Wikipedia:Reference desk/Archives/Computing/2009 July 2

= July 2 =

Boolean as tinyint(1)
Why does MySQL make columns specified as "Boolean" tinyint(1)s when bit(1) is available? Is there any way to have a group of Boolean columns stored internally as a bit field rather than giving each one its own byte? Neon Merlin  04:20, 2 July 2009 (UTC)

400 bad request
I get a 400 bad request while attempting to access YouTube with Google Chrome. Internet Explorer works fine for the site. Has anyone else noticed this? --Blue387 (talk) 05:56, 2 July 2009 (UTC)


 * Worked for me just now with Windows XP. Tempshill (talk) 06:29, 2 July 2009 (UTC)
 * Chrome works now. --Blue387 (talk) 03:56, 3 July 2009 (UTC)

New tab problems in IE8
Hi, whenever I open a new tab in Internet Explorer 8, I get a page that says

"The webpage cannot be displayed   Most likely cause: •Some content or files on this webpage require a program that you don't have installed.   What you can try:      Search online for a program you can use to view this web content.       Retype the address.       Go back to the previous page. "

and instead if "about:Tabs" being in the address bar, it says: "tbr:res?id=tabs&rep=1"

I'm running Windows XP SP3.

How can I fix this?

Thanks 144.138.21.100 (talk) 06:32, 2 July 2009 (UTC)


 * Do you have a default homepage set up? It may be that the default home page is not loading properly. ··· 日本穣 ? · 投稿  · Talk to Nihonjoe 07:33, 2 July 2009 (UTC)


 * Even if I had a default home page set up, it wouldn't open if I clicked new tab, as a new tab goes to "about:Tabs". 144.138.21.92 (talk) 21:45, 2 July 2009 (UTC)

Check "do not show this tab welcome message again" also, in tools/internet options you can set homepages for multiple tabs UNDER your initial homepage. If you do not do this, it will give you a generic welcome to tabs messageIvtv (talk) 23:09, 8 July 2009 (UTC)

Internet Connections and Topology
Hi friends, can u please describe for me the various methods in which intrnet connections are made possible? (2) what is network Topology? Bye. Kvees. —Preceding unsigned comment added by 59.178.188.221 (talk) 08:00, 2 July 2009 (UTC)
 * I think that you can search Wikipedia to find your answers. In addition, you could try to search the reference desk archives. --Andreas Rejbrand (talk) 11:39, 2 July 2009 (UTC)
 * From Network topology: "the study of the arrangement or mapping of the elements (links, nodes, etc.) of a network, especially the physical (real) and logical (virtual) interconnections between nodes." Hope this helps! Please try to use the reference desk search box at the top of this page or the Wikipedia search next time.-ankØku- (talk) 18:14, 2 July 2009 (UTC)
 * --Xp54321 (Hello! • Contribs ) 19:57, 2 July 2009 (UTC)

USB flash drive/pen drive/stick drive failures
I have a number of pen drives which I use to transfer data from one computer to another. Typical use would be to fill one up for a week with data from one computer, then move it to the other computer, copy all the data to that computer, then delete everything on the pen drive. So, I figure that's one read/write cycle per week, so it should last for years at that low usage rate. However, they don't last for years, but only months.

At first I thought I just had a defective drive, but now it's happened again. So, why might this be happening ? One theory I have is that while the data area only gets one read/write cycle per week, the index area is being updated every time I write a new sector. Since these are 32 Gb pen drives, that would mean there are some 64 million sectors at 512 bytes per sector. So, if there's an index that keeps track of the "highest occupied sector address" (or "lowest unoccupied sector address"), this might update 64 million times a week. In that case I could certainly understand why they would fail after a few months.

