Jump to content

Wikipedia:Reference desk/Computing: Difference between revisions

From Wikipedia, the free encyclopedia
Content deleted Content added
Ajax-x86 (talk | contribs)
Tags: Mobile edit Mobile web edit
Line 201: Line 201:


= December 10 =
= December 10 =

== Multiprocess webview ==

So yeah, I found multiprocess webview in settings.What can I do with it? Should I enable or disable it?[[User:Ajax-x86|Ajax-x86]] ([[User talk:Ajax-x86|talk]]) 16:22, 10 December 2018 (UTC)

Revision as of 16:22, 10 December 2018

Welcome to the computing section
of the Wikipedia reference desk.
Select a section:
Want a faster answer?

Main page: Help searching Wikipedia

   

How can I get my question answered?

  • Select the section of the desk that best fits the general topic of your question (see the navigation column to the right).
  • Post your question to only one section, providing a short header that gives the topic of your question.
  • Type '~~~~' (that is, four tilde characters) at the end – this signs and dates your contribution so we know who wrote what and when.
  • Don't post personal contact information – it will be removed. Any answers will be provided here.
  • Please be as specific as possible, and include all relevant context – the usefulness of answers may depend on the context.
  • Note:
    • We don't answer (and may remove) questions that require medical diagnosis or legal advice.
    • We don't answer requests for opinions, predictions or debate.
    • We don't do your homework for you, though we'll help you past the stuck point.
    • We don't conduct original research or provide a free source of ideas, but we'll help you find information you need.



How do I answer a question?

Main page: Wikipedia:Reference desk/Guidelines

  • The best answers address the question directly, and back up facts with wikilinks and links to sources. Do not edit others' comments and do not give any medical or legal advice.
See also:


December 4

Sources/References for licences of Unix commands to implement in Wikidata

Hello all,

I'd like to add the licenses and their references to the following Unix commands in Wikidata: List of Unix commands. The current status of the inserted (but not always referenced) licenses can be seen on my (german) talk subpage: de:Benutzer:Hundsrose/Wikidata Spielwiese. Is there a general approach to find them online and reference them afterwards in Wikidata, maybe with the alias (Unix) as example?

Many thanks for your help.

Best regards --Hundsrose (talk) 12:32, 4 December 2018 (UTC)[reply]

Well, this is a worthy endeavour, but may be a bit like the search for the holy grail. Many of the UNIX commands have been re-implemented multiple times - for most you will find at least a BSD version and a GPL version. Many have also been re-released under different licenses at different times. And maybe the original AT&T code is even in the public domain (AT&T was convinced enough to settle UNIX System Laboratories, Inc. v. Berkeley Software Design, Inc. before a formal verdict, and without getting anything out of the lawsuit). --Stephan Schulz (talk) 13:09, 4 December 2018 (UTC)[reply]
Thanks a lot for your answer, @Stephan Schulz:! I underestimated the complexity of the question and the historical background. I appreciate the clarification and I wasn't aware of the lawsuit. Concerning the BSD and GPL version: on the webpage of the Free Software Directory, I can see the licenses of GNU [1]. Using the same webpage for BSD etc. wil display a lot of software. Could you please help me to understand how to filter on this, meaning searching for a specific unix command: [2]? Your help is much appreciated. Kind regards --Hundsrose (talk) 17:28, 4 December 2018 (UTC)[reply]
Sorry, but it looks like you jumped into something a lot more complicated than you thought. History of Unix is worth a read if you haven't read it. "BSD" isn't even a single thing. Are you talking about the actual good old BSD system that was written and distributed from UC Berkeley in the 1970s-80s (which doesn't have an official webpage that I know of), or the various descendants of it such as FreeBSD and NetBSD? And what about the proprietary Unixes like HP-UX and Solaris? The problem is there's no single software program you can point to and say, "This is ping", unless you're talking about a specific historical version of ping, e.g., "This is Version 7 Unix ping". "Unix" today, to the extent it is a "thing", is a set of standards (not source code) maintained by The Open Group, which actually owns the "Unix" trademark. As long as you write code that conforms to the standards, your code can be considered at least unofficially "Unix". If I write my own standards-conformant ping and release it under my own Super Cool Awesome License, are you going to list that license? Most proprietary software isn't even released under a standardized license like the GPL; it's just under generic "all rights reserved" boilerplate. This is why articles here on the English Wikipedia for proprietary software usually just list "closed-source" or "proprietary license" as the software license (example: Microsoft Windows). --47.146.63.87 (talk) 04:58, 5 December 2018 (UTC)[reply]
Yes, I agree. If I need to know the actual license for a given piece of software, I would normally look at the package description and/or the source code package for that particular version I need. There is not really a generic answer for "ping" or "cat". --Stephan Schulz (talk) 08:54, 5 December 2018 (UTC)[reply]
Thanks a lot for your insights! I will read myself through the proposed topics and I also removed the license column from my subuserpage. Best regards --Hundsrose (talk) 19:47, 5 December 2018 (UTC)[reply]

