Wikipedia:Reference desk/Archives/Computing/2015 November 10

= November 10 =

Which was the first video recording format?
Could anyone tell which was the first video recording format?-- Jos   eph   08:42, 10 November 2015 (UTC)
 * The very first was the BBC's VERA system (1952) using steel tape, but that wasn't particularly successful. The earliest usable commercial format was Ampex 2" Quadruplex (1956).  The first domestic video recorder was the Sony CV-2000 (1965). Tevildo (talk) 09:13, 10 November 2015 (UTC)
 * And the earliest digital video format was CCIR 601 (1982). Tevildo (talk) 09:21, 10 November 2015 (UTC)
 * Thanks a lot.-- Jos   eph   12:17, 10 November 2015 (UTC)

✅

Are there different type of film magazines?
Which are the different types of film magazines..?-- Jos   eph   12:42, 10 November 2015 (UTC)
 * Do you mean publications about films, or types of camera magazine? --Canley (talk) 13:25, 10 November 2015 (UTC)
 * Talking on general. I think that will include too.-- Jos   eph   10:25, 11 November 2015 (UTC)

Slow connection in china
My access to wikipedia in Shanghai is incredibly slow. Assuming I manage to connect at all, the loading time of pages is nearly 5 minutes. Generally I can only access wikipedia with a VPN, which is inconvenient. This situation has only been occurring about a week or so. Does anybody know what might be causing this, and what I can do to resolve the issue? — Preceding unsigned comment added by 45.56.158.91 (talk) 13:20, 10 November 2015 (UTC)


 * The Great Firewall entirely blocks the Standard Chinese language Wikipedia and I think the Cantonese one too. It selectively blocks pages on the English language wikipedia (such as the Tiananmen Papers article). Further, because Wikimedia hosts all the encyclopedias on the same IP addresses (well, that's how it appears in the European cluster, I'm not sure that the Asian cluster is configured the same way) they can't do selective blocking based on IP address alone - they have to be parsing the URLs too - which is a much more resource-intensive task. If you're seeing slowdowns specific to Wikipedia, but not other international sites, it may be that the Great Firewall's filtering resources are in high demand. -- Finlay McWalterᚠTalk 13:42, 10 November 2015 (UTC)


 * If you use https, it will be even slower. The Great Firewall has to decrypt everything and then encrypt it again. 209.149.114.132 (talk) 14:09, 10 November 2015 (UTC)

Programs closing all over the place...
I was messing around in Control Panel on my Windows 10 PC the other day, and just for a laugh I attached a silly sound effect to the "Close program" event in the Sounds menu. Not surprisingly, the sound was played when I exited programs in the normal way, as I expected. However I found that it was also triggered at all sorts of random times – frequently, but with no pattern – when I wasn't actively using anything. I presume there are background tasks opening and closing all the time, but staring at Task Manager for a few minutes didn't reveal anything obvious. Any thoughts? – I'm just intrigued, rather than concerned! Cheers, Hassocks  5489 (Floreat Hova!)  13:28, 10 November 2015 (UTC)


 * I would guess that services may be starting & stopping within various instances of svchost.exe, and they're firing the "Close Program" event for the OS while the host process continues on. I would also guess that other programs may occasionally do the same/similar things. FrameDrag (talk) 20:32, 10 November 2015 (UTC)
 * That would make sense. Interestingly when I looked at the svchost.exe article, it says it was introduced with Windows 2000 – and when I used to have a Windows 95 PC which had sounds attached to close program events, the behaviour I described above didn't happen, so that would suggest it is to do with svchost.exe.  Cheers,   Hassocks  5489 (Floreat Hova!)  09:02, 11 November 2015 (UTC)
 * Windows 386-enhanced/95/98/ME was a completely different operating system from Windows NT/2000/XP/Vista/7/8/10. Consumer-oriented versions of Windows switched lineages between ME and XP. If the app-close-sound behavior changed then, it could be for any number of reasons because most of the code was rewritten. I doubt it's related to svchost.exe, which was just a performance optimization; the NT lineage always had services and the 95 lineage never had them. I doubt it's related to services at all, because they run in a separate user account that's supposed to be isolated from the GUI. My first guess would be third-party software of some sort. In this person's case it was BOINC. -- BenRG (talk) 18:41, 11 November 2015 (UTC)
 * Found this that might shed light on what exactly is making sounds, if you're still curious. I know I'm curious now... FrameDrag (talk) 21:48, 11 November 2015 (UTC)

