Wikipedia:Reference desk/Archives/Computing/2014 April 13

= April 13 =

Importance of passwords for online forum accounts
Currently my password for all online forums is the same because I don't see the benefit for someone to overtake my account on the random forums I frequent. The only motivation I see is for spammers to take advantage of my account to post links but does that actually happen? --129.215.47.59 (talk) 03:13, 13 April 2014 (UTC)


 * Yes it does actually happen. 71.3.50.250 (talk) 03:19, 13 April 2014 (UTC)


 * If this is about the Heartbleed bug, keep in mind that most web forums don't use SSL to begin with, and they often aren't careful with passwords in general. The main danger is using the same password on a forum as on some more sensitive site (financial etc) and a secondary danger is being impersonated on multiple forums if just one forum leaks the password.  I think the era of being able to remember passwords is over.  Best practice is to use a different random password for each site, with password manager software (some is built into browsers these days) to remember them.  It's more important for sensitive sites, but once you're used the software it becomes easy to use it on everything. 70.36.142.114 (talk) 04:48, 13 April 2014 (UTC)


 * But if somehow you lose access to that password manager, then you are in serious trouble. I think writing the passwords down and storing them somewhere safe is a good idea.  (Even if somebody breaks into your home, they aren't likely to be the same sort of people who engage in cybercrime.) StuRat (talk) 13:44, 13 April 2014 (UTC)


 * That and you can hide the slip/sheet of paper that your passwords are on in a book which is not a likely target for thieves. Dismas |(talk) 14:46, 13 April 2014 (UTC)


 * Most web forums and similar sites have a "reset password" function that sends a link to your registered email address. A consequence is that your passwords on those sites are all potentially compromised if your email account is compromised.  This means you should keep your email especially secure. 70.36.142.114 (talk) 23:46, 14 April 2014 (UTC)


 * On the note of writing passwords down. If you are going to do that, and I would suggest highly you don't and just use a password manager instead, I would store them in your wallet.  Storing written passwords in your wallet isn't a bad idea because you generally keep an eye on it, and if someone steals it you already know you have some work ahead of you. Zell Faze (talk) 03:26, 16 April 2014 (UTC)


 * But wallets are frequently lost or stolen. If you do write down your passwords, try to leave part of it out, like you could have a word you add to the end of each PW that's the same for every one, but never write that word down.  Hopefully you can always remember that word. StuRat (talk) 16:13, 16 April 2014 (UTC)

Scatter plot in Python
Let's say I have 10 different hashtags that I want to plot on the x-axis (no particular order) with their average sentiment score on the y-axis, and the area of each circle of the plots have to be directly proportional to the number of tweets that particular hashtag appears in. For example, the variables look something like this: Can I do this in matplotlib? If not then which package do I need to get? I've been trying, but I can't seem to get the hang of coding graphs and plots in Python. Any hekp will be much appreciated. La  Al qu im  is ta  04:29, 13 April 2014 (UTC)


 * I've never used matplotlib but have used Gnuplot for stuff like that. 70.36.142.114 (talk) 04:50, 13 April 2014 (UTC)


 * Note that the circle size introduces a potential problem, that the circles could overlap. StuRat (talk) 13:41, 13 April 2014 (UTC)


 * The matplotlib gallery has two examples very similar to what you ask. (The stateful pyplot interface is a horror; ask if you want to do it better but don't know how.)  --Tardis (talk) 03:54, 15 April 2014 (UTC)

What does the government does, when it blocks a site?
Like in the case of Turkey recently, that blocked twitter. OsmanRF34 (talk) 14:05, 13 April 2014 (UTC)


 * First they ordered Turkish ISPs to put entries into their DNS servers to redirect name lookups for twitter.com (etc.). So people started using Google Public DNS instead of their ISPs' ones. So they made the ISPs block that (by blocking its IPs), and then made the ISPs block twitter's own IP range. Folks started to use Tor (10,000 new users a day, apparently). Tor itself is hard to block, but they did block access to the Tor project's website to make it (somewhat) harder to download the Tor client. Princess Leia could have told them it would go like this. -- Finlay McWalterᚠTalk 14:29, 13 April 2014 (UTC)


 * Folks also used VPNs, which similarly experienced massive uptake of services following the block. -- Finlay McWalterᚠTalk 14:36, 13 April 2014 (UTC)


 * In other words, the Streisand effect is playing out: attempts to block internet traffic actually increased the amount of traffic to Twitter.
 * To be honest, I don't see why overbearing censors haven't wised up yet; we have over a century of scientific research on crowd psychology and mass group dynamics. Instead of blocking the traffic outright, an efficacious  censor would maliciously abuse Quality of Service to make certain internet traffic functional but unpleasantly slow.  The resulting effect would be a total lack of sensationalism: it would be difficult to promulgate the claim that the government were censoring something (any user could easily verify otherwise).  The sensational claims of censorship would be less likely to "go viral."  The service would become unpleasantly sluggish for many individuals, and in the bulk aggregate, the crowd would voluntarily reduce their usage of the service.  All the while, any individual would probably not be sufficiently motivated to work around the problem.
 * This is why crowd control barriers are often less effective against a crowd than a handful of sparsely-placed bollards, through which any individual can easily walk; but a crowd as a whole chooses not to concentrate there, because there is a perceived but not actual barrier.
 * (Conspiracy theory: this is why it is so easy to get internet news, but so difficult to get transcripts of floor proceedings or other useful information pertaining to the United States Government. For example, here's a direct link to the United States Federal Budget: Financial Disclosure Reports, part of the Open Government initiative, a news-service so factual and useful and boring that only our most disaffected citizens would read anything it publishes!  I guarantee you won't find it referenced on television or major internet news websites - even when they report news on these exact topics - but it sure as heck doesn't even need to be censored)!
 * Nimur (talk) 18:24, 13 April 2014 (UTC)