So, is this how they work ? If so, this seems to be a major flaw in their design. Is there any way to prevent them from doing this (say by having them only update that index after the copy operation is completed, instead of when each sector is written). Are there some brands that don't do this ? Are there some brands that include a small portion of more stable memory for this type of index space ? Any other solutions would be welcomed (I'm considering using a portable hard drive if I can't get the pen drives to last longer). StuRat (talk) 12:19, 2 July 2009 (UTC)


 * What file system did you use for the drive? I'd guess fat16/32 are the best for flash drives. --12:30, 2 July 2009 (UTC) —Preceding unsigned comment added by 194.197.235.36 (talk)


 * FAT32. Note that when you follow that link you get to a paragraph that says "...the boot sector uses a 32-bit field for the sector count...". If that "sector count" is updated every time a sector is written, then the low bit would be changed millions of times a week, and that may be the bit that wears out first.  So, would some other file system be better ?  I prefer a format that's usable by both Windows and Linux. Or maybe there's a way to periodically change where the "sector count" is stored so it can wear out other bits ? StuRat (talk) 12:33, 2 July 2009 (UTC)


 * Most modern flash memory controllers incorporate wear levelling into the flash controller, wherein there is a lookup table that maps logical block addresses with physical block addresses, and the PBA for a given LBA is shuffled around (on erase)(example(pdf)). So even a dumb filesystem that beats constantly on the same blocks will result in levelled wear. Now if you were using a unix filesystem with atime enabled, I could speculate that you'd be reading a lot, and updating those atimes would cause a large amount of writes you weren't immediately cognisant of. It seems FAT32 does store atime of a sort, one that (at least on linux) is honoured by the vfat driver (ref); so if I were planning on doing tons of reads on a flash drive on linux, it would make sense to make sure that it was mounted noatime.87.113.26.43 (talk) 13:23, 2 July 2009 (UTC)


 * I did some experiments: by default Ubuntu does mount fat32 volumes with vfat with atime enabled, and it shows a proper atime (ls -lu) for files, down to the second - but remount the disk and the atime is shown with the HH:MM back to (UTC) zero. It's like the ref I gave above says, FAT32 only stores the day part of access-time stamps. So whether repeatedly accessing a file on a vfat volume actually causes a torrent of writes to the disk depends on how smart the vfat driver is (and I guess the same is true for its counterpart in Windows). If the update_atime logic checks that the file's current atime(.date) is the same as that which we propose to write, and doesn't bother writing, then the wear from reading is trivial. If the driver is dumb, then what I wrote above is a real concern for read-heavy write-light flash devices. 87.113.26.43 (talk) 14:32, 2 July 2009 (UTC)


 * But, in my case, the use isn't really read-heavy or write-heavy. I write maybe 100 files containing perhaps 32 Gb of data over the course of a week, then read them perhaps a few times each, then delete them.   How can I find out if a given flash drive has the wear leveling feature, and does this apply to the index space/"sector count", as well as the data area ? StuRat (talk) 17:26, 2 July 2009 (UTC)


 * Open it up, figure out which usb<->flash controller chip it has, find its datasheet on the website of the semiconductor company that made it (lsusb -v might help, particularly for a sealed drive). As a simple block-structured device, it doesn't have any concept of the index or the fat or the data area; it's all just blocks of bytes. Failing that, I can't think of an easy way to figure it out only in software; much of the point of these controller chips is to hide the complexities of LBA<->PBA, wear levelling, and NAND flash's vexing rewrite characteristics. 87.113.26.43 (talk) 18:15, 2 July 2009 (UTC)


 * I don't think I'm willing to "open it up" quite yet, as I imagine them being even more fragile than a laptop, and I managed to sever a paper cable (yes, paper !) the last time I tried that. My Puppy Linux doesn't recognize the "lsusb" command at all.  Under Windows XP I looked for "scandisk" and the help informed me that it's no longer supported and I should run "chkdsk" instead.  I'm running that now. StuRat (talk) 23:49, 2 July 2009 (UTC)


 * quote "One theory I have is that while the data area only gets one read/cycle per week, the index area is being updated every time I write a new sector." - that would explain your problem - but would be a massively shortshighted way fo doing things - I doubt flash would have ever made it to market with algorhthyms like that - I give a better way below - but it is a guess - I'd like to know the answer too.. 83.100.250.79 (talk) 19:32, 2 July 2009 (UTC)


 * I can easily imagine it happening, say if the manufacturers only concerned themselves with hardware issues and those that set the standards for the FAT32 file system didn't think about it's use on flash drives. In such a case, there may not be anyone who actually worried about the implications of having a FAT32 file system on a flash drive. StuRat (talk) 23:49, 2 July 2009 (UTC)


