Wikipedia:Reference desk/Archives/Computing/2011 March 31

= March 31 =

Weird Xcode Problem
I'm trying to build an application in Xcode, but every time I compile it, it gives me this error: Error from Debugger: The program being debugged is not being run. So, I navigate to it in the finder, and it has a little prohibitory sign. Double clicking it, it says that the application can't be opened because it is damaged or incomplete. However, I can get it to open by opening the executable in the Package Contents>MacOS folder. Could anyone explain what is going on? --Thekmc (Leave me a message) 00:28, 31 March 2011 (UTC)


 * Sounds like your info.plist file is missing or corrupt. --Carnildo (talk) 23:40, 31 March 2011 (UTC)

How would I fix that? --Thekmc (Leave me a message) 21:15, 1 April 2011 (UTC)

Microsoft Windows Server 2008 R2 Datacenter Edition - Temporary Licensing Non-Compliance
Hello RefDeskers! If, on an eight-socket server with four installed processors running Microsoft Windows Server 2008 R2 Datacenter Edition, fifth and sixth processors were to be installed several days before two extra per-processor licenses are acquired and assigned, thereby causing license non-compliance, would the server continue to function normally or would the server stop functioning due to license non-compliance? I tried asking at Microsoft TechNet, but did not get an answer to the question, and was instead schooled on the importance of license compliance. Thanks as always.


 * There is generally a grace period for license activation, usually 30 days. Server 2008 seems to have one for retail/oem versions . Our article on Client Access Licenses says that only two products currently use technical means to enforce licensing: SBS2008 and Terminal Services, but that is uncited. If you use volume licensing, the best thing to do might be to speak to your reseller or account manager, as they would be able to give you a definitive answer. -- Kateshortforbob talk  09:13, 31 March 2011 (UTC)

Upgrading to Firefox 4 from Firefox 3.6.16
Should I save any of my bookmarks, history etc before upgrading? Will the same add-ons be installed automatically? Thanks 92.29.119.112 (talk) 09:55, 31 March 2011 (UTC)


 * I had no issues with bookmarks and history, but you should definitely keep them backed up if they are important to you (not just when you upgrade). The add-ons should remain in place if they are compatible with Firefox 4; if not, they will be disabled until new versions are available. 130.88.134.221 (talk) 15:51, 31 March 2011 (UTC)

Firefox 4
Now that I've installed it, I have a few annoying minor problems. First, I keep having to log into Wikipedia each time. Second, where did the refresh, undo, redo, etc. go, and how can I get them back? Clarityfiend (talk) 18:04, 31 March 2011 (UTC)


 * Refresh is the little circle-arrow button at the right end of the URL bar. It changes to the stop (X) button when a page is loading. I don't know what you mean by undo/redo : if you mean the back/forward arrows, I still have them to the left of the url bar. -- Finlay McWalter ☻ Talk 18:09, 31 March 2011 (UTC)


 * All of that stuff is on the "navigation toolbar", which you may (now) have turned off. -- Finlay McWalter ☻ Talk 18:10, 31 March 2011 (UTC)


 * Undo/redo let you remove/restore your last edit or edits. They used to be on a menu, but the menus seem to have all gone the way of the dodo. Clarityfiend (talk) 19:07, 31 March 2011 (UTC)


 * Oh, I see what you mean. Those are still in the EDIT menu: you need to turn on the menubar to get it back.  I'd have thought you should be able to get icons back for them from the customise option, but I can't immediately figure out how. -- Finlay McWalter ☻ Talk 19:34, 31 March 2011 (UTC)
 * Aha. I've got my menu bar back. Thanks. Clarityfiend (talk) 03:45, 1 April 2011 (UTC)

1) Make sure you're not in private browsing mode 2) ALT+v, Toolbars > Customize ¦ Reisio (talk) 00:20, 1 April 2011 (UTC)

Algorithm used in Mega Millions POS computers
I just saw a news story stating how the buyer of the ticket that recently won $319 million (before taxes) had someone cut in line and he opined that if this hadn't taken place, the other guy might have gotten the winning number. But that's assuming that the algorithm the machine uses is sequential in that the machine is seeded maybe upon startup and the time span between ticket purchases doesn't matter, i.e., the next number that comes out will be the same (even if nobody knows what it will be) whether you wait 25 ms or 3 hours. If upon each purchase the machine took a random seed like the time from the system clock, then not cutting in line almost certainly wouldn't have made the cutter get the winning number and also would have made the guy who did get the winning ticket get to the counter earlier than he did, almost certainly totally changing his numbers too. Is there any public specifications of those machines, or is that top secret information? Just curious. 76.27.175.80 (talk) 22:42, 31 March 2011 (UTC)


 * I don't know specifically which vendor(s) supply the Mega Millions lottery equipment, but some lottery system vendors use the "Trusted Draw" system, which is made by Szrek2Systems. This white paper has some (scant) details. It seems they take a realtime clock (which I guess is in the crypto card in the lottery terminal), cons that with the user data (I guess the terminal ID and stuff like that) and take a Digital Signature of that. They don't say specifically which DS scheme they use, but they say the signatures have the "cryptographic strength of 1024 bit RSA digital signatures" (there are several RSA-based DS schemes, including RSASSA-PSS and RSASSA-PKCS1-v1_5).  Assuming things are implemented properly, and the clock ticks quickly (which surely it will), and that the random number is generated when the "print ticket" button is pushed, then pushing the button even a fraction of a second sooner or later would result in entirely different numbers.   This seems like a pragmatic scheme (it's tempting to say they should use an analog random-number generator, but that gives them problems with later auditing). The simple scheme you suggest (seed a PRNG and then just print off tickets deterministically) would be very vulnerable to all kinds of abuse. -- Finlay McWalter ☻ Talk 23:16, 31 March 2011 (UTC)