Black hacking, Scammers etc., etc.
Good afternoon all. I am in my late 60's, a Briton in Britain, fairly IT Literate (as a regular user) and becoming increasingly terrified about logging into any of my very many internet accounts, whether Banks, Building Societies, Pension Providers, Amazon, Pay-Pal, Health Service providers, Store Accounts, Credit Card providers et al. I even play the National Lottry online. I buy plane and train and holiday bookings online. And I am very aware that I am not in the minority. Heck, even by posting this help for advice I will probably have made myself an attractive potential target for some unsavoury criminal. But unless I want to cancel all my online accounts and instead revert to carrying cash and keeping my nest-egg at home (under the mattress?????), how can I keep my identity and property safe. I never share my passwords, I change them frequently, and I never use the same password across my accounts. But is that enough? I have a regularly updated virus scanner and I double - lock my doors at night. Any advice would be most welcome. Thanks in anticipation. 92.236.124.188 (talk) 16:18, 10 November 2015 (UTC)


 * First fact: Nothing is secure. To make your computer secure, you must completely destroy it. Dropping it into an active volcano might do the trick. As long as there is any hint that it was once a computer, it is not secure. Second fact: There is no such thing as privacy. Everything about you is known by someone (many someones). The concept of privacy is a state of mind. Do you feel private? Feeling private does not mean you are private. So, what to do? You can make yourself less of a target by never using anything that has electricity running through it. Move to a cabin in the woods. Survive off of berries and squirrels. A more common solution is to make yourself a harder target. You are doing that. Don't sign up for every online service you see. When you do sign up, use a hard-to-crack and unique password. What many people do not do is delete the account when you don't need it anymore. I actually use a password so long that I could never ever remember it. Then, if I use the service again, I do the whole "I forgot my password" thing. For services I do use, I use strong passwords and change them every 6 months. Finally, expect to be hacked. Expect someone will steal your credit cards. Expect someone will steal your identity. What will you do once that happens? Have a plan in place. I talked to my bank and I know exactly what to do if my bank card is used by someone else. I talked to the Social Security administration and I have already gone through the paperwork required to claim someone stole my identity, just to see it work (they ended up showing that someone DID once use my SSN for a temp job in another state). So, you are doing better than most people. Most people freely share all their personal information and use passwords like "password" for every service. They use 1111 as their bank PIN. By simply being harder to hack than them, you are protecting yourself. 209.149.114.132 (talk) 17:15, 10 November 2015 (UTC)
 * Thanks for your contribution. It would be more helpful if you could provide references, as otherwise this is just WP:OR that you're asking us to trust on faith. SemanticMantis (talk) 17:17, 10 November 2015 (UTC)


 * Looks like you're using most of the best practices for internet security. Posting here won't make you any more of a target for anything. Most "hacks" and identity thefts work via botnets, i.e. they aren't after you, they're just trawling to see what comes up (things like port scanning). You will likely not be a specific target of attack unless you are a celebrity or very very rich, e.g. your threat model is probably not one of personal attack, though you may well be caught up in something like the Target security breach, but that affected people who used cards at brick-and-mortar stores too.
 * See also browser security. You might like the podcast Security Now, which discusses some of these issues. You don't mention what browser you're using, and advice will vary, but I like using NoScript with Firefox to help minimize my security risks while browsing. Also many people recommend setting the browser to use https by default, and possibly using the add-on HTTPS_Everywhere. SemanticMantis (talk) 17:17, 10 November 2015 (UTC)


 * Thanks folks. It seems that no longer can "forewarned is fore-armed" be relied upon. I was once travelling South from Scotland into Northern England when Pan Am 103 was exploded out of the sky killing 270 innocent folks (259 in the plane and 11 on the ground- -I think). And so it goes on. It missed us by a matter of minutes. You just don't know anymore. What a sad race of creatures we have become. But thanks again.Tenosynovitis (talk) 17:39, 10 November 2015 (UTC)

What's the longest a process has been waiting on a mainframe?
Obviously on an old OS that allows processes to get preempted indefinitely. 20.137.7.64 (talk) 17:22, 10 November 2015 (UTC)