 * As a related question - a perhaps close to what the OP was asking: The wear levelling of the LBA (the data) is I think easy to understand - but what of the PBA (the pointers or references to the actual data) - how are they wear leveled ? (I understand that a PBA will point to a relatively large block of data)
 * My guess: linked lists with several pointer fields per PBA that can be 'crossed out' as used seems like one option - (coupled with a binary tree referencing the PBA to prevent the flash chip turning into a vast tape drive - also with 'cross outable fields') - but how exactly in practice - that was a guess - could someone give a simple(ified) example? Thanks?83.100.250.79 (talk) 19:30, 2 July 2009 (UTC)


 * I think you've misunderstood what a PBA is, and you're conflating how flash works and how fat works. An LBA is just the linear block address of a hypothetical block; all the data on the drive is addressed using LBAs. LBAs are the chief part of the interface between drives (over SCSI,USB,SATA, or IDE), and really they're a bit abstract - you think about them actually being blocks in a row on the surface of a disk, but there's no guarantee that they are (and sometimes they aren't). IN the case of wear-levelling flash controllers, the controller device (I really mean the device, not the device driver, and certainly not the filesystem layer or OS block layer) has a translation table that maps 1-1 LBAs to addresses on the physical flash chips that only it can see. The OS/fs/blockdriver doesn't know anything about PBAs, and it can't see or manipulate that table. When filesystems manipulate storage, they do so solely by dealing with logical blocks, and they don't care if those are blocks of ram or flash or areas on a magnetic disk or blocks in a storage-area-network (wherein blocks get translated a bunch of times). So the fat driver doesn't know anything about flash (and doesn't do anything special for it) and the flash device doesn't know anything about FAT (and doesn't have any special support for it). 87.113.26.43 (talk) 20:20, 2 July 2009 (UTC)


 * Not at all - I was taking about the underlying(overlying?) structure of the data+extra stuff for wear leveliing in the flash - independent of the data structure imposed on it by the OS. You prompted me to point out that there is more than one way to do this - including a method that doesn't included separate pointers (an index) - one which springs to mind would be a doubly linked list, containing blocks of data. In either case the idea of having more than one pointer field (ie the pointer to previous and next data items in the doubly linked list case) is still valid, maybe even more so.
 * ie the idea of having more than one slot for a specific pointer per data block so that as the pointer slots are used, they can be marked 'dirty', and the next slot used - preventing the movement of whole blocks because of the change of a single pointer (particularily necessary if a doubly linked list were used).


 * I get the feeling I haven't explained my response to the OP very well - maybe someone else could give it clearer. (Did I mix up LBA and PBA in what I said - it's summer - the heat may be affecting my clarity) 83.100.250.79 (talk) 21:49, 2 July 2009 (UTC)


 * All the above doesn't explain why you are getting such poor value from your drives - you should expect that say for a 1GB drive with 10,000 average time to failure that you can write 10,000GB to it before your expectation of failure becomes 50/50. One possibility is that you are immensly unlucky. Another is failure due to static (maybe) - they have static protection built in (not sure about USB)- but if you work in a high static enviroment eg lots of polyester seat covers and carpets - that might affect the lifespan - I've no data on this - it's a unlikely explanation - unless you are constantly getting staic elctricity shocks every time you touch an earthed object.
 * Some devices do contain some normal ram that can be used to store acceleration data (either cache or lookup tables) - but that won't(shouldn't) be related to the lifetime. Are you doing standard read/write to large files eg documents/pictures - I doubt the standard thumb stick is configured for bitwise access?83.100.250.79 (talk) 20:10, 2 July 2009 (UTC)


 * No static electricity problems around here, as it's usually quite humid. The typical usage is to use Windows XP File Manager (or, as they call it, "Windows Explorer") to copy large files (maybe 500 Mb each) to the flash drive, and then later off the flash drive to another Windows XP  computer, followed by using Windows XP to delete the files.  Puppy Linux may also sometimes be used to copy or delete files. StuRat (talk) 23:49, 2 July 2009 (UTC)

In case it will help, let me describe some of the errors I get. On one flash drive I can write to about half of it, then it says "Device is write protected" and won't write any further. On another it says "File not found" when I try to write to it or read from it at all. I assumed these errors were the typical nonsense errors you get and that the actual problem is bad sectors on the flash drive. But how could I actually confirm that this is the real problem ? StuRat (talk) 23:49, 2 July 2009 (UTC)