LTE

If a smartphone doesn't include 700mhz (band 17) when a carrier says it should. Would LTE still work anyway if the frequency is higher, lets say 850mhz.Ajax-x86 (talk) 15:44, 4 December 2018 (UTC)[reply]

You should address this question to your carrier. Ruslik_Zero 20:42, 4 December 2018 (UTC)[reply]

If your phone doesn't support the band then it doesn't support the band. It doesn't matter what other bands it supports and what their frequency is. (As noted in the LTE frequency bands articles, some bands share at least part of the same frequencies. For example a phone which supports 1 and 3 is not guaranteed to support 4.) It is sometimes possible to enable unsupported bands when they are actually supported by the hardware but this is rare, complicated and depends on the specific phone and may also violate local law or similar.

If the carrier also uses other bands which the phone does support, then it will work in those areas those bands are available, but it won't work in areas where the only band is the one which isn't supported and of course it also won't be able to use the unsupported band where both are available. (So coverage may be poorer or suddenly drop out, and speeds may be reduced.)

I don't understand what you mean by "carrier says it should". If the carrier sold you the phone and told you it supports a band but it doesn't then in many developed countries and even a number of developing ones this would come under some form of false marketing and you're probably entitled to either return the phone for a refund or have the carrier fix the defect although you we can't give you legal advice on that here.

If your carrier says you need a certain band to use your network my earlier comment applies. The carrier saying you need the band means they likely won't offer any support for coverage or speed problems but it will work where and how it works in areas where other bands may be available that are supported. (And assuming it's not an LTE only network, for any non LTE bands and networks the phone and carrier use.) Precisely how well, you'll need to find out how important that band is in their coverage. Although bear in mind networks are always changing, and especially with next generation networks coming in they may switch which bands they are using. So even if it works fine in most areas that matter to you now, it doesn't mean it will 1 year from now. Especially if support for that band has always been required and offered in all phones the carrier has official supported or sold.

Nil Einne (talk) 06:57, 5 December 2018 (UTC)[reply]

December 5

Battery eliminator for smartphones?

How could a device substitute the battery of a smartphone? The first phone I'd like to use (without buying a battery for it) has 3 pins. How do you connect the third pin, so the phone 'believes' it's being powered by a battery? Would a resistor connecting the middle pin to ground be enough? Doroletho (talk) 19:34, 4 December 2018 (UTC)[reply]

It depends on what the middle pin is supposed to do.
On some batteries the middle pin is a thermistor to monitor battery temperature. In that case you could probably replace it with the appropriate value of resistor.
On some batteries the middle pin is a serial communication pin that allows the phone to talk to the circuitry in the battery. Often this is called "Battery Status Indicator" or "Battery Size Indicator", but I think it can sometimes do more than just indicate the size and charge level. The phone may refuse to work if it can't talk to that chip. (So as to avoid running on a damaged battery.)
ApLundell (talk) 23:35, 4 December 2018 (UTC)[reply]
My question is, is this really necessary? Have you looked on places like eBay for replacement batteries? They're likely cheaper than anything you'd buy or make to substitute for it, unless you already have a bunch of electronics equipment lying around. And if you want some kind of fixed, plugged-in computing device, like for a kiosk, you might be more satisfied with something like a Raspberry Pi connected to a touchscreen, or a low-end tablet. --47.146.63.87 (talk) 05:58, 5 December 2018 (UTC)[reply]
A modern Li-Ion-battery is really misnamed - the thing you change is not just a battery, but the battery with a charge management system probably much more powerful than the computer on board the Apollo 13 mission (the official unit of computing power ;-). But if the battery is near dead, you can still operate the phone when connected to the charger - so I don't see too many use cases for a battery eliminator. --Stephan Schulz (talk) 17:18, 5 December 2018 (UTC)[reply]
Does this apply to smartphones' Li-Ion batteries as well? I am aware that laptops' batteries are mini-computers, but smartphones might need something simpler.Doroletho (talk) 18:04, 6 December 2018 (UTC)[reply]
Phone batteries are usually just lipo cells, sometimes with a thermal protection circuit, and sometimes with a resistor that informs the phone of the cell's capacity. The charging circuit is inside the phone. If you know the phone make and model, look someplace like dx.com for replacement cells. Don't get the very cheapest ones since they are crap, but the ones above US $5 or so are usually decent. 173.228.123.166 (talk) 08:32, 7 December 2018 (UTC)[reply]