 * This article argues that banning social media was done for a quite different purpose, political not practical: "It is a strategy of placing social media outside the sacred sphere, as a disruption of family, as a threat to unity, as an outside blade tearing at the fabric of society." If it is right, the mechanism (and effectiveness) of the block are almost irrelevant. --ColinFine (talk) 17:44, 14 April 2014 (UTC)

internet advertizing
How do i find the best and most cost effective means of advertizing on the internet ? — Preceding unsigned comment added by 187.64.102.199 (talk) 15:11, 13 April 2014 (UTC)


 * That's a huge topic. You might want to read internet advertising (note spelling) and follow some of the many links from there.--Shantavira|feed me 15:16, 13 April 2014 (UTC)


 * Internet advertizing should certainly exist as a redirect, and I have just created it. --ColinFine (talk) 17:48, 14 April 2014 (UTC)


 * That is indeed a huge topic and this probably isn't the best place for an answer since there are so many factors involved including what you're advertising, what your target demographic is, where you are, where your product/service is available, etc. Dismas |(talk) 15:21, 13 April 2014 (UTC)


 * Target marketing is key here. If you are advertising tampons, you don't want to advertise to men, children, or post-menopausal women, as that will just waste your resources.  There are a couple ways to do this:


 * 1) Target the demographic. If you have access to info on user's gender and age, you can use that to advertise to only those in your target demographic.  This might also include things like income and education level, for other products.


 * 2) Target those who have expressed an interest. So, if somebody did a search for a product like yours recently, push ads to them.  If they've expressed an interest in your specific product, then sending them a coupon might be a good way to get them to make that purchase decision.


 * Also, make sure that if they click on your ad you aren't going to overload their computer with pointless autoplay movies, etc. Nothing annoys a consumer more than that.  And be sure to open a new window so they don't lose whatever was on that page.


 * Also, some sites allow you to scroll back through the old ads. This is very important, otherwise once the next ad pops up they can't get back to your ad.


 * And consider having some type of loyalty club, where they can sign up and get emails from you about discounts, new products, etc. Make sure they can choose exactly what info you will send them, and change it later, so they don't get overloaded with spam and decide to designate all mail from you as spam. StuRat (talk) 13:13, 14 April 2014 (UTC)

Tune played by Turbo Tax on launch
Out of curiosity, does anyone here remember which year Turbo Tax had that midi tune that played when one would launch it? I'm thinking it was the mid to late 90s or early 2000s. PCHS-NJROTC (Messages) 17:25, 13 April 2014 (UTC)

mac mini color problem
Hi. Anyone have any thoughts or links about this problem: I have a mac mini, it is connected to a HISENSE lcd television. The tv and the cable work just fine when connected to another computer, but when the mac mini is connected all black pixels wind up as dark purple instead. I tried googling this problem but did not come up with any solution. Any ideas? Thanks --Duomillia (talk) 23:11, 13 April 2014 (UTC)


 * Which model of Mac mini is it, and what kind of cable? Is there an adaptor on the cable? --Canley (talk) 03:20, 14 April 2014 (UTC)


 * Does the cable fit snugly into the Mac Mini ? When you wiggle it at the connection, do the colors change ? StuRat (talk) 13:00, 14 April 2014 (UTC)


 * Does the mac mini display properly on any monitor? In case it helps: I dropped a macbook once, and the built-in display had a similar problem: all perfect black came through as bright green, while every other color was fine. I was able to solve the problem by opening up the case and re-seating the ribbon connector that goes between the graphics board and monitor. You can probably use ifixit.com to identify what bits connect the graphics bits to the output port. This is basically the same thing as Stu's "wiggle the cord" bit, but inside the computer. SemanticMantis (talk) 14:58, 14 April 2014 (UTC)


 * It's a mac mini from 2013 running osx 10.9.2, the cable is an ordinary mid-range cable. No adapter I took the bottom off but there is no access to anything other than the memory cards without unscrewing things. Duomillia (talk) 02:00, 15 April 2014 (UTC)


 * Yes, Semantic Mantis is talking about unscrewing things, which brings with it the possibility of screwing things up. StuRat (talk) 13:50, 15 April 2014 (UTC)


 * Absolutely. In my case, I had one goal in mind, wiggling a connector, and I kept careful track of all the screws, grounded myself, etc. So I was fairly confident I wouldn't make anything worse. I later learned a large chain computer store would have charged me about $100 for the fix. OP can make up their own mind regarding risks/benefits. I do recommend they look into ifixit.com for their specific model, they have plenty of good pictures, and I strongly suspect the issue is the same. But I don't know how the graphics card is wired to the graphics output port, there may not be any physical connector to wiggle. SemanticMantis (talk) 15:36, 15 April 2014 (UTC)