OK, "chkdsk" completed on the "file not found" flash drive, after a very long time (good flash drives are usually very quick). No errors were found except that it said "Cannot write boot sector". I take that to mean that the boot sector is bad. What would cause the boot sector to fail before the rest of the flash drive ? Is there any way to designate a new sector to be used as the boot sector ? StuRat (talk) 01:20, 3 July 2009 (UTC)


 * Maybe more info might help - is the card totally broken - is it possible to re format it and get it to work again (ie corrupted or just bust), vista? (sorry ms), works on other OS?, also card make - internet suggests FAT corruptions is commonplace on certain devices.83.100.250.79 (talk) 13:46, 3 July 2009 (UTC)
 * chkdsk has issues of it's own with boot sector (see search) - also if you are using chkdsk to do write read tests to check for bad bits on the disk this would just hammer the flash data structure.. and wouldn't detect a bad bit properly.
 * (Some computers automatically perform a defragment every day - is there a chance that the OS has been doing this to the flash drive if it's left plugged in?)83.100.250.79 (talk) 14:00, 3 July 2009 (UTC)
 * There's supposed to be a backup boot sector "If you have a corrupted primary boot sector..", but see also  83.100.250.79 (talk) 14:28, 3 July 2009 (UTC)


 * Reformatting the flash drive doesn't work. That info about the backup boot sector is interesting, but unfortunately it looks like it just copies the data from the backup boot sector to the primary and continues to use the primary boot sector to boot.  This won't help if the sector itself is bad, as opposed to the data stored on it.  What I really need is the ability to designate some sector other than sector 0 as the boot sector, since sector 0 is apparently bad.  Also, your links are for Windows 2000 and I use Windows XP.  Does this info also apply to XP ?  Defragmentation definitely doesn't happen, since the flash drives are only plugged in when the data is being written or read, then are immediately unplugged. As for the brand names, one is Micro Center, and I can't tell what the other one is. StuRat (talk) 19:19, 3 July 2009 (UTC)


 * Sorry about the 2000 didn't notice- don't know if it's the same in XP (ie has it been fixed?) - but as a format doesn't fix the drive I suppose it is irrelevent.
 * If the microcentre drive is like this one you've got a 3 year warranty - perhaps you will enjoy knowing that "We use only the highest quality components from the leading manufacturers to give you many years of worry free use"
 * Sorry couldn't resist. Honestly I'd be blaiming the drive and returning for a full refund, receipts??. I've no reason to believe that USB drives are fundamentally short lived. It sounds like you are nowhere near to the level of use for failure.83.100.250.79 (talk) 20:12, 3 July 2009 (UTC)


 * Thanks. It says a "limited warranty", which might mean it excludes "normal wear and tear", which they would claim caused it to fail.  I don't know how to prove it was only lightly used. StuRat (talk) 11:37, 4 July 2009 (UTC)


 * I think you're still within 'normal wear and tear' - unless the dog chewed it, or it was struck by lightning, or you ran over it with a truck..


 * I am within normal wear and tear, yes, but how would I prove I am ? Won't they just say I used it constantly until it failed ? StuRat (talk) 02:45, 5 July 2009 (UTC)


 * I wouldn't specifically be worried about this. If the manufacturer is at all ethical, they can't simply deny all warranty claims on the basis that "there's no proof that this unit was not used heavily". --Mathew5000 (talk) 01:06, 6 July 2009 (UTC)


 * By the way this discussion raises the issue of static vs. dynamic wear leveling - which could cause the drive to wear out faster if you have it full to 95% of files, and are only changing a few files a lot. It's not an issue if the drive occasionally has a lot of files deleted.83.100.250.79 (talk) 17:57, 4 July 2009 (UTC)


 * No, I pretty much fill it up (or half up), then copy everything to another computer and delete everything. StuRat (talk) 02:45, 5 July 2009 (UTC)

How do I make my USB flash drive last as long as possible ?
As I described in the section above, my flash drives die at an unacceptable rate. So, how can I make them last longer ? Should I use something other than the FAT32 file system ? (Remember that I need to access them in Windows XP and would also prefer to be able to access them in Linux.)  Could heat cause them to wear out prematurely ? Would pointing a fan at them while in use help ? StuRat (talk) 19:52, 3 July 2009 (UTC)