Yahoo! Powered

How do I uninstall the damn thing? Answers.microsoft.com is singularly unhelpful: I tried removing it using control panel, but nothing happened. There are a bunch of other, more elaborate procedures out there, but I'm not sure which I can trust. Clarityfiend (talk) 06:38, 5 December 2018 (UTC)[reply]

From some quick searches, it's probably worth trying Malwarebytes first. Nil Einne (talk) 16:27, 5 December 2018 (UTC)[reply]

Any website to test Wiki Admin

Is there any website where you can test admin stuff like protecting articles — Preceding unsigned comment added by 71.161.214.1 (talk) 16:29, 5 December 2018 (UTC)[reply]

There is https://test.wikipedia.org/wiki/Main_Page in particular https://test.wikipedia.org/wiki/Wikipedia:Requests/Tools and https://testwiki.wiki/. However you may wish to install your own copy of MediaWiki and play around yourself. Graeme Bartlett (talk) 02:14, 6 December 2018 (UTC)[reply]
If you want to install MediaWiki, here's a guide: https://www.mediawiki.org/wiki/Manual:Running_MediaWiki_on_Debian_or_Ubuntu. However, it can be rather a time-consuming lengthy process. Doroletho (talk) 00:39, 7 December 2018 (UTC)[reply]

1st computer components?

I liked the discussion on "how did the 1st compiler get compiled, and who compiled it?" I think the philosophical answer revolves around "it was something that was not quite a compiler." I wonder if I can extend some discussions on other 1sts in computers, cuz I'm still puzzled on how the 1st programs were made. Some trivia I have here is (and wondering if anyone can expand on)

  • -IBM wrote the 1st BIOS chip, then others get IBM assembler codes, downloaded the bits.
  • -Xerox was the 1st company to have a graphical operating system.

Who wrote the 1st motherboard? 67.175.224.138 (talk) 19:21, 5 December 2018 (UTC).[reply]

  • In the IBM PC era, much of such work in large, well-financed companies was done with cross compilers. You run the tools on an existing computer, and target them to the new computer. This is easier if they use the same CPU family, as the cross compiler is then close to, or the same as, the compiler on the existing computer. Operating systems like CP/M (on the 8080 or Z80) were well established by the time of the IBM PC and could be used for this. IBM probably used one of their own larger machines for doing it. The Intel "Blue Cube" (see ISIS for a photo) was used for a lot of such work.
The original IBM PC manual was in the typical squat, floppy disk-sized, hard ringbinder and the original ones included a BIOS source listing. You also needed a Pink Shirt Book in order to program one. Andy Dingley (talk) 21:43, 5 December 2018 (UTC)[reply]
The article Motherboard might tell you what you want to know about them, though I'm not very sure what you're trying to get at. Dmcq (talk) 22:58, 5 December 2018 (UTC)[reply]
Perhaps the rather Heath-Robinson machine shown at Colossus computer might indicate to you how computer hardware started off. Dmcq (talk) 23:02, 5 December 2018 (UTC)[reply]
And if you are interested in the days before there was a BIOS or ROM, to start up some computers, such as PDP 8, you had to use switches on the front panel to load instructions into memory, which would load the first bootstrap program into memory, and then transfer control to it. On IBM 370 mainframes you could set the boot address with dials on the front. If you had no bootable disk, you could use an IPL magnetic tape to load up an operating system to a disk. Graeme Bartlett (talk) 00:39, 6 December 2018 (UTC)[reply]
I think that on the IBM 1400 series or [[IBM 1620], you put one card in the card reader, pressed a button to read that card, and that booted it. Bubba73 You talkin' to me? 01:13, 6 December 2018 (UTC)[reply]
Nitpick: you didn't have to toggle in a boot program, unless for some reason you were without any working I/O peripherals. Typical procedure was to load a boot program from paper or magnetic tape. Sometimes "toggling in" was done for fun, experiment, or to impress onlookers. --47.146.63.87 (talk) 05:55, 6 December 2018 (UTC)[reply]
The successful IBM PC clones very notably did not copy the BIOS. IBM would not license the BIOS to anyone else. This was the main impediment to legally producing PC clones. Copying the software without permission is copyright infringement, and got you sued by IBM (as was the case with other computers). Phoenix Technologies reverse-engineered the PC BIOS using clean room design: a team analyzed the BIOS, including the source code provided in IBM's PC manuals, and wrote a specification for a compatible BIOS. A separate team then wrote a BIOS from scratch based on the specification, without looking at any IBM code.
Motherboards are younger than electronic computers. Early computers were wired together by hand. Later more standardized methods like wire-wrapping became common. The idea of a "motherboard" originated with early microcomputers. Minicomputers and mainframes took up at least a large cabinet; there was no single circuit board that all the components were on. --47.146.63.87 (talk) 05:55, 6 December 2018 (UTC)[reply]
I thought Phoenix Technologies or whoever, had to take out all the 0's and 1's from the BIOS chip in order to make their own version of it? 67.175.224.138 (talk) 00:35, 7 December 2018 (UTC).[reply]

