Jump to content

Wikipedia:Reference desk/Archives/Computing/2009 March 27

From Wikipedia, the free encyclopedia
Computing desk
< March 26 << Feb | March | Apr >> March 28 >
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.


March 27

[edit]

A way to guard against identity theft?

[edit]

Reading through Template talk:User committed identity, I realized it isn't what I was looking for. Namely, I would like to post on Wikipedia and elsewhere using my real-life name. However, what stops me from doing so is the risk that someone could impersonate me elsewhere on the Internet using my name and post ugly things under that name. Then any time someone googles my name, they would find those posts.

Are there any schemes that would ensure that postings using my name can be independently verified that they come from me and no one else?

My initial very crude idea would be a website where you could download a "scratch pad" with a list of one-time codes, similar to the transaction numbers used in online banking. Every code could be used only once. For every post I make anywhere on the Web, I would copy one code from the list into the post and then strike it off my list. A reader could then click on the code, which would lead to the repository and allow verification.

It sounds cumbersome and it is. My question is, would it be workable at all? Has it been implemented anywhere yet? And is there a less cumbersome method to accomplish the same goal?

There are other concerns besides identity theft when posting under a real-life name. For example, verification would not protect against stalkers harrassing me in real life, but I would be wiling to take that chance.--Goodmorningworld (talk) 02:35, 27 March 2009 (UTC)[reply]

I believe you are looking for digital signatures, which are available in many public-key cryptography applications. – 74  03:58, 27 March 2009 (UTC)[reply]
Thank you. Here is the first paragraph from public-key cryptography:
Public-key cryptography is a method for secret communication between two parties without requiring an initial exchange of secret keys. It can also be used to create digital signatures. Public key cryptography is a fundamental and widely used technology around the world, and enables secure transmission of information on the Internet.
I'm not sure this helps me as I wasn't looking for "secret communication between two parties" but a way to post anywhere in a public place – message board, discussion forum, Wikipedia – such that anyone reading my posting can easily verify that it came from me.--Goodmorningworld (talk) 04:14, 27 March 2009 (UTC)[reply]
Digital signatures can do what you want, but it might be a lot of effort, and people who wanted to verify your posts would have to install the relevant software to do so. A lower-tech solution would be simply to maintain (on you Wikipedia user page, say, or some other personal page more under your control) a list of all the websites on which you post and the usernames you use, along with an explicit disclaimer of any posts made under your name on any other site. Algebraist 04:42, 27 March 2009 (UTC)[reply]
Thank you. I just realized a major flaw in my idea. A reader could not know with certainty that a posting made under my name but without the special authentication code was in fact not made by me; perhaps I left out the code on purpose. In other words, even if I can ensure that no one but me can add an authentication traceabe only to me, I haven't solved the issue of getting the public to trust that I would never make postings without it. Of course this problem applies also to your otherwise useful low-tech idea. Hmmm…--Goodmorningworld (talk) 05:15, 27 March 2009 (UTC)[reply]
I don't see how and why it matters that people can't verify posts made under your name but without your signature weren't made by you any more then they can know whether a post made under Nil Einne was actually made by you. People who trust you, will only trust posts made by you and signed. If you forget to sign, they will presume that it's not you or ask you in which case you can clarify with a signed post if you forgot. People will grow to trust your online identity if they feel there are reasons to trust it. They're obviously not going to presume they definitely know everything about you from your online identity just as knowing someone in real life doesn't mean you know everything about them, haven't you heard how many times people describe how criminals "seemed such a nice person". Nil Einne (talk) 11:07, 28 March 2009 (UTC)[reply]
Just be anonymous —Preceding unsigned comment added by 194.80.240.66 (talk) 10:55, 27 March 2009 (UTC)[reply]
My true name is widely scattered on the Net since about 1990. In 1996 when I went out of my way to annoy Krazy Kevin he put my address on a batch of spam, so I got four complaints. That's about it. —Tamfang (talk) 18:40, 28 March 2009 (UTC)[reply]
Somebody who thinks badly of you because they saw your name attached to something ugly on the internet, is not going to bother checking any kind of technical authentication scheme. The scheme will be ignored no matter how secure it is. 75.62.6.87 (talk) 04:58, 1 April 2009 (UTC)[reply]

Nag, nag, nag

[edit]

My Vista PC wants me to create some recovery disks. Should I use DVD-R or DVD+R or does it not really matter? Clarityfiend (talk) 04:51, 27 March 2009 (UTC)[reply]

