Jump to content

Wikipedia:Reference desk/Archives/Computing/2009 June 22

From Wikipedia, the free encyclopedia
Computing desk
< June 21 << May | June | Jul >> June 23 >
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.


June 22

[edit]

Facebook imports gmail contacts without knowing the password. How?

[edit]

I just joined facebook. I provided it with a working email address, my gmail address. I did not give it my gmail password (and my facebook password is different from my gmail password). Nevertheless it suggested a "friend list" coinciding almost exactly with my gmail contacts list. As far as I know, there are no other data sources from which it could have obtained this list.

So how did it do this? Does facebook have access to gmail data? Might it have something to do with the confirmarion email? —Preceding unsigned comment added by 81.11.170.162 (talk) 07:51, 22 June 2009 (UTC)[reply]

I've observed a similar thing, although only one unexpected friend was suggested. In that case, someone who ought to know better (a very tech-savvy person) evidently had given Facebook his gmail password. Facebook finds my gmail address in his contact list, and suggests him as a friend to me. --NorwegianBlue talk 08:06, 22 June 2009 (UTC)[reply]
Google Contacts API? Though I'd expect it to need to somehow confirm that you are who you say you are. --98.217.14.211 (talk) 02:12, 26 June 2009 (UTC)[reply]
[edit]

Hi! I was just wondering if anyone could help me with a little query I have. I have reated a product guide/review site and want to use images of the products I am reviewing. A link to a online store that sells any paticular product will be claerly visiable next to any images I use. Is this copyright infinigment? I'm using the product image to facilitate the sale of the product in a way I guess.... —Preceding unsigned comment added by StecenSchiffmeister (talkcontribs) 08:41, 22 June 2009 (UTC)[reply]

Yep. It'd probably be copyright infringment since you're using the image without their permission (unless there are allowances for it on the site), and it would be prudent to email the site and ask, although (not legal advice, just from personal experience) they probably won't do anything about it. Gunrun (talk) 09:27, 22 June 2009 (UTC)[reply]
As Gunrun stated, we can not give legal advice. However, in certain circumstances, the use of copyrighted material without permission may be tenable under the fair use doctrine. For example, you will find that many of Wikipedia's articles make limited use of copyrighted media under a claim of fair use. decltype (talk) 12:40, 22 June 2009 (UTC)[reply]
There's something not clear about your question - whether or not a link to a store is next to the image is irrelevent, as is whether you are facilitating a sale or not. If the images you are using of the product are made by you - ie with a camera - then you have no copyright issues, if the images are from another site then you almost certainly will be abusing someones copyright. Why not just take photos of the things that you are reviewing like other amateur and pro review sites do?83.100.250.79 (talk) 13:29, 22 June 2009 (UTC)[reply]
Why not ask the manufacturers whether you can use their product photos? They normally have product photos specially available on their web sites as sales material for news and sales uses and often have a page describing how their logo may be used also if you wish to show that. Dmcq (talk) 13:32, 22 June 2009 (UTC)[reply]
There is a section of US copyright law that covers this specifically: "In the case of a work lawfully reproduced in useful articles that have been offered for sale or other distribution to the public, copyright does not include any right to prevent the making, distribution, or display of pictures or photographs of such articles in connection with advertisements or commentaries related to the distribution or display of such articles, or in connection with news reports." Note that this refers to your own photographs of the product, not someone else's, which are almost certainly covered by copyright. --98.217.14.211 (talk) 16:08, 22 June 2009 (UTC)[reply]

InstantAtlas

[edit]

Where can I find a good guide to help create flash mapping with InstantAtlas. I'm looking for help in laying out the information in a spreadsheet firstly, maybe a template which can be downloaded or full example of the spreadsheet and end result.

Thanks —Preceding unsigned comment added by 195.49.180.146 (talk) 11:33, 22 June 2009 (UTC)[reply]

The InstantAtlas site has downloadable User Guides as well as FAQs and templates for some common mapping applications. - KoolerStill (talk) 03:10, 23 June 2009 (UTC)[reply]

Not shutting down programs can corrupt the hard drive?