 * I think you're going to discover that this is not a well-defined question: for example, are you including software that intentionally runs on a schedule? Surely, somewhere somebody has a software system that schedules work for many years in the future.  Will you consider that to be "the same" process, if it has ever been virtualized and migrated to new hardware?  Are you only counting work that has actually persisted on a single incarnation of a single piece of computer hardware?  Is it "the same" process, in the technical sense, if the computer's system software has assigned the same work-load to a new process?
 * We can try to upper-bound this problem by seeking out the system with the largest uptime; but as you dive deep into the gory details of historical computers and installed infrastructure with complicate schemes for redundancy, you may even discover that "uptime" is not as well-defined as you think it is.
 * Nimur (talk) 17:43, 10 November 2015 (UTC)


 * The longest a process has been waiting on a mainframe would be Ada Lovelace's program to calculate Bernoulli numbers on the Analytical Engine. Program written in 1843, still waiting on the Analytical Engine being built. They built a Difference Engine in the 1990s, so all hope is not lost. Some would say that the program at [ https://www.fourmilab.ch/babbage/contents.html ] meets the criteria. See [ https://www.fourmilab.ch/babbage/authentic.html ].


 * (Note clever use of the fact that "waiting on a mainframe" can be read two ways...) --Guy Macon (talk) 18:20, 10 November 2015 (UTC)


 * I once optimized a mainframe program that was originally a batch job, since it took so long to complete. I got the time down to where it was fast enough to process interactively.  When I showed it to a user, he wrote down the starting time, grabbed his coffee cup, stood up to go, then was shocked to see the results immediately pop up.  But then his shock turned to disappointment, as he realized he'd have no more "bonus coffee breaks" in his day. StuRat (talk) 19:28, 10 November 2015 (UTC)


 * I found the story I was thinking of in my copy of Operating System Concepts, Sixth Edition, by Silberschatz, Galvin, and Gagne, page 162, though it starts with "Rumor has it..." Rumor has it that, when they shut down the IBM 7094 at MIT in 1973, they found a low-priority process that had been submitted in 1967 and had not yet been run. 20.137.7.64 (talk) 19:56, 10 November 2015 (UTC)
 * I don't think I believe it...http://www.cs.tau.ac.il/~nachumd/horror.html (item number 85) relates the story in the exact words you quoted - but let's look at the whole thing: "In a heavily loaded computer system, a steady stream of high-priority processes can prevent a low-priority process from ever getting resources. Generally, one of two things will happen. Either the process will eventually be run (at 2 A.M. Sunday, when the system is finally lightly loaded), or the computer system will eventually crash and lose all unfinished low-priority processes.... Rumor has it that, when they shut down the IBM 7094 at MIT in 1973, they found a low-priority process that had been submitted in 1967 and had not yet been run. From Silbershatz and Galvin, pp. 142-143."...so "either the process will eventually be run...or the computer system will eventually crash and lose all unfinished low priority processes". So for this to be true, we're expected to believe that the computer never once crashed from 1967 until 1973.  That would be fairly remarkable - but not impossible.  Now, let's look at this: http://www.multicians.org/thvv/7094.html - another remeniscience of that same machine.  It recalls that the computer was used for Multics operating system development - and had switched OS's several times over the years.  There is NO WAY for an old task to survive an OS switch.  The story simply isn't credible. SteveBaker (talk) 23:19, 10 November 2015 (UTC)


 * Not necessarily. Multics had dynamic reconfiguration, and you could do things like adding and removing CPUs, memory, and I/O controllers while the system was running. Not the last CPU or last bit of memory, of course. That would require magic. Only if you upgraded the hardcore supervisor did you have to kill all processes and restart the OS, but big chunks of Multics could be replaced with newer versions by leaving old modules on disk when installing new versions. New processes would use the new modules, while processes that had already linked to the old versions would continue to use those.


 * This is mentioned in [ http://queue.acm.org/detail.cfm?id=1640399 ]:


 * "The real continuous approach comes from Multics, the machine that was never supposed to shut down and that used controlled, transparent change. The developers understood that the only constant is change and that migration for hardware, software, and function during system operation is necessary. Therefore, the ability to change was designed in from the very beginning."