 * You're experiencing an unusually high rate of failures. Is it possible that some environmental factor is causing them to fail? Are you leaving them on the dash of your car? Or near the ocean, or some other unfriendly-to-electronics thing?
 * I've heard a lot of anecdotal claims that certain types of memory can be effected by being in close contact with cell-phones. I've never understood the mechanism that would cause this, however. (This would make a decent science fair project.)APL (talk) 18:39, 7 July 2009 (UTC)


 * Radio frequency effects possibly, also aren't some Mobile's verging on

microwave technology? Sfan00 IMG (talk) 16:52, 8 July 2009 (UTC)

Ajax with Scripted Content
Is there a "standard" method for handling the following problem: Going modern, you decide to pull page content using Ajax. You place the content in some div's innerHTML. The content you pull is dependent on some JavaScript. However, the page is loaded. So, the browser won't load anymore scripts automatically. The content fails unless you load the script when the page loads, before the Ajax-pulled content is loaded. In other words, if you have 30 content pages, each with a unique script, you have to load all 30 scripts on page load just in case the user requests to pull one of the content pages. -- k a i n a w &trade; 12:33, 2 July 2009 (UTC)


 * I guess I should point out what I am doing... I'm using regex to locate instances of the script src=.... tag. Then, I have an array of scripts I've loaded in the past.  If the script in script tag is not in the array, I create a script node, set the src, and insert it into the head element.  This functions but has the problem that once you hit a page with a script, the script is permanently loaded until you close the website. -- k a i n a w &trade; 14:35, 2 July 2009 (UTC)
 * When the user clicks to pull a page, set the head element and auto-refresh the page to run that script, whether it's a separate one or in the array. Each call for a new page should overwrite the src in the head; perhaps a graceful closing of the previous script would also be in order. - KoolerStill (talk) 15:10, 2 July 2009 (UTC)


 * Well, this project is doomed to failure. The scripts, it turns out, are not all in separate files.  Some is embedded in the HTML.  Also, some of it must be run when the page loads (body onload option) to function at all.  So, Ajax appears to be a bad tool.  I'm suggesting they just load the whole page (content and all) per link instead of just trying to load the content. -- k <font color='#cc0033'>a <font color='#990066'>i <font color='#660099'>n <font color='#3300cc'>a <font color='#0000ff'>w &trade; 19:40, 2 July 2009 (UTC)


 * I think the crucial point you've hit at here is that you really, really need to think through the overall organization of something like this. In any case in your scenario it is hard for me to understand why you are using AJAX in this instance ("going modern" is the worst reason ever—doing it just because it is current is an awful idea, you should do it because it works better for your situation), but I understand you haven't fully described that end of things. AJAX should not be used to just load pages that could otherwise just be served up regularly. --98.217.14.211 (talk) 13:59, 3 July 2009 (UTC)


 * The product functions fine. Management got it in their head that using Ajax will reduce data transfer, decreasing network costs, increasing profits.  So, I was asked to make it Ajax-based.  The deeper I delved into it, I realized how extensive the javascript was.  So, I use Ajax to dynamically change content on non-scripted pages, but loaded the scripted pages directly. -- <font color='#ff0000'>k <font color='#cc0033'>a <font color='#990066'>i <font color='#660099'>n <font color='#3300cc'>a <font color='#0000ff'>w &trade; 14:36, 4 July 2009 (UTC)

Data transmission speeds
I've been trying to understand how data is transmitted via modulation. Am I right in thinking that for short-distance transfer as in computers and USB (and ISDN for some reason), data are transmitted by serial/parallel digital communication - simply on-off-off-on etc.? But telephones, DSL, cable, optical fibres and radio all use modulation?

Am I then right in saying that DSL, for example, uses Discrete Multi-Tone Modulation and that this works in a way similar to that mentioned under Modulation?