[edit]

Necessary programs such as antivirus , firewall etc and useful programs such as a software that makes volume control easier load automatically since they are in the start up list. When I shut down windows XP, I do not quit these softwares. In addition to that, I don't disconnect the internet either. Will this corrupt the hard drive?. —Preceding unsigned comment added by 131.220.46.25 (talk) 12:22, 22 June 2009 (UTC)[reply]

When XP shuts down, it shuts down all programs and disconnects from the Internet. So, by shutting down XP, you are shutting down all your programs and disconnecting from the Internet. -- kainaw 12:26, 22 June 2009 (UTC)[reply]

thanks!

Resolved

Joining together a modem, a Windows computer, and a Linux computer

[edit]

Currently I have a Windows XP computer linked to an external modem through an ethernet cable. I'm intending to convert an old computer to Linux - probably Ubantu. My broadband ISP does not offer any support for Linux, only Windows. I would like to be able to access the internet through either computer. And I would like to backup data between computers. What would be the best way of doing all this please? 78.147.250.114 (talk) 16:35, 22 June 2009 (UTC)[reply]

Use a router. There are many brands and models for sale at all electronics stores. The router accesses the Internet through the modem. The computers connect to the router. The specific setup is dependent on the router you decide to purchase. -- kainaw 16:45, 22 June 2009 (UTC)[reply]

A router seems to be like a computer. Isnt there an easier way to do this with ethernet? I thought that was what ethernet was for, not just the wire between the computer and the external modem. 78.144.243.58 (talk) 18:56, 22 June 2009 (UTC)[reply]

The most typical way of doing what you want is to purchase a router that was designed to do this. (Technically it's a network switch, but they are usually marketed as "routers".) Googling "cable modem router", here is an example from Linksys. Some of these are wireless, with an antenna plus usually 4 wired ports in the back; other routers are just wired, with 4 or 8 ports in the back. You will run an Ethernet cable from the cable modem to your router. Then, for a wired router, you will run one Ethernet cable from the router to computer #1, and one Ethernet cable from the router to computer #2. (If, alternatively, you get a wireless router, you can use each computer's wireless networking, and not run any cables.) You can thusly share your Internet connection between usually 4 computers via the Ethernet cables, or if you're using the wireless connection, the limit is probably 253 or so. A big bonus of using a router is that it will act as a network firewall for you. These routers aren't big computers; they are small, about the size of your cable modem. Tempshill (talk) 19:16, 22 June 2009 (UTC)[reply]
No, a router is NOT the same as a switch. Some modems designed for single computer use will operate in bridge mode and assign the computer the external IP. If you try using a switch and connect 2 computers it will give that IP to the first computer to send a DHCP request and the other one is left without any access. A router overcomes this by doing NAT, taking the external IP given by the modem and assign LAN IP's to the computers. Although commonly integrated with a switch it doesn't have to; mine just has a WAN port and a LAN port, which goes to a separate switch. --antilivedT | C | G 09:00, 23 June 2009 (UTC)[reply]
I stand corrected on the terminology. Tempshill (talk) 17:43, 23 June 2009 (UTC)[reply]
I think you have it backwards, a router IS the easy way of doing it. You can get a basic one for probably less then $40 these days. If you were going to try to do it without a rounter you'd probably need a second NIC in your windows machine and then muck around with ad hoc networks and itnernet connection sharing, which would probably be a major pain to set up with a linux machine, also then your linux machine won't have internet when the windows computer is not turned on. Vespine (talk) 22:48, 22 June 2009 (UTC)[reply]
Most people in the UK with broadband, connect through a combined modem/router - the USB modem has very much fallen out of favour, and I don't think any major ISP still supplies them. The modem/router is either supplied by your ISP (usually free, eg: a BT Home Hub) or you can buy one from stores like PC World and Comet and independents in many towns across the country. All the modem/routers I've seen have a connection for the phone line (or cable if your ISP is a cable company), one or more Ethernet RJ45 ports, and often an aerial for the wireless connection. You simply connect each computer the modem/router using an ethernet cable (one is usually supplied with the modem/router, so you might have to buy second one).
As your Windows XP machine is working fine, you don't have to do anything there. As for Linux through the wired ethernet connection, in my experience it connects to the internet straight away with no setup needing to be done. The issue of lack of support from you ISP for Linux is not such a big deal - they simply can't help you set up Linux or Firefox because the dummies they employ on telephone support only have a "script" with instructions for Windows and Internet Explorer - it does not mean that they detect your linux machine and somehow stop it using the internet. Astronaut (talk) 20:24, 23 June 2009 (UTC)[reply]

