factual sources? 99.199.4.69 (talk) 17:37, 13 August 2013 (UTC)[reply]

I don't think there exists such a process. There are some search-engine tips here but I'm not sure if they're totally relevant. If you have a general idea of reputable sites for the query (ie. National Geographic for something on nature etc) then you can limit the search to such sites. Also using keywords like "news" and "journal" can help get factual information. --.Yellow1996.^(ЬMИED¡) 17:42, 13 August 2013 (UTC)[reply]

Point of clarification: consider the phrase "Mickey Mouse has two ears." Is this a fictional fact, or a non-fictional fact? SemanticMantis (talk) 18:15, 13 August 2013 (UTC)[reply]

SPSS v. 21: Python Will Not Run in Syntax Block

I installed the Python Essentials extension for SPSS v. 21, and I have been able to compile and run Python programs outside of SPSS. However, when I try to run a Python program within a BEGIN PROGRAM PYTHON syntax block, I get the following error:

The Integration Plug-in for Python is configured to work with a different version of SPSS Statistics. Please adjust the spss.pth and SpssClient.pth files so that the first line in each file specifies this version of SPSS Statistics. SpssClient.pth and spss.pth are located in the Python site-packages directory.

I've checked both of these files, and the version is set as 21. Does anyone know what might be causing this error?

Thanks! OldTimeNESter (talk) 19:50, 13 August 2013 (UTC)[reply]

LaTeX question

(moved from the maths desk) I want to force LaTeX to typeset each line at 0.99 the size of the previous one. Or maybe 0.95, haven't decided the precise value yet. So the lines get smaller and smaller. But I want the resulting document to "look right" locally, so the inter-line spacing must be right. I don't know the proper terms to use. Can anyone help? Or is there a better way than latex? Thanks, Robinh (talk) 19:59, 13 August 2013 (UTC)[reply]

You could try setting linespacing with \begin{spacing}{2.5} This paragraph has \\ huge gaps \\ between lines.\end{spacing} or insert a fixed-length space \vspace{length}. See b:LaTeX/Text_Formatting and b:LaTeX/Lengths.--Salix (talk): 05:44, 14 August 2013 (UTC)[reply]

Looking at b:LaTeX/Fonts#Arbitrary font size there is \fontsize{<size>}{<line space>}. So it looks like you can specify the font size and desired spacing in one go.--Salix (talk): 05:51, 14 August 2013 (UTC)[reply]

Thanks guys, but how do I synchronize text size change with newlines? I want right justified text if possible. Robinh (talk) 07:33, 14 August 2013 (UTC)[reply]

You can’t do that. Line breaks are chosen after the whole paragraph is typeset, so at that point, font sizes must be already fixed. (Which is perfectly reasonable: how could the computer determine where to break the lines if it didn’t know how wide are the words going to be?) You’ll have to break the text into lines manually.—Emil J. 11:19, 14 August 2013 (UTC)[reply]

Or at least, you can’t do that easily. What you could do in principle is to (1) typeset a whole paragraph (in normal size) inside a \vbox, using \parshape to make each line 100/99-times longer than the previous one, (2) peel off the lines one by one from the \vbox in a loop using \vsplit or \unvbox/\lastbox, and insert Postscript specials around each to scale it down to the desired size. This assumes you only have a relatively short piece of text with at least approximately known number of lines (because the \parshape has bounded length), it won’t work across pages. A similar (but easier) problem is solved on tex.stackexchange using such techniques, so you can take that as an inspiration. You can also try to ask there directly.—Emil J. 11:56, 14 August 2013 (UTC)[reply]

Thank you EmilJ for this. I guess it's harder than I thought. I was hoping for some argument along the lines of \setsequentialtypesettingsize{0.977} but this clearly does not exist. But it is now easier for me to use some clumsy wysiwyg system in the knowledge that TeX really isn't the correct tool. Best wishes, thanks everyone, Robinh (talk) 08:38, 15 August 2013 (UTC)[reply]

Agreed that it would be very difficult using only LaTeX. If it were only a few dozen lines or so, I'd link LaTeX source to an Inkscape plugin (depending on your OS and Inkscape version, there may be native TeX support). That way, you'd have high-quality vector output, that can be smoothly scaled and played with in an Inkscape svg. SemanticMantis (talk) 15:18, 15 August 2013 (UTC)[reply]

Thanks, SM. I do use inkscape (on all of windows/linux/macosx). Can you elaborate what you mean by 'link LaTeX source' please? Robinh (talk) 20:51, 15 August 2013 (UTC)[reply]