The above statement ("Nitpick: you didn't have to toggle in a boot program, unless for some reason you were without any working I/O peripherals. Typical procedure was to load a boot program from paper or magnetic tape.") is completely wrong. The first PDP-8, for example, had no software in it when first powered on -- and this includes having no software to read a paper tape or tape drive. The memory was completely blank. It had switches on the front panel that would allow you to put in enough 1s and 0s so that it could read the paper tape. the paper tape would in turn have a program that told the PDP8 how to read a magnetic tape drive. See PDP-8#Programming facilities and How do I boot a pdp8?. --Guy Macon (talk) 08:37, 6 December 2018 (UTC)[reply]

I should have worded that differently, and thought so after saving the edit, but it was late. I probably shouldn't edit late at night. What I meant was you didn't have to punch in a full OS. As you note, you had to set some instructions to tell the machine to read a program from the tape and run it. --47.146.63.87 (talk) 21:57, 6 December 2018 (UTC)[reply]
Which is why we call it "booting". You are simply telling the computer to read and execute a program that in turn will load the full OS (or there might be multiple stages, in fact; the first program might read a single block of disk or tape, and this in turn would include the program to read the whole OS from disk or tape). The initial instruction tells the machine to bring itself up "by its own bootstraps", an reference to the seeming impossibility of the feat. And so the initial program was called a "bootstrap loader" and the process was "bootstrapping". The short form "boot" derives from that. --76.69.46.228 (talk) 06:19, 7 December 2018 (UTC)[reply]
Not early computer, but early microcomputer: I still have the bootstrap instructions for the IMSAI 8080 (from my first job out of college, 1978) wired into my fingers. 21 00 01 DB FF E6 7F C2 02 00 DB FE 77 23 C3 02 00. This would simply read a slightly more sophisticated loader from (in our case) the paper tape reader on a TTY; that loader knew about checksums. I sure miss switches and flashy lights -- computers were still slow enough in the late '70s that debugging stuff could involve noticing anomalous light patterns. --jpgordon𝄢𝄆 𝄐𝄇 06:47, 7 December 2018 (UTC)[reply]
You can buy a "so good that an experienced user can't tell them apart without opening the case" Altair 8080 replica at http://altairclone.com/ -- and the price is the same as it was in 1976![3]
The IMSAI was a basically a clone of the Altair. In theory you could modify the Altair clone to be an IMSAI clone, but you would have to find a source for those paddle switches. --Guy Macon (talk) 08:19, 7 December 2018 (UTC)[reply]
Altair: http://oldcomputers.net/altair-8800.html
IMSAI: http://oldcomputers.net/imsai8080.html
--Guy Macon (talk) 08:23, 7 December 2018 (UTC)[reply]

December 6

Limitations on Running Very Large Number of Chrome Windows simultaneously ?