Processing Comma-separated values files

[edit]

My bank statement is available as a CSV file. What language would it be best to use to make a program that would check automatically if for example my salary had been received? (I am only a hobby programmer). The date of arrival of the salary is going to fluctuate by two or three days each month. The salary would be principally recognised by the amount, and also by the description of the payment. I want to create a report that lists salary payments I have received over a year. I want to do something similar with other not-quite periodic payments I get. I do not want to use a spreadsheet for various reasons. Thanks 78.147.250.114 (talk) 16:48, 22 June 2009 (UTC)[reply]

Excel imports comma-separated values through its text-to-columns function. It works extremely well for most applications of CSV that i've come across (which are mostly uncomplicated data extracts). You can then do a pivot-table on the results very quickly and see the data in a 'report', chart or table. Alternatively MS Access also allows importation of data from CSV and that's worked fine for me each time too. ny156uk (talk) 16:55, 22 June 2009 (UTC)[reply]

MS Access is a bit overkill for something as simple as this... even Excel is, if all you want to know is if the salary has been received. Presumably the poster knows how to check the bank records and is specifically asking about automatic ways to have a program alert him when the salary has been deposited. --98.217.14.211 (talk) 16:57, 22 June 2009 (UTC)[reply]
The programming question here is about how you are regularly getting the CSV files, not how to parse them. The parsing them is easy once you have them. Almost any language has CSV libraries that are quite easy to use.I am betting a moderately experienced bash programmer could come up with a one-line script that would scan a given CSV file for the right values. --98.217.14.211 (talk) 16:56, 22 June 2009 (UTC)[reply]
Grep will easily parse a CSV file. For example, if I have a collection of CSV files called bank-jan.csv, bank-fed.csv, etc... and I get paid $2,19?.?? a month (the last 3 digits are variable), I could use: "grep 219.... bank-*.csv" The result will be a list of lines containing payments of the amount I asked for. -- kainaw 16:58, 22 June 2009 (UTC)[reply]
When you say grep, I suppose you mean Perl. Could Python or SmallBasic do this? 78.147.137.38 (talk) 23:51, 22 June 2009 (UTC)[reply]
I'd guess when kainaw says "grep", he means grep, which is a fairly simple program for matching lines of a file against a regex. However, python does have a module for parsing CSV files.--Taejo|대조 19:46, 25 June 2009 (UTC)[reply]

Thanks. What I'm really after is something that will sound an alarm if a payment is unusually late, and tell me what the missing payment is. 78.144.243.58 (talk) 18:51, 22 June 2009 (UTC)[reply]

The "if a payment is unusually late" is the problem. You have to turn that into a very specific algorithm, such as "if the current date is after the 5th of the month and there is no payment for the current month, then the payment is late." Once you do that, then it is just a simple script: On the "late day", get data. Grep last payment. If last payment is not from current month, beep/send email/play song/whatever. -- kainaw 19:27, 22 June 2009 (UTC)[reply]

Thanks. Can any accounting or bookkeeping software do this? 78.147.137.38 (talk) 23:01, 22 June 2009 (UTC)[reply]