The short answer is: "It doesn't matter". At one time +R and -R were not always available to all DVD writers, but today most (if not all) DVD writers will accept either disk. I would suggest getting decent quality DVD blanks for this particular chore though, rather than the cheapies. You never know when you may need to use them to restore a computer, or replace a HD - also, some PC's only allow 1 backup copy to be made. — Ched ~ (yes?)/© 07:34, 27 March 2009 (UTC)[reply]
Using Clonezilla, make as many backups as you want, and whenever, from wherever, and to wherever you want. The price is unobjectionable too. -- Hoary (talk) 07:48, 28 March 2009 (UTC)[reply]
I'd personally recommend clonezilla to clone your drive after a clean install. Once you do that, you always have a quick and easy process to revert to "start" without going though the hassle of doing a full install again. ---J.S (T/C/WRE) 15:49, 28 March 2009 (UTC)[reply]
I usually use R-Drive image to make "images"/backups of my system 85.241.115.243 (talk) 03:07, 29 March 2009 (UTC)[reply]
Our DVD+R and DVD-R articles seem to be more enthusiastic about DVD+R for alleged reliability reasons. Tempshill (talk) 20:48, 30 March 2009 (UTC)[reply]

JDBC

[edit]

How do you set up JDBC to work with SQL Server 2008? Black Carrot (talk) 06:17, 27 March 2009 (UTC)[reply]

ISP

[edit]

Hiya , I'm looking for a good isp to go with in the uk. i currently do not have the internet at home, or a bt landline phone. I've been looking at Tiscali broadband and it seems to offer what i want, especially their fair use policy of 100GB a month.

My question is, for a heavy internet downloader such as myself, what is the best isp to choose in england? Is it Tiscali? I don't want cable television or phone calls, just the internet. Thanks for your help and any suggestions you may have for me. —Preceding unsigned comment added by 194.80.240.66 (talk) 09:08, 27 March 2009 (UTC)[reply]

I use 'Freedom2surf' and find they are good, they offer it without a 12-month contract and the prices are quite reasonable: around £16 for unlimited 'up to 16mb' downloads (1mb uploads) or around £23 for their '50kbps download guaranteed, up to 16mb' product. Both of which state usage is 'unlimited'. Personally From a really quick scan I don't see anything in the T&Cs about a cap on that 'unlimited' but in reality i'd be amazed if there isn't. Honestly though 100gb? Do you just spend the entire month seeding/leeching on Torrent files? It always amazes me - what do people do that uses that much per month! I feel like a heavy internet user and I get through maybe 20gb a month. 194.221.133.226 (talk) 11:01, 27 March 2009 (UTC)[reply]

I really do need 100GB a month. In one day alone I've downloaded that much, but that was on a public computer with an extremely fast connection. Since I'm not going to have access to that computer any more, I need an alternative home internet solution. The 100GB limit was the highest I could find in my searches, but any less will simply not be enough. Download speed is not so much of an issue as I can leave the computer on overnight. —Preceding unsigned comment added by 194.80.240.66 (talk) 12:00, 27 March 2009 (UTC)[reply]

Out of interest...what do you download? I just can't imagine downloading that much stuff and actually ever using the stuff. Anyhoo as I noted Freedom2Surf don't offer an upper-limit on their account, you could probably contact them to see whether or not they have a 'fair use' policy. To be honest the amount of bandwidth you're using i'd expect is getting towards business-lines usage from a ISPs perspective. 194.221.133.226 (talk) 12:07, 27 March 2009 (UTC)[reply]

100GB a day? That seems almost implausible to me from a bandwidth standpoint of any public computer I've used (lets say you get the full 24 hours -- you're really getting a full 1MBps consistently? Much less in a less amount of time?). Are you sure you have your MB and GB right? --98.217.14.211 (talk) 16:01, 27 March 2009 (UTC)[reply]
I've downloaded 70GB in a day, mostly from megaupload. But that was a one off and actually got me banned for a week :( Also I don't think OP wants 100GB a day, he was just giving that as an example. —Preceding unsigned comment added by 82.43.88.87 (talk) 18:51, 27 March 2009 (UTC)[reply]
Without going in to legalities and practicalities, I can easily see 100GB/day if you download several HD movies. Actually if you are talking about entire Bluray disks unrecompressed etc it may only be 3 or so. 100GB/month is definitely resonable without even going in to HD movies. Presuming you are referring to P2P and you are a fair downloader, that means only 50GB of content. Less when you consider overhead as well as the fact your likely to have general web browsing, you tube videos and perhaps also streaming TV, windows updates, driver updates et al. Let alone if you are a generous uploader (i.e. upload more then you download). 50GB/month isn't really that much particularly if there are multiple people involved. It may not be every month but having the option is likely to be useful Nil Einne (talk) 10:10, 28 March 2009 (UTC)[reply]