Is there a practical or theoretical limit on running a very large number of Google Chrome windows simultaneously under Windows 10 on a desktop machine with 12 Gb of main memory? Does running a very large number of Google Chrome windows simultaneously lead to a risk that Chrome can shut down suddenly? Robert McClenon (talk) 05:16, 6 December 2018 (UTC)[reply]

Are we talking about a couple of hundred tabs or a couple of thousand? And do you have google chrome background tab throttling[4] enabled or disabled? --Guy Macon (talk) 08:46, 6 December 2018 (UTC)[reply]
Here[5] is a fellow who did some tests and found:
  • A typical 8GB RAM and i7 Processor can crash out your google chrome for around ~9600 tabs.
  • If you have 4GB RAM and cheap processor than it can crash out at ~8000 tabs.
  • After I pass about 50 tabs, I can no longer see the favicons that help me distinguish tabs apart.
  • When nearing 70 tabs, I can no longer see any text from the page title; all I see is 70 little bumps.
  • The first 3k can be opened in less than a minute, but as you get close to 10k it takes about a second per tab.
--Guy Macon (talk) 08:53, 6 December 2018 (UTC)[reply]
This might not be entirely representative, the guy just opened google.com 10,000 times. Chrome might have parallelized some of this under the hood, and Google's homepage is much less demanding than the average website. I think if you opened something very simple, like a standard 404 page on Apache, you could probably get even more tabs.
Well, first, you can see what tab you are on by hovering over it. Second, the slowing down will predictably occur when the browser frames can no longer all reside in physical memory. Robert McClenon (talk) 02:00, 8 December 2018 (UTC)[reply]
FWIW, not what you're asking, but I use Waterfox and I've found that 1,000-2,000 tabs (of different webpages) work just fine on 16 GB and probably a lot more would work too but I have to restart it occasionally because it tends to fill up the RAM over the course of a week or two. 93.136.105.185 (talk) 09:56, 6 December 2018 (UTC)[reply]
I am using about 50 tabs, not hundreds or thousands. I wasn't aware of the background tab throttling feature, so I am using whatever is the default. The Chrome crashes to which I am referring happen suddenly and seemingly randomly. When I launch Chrome again, it restores the tabs. Chrome isn't always able to restore the tabs if Windows crashes, but that is a different question. (As to telling the tabs apart, they identify themselves when hovering with the mouse cursor.) That is, with about 50 tabs, the performance is typically fine, and then Chrome suddenly disappears. Robert McClenon (talk) 00:53, 7 December 2018 (UTC)[reply]
Sorry for more unwanted Mozilla-based browser promotion, but Firefox/Waterfox/Pale Moon with Session Manager addon has always managed to restore the tabs for me, even if power failed, and I never get crashes any more (used to have a lot with Firefox until a year or two ago, probably when I switched to Firefox ESR and then Waterfox). Maybe Chrome also has an "extended support release" program. I find staying at the bleeding edge is rarely worth it with browsers. 93.136.49.90 (talk) 06:37, 7 December 2018 (UTC)[reply]
What is happening to me is that, apparently randomly, Chrome just disappears suddenly. That does seem to give back the memory. (I know that it should give back the memory. I also know that memory is not always as neatly managed as it should be.) Robert McClenon (talk) 02:00, 8 December 2018 (UTC)[reply]

Removing apps that can not be un-installed from Windows 10 "Apps & Features"

My "Apps & Features" include such "dead" apps.

I have tried the regedit method ( HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Uninstall) but none of these apps appear there.

אילן שמעוני (talk) 08:55, 6 December 2018 (UTC)[reply]

I just did something I loath to do - going over the entire registry and deleting all keys that include the dead apps names. It did remove them from the uninstall list. I hope I didn't do any damage.
אילן שמעוני (talk) 21:25, 6 December 2018 (UTC)[reply]
You could always just ignore them. You definitely should back up the Registry before editing it. In Windows 10, you could also try the "Windows Reset" feature, which reinstalls Windows without deleting your personal files. (You should still back them up anyway, and make regular backups as a matter of course, since plenty of things can cause data loss.) --47.146.63.87 (talk) 22:01, 6 December 2018 (UTC)[reply]

December 7

In believe that such a website / tool might be of great help for people who want to learn a completley new language as they would be able to begin by learning the most common words in that language (in this instance Norwegian, but it could be any other language that has a Wikipedia site) and how to spell those most used words according to the most common spelling.