Assuming you use Windows, there appears to exist [GREP for Windows]. Make a .bat file that uses Grep to find the line as discussed above, then uses sed (which is also available for win32) to get the month from that string (or just use sed for everything), then compare that month (using if in the bat file) to a sed-ed month from the DATE command, then set the program to pause if the months are different (if the months are equal, just quit the bat file). Then set up Windows Scheduler to run the .bat file on the "late" day of each month. Jørgen (talk) 13:17, 23 June 2009 (UTC)[reply]
Excel is probably the easiest to use to manage your bank account using the csv files from your bank. If you want to write your own program, Python has a csv module in the standard library, while Small Basic provides the split command to split a long sting into shorter strings using a delimiter such as comma. In other languages, I've seen specific provision to handle csv files in C#, but in standard C, and pretty much every other language I've used you either have to rely on a basic tokenizer functionality, or write your own string parser to extract text between delimiters. Astronaut (talk) 20:24, 23 June 2009 (UTC)[reply]

I think it is going to be difficult to identify the paymemts. I'm thinking of using some sort of scoring system that rates how much the description, amount, and date, match those expected. 89.243.201.240 (talk) 20:47, 26 June 2009 (UTC)[reply]

Facebook unknown "friends" who only have 4 or 5 friends

[edit]

I've recently started getting lots of random people adding me as their friends on facebook, people who I've never met, have no obvious connections to (they are not friends of friends or people who would know my name at all), and these new "friends" all have about 5 friends of their own total. I suspect it must be some sort of marketing campaign or scam or something, yes? But I haven't "friended" any of them so I don't really know. Does anyone have any experience with this? They seem to come and go in waves, these random "friends" who don't appear to be real people's accounts at all. I had one wave of such people who were all supposedly blond girls with similar fake-sounding names, who again only had maybe four other friends and were trying to be mine. What's going on here? What's the angle? --98.217.14.211 (talk) 17:13, 22 June 2009 (UTC)[reply]

Chances are they are just as you suspect. They probably try to add hundreds of people. Then when they get a response from someone, they know they have made contact with someone who is a potential target for whatever scheme they have. It could be marketing cheap Viagra rip-offs or a pyramid scheme who knows, not me.Popcorn II (talk) 12:57, 23 June 2009 (UTC)[reply]

Recording sounds in Windows

[edit]

Is there any Windows software that allows the user digitally to record the sound he hears on the computer? There are tons of applications that are able to record from external sources (via line-in, microphone etc.), but I want to record the sound produced on my own computer, e.g. in Windows. --81.227.66.246 (talk) 17:25, 22 June 2009 (UTC)[reply]

The program I'm familiar with that can do this is VirtualDub, which can capture an AVI of your screen activity along with the sound that your computer generates. You could then use movie-editing software like Adobe Premiere - or maybe VirtualDub itself - to save off the audio as a separate .wav file. Tempshill (talk) 17:39, 22 June 2009 (UTC)[reply]
Sound Recorder. Go to the recording properties and set it to steromix or whatever name it gives your sound card system
But that will not work in Windows Vista (at least it will not on my Vista machine). --Andreas Rejbrand (talk) 00:34, 23 June 2009 (UTC)[reply]
I seem to recall that this doesn't work on all computers, because not all sound cards support recording in this manner. Specifically, newer sound cards are less likely to have this ability! Something about the recording industry demanding that manufacturers remove the functionality since it was encouraging piracy... « Aaron Rotenberg « Talk « 05:23, 23 June 2009 (UTC)[reply]
Some sound cards (for example, most of those produced by Creative) have a dummy input (called "What U Hear" on Creative cards) that allows you to record the audio currently being produced by the sound card. Most decent audio recording programs will be able to select this device as an input source. Audacity, which is free, has this functionality (select "What U Hear" from the drop-down menu in the top right before recording).