So does anyone have some suggestions of ISPs in England that would be good for me?? 194.80.240.66 (talk) 12:38, 28 March 2009 (UTC)[reply]

There is no reason for an abusive edit summary. The first volunteer who responded gave you a positive suggestion. BrainyBabe (talk) 12:53, 28 March 2009 (UTC)[reply]
No you misunderstand, that summary was meant for signbot who immediately jumped in before i had a chance to correct my mistake. Although I should not have cursed anyway, so I'm sorry for that. I'm thankful for the replies so far but i really would appreciate some sort of list of good isps to look through. Thanks for the help anyway [edit] See he did it again http://en.wikipedia.org/w/index.php?title=Wikipedia%3AReference_desk%2FComputing&diff=280220150&oldid=280220131 I've already ask him several times to stop and he won't. Anyway, again thanks for the replies 194.80.240.66 (talk) 14:53, 28 March 2009 (UTC)[reply]
Signbot is a bot and does not understand your edit summaries. ---J.S (T/C/WRE) 15:47, 28 March 2009 (UTC)[reply]

Be might work for you. They have a fair usage limit but they're very cheap for their speeds. It MIGHT not be avalible in your area but you can check on their site. Also I have a friend who works for them and has told me they have hardly ever acted on the fair usage thing and they don't block ports or anything like that. Gunrun (talk) 10:44, 30 March 2009 (UTC)[reply]

Black bar appearing either side of you tube videos!

[edit]

I seem to be getting not one but two black bars, either side my youtube videos. I assume one is by youtube the other by windows movie maker, but how can I avoid one (or both!) of these black bars. When saving it with wmv I clicked Widescreen which said suitable for 16:9 which in turn is suitable for youtube (or so I was told)... I'd be really grateful for any help on the matter! ; - ) --217.227.121.117 (talk) 18:15, 27 March 2009 (UTC)[reply]

Youtube have recently (November, according to YouTube) changed the size of their video player to be in 16:9 format, rather than 4:3. The black bars allow 4:3 videos to be played in the 16:9-size player. Use 16:9 to avoid black bars. --h2g2bob (talk) 02:02, 28 March 2009 (UTC)[reply]

SQL optimisation

[edit]

If I want to run a count, is "select count(col1) from table2 where col2='x'" the quickest way of doing things? I've now got access to a very large read-only table, and this seems to be grinding to a halt. Thanks! - Jarry1250 (t, c) 18:23, 27 March 2009 (UTC)[reply]

If there is no specific need to COUNT col1, you can COUNT(1) and that helps things a little bit (it doesn't have to retrieve any data for that, and assuming you are not worried about NULL fields it will count exactly the same way as putting a variable in there). But other than I think your options are limited unless you can do tweaks to the database itself, e.g. having indices of the right length and etc. --140.247.240.77 (talk) 20:16, 27 March 2009 (UTC)[reply]
I agree with the above comments, provided you are only doing the single query shown. However, if you are also planning on doing a count for col2='y' and col2='z' and col2 = every other value present, then there are much faster ways to do that. StuRat (talk) 03:20, 28 March 2009 (UTC)[reply]
As .77 mentioned, an index on col2 is the first tool of efficiency. Not only does it quickly identify what records need to be accessed (eliminating the need to scan the hole table), but since the only column you are referencing is already in the index, the index functions as a covering index, and there is no need to access the main table at all. As an analogy, suppose you were given a list of 1,000,000 phone numbers and had to find out how many were owned by households with last name of Wales. That would be very time consuming. Now what if you had a phone book? Looking up Wales would allow you to significantly trim your list. Better yet, you don't even need to call any of the numbers because the phone book already tells you the household name. That's what a covering index gives you. If you wanted to count the number of Wales households that lived on Wikipedia lane, you would want a phone book (index) that had both name and address.
The other thing you can do is to ignore all database locks when executing the query. Do this only if you are not concerned with transient records that may be added, updated, or removed while the count is being performed. The syntax may be different for different database systems. For Microsoft SQL server, you can either execute the command "SET TRANSACTION ISOLATION LEVEL READ UNCOMMITTED" before the query, or you can add "WITH(NOLOCK)" after the table name. Normally, database management systems are very careful to use various locking mechanisms to ensure the accuracy and consistency results. This generally means locking the records or tables being accessed so that nothing changes in the middle of a query. If some other process has a lock (such as an uncommitted insert) your query may wait. All this has some degree of lock-management overhead and causes delays (and in some cases deadlocks) when other processes are manipulating the same data. Disabling locking effectively says "I don't care what else is going on, just give me what you have as fast as you can, even if it's changing." It also prevents you from unnecessarily delaying other processes. For example, if you actually ended up calling 1,000,000 phone numbers, it would be unreasonable to demand that nobody change their phone number during your three-year survey.
One last note: Make sure that the data types are consistent in your query. For example if col2 is an integer and you search for the text string "42", the database server may have to convert every value checked to a string before comparing, making the indexes virtually useless. Better to search for the number 42. -- Tcncv (talk) 03:53, 28 March 2009 (UTC)[reply]
Also, if the number of such count operations greatly outnumbers the number of writes to the table, you could consider using database triggers to keep a running total of the counts. --Sean 21:51, 28 March 2009 (UTC)[reply]