I tried googling if such a program / tool exists and all I have found so far is that there is such a tool at github for the English Wikipedia but I'm not sure if this particular tool could be fixed so that it would do the same thing but with a different Wikipedia (the Norwegian Wikipedia in this instance) + I am not sure how to run github repositories. Do any of you know a simpler solution? (maybe there is some other easier tool / website?) ויקיג'אנקי (talk) 21:23, 7 December 2018 (UTC)[reply]

I'm sorry, I don't know. However, two points: (i) Norwegian orthography is unusually problematic. (ii) The description that you point to says nothing about recognizing two or more strings of letters as alternative spellings of the same word. It's not easy to recognize that "traveler" and "traveller" (or indeed "jail" and "gaol") are a pair of alternatives. -- Hoary (talk) 00:34, 8 December 2018 (UTC)[reply]
The unusual problem with Norwegian that Hoary mentions is so great that there are actually two Norwegian Wikipedias. Please see no.wikipedia.org for the Bokmål version and nn.wikipedia.org for the Nynorsk version. Nyttend (talk) 19:30, 9 December 2018 (UTC)[reply]

December 8

Path to Microsoft Word Autosaves

Can someone please tell me what the path of directories is (with any fill-ins) where Microsoft Word saves its autosave information every N (default 10) minutes? I would like to see if I can find a work in progress that wasn't picked up by autorecovery. Thank you in advance. By the way, I am using Microsoft Office 365 (there goes Microsoft changing the numbers in a non-sequential fashion) and Windows 10. Robert McClenon (talk) 01:55, 8 December 2018 (UTC)[reply]

This can vary with the installation, but try C:\Users\user\AppData\Roaming\Microsoft\Word\
or C:\Users\Your_username\AppData\Local\Microsoft\Word
or C:\Users\Your_username\AppData\Local\Temp
Alternatively, search your home drive (assumed to be C: above) for file names beginning with "Backup of " or with extension .wbk Dbfirs 12:18, 8 December 2018 (UTC)[reply]

Xubuntu: Find and remove conflicting keyboard shortcuts that aren't in Settings

On my PC running Xubuntu 18.04, IntelliJ IDEA isn't responding to Shift-F6, which it uses as the Rename shortcut. (No idea what Shift-F6 does, except that I can't detect any effect on my typical usage. The symptom is what Shift-F6 isn't doing: it's not reaching IntelliJ and activating the Rename command.) But nothing is listed for Shift-F6 under Settings > Keyboard > Application Shortcuts or Settings > Window Manager > Keyboard. I've checked that Shift-F6 is indeed Rename in IntelliJ, so something at the desktop-environment level or lower must be intercepting it. How do I find what component of my system is intercepting Shift-F6, and how do I turn it off so that IntelliJ can receive Shift-F6? Also, are there any tools that can help me answer this question about other keyboard-shortcut conflicts, in case I encounter more that aren't in either menu? NeonMerlin 03:02, 8 December 2018 (UTC)[reply]

Note that this involves using a terminal. Step one: xev. Run it, press ⇧ Shift+F6, and see if the window receives it. If not, something in the windowing environment is grabbing that key combo. In that case, the next step is process of elimination. You could just start killing processes, retrying the xev test after each time. Make sure you close any other applications first of course.
If a bare xev window does receive the key combo, your next step should be to run IntelliJ, and then run xev and attach it to the IntelliJ window. This will show you whether IntelliJ is getting the key combo. To get the window ID of the IntelliJ window, use xwininfo. --47.146.63.87 (talk) 09:54, 9 December 2018 (UTC)[reply]
I should have asked this first: does ⇧ Shift+F6 work in other programs? If you don't have another GUI program to try it in easily, the best thing to do is exit X to a terminal, run showkey, and see if ⇧ Shift+F6 works in the terminal. If the keypress isn't detected in the terminal, it's either a hardware or OS problem. I have no experience with Xubuntu, but it looks like you can use the standard virtual console switch method: Ctrl+Alt+F1 will switch you to a terminal. Enter your username and password to log in. If the keypress doesn't work in a terminal, see if you can get another keyboard to try. --47.146.63.87 (talk) 10:12, 9 December 2018 (UTC)[reply]

Is Driverwhiz.exe safe software?

I am trying to download and install TWAIN for image manipulation. One way to do it is via TWAIN and that in tern led me to the above exec.