Mobile broadband disconnecting

[edit]

I have mobile broadband and I'm having som problems with it. I've had it three years and it's always worked fine, though admittedly I've had lots of problems with coverage and speed over that time. Recently I brought two usb extension leads and put it over by the window. The reception has improved vastly, but it keeps disconnecting so I have to recconect again. Is this constant disconnecting because of the 2 usb extensions plugged into each other is it bad coverage (it's very hot and dry though quite cloudy)?--92.251.208.53 (talk) 18:10, 22 June 2009 (UTC)[reply]

The signal strength may have improved. But the problems usually stem from poor signal 'quality', which can be affected by various things, including cloudy conditions. Find out where your nearest tower is located, and position the modem where there are the fewest tall buildings or trees in the way. Ask your provider to latch you to another tower; a slightly more distant tower with better line of sight will often give a better quality of signal. There should be a reception utility with your system; I've found a lowish (25% level) steady quality reading gives better results than a high but fluctuating signal. Also, strangely, 'below' and slightly back from a window seems better than level with it. Your extensions shouldn't be affecting it, as long as they are clean and firmly plugged together.- KoolerStill (talk) 15:57, 23 June 2009 (UTC)[reply]
Ah right. Well there are quite a few trees between here and the tower, and the next nearest tower is a long way away. Would I be better off putting it in the attic?--92.251.131.227 (talk) 16:30, 23 June 2009 (UTC)[reply]

wireless card for Apple Power Mac G5

[edit]

I need to install a wireless card in a Power Mac G5. What do I need to get and how hard is it to do? I'm not too competent. Thanks. --Halcatalyst (talk) 18:30, 22 June 2009 (UTC)[reply]

This Apple support article may interest you. The card discussed in the article is not user-installable, you need to take the machine in for service. If you have a spare network port you might consider a separate Airport Extreme instead. That would not require a professional installation, just the usual networking frazzle. Try to get support for 802.11n if you can. EdJohnston (talk) 18:59, 22 June 2009 (UTC)[reply]

directX function - openGL equivalent

[edit]

I'm trying to remember the name of a new-ish directX function - one that tests a polygon (presumably a triangle) against the z-buffer and return a value if the polygon is not totally occluded (and another value if a part of it is visible) - obviously useful if doing visibility tests for box bounded objects etc. It was mentioned fairly extensively in MS powerpoint presentations at the time as being 'important' - but I've totally forgotten the name - and so is having difficulty finding it again. Anyone remember. Thanks.83.100.250.79 (talk) 18:37, 22 June 2009 (UTC) Think I've found it - D3D10_QUERY_OCCLUSION_PREDICATE[reply]

New question is there an openGL method or equivalent of this. Thanks83.100.250.79 (talk) 19:40, 22 June 2009 (UTC)[reply]

YouTube video ratio

[edit]

If my videos are still using the old ratio can I somehow convert them to make them fit into the widescreen without having to remake the videos. I recently noticed black and white films being uploaded as widescreens you see so it must be possible somehow... --217.84.174.131 (talk) 19:40, 22 June 2009 (UTC)[reply]

There are many methods for converting 4:3 to 16:9. The simplest is to crop off the top and bottom of the video, which makes the assumption that what is on the top and bottom won't be missed. A better method is commonly called "pan and scan" - which actually refers to a method of converting 16:9 to 4:3, but is the same. You manually move the 16:9 cropping rectangle up and down inside the 4:3 video to ensure nothing important is lost. Another automatic method is to squish the 4:3 video, making all the people look short and fat. To avoid that, another automatic method uses a logarithmic algorithm to stretch out the sides of the 4:3 but not he middle. Consider the video to be a bunch of vertical lines. The further from center, the more the line is stretched to make it wider. The result looks acceptable in a still shot but has a fish-eye lens effect when the camera moves across the background. Those are just a few options - there are a lot more. -- kainaw 19:55, 22 June 2009 (UTC)[reply]
Thanks for your interesting and detailed reply. Could you recommend any programmes to apply the above mentioned techniques. The cropping method is tempting as it involves the least work, however I would be willing to give the pan and scan method a try. --217.227.88.30 (talk) 20:40, 22 June 2009 (UTC)[reply]
We've got an article on it, of course — Pan and scan, which has the nice animated GIF that I just added here. Tempshill (talk) 22:02, 22 June 2009 (UTC)[reply]
This is the "normal" pan-and-scan technique, converting from a widescreen ratio to 4:3. The suggestion of kainaw is a little different, but it means some content will be lost, just as it is here...
Unfortunately, we can't use fair use images in this namespace. I made it a link. --Tardis (talk) 23:37, 22 June 2009 (UTC)[reply]