.net on mac

[edit]

Will windows programs written in the .net framework run on a mac with Mono installed? —Preceding unsigned comment added by 82.43.88.87 (talk) 18:48, 27 March 2009 (UTC)[reply]

No takers? Oh well. I had a read of the mono forums and it seems exe's might work on macs but they'll be buggy as hell. —Preceding unsigned comment added by 82.43.88.87 (talk) 18:44, 28 March 2009 (UTC)[reply]

Electronics: Logic processing: How to feed D type flip flops from parallel ...

[edit]

I am creating a complex logic system, part of that system takes an 8 bit boolean value stored in 8 D type flip flops.

The Q gate of each flip flop is linked to an AND gate. Each of the eight AND gates are also linked (in parallel) to another trigger (which i will refer to as a button), the AND gates output into the 'set' inputs of another 8 D types, which are aranged to count down.

The system is designed to take the 8 bit number, move it to the seperate D flip flops, and subtract another number from it, as the flip flops are aranged to count down, and then replace the origional 8 bit figure in the first 8 flip flops.

The flip flops count down as they are aranged as displayed in this image: http://www.softwareforeducation.com/electronics/notes/AS/ttl/binary-down-counter.gif

However, when I set the second set of flip flops, the set causes pin Q to be taken high in relevent flip flops, which is counted as a clock pulse by the next flip flop in the counting set, causing unwanted bit generaton.

I cannot fathom a way in which to stop this happening and retain the ability to reduce the value stored in the second set. —Preceding unsigned comment added by 82.45.1.217 (talk) 20:34, 27 March 2009 (UTC)[reply]

I'm no expert and it's been a while since I looked at this stuff, but I can't think of a way to do this using D-type flip flops. However, I think you can make a working circuit using JK flip-flops. This article show how they can be wired up to form the count-up circuit. Count down would be similar. You can wire up additional logic to transfer data between the count-up and count-down circuits using the S and R inputs. (Extra credit if you can figure out how to wire a single set of JK flip-flops to perform both the count-up and count-down functions.) -- Tcncv (talk) 04:29, 28 March 2009 (UTC)[reply]

The following 2 responses were originally added to the WP:RD/S and were moved here as the question was cross posted. Nil Einne (talk) 10:01, 28 March 2009 (UTC)[reply]

I am not implying this is at all nonsense, but this is an awesome example of jargonese, so extreme, that to some people like me, it's impossible to tell whether it's nonsense or not, and that's why I'm copying this and going to use it as an example, and yes, I know this is a run on sentence. 173.52.36.16 (talk) 21:27, 27 March 2009 (UTC)[reply]
I have only studied electronics at a quite elementary level, and I recognize all phrases used, so it is not particularly extreme. Also, that is not a run-on sentence. However, if you had removed any "and" after the commas, then it had been a run-on sentence. --Andreas Rejbrand (talk) 10:45, 28 March 2009 (UTC)[reply]

(The terms that the OP uses are comprehensible to any electronic designer.) You mention 16 flip-flops and you show a diagram of 3 flip-flops connected as a ripple-through binary counter. In the counter the S and R (set and reset) inputs can overide the other inputs (D and CLK) so you can force the counter to any state you want by controlling all the S and R's You will need some gates and inverters to drive high only the right S and R's. Catalogs of standard logic integrated circuits are a source of diagrams of many varieties of presettable counters, which seems to be what you want to construct. Cuddlyable3 (talk) 21:46, 27 March 2009 (UTC)[reply]

I have removed your post to the science desk. Please do not post the same question in two different places. If you are unsure which desk is the best, just choose one at random or whatever Nil Einne (talk) 10:01, 28 March 2009 (UTC)[reply]