Why then is this not much faster than serial communication? Why is USB2 so much faster than ADSL (and even Fibre-to-the-home) when each symbol in ADSL can represent many bits compared to one bit in serial? And this is without going into the complexities of Orthogonal frequency-division multiplexing and Quadrature amplitude modulation. Why not have a serial internet given that USB3 is 5Gbit/s and presumably a robust signal as it is only binary? ASmartKid (talk) 15:28, 2 July 2009 (UTC)


 * The trick is, you can send data as fast as you want - but if the signal gets corrupted, you can't receive it. USB can make some assumptions about the signal integrity - the cable will only be a few meters long, and the wire will be inside a house with reasonable, homogeneous temperature; etc. etc.   All these mean that there is an implicit "guarantee" that a certain quality of signal will be received on the other end.  (Specifically, high frequencies will not be attenuated).  DSL, however, must travel for kilometers over an uncontrolled, outdoor cable.  Along this signal path, the analog details start to matter, and the signal will become distorted.  (It's going to get distorted; the only question is - how much?)  So, the engineers must be a little more conservative with the amount of data that they put on the long-distance cables, given that they have an expected signal integrity.  Depending on the particular transmission method, the signal degradation might be quantified as a total power loss ("dB of attenuation") or it may be a particular frequency-response.  This analog error creates a loss of digital information.  One option is to slow the amount of data so it is always below the distortion threshold.  This severely throttles the network speed, though.  Another option is to push data as fast as possible, right on the edge of the "lossy" part - so that there's a particular statistical likelihood that the data gets through undistorted.  If there is an error, the network controlling software and hardware must request re-transmissions (which of course slows things down a little).  If properly designed, based on a realistic bit error rate, this is the fastest way to transmit data over an analog channel with a known amount of expected distortion.  This is the reason for the OSI model of networks - where a variety of interoperating abstraction layers all have certain understandings and assumptions about the end-to-end connection.  Nimur (talk) 16:22, 2 July 2009 (UTC)


 * Thinking back to my training as a Radio tech, it was initially explained simply as: modulation is used in analogue transmission, but not in digital transmission. So for AM radio, FM Radio, modems, S-Video etc there is some form of modulation. For transmission that is digital, SATA, Parallel ATA, USB, HDMI etc. then it is bitstream of on/off. It is not really a matter of short vs. long distance, but analogue vs. digital. I suppose I'm about to confuse the issue, since there is also modulation of digital signals used to increase the amount of data sent per cycle. Although one could say this is analogue to digital conversion it is not really considered to be so. <font style="color:#3300FF;">Lantrix  ::<font style="color:#FF3300;">Talk ::<font style="color:#009933;">Contrib :: 15:04, 8 August 2009 (UTC)

SQL Server 2000 & 2005 Residing on the same server
Does anyone know if it's possible to run two different versions of SQL Server on the same physical server if they are installed in different directories? Thanks in advance for your help.
 * Yes it'll work just fine (you shouldn't even need to specify a different directory as the default locations are already slightly different). Install the 2000 server first AND updating it to service pack 4, but after that just install 2005 as normal. From memory it'll either ask you if you want to upgrade the 2000 server (if it can) or if you want to install a new instance and you just want the new instance option. <span style="font-size:14px;font-weight:bold;font-family:verdana;color:#800040;filter:Shadow(color=#000000,strength=1);height:15px;">ZX81  talk  19:51, 2 July 2009 (UTC)
 * Yes, I'm currently running both versions on the same machine at the same time. Yes, they are installed in different directories.  A Quest For Knowledge (talk) 02:38, 3 July 2009 (UTC)

Linux keyboard issue
I have a Linux home computer (not sure about details) and the backspace key has started making this symbol ( ç ) but the keyboard configuration says it should be backspace... If anyone could shed some light on the matter that would be greatly appreciated. 83.50.42.15 (talk) 21:59, 2 July 2009 (UTC)


 * It sounds to me like Linux may think you have a foreign language keyboard, perhaps Spanish or Portuguese. Hopefully others can provide details of how to check and change that setting. StuRat (talk) 00:46, 3 July 2009 (UTC)

Yeah it is set in Spanish (I live in Spain) but it used to be absolutely fine and it just decided to go haywire with the backspace key for unknown reason 81.32.39.224 (talk) 11:32, 3 July 2009 (UTC)


 * What distribution of Linux are you using? If you are not sure, open up a terminal and type uname -a and tell us what the output is. - Akamad (talk) 22:51, 3 July 2009 (UTC)

Done.. The message that comes up is Linux casacaracol-desktop 2.6.27-9-generic #1 SMP Thu Nov 20 21:57:00 UTC 2008 i686 GNU/Linux  81.39.56.119 (talk) 10:51, 4 July 2009 (UTC)


 * That didn't help unfortunately. What's the output from cat /etc/*-release? - Akamad (talk) 00:18, 5 July 2009 (UTC)