U+2308, U+2309, U+230A, and U+230B

[edit]

Why are the Unicode characters U+2308 (⌈, LEFT CEILING), U+2309 (⌉, RIGHT CEILING), U+230A (⌊, LEFT FLOOR), and U+230B (⌋, RIGHT FLOOR) beloning to block "Miscellaneous Technical", and not to "Mathematical Operators"? --Andreas Rejbrand (talk) 19:49, 22 June 2009 (UTC)[reply]

They are in the 'category:math synbols' - and listed as such (ceilings and floors) in the misc. technical section http://www.unicode.org/charts/PDF/U2300.pdf (page 210 ie second sheet), also listed under 'mathematical operators' at http://www.unicode.org/charts/symbols.html
Maybe they decided they weren't sufficiently standard for inclusion in the normal list - is there a 'International standards organisation of mathematic notation' - maybe they would be able to confirm this?83.100.250.79 (talk) 20:51, 22 June 2009 (UTC)[reply]
Proper answer - Unicode is divided into blocks of 128 or 256 characters - it's seems that these characters would have required an extra block of purely mathematical symbols, with the consequence that the vast majority of that block being wasted - thus they find themselves in a misc. section.
In particular the code blocks do not signify anything particular in themselves, though often than not a whole code block will be used for a given type of glyph - see http://www.unicode.org/reports/tr25/ (section 2.7 page 14) quote "The allocation of any operator to a particular block is rarely significant"
Hope that answers it. (There is a link in the the above pdf for feedback and comments if you need it, it's possible that a fuller explanation might be got, though I would imagine they would only be guaranteed to respond to gross errors in the documentation/implementation.)83.100.250.79 (talk) 21:09, 22 June 2009 (UTC)[reply]
(EDIT) Probably Irrelevent or just plain ignorant : There's one additional complication - in that floor(1.0) as a mathematical function actually evaluates to 0.5 (ie if calculated with floor defined using the Sawtooth wave using an infinte polynomial representation of the sawtooth function ie the infinite fourier series of the sawtooth function), whereas the unicode characters you mention have their origins in computer science, and so I would expect that floorcomputer function(1.0) would evaluate to 1 (integer), and in general always evaluate to an integer - giving a subtly different meaning to the purely mathematical definition - it's my speculation to suggest that this could be a reason why those glyphs don't find themselves in the standard maths codeblock - it might be interesting to ask (on the maths desk) for their interpretation of what floor(x) does, and compare that with what a computer sciencetist would say floor(x) does when x is an integer...83.100.250.79 (talk) 21:29, 22 June 2009 (UTC)[reply]
Not sure what you'reks going on about there. The floor function used in the sawtooth function article is the normal ordinary floor as used in maths and I haven't heard of others. Dmcq (talk) 09:15, 23 June 2009 (UTC)[reply]
I think you're right as to standard use and definition - the explanation is in the above paragraph in the bracketed comment as to the potential for misinterpretation as to what floor does.
The floor of 1 is 1. Deficiencies of Fourier-series representations are neither here nor there. Algebraist 12:53, 23 June 2009 (UTC)[reply]
Thanks for picking me up on that.83.100.250.79 (talk) 13:48, 23 June 2009 (UTC)[reply]
Thanks. I see now what he was up to. Dmcq (talk) 19:33, 23 June 2009 (UTC)[reply]
I probably should scrub the last paragraph then, made it small anyway..83.100.250.79 (talk) 15:30, 24 June 2009 (UTC)[reply]

Characters after "z" in Mac OS X

[edit]