Thanks AboutFace 22 (talk) 23:54, 8 December 2018 (UTC)[reply]

TWAIN is not, itself, a thing you install. It's an API that programmers write software to use. What exactly are you trying to do? Details such as your platform, operating system, and whatever hardware (e.g., scanner, digital camera) you're trying to work with would help. If trying to use something like a scanner, did you install whatever drivers and software were provided by the manufacturer, and are you following the manufacturer's instructions? If you don't have an installation disk, go to the manufacturer's website and see if the software is available there.
Randomly Googling for software is generally not a great idea. Symantec says that DriverWhiz is "scamware": it tells you things are wrong with your computer and tries to get you to pay money to "fix" it. Paying money won't actually accomplish anything. If you've installed it, try to uninstall it normally. Then run a full system scan with an antivirus program. If you're not using a third-party program, Microsoft provides Windows Defender or Microsoft Security Essentials with Windows (depending on version). (This is assuming you're using Windows and the "DriverWhiz" program is the same thing as in that Symantec bulletin.) --47.146.63.87 (talk) 09:32, 9 December 2018 (UTC)[reply]

Thank you for the answer. Mine is a visual recognition application that I am trying to implement. My hardware now is HP Pavilion With Windows 10 OS 64 bit. The plan for this application is to take images directly off the webcam and transform them into some other derivatives and the latter will be stored in SQL Server. That will not be the end of the process but a very important ingredient. From what I've read about TWAIN it seems it provides a user interface, a GUI which I would prefer to avoid. I downloaded driverwhiz but haven't installed it yet.

In lieu pf what you've said about the driverwhiz what other options exist for me? In principle I can write a totally independent C++ application that will read the webcam images one by one, it is what I need. I need 10 such images per second. Any suggestion will be appreciated.

Thank you, AboutFace 22 (talk) 15:27, 9 December 2018 (UTC)[reply]

On new versions of Windows, including Windows 10, you should learn and use the MediaCapture interface to retrieve images from your web-camera. Here is the official introductory documentation: Basic Photo and Video Capture... from Microsoft.
Don't waste time learning obsolete technologies; don't install third-party software; image capture is a standard task that can be done using standard tools built in to the default Windows development environment - as long as you carefully follow Microsoft's documentation.
Nimur (talk) 16:05, 10 December 2018 (UTC)[reply]

December 9

Android developer setting

Good afternoon folks, what does the enable view attribute inspection mean? — Preceding unsigned comment added by Ajax-x86 (talkcontribs) 17:40, 9 December 2018 (UTC)[reply]

Moving windows with a single keystroke (Windows 10)

Hello, I've looked around online but found nothing; probably I'm missing the right keywords.

I use keyboard shortcuts extensively, and I'm familiar with many of them to the point that I don't look at the keyboard at all when using them. Problem is, sometimes I end up doing something unintentionally, and I can't undo it without restarting the computer. In this case, there's some way whereby typing a few keystrokes causes Windows to do a completely different set of commands: down arrow minimises, up arrow maximises, left and right arrows un-maximise and align to one side. (The effect with these is the same as holding down the start button and hitting an arrow.) Other key commands do yet other things, e.g. Ctrl+S brings up Speech Recognition (again, the same as Win+Ctrl+S), and it's virtually impossible to use the keyboard for normal purposes. The last time this happened (earlier today), I'd been using my left hand to operate a bunch of commands, particularly Ctrl+H, Ctrl+V, and Alt+A. I'm on a QWERTY laptop with bottom-left Windows and Function keys.

How does one bring up this state of affairs, and more importantly, how does one end it without restarting? Nyttend (talk) 20:48, 9 December 2018 (UTC)[reply]

That sure sounds like the windows key is getting stuck. Does pressing the windows key again fix it? Or does fn-win do something special on your laptop? Also worth making sure that stickykeys isn't enabled in your accessibility settings. Egglz (talk) 15:11, 10 December 2018 (UTC)[reply]
It's definitely not getting stuck (if it were, restarting wouldn't work), and pressing it again doesn't seemingly have an effect if I remember rightly. Fn+Win doesn't seem to have an effect. I'd thought of Sticky Keys, but when I trigger it with any other key, I get a message that's basically "Do you really want to trigger Sticky Keys?" No message with this. Nyttend (talk) 15:22, 10 December 2018 (UTC)[reply]

December 10

Multiprocess webview

So yeah, I found multiprocess webview in settings.What can I do with it? Should I enable or disable it?Ajax-x86 (talk) 16:22, 10 December 2018 (UTC)[reply]