 * Another common use of this method was running production and development as subsystems on the same machine. You could (usually, unless you needed to upgrade the hardcore supervisor) intall and test several new versions on development without disturbing the production side. And now we have gone full circle: you can do the same with virtualization on a modern PC.


 * On the other hand, if you look at [ http://www.multicians.org/mult-sched.html ] it is hard to see how a process could be starved forever. I also find it hard to believe that anyone would tolerate an overloaded Multics system for long. So I think it's possible but I don't buy it actually being true.--Guy Macon (talk) 10:31, 11 November 2015 (UTC)


 * It's inconceivable that over years of operating-system development, they never crashed the machine or were forced to reboot it. I might buy that an uber-reliable machine with a reconfigurable operating system might continue to run that long - but not if people were using it to develop the OS and had to test buggy kernel modules.  Clearly this statement is meant as a kind of joke - based on the supposed reliability of OS and hardware and on the fact that the machine was so often heavily loaded...but it's certainly not plausible for a machine that was used to develop the very OS it was running. SteveBaker (talk) 19:12, 11 November 2015 (UTC)


 * As far as I can tell, MIT's IBM 7094 ran only CTSS. It was used to develop Multics, not run it. The "process" in question might have been a job of some sort that survived reboots. Still, the rumor is most likely a joke.
 * Regardless, the idea of processes that can survive OS upgrades is perfectly reasonable, even if it's not supported by any widely used OS today. It's a type of process migration. -- BenRG (talk) 22:09, 11 November 2015 (UTC)
 * The authors definitely infused humor into the book at points. From section 1.4, Operating System Structure, speaking of multiprogramming, "As long as at least one job needs to execute, the CPU is never idle. This idea is common in other life situations. A lawyer does not work on only one client at a time, for example. While one case is waiting to go to trial or have papers typed, the lawyer can work on another case. If he has enough clients, the lawyer will never be idle for lack of work. (Idle lawyers tend to become politicians, so there is a certain social value in keeping lawyers busy.)" 75.75.42.89 (talk) 03:23, 12 November 2015 (UTC)

SQL parametrisation question
It turns out that I have to maintain yet another project at work. The project in question is a tool that can make SQL queries to different kinds of databases, such as Oracle, MySQL and Microsoft SQL. There already seems to be an abstraction of all these in C#, so all I have to do is to find out the specific database type and construct a database connection based on it, and then I can access all of them through a unified interface as I please. But how do I do parametrised queries? The ways to mark the parameters in the queries can differ. For example, Oracle seems to use the colon (":") to mark a parameter, but the other databases might not necessarily do. Is there some unified way which would work for all of these, or will I have to resort to a hack to find the parameter character in the query and change it depending on the database? J I P &#124; Talk 20:16, 10 November 2015 (UTC)


 * Does the tool have you make a selection as to which kind of database you intend to query? FrameDrag (talk) 13:23, 11 November 2015 (UTC)


 * Yes, it's set in a configuration file. J I P  &#124; Talk 16:49, 11 November 2015 (UTC)


 * In that case, I think the best course would be to use some other delimiting character and then, depending on that config option, convert it to the proper delimiter. There's not really a "unified" way to do it since they're all oh-so-slightly different dialects. You can make it appear unified, but behind-the-scenes the logic has to be there to interpret your unified notation to Oracle/MySQL/SQL Server.


 * The other option is to assume that the user knows the parameter delimiting character, but that presumes certain things about the knowledge level of the user, and depends on whether or not you're willing/able to reveal which type of database they're connected to. Though, with this option, you don't have to add a large amount of conversion logic. FrameDrag (talk) 19:57, 11 November 2015 (UTC)


 * I like the first option better, as it makes transition to a new database system far easier. StuRat (talk) 20:59, 11 November 2015 (UTC)


 * OK, so now I need to know what are the parameter delimiter characters for MS SQL, MySQL, PostgreSQL and ODBC. I already know Oracle uses the colon (":"). Then I'm all set for implementing User:FrameDrag's idea. J I P  &#124; Talk 07:33, 12 November 2015 (UTC)

Detail(s) Handling software sought
Is there software available like MoneyManagerEX that can be synchronised with PC to phone and from phone to PC, for ‘Customer/Client/Company details’ handlings? An open source software is sought. -- Space Ghost (talk) 20:25, 10 November 2015 (UTC)