I'm trying to alphabetize some folders. But I can't seem to find a character that gets sorted after "z". I've tried •»>#×−→§↔•♦¤, but they all get sorted before numbers. Ideally, I'd like an ASCII character (for compatibility with older apps), but any unicode glyph could work. --70.167.58.6 (talk) 22:32, 22 June 2009 (UTC)[reply]

A space? I think a blank space might get sorted low down in alphabetization. Bus stop (talk) 22:36, 22 June 2009 (UTC)[reply]
in acsii } { ~ and | all come after z, if it will let you use those.83.100.250.79 (talk) 22:42, 22 June 2009 (UTC)[reply]
I guess OS X is trying to be extra helpful, but its seems to sort all symbols before zeros, then letters. So {}~ all get sorted on top. --70.167.58.6 (talk) 22:59, 22 June 2009 (UTC)[reply]
In ascii there is nothing else except those above z.. (see below)
if the os sorts by, and allows extended unicode then greek, russian, etc all should come after a to z.83.100.250.79 (talk) 22:44, 22 June 2009 (UTC)[reply]
Are there any interesting symbols? I'd rather not use foreign characters. --70.167.58.6 (talk) 22:59, 22 June 2009 (UTC)[reply]
Pound sign, euro, yen spring to mind which you might be easily be able to access from the keyboard using a 'shift graphics' key - whatever is on the mac. Apparently what you need to use is "KeyCaps" - somesort of application on the mac. Also copyright, TM,,, maybe this [1] From what you've said I'm guessing that the mac OS might be sorting everything except 0..9 a..z and A..Z to be before those characters - in other words sorting in unicode order, but making 0..9 and the alphabet come after everything else. Is the machine ignoring punctuation completely? I wonder if it's not possible - but since macs use UNIX it might be possible to change the sort algorhythm.
Its possible that the OS is using the UNIX sort command eg see [2] some-one might know how to change it's behaviour with OSX, if so it looks like the -f -i and -n modifiers are set , or maybe -d and -n (see link).
though it's the opposite solution - you could try files names such as -100abd , -101abd , -99abd since I think it will sort these as negative numbers ascending.. maybe you could invent a work around using that??? I can't think of anything else.83.100.250.79 (talk) 01:37, 23 June 2009 (UTC)[reply]
I understand that apple are responsive to requests - so possibly you should email their customer support requesting sort by unicode value or something - it's possible that they might do it, albeit a few years too late..83.100.250.79 (talk) 01:40, 23 June 2009 (UTC)[reply]

THis: [3] also http://developer.apple.com/documentation/MacOSX/Conceptual/BPFileSystem/Articles/SortingRules.html#//apple_ref/doc/uid/20002377 it doesn't mention anything about being able to change the sort order- though does mention "In non-Unicode applications, sort order is also determined in System Preferences/International/Language/Customize Sorting." - that might be your only hope. —Preceding unsigned comment added by 83.100.250.79 (talk) 01:45, 23 June 2009 (UTC)[reply]

Unicode 01B6 (z with a line through might sort after z) —Preceding unsigned comment added by 83.100.250.79 (talk) 02:22, 23 June 2009 (UTC)[reply]

There are plenty of interesting extended ASCII symbols to choose from, with the advantage of being able to write them with simple keyboard codes (Alt and 3 numbers). They would make your file names hard to pronounce though.- KoolerStill (talk) 04:56, 23 June 2009 (UTC)[reply]
This question came up for PC or Unix some time ago, don't know if it was you. I suggested æøå then, certainly sort after Z in Norwegian, but don't know how it works on English settings, and on Mac some of them might even be available on option+o, option+a or something. Jørgen (talk) 13:10, 23 June 2009 (UTC)[reply]
Why not just use "zzz_"? --140.247.42.218 (talk) 18:26, 23 June 2009 (UTC)[reply]

Don't know about ASCII characters, but I've found that file names starting with the Greek letter Omega ("Ω"), which you can get by pressing Option+Z, will be sorted after files beginning with Z.--CalusReyma (talk) 03:39, 26 June 2009 (UTC)[reply]